diff options
Diffstat (limited to 'tools/perf/pmu-events')
84 files changed, 11438 insertions, 6444 deletions
diff --git a/tools/perf/pmu-events/arch/arm64/ampere/ampereone/branch.json b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/branch.json new file mode 100644 index 000000000000..c751d57f2e19 --- /dev/null +++ b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/branch.json @@ -0,0 +1,17 @@ +[ + { + "ArchStdEvent": "BR_IMMED_SPEC" + }, + { + "ArchStdEvent": "BR_RETURN_SPEC" + }, + { + "ArchStdEvent": "BR_INDIRECT_SPEC" + }, + { + "ArchStdEvent": "BR_MIS_PRED" + }, + { + "ArchStdEvent": "BR_PRED" + } +] diff --git a/tools/perf/pmu-events/arch/arm64/ampere/ampereone/bus.json b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/bus.json new file mode 100644 index 000000000000..8623be121818 --- /dev/null +++ b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/bus.json @@ -0,0 +1,32 @@ +[ + { + "ArchStdEvent": "CPU_CYCLES" + }, + { + "ArchStdEvent": "BUS_CYCLES" + }, + { + "ArchStdEvent": "BUS_ACCESS_RD" + }, + { + "ArchStdEvent": "BUS_ACCESS_WR" + }, + { + "ArchStdEvent": "BUS_ACCESS_SHARED" + }, + { + "ArchStdEvent": "BUS_ACCESS_NOT_SHARED" + }, + { + "ArchStdEvent": "BUS_ACCESS_NORMAL" + }, + { + "ArchStdEvent": "BUS_ACCESS_PERIPH" + }, + { + "ArchStdEvent": "BUS_ACCESS" + }, + { + "ArchStdEvent": "CNT_CYCLES" + } +] diff --git a/tools/perf/pmu-events/arch/arm64/ampere/ampereone/cache.json b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/cache.json new file mode 100644 index 000000000000..fc0633054211 --- /dev/null +++ b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/cache.json @@ -0,0 +1,104 @@ +[ + { + "ArchStdEvent": "L1D_CACHE_RD" + }, + { + "ArchStdEvent": "L1D_CACHE_WR" + }, + { + "ArchStdEvent": "L1D_CACHE_REFILL_RD" + }, + { + "ArchStdEvent": "L1D_CACHE_INVAL" + }, + { + "ArchStdEvent": "L1D_TLB_REFILL_RD" + }, + { + "ArchStdEvent": "L1D_TLB_REFILL_WR" + }, + { + "ArchStdEvent": "L2D_CACHE_RD" + }, + { + "ArchStdEvent": "L2D_CACHE_WR" + }, + { + "ArchStdEvent": "L2D_CACHE_REFILL_RD" + }, + { + "ArchStdEvent": "L2D_CACHE_REFILL_WR" + }, + { + "ArchStdEvent": "L2D_CACHE_WB_VICTIM" + }, + { + "ArchStdEvent": "L2D_CACHE_WB_CLEAN" + }, + { + "ArchStdEvent": "L2D_CACHE_INVAL" + }, + { + "ArchStdEvent": "L1I_CACHE_REFILL" + }, + { + "ArchStdEvent": "L1I_TLB_REFILL" + }, + { + "ArchStdEvent": "L1D_CACHE_REFILL" + }, + { + "ArchStdEvent": "L1D_CACHE" + }, + { + "ArchStdEvent": "L1D_TLB_REFILL" + }, + { + "ArchStdEvent": "L1I_CACHE" + }, + { + "ArchStdEvent": "L2D_CACHE" + }, + { + "ArchStdEvent": "L2D_CACHE_REFILL" + }, + { + "ArchStdEvent": "L2D_CACHE_WB" + }, + { + "ArchStdEvent": "L1D_TLB" + }, + { + "ArchStdEvent": "L1I_TLB" + }, + { + "ArchStdEvent": "L2D_TLB_REFILL" + }, + { + "ArchStdEvent": "L2I_TLB_REFILL" + }, + { + "ArchStdEvent": "L2D_TLB" + }, + { + "ArchStdEvent": "L2I_TLB" + }, + { + "ArchStdEvent": "DTLB_WALK" + }, + { + "ArchStdEvent": "ITLB_WALK" + }, + { + "ArchStdEvent": "L1D_CACHE_LMISS_RD" + }, + { + "ArchStdEvent": "L1D_CACHE_LMISS" + }, + { + "ArchStdEvent": "L1I_CACHE_LMISS" + }, + { + "ArchStdEvent": "L2D_CACHE_LMISS_RD" + } +] diff --git a/tools/perf/pmu-events/arch/arm64/ampere/ampereone/core-imp-def.json b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/core-imp-def.json new file mode 100644 index 000000000000..95c30243f2b2 --- /dev/null +++ b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/core-imp-def.json @@ -0,0 +1,698 @@ +[ + { + "PublicDescription": "Level 2 prefetch requests, refilled to L2 cache", + "EventCode": "0x10A", + "EventName": "L2_PREFETCH_REFILL", + "BriefDescription": "Level 2 prefetch requests, refilled to L2 cache" + }, + { + "PublicDescription": "Level 2 prefetch requests, late", + "EventCode": "0x10B", + "EventName": "L2_PREFETCH_UPGRADE", + "BriefDescription": "Level 2 prefetch requests, late" + }, + { + "PublicDescription": "Predictable branch speculatively executed that hit any level of BTB", + "EventCode": "0x110", + "EventName": "BPU_HIT_BTB", + "BriefDescription": "Predictable branch speculatively executed that hit any level of BTB" + }, + { + "PublicDescription": "Predictable conditional branch speculatively executed that hit any level of BTB", + "EventCode": "0x111", + "EventName": "BPU_CONDITIONAL_BRANCH_HIT_BTB", + "BriefDescription": "Predictable conditional branch speculatively executed that hit any level of BTB" + }, + { + "PublicDescription": "Predictable taken branch speculatively executed that hit any level of BTB that access the indirect predictor", + "EventCode": "0x112", + "EventName": "BPU_HIT_INDIRECT_PREDICTOR", + "BriefDescription": "Predictable taken branch speculatively executed that hit any level of BTB that access the indirect predictor" + }, + { + "PublicDescription": "Predictable taken branch speculatively executed that hit any level of BTB that access the return predictor", + "EventCode": "0x113", + "EventName": "BPU_HIT_RSB", + "BriefDescription": "Predictable taken branch speculatively executed that hit any level of BTB that access the return predictor" + }, + { + "PublicDescription": "Predictable unconditional branch speculatively executed that did not hit any level of BTB", + "EventCode": "0x114", + "EventName": "BPU_UNCONDITIONAL_BRANCH_MISS_BTB", + "BriefDescription": "Predictable unconditional branch speculatively executed that did not hit any level of BTB" + }, + { + "PublicDescription": "Predictable branch speculatively executed, unpredicted", + "EventCode": "0x115", + "EventName": "BPU_BRANCH_NO_HIT", + "BriefDescription": "Predictable branch speculatively executed, unpredicted" + }, + { + "PublicDescription": "Predictable branch speculatively executed that hit any level of BTB that mispredict", + "EventCode": "0x116", + "EventName": "BPU_HIT_BTB_AND_MISPREDICT", + "BriefDescription": "Predictable branch speculatively executed that hit any level of BTB that mispredict" + }, + { + "PublicDescription": "Predictable conditional branch speculatively executed that hit any level of BTB that (direction) mispredict", + "EventCode": "0x117", + "EventName": "BPU_CONDITIONAL_BRANCH_HIT_BTB_AND_MISPREDICT", + "BriefDescription": "Predictable conditional branch speculatively executed that hit any level of BTB that (direction) mispredict" + }, + { + "PublicDescription": "Predictable taken branch speculatively executed that hit any level of BTB that access the indirect predictor that mispredict", + "EventCode": "0x118", + "EventName": "BPU_INDIRECT_BRANCH_HIT_BTB_AND_MISPREDICT", + "BriefDescription": "Predictable taken branch speculatively executed that hit any level of BTB that access the indirect predictor that mispredict" + }, + { + "PublicDescription": "Predictable taken branch speculatively executed that hit any level of BTB that access the return predictor that mispredict", + "EventCode": "0x119", + "EventName": "BPU_HIT_RSB_AND_MISPREDICT", + "BriefDescription": "Predictable taken branch speculatively executed that hit any level of BTB that access the return predictor that mispredict" + }, + { + "PublicDescription": "Predictable taken branch speculatively executed that hit any level of BTB that access the overflow/underflow return predictor that mispredict", + "EventCode": "0x11a", + "EventName": "BPU_MISS_RSB_AND_MISPREDICT", + "BriefDescription": "Predictable taken branch speculatively executed that hit any level of BTB that access the overflow/underflow return predictor that mispredict" + }, + { + "PublicDescription": "Predictable branch speculatively executed, unpredicted, that mispredict", + "EventCode": "0x11b", + "EventName": "BPU_NO_PREDICTION_MISPREDICT", + "BriefDescription": "Predictable branch speculatively executed, unpredicted, that mispredict" + }, + { + "PublicDescription": "Predictable branch speculatively executed, unpredicted, that mispredict", + "EventCode": "0x11c", + "EventName": "BPU_BTB_UPDATE", + "BriefDescription": "Predictable branch speculatively executed, unpredicted, that mispredict" + }, + { + "PublicDescription": "Count predict pipe stalls due to speculative return address predictor full", + "EventCode": "0x11d", + "EventName": "BPU_RSB_FULL_STALL", + "BriefDescription": "Count predict pipe stalls due to speculative return address predictor full" + }, + { + "PublicDescription": "Macro-ops speculatively decoded", + "EventCode": "0x11f", + "EventName": "ICF_INST_SPEC_DECODE", + "BriefDescription": "Macro-ops speculatively decoded" + }, + { + "PublicDescription": "Flushes", + "EventCode": "0x120", + "EventName": "GPC_FLUSH", + "BriefDescription": "Flushes" + }, + { + "PublicDescription": "Flushes due to memory hazards", + "EventCode": "0x121", + "EventName": "BPU_FLUSH_MEM_FAULT", + "BriefDescription": "Flushes due to memory hazards" + }, + { + "PublicDescription": "ETM extout bit 0", + "EventCode": "0x141", + "EventName": "MSC_ETM_EXTOUT0", + "BriefDescription": "ETM extout bit 0" + }, + { + "PublicDescription": "ETM extout bit 1", + "EventCode": "0x142", + "EventName": "MSC_ETM_EXTOUT1", + "BriefDescription": "ETM extout bit 1" + }, + { + "PublicDescription": "ETM extout bit 2", + "EventCode": "0x143", + "EventName": "MSC_ETM_EXTOUT2", + "BriefDescription": "ETM extout bit 2" + }, + { + "PublicDescription": "ETM extout bit 3", + "EventCode": "0x144", + "EventName": "MSC_ETM_EXTOUT3", + "BriefDescription": "ETM extout bit 3" + }, + { + "PublicDescription": "Bus request sn", + "EventCode": "0x156", + "EventName": "L2C_SNOOP", + "BriefDescription": "Bus request sn" + }, + { + "PublicDescription": "L2 TXDAT LCRD blocked", + "EventCode": "0x169", + "EventName": "L2C_DAT_CRD_STALL", + "BriefDescription": "L2 TXDAT LCRD blocked" + }, + { + "PublicDescription": "L2 TXRSP LCRD blocked", + "EventCode": "0x16a", + "EventName": "L2C_RSP_CRD_STALL", + "BriefDescription": "L2 TXRSP LCRD blocked" + }, + { + "PublicDescription": "L2 TXREQ LCRD blocked", + "EventCode": "0x16b", + "EventName": "L2C_REQ_CRD_STALL", + "BriefDescription": "L2 TXREQ LCRD blocked" + }, + { + "PublicDescription": "Early mispredict", + "EventCode": "0xD100", + "EventName": "ICF_EARLY_MIS_PRED", + "BriefDescription": "Early mispredict" + }, + { + "PublicDescription": "FEQ full cycles", + "EventCode": "0xD101", + "EventName": "ICF_FEQ_FULL", + "BriefDescription": "FEQ full cycles" + }, + { + "PublicDescription": "Instruction FIFO Full", + "EventCode": "0xD102", + "EventName": "ICF_INST_FIFO_FULL", + "BriefDescription": "Instruction FIFO Full" + }, + { + "PublicDescription": "L1I TLB miss", + "EventCode": "0xD103", + "EventName": "L1I_TLB_MISS", + "BriefDescription": "L1I TLB miss" + }, + { + "PublicDescription": "ICF sent 0 instructions to IDR this cycle", + "EventCode": "0xD104", + "EventName": "ICF_STALL", + "BriefDescription": "ICF sent 0 instructions to IDR this cycle" + }, + { + "PublicDescription": "PC FIFO Full", + "EventCode": "0xD105", + "EventName": "ICF_PC_FIFO_FULL", + "BriefDescription": "PC FIFO Full" + }, + { + "PublicDescription": "Stall due to BOB ID", + "EventCode": "0xD200", + "EventName": "IDR_STALL_BOB_ID", + "BriefDescription": "Stall due to BOB ID" + }, + { + "PublicDescription": "Dispatch stall due to LOB entries", + "EventCode": "0xD201", + "EventName": "IDR_STALL_LOB_ID", + "BriefDescription": "Dispatch stall due to LOB entries" + }, + { + "PublicDescription": "Dispatch stall due to SOB entries", + "EventCode": "0xD202", + "EventName": "IDR_STALL_SOB_ID", + "BriefDescription": "Dispatch stall due to SOB entries" + }, + { + "PublicDescription": "Dispatch stall due to IXU scheduler entries", + "EventCode": "0xD203", + "EventName": "IDR_STALL_IXU_SCHED", + "BriefDescription": "Dispatch stall due to IXU scheduler entries" + }, + { + "PublicDescription": "Dispatch stall due to FSU scheduler entries", + "EventCode": "0xD204", + "EventName": "IDR_STALL_FSU_SCHED", + "BriefDescription": "Dispatch stall due to FSU scheduler entries" + }, + { + "PublicDescription": "Dispatch stall due to ROB entries", + "EventCode": "0xD205", + "EventName": "IDR_STALL_ROB_ID", + "BriefDescription": "Dispatch stall due to ROB entries" + }, + { + "PublicDescription": "Dispatch stall due to flush (6 cycles)", + "EventCode": "0xD206", + "EventName": "IDR_STALL_FLUSH", + "BriefDescription": "Dispatch stall due to flush (6 cycles)" + }, + { + "PublicDescription": "Dispatch stall due to WFI", + "EventCode": "0xD207", + "EventName": "IDR_STALL_WFI", + "BriefDescription": "Dispatch stall due to WFI" + }, + { + "PublicDescription": "Number of SWOB drains triggered by timeout", + "EventCode": "0xD208", + "EventName": "IDR_STALL_SWOB_TIMEOUT", + "BriefDescription": "Number of SWOB drains triggered by timeout" + }, + { + "PublicDescription": "Number of SWOB drains triggered by system register or special-purpose register read-after-write or specific special-purpose register writes that cause SWOB drain", + "EventCode": "0xD209", + "EventName": "IDR_STALL_SWOB_RAW", + "BriefDescription": "Number of SWOB drains triggered by system register or special-purpose register read-after-write or specific special-purpose register writes that cause SWOB drain" + }, + { + "PublicDescription": "Number of SWOB drains triggered by system register write when SWOB full", + "EventCode": "0xD20A", + "EventName": "IDR_STALL_SWOB_FULL", + "BriefDescription": "Number of SWOB drains triggered by system register write when SWOB full" + }, + { + "PublicDescription": "Dispatch stall due to L1 instruction cache miss", + "EventCode": "0xD20B", + "EventName": "STALL_FRONTEND_CACHE", + "BriefDescription": "Dispatch stall due to L1 instruction cache miss" + }, + { + "PublicDescription": "Dispatch stall due to L1 instruction TLB miss", + "EventCode": "0xD20C", + "EventName": "STALL_FRONTEND_TLB", + "BriefDescription": "Dispatch stall due to L1 instruction TLB miss" + }, + { + "PublicDescription": "Dispatch stall due to L1 data cache miss", + "EventCode": "0xD20D", + "EventName": "STALL_BACKEND_CACHE", + "BriefDescription": "Dispatch stall due to L1 data cache miss" + }, + { + "PublicDescription": "Dispatch stall due to L1 data TLB miss", + "EventCode": "0xD20E", + "EventName": "STALL_BACKEND_TLB", + "BriefDescription": "Dispatch stall due to L1 data TLB miss" + }, + { + "PublicDescription": "Dispatch stall due to lack of any core resource", + "EventCode": "0xD20F", + "EventName": "STALL_BACKEND_RESOURCE", + "BriefDescription": "Dispatch stall due to lack of any core resource" + }, + { + "PublicDescription": "Instructions issued by the scheduler", + "EventCode": "0xD300", + "EventName": "IXU_NUM_UOPS_ISSUED", + "BriefDescription": "Instructions issued by the scheduler" + }, + { + "PublicDescription": "Any uop issued was canceled for any reason", + "EventCode": "0xD301", + "EventName": "IXU_ISSUE_CANCEL", + "BriefDescription": "Any uop issued was canceled for any reason" + }, + { + "PublicDescription": "A load wakeup to the scheduler has been cancelled", + "EventCode": "0xD302", + "EventName": "IXU_LOAD_CANCEL", + "BriefDescription": "A load wakeup to the scheduler has been cancelled" + }, + { + "PublicDescription": "The scheduler had to cancel one slow Uop due to resource conflict", + "EventCode": "0xD303", + "EventName": "IXU_SLOW_CANCEL", + "BriefDescription": "The scheduler had to cancel one slow Uop due to resource conflict" + }, + { + "PublicDescription": "Uops issued by the scheduler on IXA", + "EventCode": "0xD304", + "EventName": "IXU_IXA_ISSUED", + "BriefDescription": "Uops issued by the scheduler on IXA" + }, + { + "PublicDescription": "Uops issued by the scheduler on IXA Par 0", + "EventCode": "0xD305", + "EventName": "IXU_IXA_PAR0_ISSUED", + "BriefDescription": "Uops issued by the scheduler on IXA Par 0" + }, + { + "PublicDescription": "Uops issued by the scheduler on IXA Par 1", + "EventCode": "0xD306", + "EventName": "IXU_IXA_PAR1_ISSUED", + "BriefDescription": "Uops issued by the scheduler on IXA Par 1" + }, + { + "PublicDescription": "Uops issued by the scheduler on IXB", + "EventCode": "0xD307", + "EventName": "IXU_IXB_ISSUED", + "BriefDescription": "Uops issued by the scheduler on IXB" + }, + { + "PublicDescription": "Uops issued by the scheduler on IXB Par 0", + "EventCode": "0xD308", + "EventName": "IXU_IXB_PAR0_ISSUED", + "BriefDescription": "Uops issued by the scheduler on IXB Par 0" + }, + { + "PublicDescription": "Uops issued by the scheduler on IXB Par 1", + "EventCode": "0xD309", + "EventName": "IXU_IXB_PAR1_ISSUED", + "BriefDescription": "Uops issued by the scheduler on IXB Par 1" + }, + { + "PublicDescription": "Uops issued by the scheduler on IXC", + "EventCode": "0xD30A", + "EventName": "IXU_IXC_ISSUED", + "BriefDescription": "Uops issued by the scheduler on IXC" + }, + { + "PublicDescription": "Uops issued by the scheduler on IXC Par 0", + "EventCode": "0xD30B", + "EventName": "IXU_IXC_PAR0_ISSUED", + "BriefDescription": "Uops issued by the scheduler on IXC Par 0" + }, + { + "PublicDescription": "Uops issued by the scheduler on IXC Par 1", + "EventCode": "0xD30C", + "EventName": "IXU_IXC_PAR1_ISSUED", + "BriefDescription": "Uops issued by the scheduler on IXC Par 1" + }, + { + "PublicDescription": "Uops issued by the scheduler on IXD", + "EventCode": "0xD30D", + "EventName": "IXU_IXD_ISSUED", + "BriefDescription": "Uops issued by the scheduler on IXD" + }, + { + "PublicDescription": "Uops issued by the scheduler on IXD Par 0", + "EventCode": "0xD30E", + "EventName": "IXU_IXD_PAR0_ISSUED", + "BriefDescription": "Uops issued by the scheduler on IXD Par 0" + }, + { + "PublicDescription": "Uops issued by the scheduler on IXD Par 1", + "EventCode": "0xD30F", + "EventName": "IXU_IXD_PAR1_ISSUED", + "BriefDescription": "Uops issued by the scheduler on IXD Par 1" + }, + { + "PublicDescription": "Uops issued by the FSU scheduler", + "EventCode": "0xD400", + "EventName": "FSU_ISSUED", + "BriefDescription": "Uops issued by the FSU scheduler" + }, + { + "PublicDescription": "Uops issued by the scheduler on pipe X", + "EventCode": "0xD401", + "EventName": "FSU_FSX_ISSUED", + "BriefDescription": "Uops issued by the scheduler on pipe X" + }, + { + "PublicDescription": "Uops issued by the scheduler on pipe Y", + "EventCode": "0xD402", + "EventName": "FSU_FSY_ISSUED", + "BriefDescription": "Uops issued by the scheduler on pipe Y" + }, + { + "PublicDescription": "Uops issued by the scheduler on pipe Z", + "EventCode": "0xD403", + "EventName": "FSU_FSZ_ISSUED", + "BriefDescription": "Uops issued by the scheduler on pipe Z" + }, + { + "PublicDescription": "Uops canceled (load cancels)", + "EventCode": "0xD404", + "EventName": "FSU_CANCEL", + "BriefDescription": "Uops canceled (load cancels)" + }, + { + "PublicDescription": "Count scheduler stalls due to divide/sqrt", + "EventCode": "0xD405", + "EventName": "FSU_DIV_SQRT_STALL", + "BriefDescription": "Count scheduler stalls due to divide/sqrt" + }, + { + "PublicDescription": "Number of SWOB drains", + "EventCode": "0xD500", + "EventName": "GPC_SWOB_DRAIN", + "BriefDescription": "Number of SWOB drains" + }, + { + "PublicDescription": "GPC detected a Breakpoint instruction match", + "EventCode": "0xD501", + "EventName": "BREAKPOINT_MATCH", + "BriefDescription": "GPC detected a Breakpoint instruction match" + }, + { + "PublicDescription": "L1D TLB miss", + "EventCode": "0xD600", + "EventName": "L1D_TLB_MISS", + "BriefDescription": "L1D TLB miss" + }, + { + "PublicDescription": "OFB full cycles", + "EventCode": "0xD601", + "EventName": "OFB_FULL", + "BriefDescription": "OFB full cycles" + }, + { + "PublicDescription": "Load satisified from store forwarded data", + "EventCode": "0xD605", + "EventName": "LD_FROM_ST_FWD", + "BriefDescription": "Load satisified from store forwarded data" + }, + { + "PublicDescription": "L1 prefetcher, load prefetch requests generated", + "EventCode": "0xD606", + "EventName": "L1_PFETCH_LD_GEN", + "BriefDescription": "L1 prefetcher, load prefetch requests generated" + }, + { + "PublicDescription": "L1 prefetcher, load prefetch fills into the L1 cache", + "EventCode": "0xD607", + "EventName": "L1_PFETCH_LD_FILL", + "BriefDescription": "L1 prefetcher, load prefetch fills into the L1 cache" + }, + { + "PublicDescription": "L1 prefetcher, load prefetch to L2 generated", + "EventCode": "0xD608", + "EventName": "L1_PFETCH_L2_REQ", + "BriefDescription": "L1 prefetcher, load prefetch to L2 generated" + }, + { + "PublicDescription": "L1 prefetcher, distance was reset", + "EventCode": "0xD609", + "EventName": "L1_PFETCH_DIST_RST", + "BriefDescription": "L1 prefetcher, distance was reset" + }, + { + "PublicDescription": "L1 prefetcher, distance was increased", + "EventCode": "0xD60A", + "EventName": "L1_PFETCH_DIST_INC", + "BriefDescription": "L1 prefetcher, distance was increased" + }, + { + "PublicDescription": "L1 prefetcher, table entry is trained", + "EventCode": "0xD60B", + "EventName": "L1_PFETCH_ENTRY_TRAINED", + "BriefDescription": "L1 prefetcher, table entry is trained" + }, + { + "PublicDescription": "Store retirement pipe stall", + "EventCode": "0xD60C", + "EventName": "LSU_ST_RETIRE_STALL", + "BriefDescription": "Store retirement pipe stall" + }, + { + "PublicDescription": "LSU detected a Watchpoint data match", + "EventCode": "0xD60D", + "EventName": "WATCHPOINT_MATCH", + "BriefDescription": "LSU detected a Watchpoint data match" + }, + { + "PublicDescription": "L2 pipeline replay", + "EventCode": "0xD700", + "EventName": "L2C_PIPE_REPLAY", + "BriefDescription": "L2 pipeline replay" + }, + { + "PublicDescription": "L2 refill from I-side miss", + "EventCode": "0xD701", + "EventName": "L2C_INST_REFILL", + "BriefDescription": "L2 refill from I-side miss" + }, + { + "PublicDescription": "L2 refill from D-side miss", + "EventCode": "0xD702", + "EventName": "L2C_DATA_REFILL", + "BriefDescription": "L2 refill from D-side miss" + }, + { + "PublicDescription": "L2 prefetcher, load prefetch requests generated", + "EventCode": "0xD703", + "EventName": "L2_PREFETCH_REQ", + "BriefDescription": "L2 prefetcher, load prefetch requests generated" + }, + { + "PublicDescription": "L2D OTB allocate", + "EventCode": "0xD800", + "EventName": "MMU_D_OTB_ALLOC", + "BriefDescription": "L2D OTB allocate" + }, + { + "PublicDescription": "DTLB Translation cache hit on S1L2 walk cache entry", + "EventCode": "0xD801", + "EventName": "MMU_D_TRANS_CACHE_HIT_S1L2_WALK", + "BriefDescription": "DTLB Translation cache hit on S1L2 walk cache entry" + }, + { + "PublicDescription": "DTLB Translation cache hit on S1L1 walk cache entry", + "EventCode": "0xD802", + "EventName": "MMU_D_TRANS_CACHE_HIT_S1L1_WALK", + "BriefDescription": "DTLB Translation cache hit on S1L1 walk cache entry" + }, + { + "PublicDescription": "DTLB Translation cache hit on S1L0 walk cache entry", + "EventCode": "0xD803", + "EventName": "MMU_D_TRANS_CACHE_HIT_S1L0_WALK", + "BriefDescription": "DTLB Translation cache hit on S1L0 walk cache entry" + }, + { + "PublicDescription": "DTLB Translation cache hit on S2L2 walk cache entry", + "EventCode": "0xD804", + "EventName": "MMU_D_TRANS_CACHE_HIT_S2L2_WALK", + "BriefDescription": "DTLB Translation cache hit on S2L2 walk cache entry" + }, + { + "PublicDescription": "DTLB Translation cache hit on S2L1 walk cache entry", + "EventCode": "0xD805", + "EventName": "MMU_D_TRANS_CACHE_HIT_S2L1_WALK", + "BriefDescription": "DTLB Translation cache hit on S2L1 walk cache entry" + }, + { + "PublicDescription": "DTLB Translation cache hit on S2L0 walk cache entry", + "EventCode": "0xD806", + "EventName": "MMU_D_TRANS_CACHE_HIT_S2L0_WALK", + "BriefDescription": "DTLB Translation cache hit on S2L0 walk cache entry" + }, + { + "PublicDescription": "D-side S1 Page walk cache lookup", + "EventCode": "0xD807", + "EventName": "MMU_D_S1_WALK_CACHE_LOOKUP", + "BriefDescription": "D-side S1 Page walk cache lookup" + }, + { + "PublicDescription": "D-side S1 Page walk cache refill", + "EventCode": "0xD808", + "EventName": "MMU_D_S1_WALK_CACHE_REFILL", + "BriefDescription": "D-side S1 Page walk cache refill" + }, + { + "PublicDescription": "D-side S2 Page walk cache lookup", + "EventCode": "0xD809", + "EventName": "MMU_D_S2_WALK_CACHE_LOOKUP", + "BriefDescription": "D-side S2 Page walk cache lookup" + }, + { + "PublicDescription": "D-side S2 Page walk cache refill", + "EventCode": "0xD80A", + "EventName": "MMU_D_S2_WALK_CACHE_REFILL", + "BriefDescription": "D-side S2 Page walk cache refill" + }, + { + "PublicDescription": "D-side Stage1 tablewalk fault", + "EventCode": "0xD80B", + "EventName": "MMU_D_S1_WALK_FAULT", + "BriefDescription": "D-side Stage1 tablewalk fault" + }, + { + "PublicDescription": "D-side Stage2 tablewalk fault", + "EventCode": "0xD80C", + "EventName": "MMU_D_S2_WALK_FAULT", + "BriefDescription": "D-side Stage2 tablewalk fault" + }, + { + "PublicDescription": "D-side Tablewalk steps or descriptor fetches", + "EventCode": "0xD80D", + "EventName": "MMU_D_WALK_STEPS", + "BriefDescription": "D-side Tablewalk steps or descriptor fetches" + }, + { + "PublicDescription": "L2I OTB allocate", + "EventCode": "0xD900", + "EventName": "MMU_I_OTB_ALLOC", + "BriefDescription": "L2I OTB allocate" + }, + { + "PublicDescription": "ITLB Translation cache hit on S1L2 walk cache entry", + "EventCode": "0xD901", + "EventName": "MMU_I_TRANS_CACHE_HIT_S1L2_WALK", + "BriefDescription": "ITLB Translation cache hit on S1L2 walk cache entry" + }, + { + "PublicDescription": "ITLB Translation cache hit on S1L1 walk cache entry", + "EventCode": "0xD902", + "EventName": "MMU_I_TRANS_CACHE_HIT_S1L1_WALK", + "BriefDescription": "ITLB Translation cache hit on S1L1 walk cache entry" + }, + { + "PublicDescription": "ITLB Translation cache hit on S1L0 walk cache entry", + "EventCode": "0xD903", + "EventName": "MMU_I_TRANS_CACHE_HIT_S1L0_WALK", + "BriefDescription": "ITLB Translation cache hit on S1L0 walk cache entry" + }, + { + "PublicDescription": "ITLB Translation cache hit on S2L2 walk cache entry", + "EventCode": "0xD904", + "EventName": "MMU_I_TRANS_CACHE_HIT_S2L2_WALK", + "BriefDescription": "ITLB Translation cache hit on S2L2 walk cache entry" + }, + { + "PublicDescription": "ITLB Translation cache hit on S2L1 walk cache entry", + "EventCode": "0xD905", + "EventName": "MMU_I_TRANS_CACHE_HIT_S2L1_WALK", + "BriefDescription": "ITLB Translation cache hit on S2L1 walk cache entry" + }, + { + "PublicDescription": "ITLB Translation cache hit on S2L0 walk cache entry", + "EventCode": "0xD906", + "EventName": "MMU_I_TRANS_CACHE_HIT_S2L0_WALK", + "BriefDescription": "ITLB Translation cache hit on S2L0 walk cache entry" + }, + { + "PublicDescription": "I-side S1 Page walk cache lookup", + "EventCode": "0xD907", + "EventName": "MMU_I_S1_WALK_CACHE_LOOKUP", + "BriefDescription": "I-side S1 Page walk cache lookup" + }, + { + "PublicDescription": "I-side S1 Page walk cache refill", + "EventCode": "0xD908", + "EventName": "MMU_I_S1_WALK_CACHE_REFILL", + "BriefDescription": "I-side S1 Page walk cache refill" + }, + { + "PublicDescription": "I-side S2 Page walk cache lookup", + "EventCode": "0xD909", + "EventName": "MMU_I_S2_WALK_CACHE_LOOKUP", + "BriefDescription": "I-side S2 Page walk cache lookup" + }, + { + "PublicDescription": "I-side S2 Page walk cache refill", + "EventCode": "0xD90A", + "EventName": "MMU_I_S2_WALK_CACHE_REFILL", + "BriefDescription": "I-side S2 Page walk cache refill" + }, + { + "PublicDescription": "I-side Stage1 tablewalk fault", + "EventCode": "0xD90B", + "EventName": "MMU_I_S1_WALK_FAULT", + "BriefDescription": "I-side Stage1 tablewalk fault" + }, + { + "PublicDescription": "I-side Stage2 tablewalk fault", + "EventCode": "0xD90C", + "EventName": "MMU_I_S2_WALK_FAULT", + "BriefDescription": "I-side Stage2 tablewalk fault" + }, + { + "PublicDescription": "I-side Tablewalk steps or descriptor fetches", + "EventCode": "0xD90D", + "EventName": "MMU_I_WALK_STEPS", + "BriefDescription": "I-side Tablewalk steps or descriptor fetches" + } +] diff --git a/tools/perf/pmu-events/arch/arm64/ampere/ampereone/exception.json b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/exception.json new file mode 100644 index 000000000000..ada052e19632 --- /dev/null +++ b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/exception.json @@ -0,0 +1,44 @@ +[ + { + "ArchStdEvent": "EXC_UNDEF" + }, + { + "ArchStdEvent": "EXC_SVC" + }, + { + "ArchStdEvent": "EXC_PABORT" + }, + { + "ArchStdEvent": "EXC_DABORT" + }, + { + "ArchStdEvent": "EXC_IRQ" + }, + { + "ArchStdEvent": "EXC_FIQ" + }, + { + "ArchStdEvent": "EXC_HVC" + }, + { + "ArchStdEvent": "EXC_TRAP_PABORT" + }, + { + "ArchStdEvent": "EXC_TRAP_DABORT" + }, + { + "ArchStdEvent": "EXC_TRAP_OTHER" + }, + { + "ArchStdEvent": "EXC_TRAP_IRQ" + }, + { + "ArchStdEvent": "EXC_TRAP_FIQ" + }, + { + "ArchStdEvent": "EXC_TAKEN" + }, + { + "ArchStdEvent": "EXC_RETURN" + } +] diff --git a/tools/perf/pmu-events/arch/arm64/ampere/ampereone/instruction.json b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/instruction.json new file mode 100644 index 000000000000..18d1f2f76a23 --- /dev/null +++ b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/instruction.json @@ -0,0 +1,89 @@ +[ + { + "ArchStdEvent": "SW_INCR" + }, + { + "ArchStdEvent": "ST_RETIRED" + }, + { + "ArchStdEvent": "OP_SPEC" + }, + { + "ArchStdEvent": "LD_SPEC" + }, + { + "ArchStdEvent": "ST_SPEC" + }, + { + "ArchStdEvent": "LDST_SPEC" + }, + { + "ArchStdEvent": "DP_SPEC" + }, + { + "ArchStdEvent": "ASE_SPEC" + }, + { + "ArchStdEvent": "VFP_SPEC" + }, + { + "ArchStdEvent": "PC_WRITE_SPEC" + }, + { + "ArchStdEvent": "BR_IMMED_RETIRED" + }, + { + "ArchStdEvent": "BR_RETURN_RETIRED" + }, + { + "ArchStdEvent": "CRYPTO_SPEC" + }, + { + "ArchStdEvent": "ISB_SPEC" + }, + { + "ArchStdEvent": "DSB_SPEC" + }, + { + "ArchStdEvent": "DMB_SPEC" + }, + { + "ArchStdEvent": "RC_LD_SPEC" + }, + { + "ArchStdEvent": "RC_ST_SPEC" + }, + { + "ArchStdEvent": "INST_RETIRED" + }, + { + "ArchStdEvent": "CID_WRITE_RETIRED" + }, + { + "ArchStdEvent": "PC_WRITE_RETIRED" + }, + { + "ArchStdEvent": "INST_SPEC" + }, + { + "ArchStdEvent": "TTBR_WRITE_RETIRED" + }, + { + "ArchStdEvent": "BR_RETIRED" + }, + { + "ArchStdEvent": "BR_MIS_PRED_RETIRED" + }, + { + "ArchStdEvent": "OP_RETIRED" + }, + { + "ArchStdEvent": "OP_SPEC" + }, + { + "PublicDescription": "Operation speculatively executed, NOP", + "EventCode": "0x100", + "EventName": "NOP_SPEC", + "BriefDescription": "Speculatively executed, NOP" + } +] diff --git a/tools/perf/pmu-events/arch/arm64/ampere/ampereone/intrinsic.json b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/intrinsic.json new file mode 100644 index 000000000000..7ecffb989ae0 --- /dev/null +++ b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/intrinsic.json @@ -0,0 +1,14 @@ +[ + { + "ArchStdEvent": "LDREX_SPEC" + }, + { + "ArchStdEvent": "STREX_PASS_SPEC" + }, + { + "ArchStdEvent": "STREX_FAIL_SPEC" + }, + { + "ArchStdEvent": "STREX_SPEC" + } +] diff --git a/tools/perf/pmu-events/arch/arm64/ampere/ampereone/memory.json b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/memory.json new file mode 100644 index 000000000000..0711782bfa6b --- /dev/null +++ b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/memory.json @@ -0,0 +1,44 @@ +[ + { + "ArchStdEvent": "LD_RETIRED" + }, + { + "ArchStdEvent": "MEM_ACCESS_RD" + }, + { + "ArchStdEvent": "MEM_ACCESS_WR" + }, + { + "ArchStdEvent": "UNALIGNED_LD_SPEC" + }, + { + "ArchStdEvent": "UNALIGNED_ST_SPEC" + }, + { + "ArchStdEvent": "UNALIGNED_LDST_SPEC" + }, + { + "ArchStdEvent": "LD_ALIGN_LAT" + }, + { + "ArchStdEvent": "ST_ALIGN_LAT" + }, + { + "ArchStdEvent": "MEM_ACCESS" + }, + { + "ArchStdEvent": "MEMORY_ERROR" + }, + { + "ArchStdEvent": "LDST_ALIGN_LAT" + }, + { + "ArchStdEvent": "MEM_ACCESS_CHECKED" + }, + { + "ArchStdEvent": "MEM_ACCESS_CHECKED_RD" + }, + { + "ArchStdEvent": "MEM_ACCESS_CHECKED_WR" + } +] diff --git a/tools/perf/pmu-events/arch/arm64/ampere/ampereone/pipeline.json b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/pipeline.json new file mode 100644 index 000000000000..f9fae15f7555 --- /dev/null +++ b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/pipeline.json @@ -0,0 +1,23 @@ +[ + { + "ArchStdEvent": "STALL_FRONTEND" + }, + { + "ArchStdEvent": "STALL_BACKEND" + }, + { + "ArchStdEvent": "STALL" + }, + { + "ArchStdEvent": "STALL_SLOT_BACKEND" + }, + { + "ArchStdEvent": "STALL_SLOT_FRONTEND" + }, + { + "ArchStdEvent": "STALL_SLOT" + }, + { + "ArchStdEvent": "STALL_BACKEND_MEM" + } +] diff --git a/tools/perf/pmu-events/arch/arm64/ampere/ampereone/spe.json b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/spe.json new file mode 100644 index 000000000000..20f2165c85fe --- /dev/null +++ b/tools/perf/pmu-events/arch/arm64/ampere/ampereone/spe.json @@ -0,0 +1,14 @@ +[ + { + "ArchStdEvent": "SAMPLE_POP" + }, + { + "ArchStdEvent": "SAMPLE_FEED" + }, + { + "ArchStdEvent": "SAMPLE_FILTRATE" + }, + { + "ArchStdEvent": "SAMPLE_COLLISION" + } +] diff --git a/tools/perf/pmu-events/arch/arm64/hisilicon/hip08/metrics.json b/tools/perf/pmu-events/arch/arm64/hisilicon/hip08/metrics.json index 6443a061e22a..6463531b9941 100644 --- a/tools/perf/pmu-events/arch/arm64/hisilicon/hip08/metrics.json +++ b/tools/perf/pmu-events/arch/arm64/hisilicon/hip08/metrics.json @@ -3,28 +3,32 @@ "MetricExpr": "FETCH_BUBBLE / (4 * CPU_CYCLES)", "PublicDescription": "Frontend bound L1 topdown metric", "BriefDescription": "Frontend bound L1 topdown metric", - "MetricGroup": "TopDownL1", + "DefaultMetricgroupName": "TopDownL1", + "MetricGroup": "Default;TopDownL1", "MetricName": "frontend_bound" }, { "MetricExpr": "(INST_SPEC - INST_RETIRED) / (4 * CPU_CYCLES)", "PublicDescription": "Bad Speculation L1 topdown metric", "BriefDescription": "Bad Speculation L1 topdown metric", - "MetricGroup": "TopDownL1", + "DefaultMetricgroupName": "TopDownL1", + "MetricGroup": "Default;TopDownL1", "MetricName": "bad_speculation" }, { "MetricExpr": "INST_RETIRED / (CPU_CYCLES * 4)", "PublicDescription": "Retiring L1 topdown metric", "BriefDescription": "Retiring L1 topdown metric", - "MetricGroup": "TopDownL1", + "DefaultMetricgroupName": "TopDownL1", + "MetricGroup": "Default;TopDownL1", "MetricName": "retiring" }, { "MetricExpr": "1 - (frontend_bound + bad_speculation + retiring)", "PublicDescription": "Backend Bound L1 topdown metric", "BriefDescription": "Backend Bound L1 topdown metric", - "MetricGroup": "TopDownL1", + "DefaultMetricgroupName": "TopDownL1", + "MetricGroup": "Default;TopDownL1", "MetricName": "backend_bound" }, { diff --git a/tools/perf/pmu-events/arch/arm64/mapfile.csv b/tools/perf/pmu-events/arch/arm64/mapfile.csv index 9d400785d195..32674ddd2b63 100644 --- a/tools/perf/pmu-events/arch/arm64/mapfile.csv +++ b/tools/perf/pmu-events/arch/arm64/mapfile.csv @@ -41,3 +41,4 @@ 0x00000000460f0010,v1,fujitsu/a64fx,core 0x00000000480fd010,v1,hisilicon/hip08,core 0x00000000500f0000,v1,ampere/emag,core +0x00000000c00fac30,v1,ampere/ampereone,core diff --git a/tools/perf/pmu-events/arch/arm64/sbsa.json b/tools/perf/pmu-events/arch/arm64/sbsa.json index f678c37ea9c3..f90b338261ac 100644 --- a/tools/perf/pmu-events/arch/arm64/sbsa.json +++ b/tools/perf/pmu-events/arch/arm64/sbsa.json @@ -2,28 +2,32 @@ { "MetricExpr": "stall_slot_frontend / (#slots * cpu_cycles)", "BriefDescription": "Frontend bound L1 topdown metric", - "MetricGroup": "TopdownL1", + "DefaultMetricgroupName": "TopdownL1", + "MetricGroup": "Default;TopdownL1", "MetricName": "frontend_bound", "ScaleUnit": "100%" }, { "MetricExpr": "(1 - op_retired / op_spec) * (1 - stall_slot / (#slots * cpu_cycles))", "BriefDescription": "Bad speculation L1 topdown metric", - "MetricGroup": "TopdownL1", + "DefaultMetricgroupName": "TopdownL1", + "MetricGroup": "Default;TopdownL1", "MetricName": "bad_speculation", "ScaleUnit": "100%" }, { "MetricExpr": "(op_retired / op_spec) * (1 - stall_slot / (#slots * cpu_cycles))", "BriefDescription": "Retiring L1 topdown metric", - "MetricGroup": "TopdownL1", + "DefaultMetricgroupName": "TopdownL1", + "MetricGroup": "Default;TopdownL1", "MetricName": "retiring", "ScaleUnit": "100%" }, { "MetricExpr": "stall_slot_backend / (#slots * cpu_cycles)", "BriefDescription": "Backend Bound L1 topdown metric", - "MetricGroup": "TopdownL1", + "DefaultMetricgroupName": "TopdownL1", + "MetricGroup": "Default;TopdownL1", "MetricName": "backend_bound", "ScaleUnit": "100%" } diff --git a/tools/perf/pmu-events/arch/x86/alderlake/adl-metrics.json b/tools/perf/pmu-events/arch/x86/alderlake/adl-metrics.json index 75d80e70e5cd..85fb975b6f56 100644 --- a/tools/perf/pmu-events/arch/x86/alderlake/adl-metrics.json +++ b/tools/perf/pmu-events/arch/x86/alderlake/adl-metrics.json @@ -71,7 +71,7 @@ }, { "BriefDescription": "Uncore frequency per die [GHZ]", - "MetricExpr": "tma_info_socket_clks / #num_dies / duration_time / 1e9", + "MetricExpr": "tma_info_system_socket_clks / #num_dies / duration_time / 1e9", "MetricGroup": "SoC", "MetricName": "UNCORE_FREQ" }, @@ -120,7 +120,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to certain allocation restrictions.", - "MetricExpr": "TOPDOWN_BE_BOUND.ALLOC_RESTRICTIONS / tma_info_slots", + "MetricExpr": "TOPDOWN_BE_BOUND.ALLOC_RESTRICTIONS / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_resource_bound_group", "MetricName": "tma_alloc_restriction", "MetricThreshold": "tma_alloc_restriction > 0.1", @@ -129,46 +129,53 @@ }, { "BriefDescription": "Counts the total number of issue slots that were not consumed by the backend due to backend stalls", - "MetricExpr": "TOPDOWN_BE_BOUND.ALL / tma_info_slots", - "MetricGroup": "TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "TOPDOWN_BE_BOUND.ALL / tma_info_core_slots", + "MetricGroup": "Default;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.1", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "Counts the total number of issue slots that were not consumed by the backend due to backend stalls. Note that uops must be available for consumption in order for this event to count. If a uop is not available (IQ is empty), this event will not count. The rest of these subevents count backend stalls, in cycles, due to an outstanding request which is memory bound vs core bound. The subevents are not slot based events and therefore can not be precisely added or subtracted from the Backend_Bound_Aux subevents which are slot based.", "ScaleUnit": "100%", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the total number of issue slots that were not consumed by the backend due to backend stalls", + "DefaultMetricgroupName": "TopdownL1", "MetricExpr": "tma_backend_bound", - "MetricGroup": "TopdownL1;tma_L1_group", + "MetricGroup": "Default;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound_aux", "MetricThreshold": "tma_backend_bound_aux > 0.2", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "Counts the total number of issue slots that were not consumed by the backend due to backend stalls. Note that UOPS must be available for consumption in order for this event to count. If a uop is not available (IQ is empty), this event will not count. All of these subevents count backend stalls, in slots, due to a resource limitation. These are not cycle based events and therefore can not be precisely added or subtracted from the Backend_Bound subevents which are cycle based. These subevents are supplementary to Backend_Bound and can be used to analyze results from a resource perspective at allocation.", "ScaleUnit": "100%", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the total number of issue slots that were not consumed by the backend because allocation is stalled due to a mispredicted jump or a machine clear", - "MetricExpr": "(tma_info_slots - (TOPDOWN_FE_BOUND.ALL + TOPDOWN_BE_BOUND.ALL + TOPDOWN_RETIRING.ALL)) / tma_info_slots", - "MetricGroup": "TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "(tma_info_core_slots - (cpu_atom@TOPDOWN_FE_BOUND.ALL@ + cpu_atom@TOPDOWN_BE_BOUND.ALL@ + cpu_atom@TOPDOWN_RETIRING.ALL@)) / tma_info_core_slots", + "MetricGroup": "Default;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "Counts the total number of issue slots that were not consumed by the backend because allocation is stalled due to a mispredicted jump or a machine clear. Only issue slots wasted due to fast nukes such as memory ordering nukes are counted. Other nukes are not accounted for. Counts all issue slots blocked during this recovery window including relevant microcode flows and while uops are not yet available in the instruction queue (IQ). Also includes the issue slots that were consumed by the backend but were thrown away because they were younger than the mispredict or machine clear.", "ScaleUnit": "100%", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of uops that are not from the microsequencer.", - "MetricExpr": "(TOPDOWN_RETIRING.ALL - UOPS_RETIRED.MS) / tma_info_slots", + "MetricExpr": "(cpu_atom@TOPDOWN_RETIRING.ALL@ - cpu_atom@UOPS_RETIRED.MS@) / tma_info_core_slots", "MetricGroup": "TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_base", "MetricThreshold": "tma_base > 0.6", + "MetricgroupNoGroup": "TopdownL2", "ScaleUnit": "100%", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to BACLEARS, which occurs when the Branch Target Buffer (BTB) prediction or lack thereof, was corrected by a later branch predictor in the frontend", - "MetricExpr": "TOPDOWN_FE_BOUND.BRANCH_DETECT / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.BRANCH_DETECT / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_detect", "MetricThreshold": "tma_branch_detect > 0.05", @@ -178,16 +185,17 @@ }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to branch mispredicts.", - "MetricExpr": "TOPDOWN_BAD_SPECULATION.MISPREDICT / tma_info_slots", + "MetricExpr": "TOPDOWN_BAD_SPECULATION.MISPREDICT / tma_info_core_slots", "MetricGroup": "TopdownL2;tma_L2_group;tma_bad_speculation_group", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.05", + "MetricgroupNoGroup": "TopdownL2", "ScaleUnit": "100%", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to BTCLEARS, which occurs when the Branch Target Buffer (BTB) predicts a taken branch.", - "MetricExpr": "TOPDOWN_FE_BOUND.BRANCH_RESTEER / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.BRANCH_RESTEER / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteer", "MetricThreshold": "tma_branch_resteer > 0.05", @@ -196,7 +204,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to the microcode sequencer (MS).", - "MetricExpr": "TOPDOWN_FE_BOUND.CISC / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.CISC / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_cisc", "MetricThreshold": "tma_cisc > 0.05", @@ -209,12 +217,13 @@ "MetricGroup": "TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1", + "MetricgroupNoGroup": "TopdownL2", "ScaleUnit": "100%", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to decode stalls.", - "MetricExpr": "TOPDOWN_FE_BOUND.DECODE / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.DECODE / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_decode", "MetricThreshold": "tma_decode > 0.05", @@ -223,7 +232,7 @@ }, { "BriefDescription": "Counts the number of machine clears relative to the number of nuke slots due to memory disambiguation.", - "MetricExpr": "tma_nuke * (MACHINE_CLEARS.DISAMBIGUATION / MACHINE_CLEARS.SLOW)", + "MetricExpr": "tma_nuke * (cpu_atom@MACHINE_CLEARS.DISAMBIGUATION@ / cpu_atom@MACHINE_CLEARS.SLOW@)", "MetricGroup": "TopdownL4;tma_L4_group;tma_nuke_group", "MetricName": "tma_disambiguation", "MetricThreshold": "tma_disambiguation > 0.02", @@ -233,7 +242,7 @@ { "BriefDescription": "Counts the number of cycles the core is stalled due to a demand load miss which hit in DRAM or MMIO (Non-DRAM).", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEM_BOUND_STALLS.LOAD_DRAM_HIT / tma_info_clks - MEM_BOUND_STALLS_AT_RET_CORRECTION * MEM_BOUND_STALLS.LOAD_DRAM_HIT / MEM_BOUND_STALLS.LOAD", + "MetricExpr": "cpu_atom@MEM_BOUND_STALLS.LOAD_DRAM_HIT@ / tma_info_core_clks - max((cpu_atom@MEM_BOUND_STALLS.LOAD@ - cpu_atom@LD_HEAD.L1_MISS_AT_RET@) / tma_info_core_clks, 0) * cpu_atom@MEM_BOUND_STALLS.LOAD_DRAM_HIT@ / cpu_atom@MEM_BOUND_STALLS.LOAD@", "MetricGroup": "TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1", @@ -242,7 +251,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to a machine clear classified as a fast nuke due to memory ordering, memory disambiguation and memory renaming.", - "MetricExpr": "TOPDOWN_BAD_SPECULATION.FASTNUKE / tma_info_slots", + "MetricExpr": "TOPDOWN_BAD_SPECULATION.FASTNUKE / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_machine_clears_group", "MetricName": "tma_fast_nuke", "MetricThreshold": "tma_fast_nuke > 0.05", @@ -251,25 +260,27 @@ }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to frontend bandwidth restrictions due to decode, predecode, cisc, and other limitations.", - "MetricExpr": "TOPDOWN_FE_BOUND.FRONTEND_BANDWIDTH / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.FRONTEND_BANDWIDTH / tma_info_core_slots", "MetricGroup": "TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_bandwidth", "MetricThreshold": "tma_fetch_bandwidth > 0.1", + "MetricgroupNoGroup": "TopdownL2", "ScaleUnit": "100%", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to frontend bandwidth restrictions due to decode, predecode, cisc, and other limitations.", - "MetricExpr": "TOPDOWN_FE_BOUND.FRONTEND_LATENCY / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.FRONTEND_LATENCY / tma_info_core_slots", "MetricGroup": "TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.15", + "MetricgroupNoGroup": "TopdownL2", "ScaleUnit": "100%", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of machine clears relative to the number of nuke slots due to FP assists.", - "MetricExpr": "tma_nuke * (MACHINE_CLEARS.FP_ASSIST / MACHINE_CLEARS.SLOW)", + "MetricExpr": "tma_nuke * (cpu_atom@MACHINE_CLEARS.FP_ASSIST@ / cpu_atom@MACHINE_CLEARS.SLOW@)", "MetricGroup": "TopdownL4;tma_L4_group;tma_nuke_group", "MetricName": "tma_fp_assist", "MetricThreshold": "tma_fp_assist > 0.02", @@ -278,7 +289,7 @@ }, { "BriefDescription": "Counts the number of floating point divide operations per uop.", - "MetricExpr": "UOPS_RETIRED.FPDIV / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.FPDIV / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_base_group", "MetricName": "tma_fpdiv_uops", "MetricThreshold": "tma_fpdiv_uops > 0.2", @@ -287,16 +298,18 @@ }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to frontend stalls.", - "MetricExpr": "TOPDOWN_FE_BOUND.ALL / tma_info_slots", - "MetricGroup": "TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "TOPDOWN_FE_BOUND.ALL / tma_info_core_slots", + "MetricGroup": "Default;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1;Default", "ScaleUnit": "100%", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to instruction cache misses.", - "MetricExpr": "TOPDOWN_FE_BOUND.ICACHE / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.ICACHE / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_icache_misses", "MetricThreshold": "tma_icache_misses > 0.05", @@ -304,246 +317,220 @@ "Unit": "cpu_atom" }, { - "BriefDescription": "Percentage of total non-speculative loads with a address aliasing block", - "MetricExpr": "100 * LD_BLOCKS.4K_ALIAS / MEM_UOPS_RETIRED.ALL_LOADS", - "MetricName": "tma_info_address_alias_blocks", - "Unit": "cpu_atom" - }, - { - "BriefDescription": "Ratio of all branches which mispredict", - "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": " ", - "MetricName": "tma_info_branch_mispredict_ratio", - "Unit": "cpu_atom" - }, - { - "BriefDescription": "Ratio between Mispredicted branches and unknown branches", - "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / BACLEARS.ANY", - "MetricGroup": " ", - "MetricName": "tma_info_branch_mispredict_to_unknown_branch_ratio", - "Unit": "cpu_atom" - }, - { "BriefDescription": "", - "MetricExpr": "CPU_CLK_UNHALTED.CORE", - "MetricGroup": " ", - "MetricName": "tma_info_clks", + "MetricExpr": "cpu_atom@CPU_CLK_UNHALTED.CORE@", + "MetricName": "tma_info_core_clks", "Unit": "cpu_atom" }, { "BriefDescription": "", - "MetricExpr": "CPU_CLK_UNHALTED.CORE_P", - "MetricGroup": " ", - "MetricName": "tma_info_clks_p", + "MetricExpr": "cpu_atom@CPU_CLK_UNHALTED.CORE_P@", + "MetricName": "tma_info_core_clks_p", "Unit": "cpu_atom" }, { "BriefDescription": "Cycles Per Instruction", - "MetricExpr": "tma_info_clks / INST_RETIRED.ANY", - "MetricGroup": " ", - "MetricName": "tma_info_cpi", + "MetricExpr": "tma_info_core_clks / INST_RETIRED.ANY", + "MetricName": "tma_info_core_cpi", "Unit": "cpu_atom" }, { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": " ", - "MetricName": "tma_info_cpu_utilization", + "BriefDescription": "Instructions Per Cycle", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricName": "tma_info_core_ipc", "Unit": "cpu_atom" }, { - "BriefDescription": "Cycle cost per DRAM hit", - "MetricExpr": "MEM_BOUND_STALLS.LOAD_DRAM_HIT / MEM_LOAD_UOPS_RETIRED.DRAM_HIT", - "MetricGroup": " ", - "MetricName": "tma_info_cycles_per_demand_load_dram_hit", + "BriefDescription": "", + "MetricExpr": "5 * tma_info_core_clks", + "MetricName": "tma_info_core_slots", "Unit": "cpu_atom" }, { - "BriefDescription": "Cycle cost per L2 hit", - "MetricExpr": "MEM_BOUND_STALLS.LOAD_L2_HIT / MEM_LOAD_UOPS_RETIRED.L2_HIT", - "MetricGroup": " ", - "MetricName": "tma_info_cycles_per_demand_load_l2_hit", + "BriefDescription": "Uops Per Instruction", + "MetricExpr": "UOPS_RETIRED.ALL / INST_RETIRED.ANY", + "MetricName": "tma_info_core_upi", "Unit": "cpu_atom" }, { - "BriefDescription": "Cycle cost per LLC hit", - "MetricExpr": "MEM_BOUND_STALLS.LOAD_LLC_HIT / MEM_LOAD_UOPS_RETIRED.L3_HIT", - "MetricGroup": " ", - "MetricName": "tma_info_cycles_per_demand_load_l3_hit", + "BriefDescription": "Percent of instruction miss cost that hit in DRAM", + "MetricExpr": "100 * cpu_atom@MEM_BOUND_STALLS.IFETCH_DRAM_HIT@ / cpu_atom@MEM_BOUND_STALLS.IFETCH@", + "MetricName": "tma_info_frontend_inst_miss_cost_dramhit_percent", "Unit": "cpu_atom" }, { - "BriefDescription": "Percentage of all uops which are FPDiv uops", - "MetricExpr": "100 * UOPS_RETIRED.FPDIV / UOPS_RETIRED.ALL", - "MetricGroup": " ", - "MetricName": "tma_info_fpdiv_uop_ratio", + "BriefDescription": "Percent of instruction miss cost that hit in the L2", + "MetricExpr": "100 * cpu_atom@MEM_BOUND_STALLS.IFETCH_L2_HIT@ / cpu_atom@MEM_BOUND_STALLS.IFETCH@", + "MetricName": "tma_info_frontend_inst_miss_cost_l2hit_percent", "Unit": "cpu_atom" }, { - "BriefDescription": "Percentage of all uops which are IDiv uops", - "MetricExpr": "100 * UOPS_RETIRED.IDIV / UOPS_RETIRED.ALL", - "MetricGroup": " ", - "MetricName": "tma_info_idiv_uop_ratio", + "BriefDescription": "Percent of instruction miss cost that hit in the L3", + "MetricExpr": "100 * cpu_atom@MEM_BOUND_STALLS.IFETCH_LLC_HIT@ / cpu_atom@MEM_BOUND_STALLS.IFETCH@", + "MetricName": "tma_info_frontend_inst_miss_cost_l3hit_percent", "Unit": "cpu_atom" }, { - "BriefDescription": "Percent of instruction miss cost that hit in DRAM", - "MetricExpr": "100 * MEM_BOUND_STALLS.IFETCH_DRAM_HIT / MEM_BOUND_STALLS.IFETCH", - "MetricGroup": " ", - "MetricName": "tma_info_inst_miss_cost_dramhit_percent", + "BriefDescription": "Ratio of all branches which mispredict", + "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.ALL_BRANCHES", + "MetricName": "tma_info_inst_mix_branch_mispredict_ratio", "Unit": "cpu_atom" }, { - "BriefDescription": "Percent of instruction miss cost that hit in the L2", - "MetricExpr": "100 * MEM_BOUND_STALLS.IFETCH_L2_HIT / MEM_BOUND_STALLS.IFETCH", - "MetricGroup": " ", - "MetricName": "tma_info_inst_miss_cost_l2hit_percent", + "BriefDescription": "Ratio between Mispredicted branches and unknown branches", + "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / BACLEARS.ANY", + "MetricName": "tma_info_inst_mix_branch_mispredict_to_unknown_branch_ratio", "Unit": "cpu_atom" }, { - "BriefDescription": "Percent of instruction miss cost that hit in the L3", - "MetricExpr": "100 * MEM_BOUND_STALLS.IFETCH_LLC_HIT / MEM_BOUND_STALLS.IFETCH", - "MetricGroup": " ", - "MetricName": "tma_info_inst_miss_cost_l3hit_percent", + "BriefDescription": "Percentage of all uops which are FPDiv uops", + "MetricExpr": "100 * cpu_atom@UOPS_RETIRED.FPDIV@ / UOPS_RETIRED.ALL", + "MetricName": "tma_info_inst_mix_fpdiv_uop_ratio", "Unit": "cpu_atom" }, { - "BriefDescription": "Instructions per Branch (lower number means higher occurance rate)", - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": " ", - "MetricName": "tma_info_ipbranch", + "BriefDescription": "Percentage of all uops which are IDiv uops", + "MetricExpr": "100 * cpu_atom@UOPS_RETIRED.IDIV@ / UOPS_RETIRED.ALL", + "MetricName": "tma_info_inst_mix_idiv_uop_ratio", "Unit": "cpu_atom" }, { - "BriefDescription": "Instructions Per Cycle", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": " ", - "MetricName": "tma_info_ipc", + "BriefDescription": "Instructions per Branch (lower number means higher occurance rate)", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", + "MetricName": "tma_info_inst_mix_ipbranch", "Unit": "cpu_atom" }, { "BriefDescription": "Instruction per (near) call (lower number means higher occurance rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.CALL", - "MetricGroup": " ", - "MetricName": "tma_info_ipcall", + "MetricName": "tma_info_inst_mix_ipcall", "Unit": "cpu_atom" }, { "BriefDescription": "Instructions per Far Branch", - "MetricExpr": "INST_RETIRED.ANY / (BR_INST_RETIRED.FAR_BRANCH / 2)", - "MetricGroup": " ", - "MetricName": "tma_info_ipfarbranch", + "MetricExpr": "INST_RETIRED.ANY / (cpu_atom@BR_INST_RETIRED.FAR_BRANCH@ / 2)", + "MetricName": "tma_info_inst_mix_ipfarbranch", "Unit": "cpu_atom" }, { "BriefDescription": "Instructions per Load", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_LOADS", - "MetricGroup": " ", - "MetricName": "tma_info_ipload", + "MetricName": "tma_info_inst_mix_ipload", "Unit": "cpu_atom" }, { "BriefDescription": "Instructions per retired conditional Branch Misprediction where the branch was not taken", - "MetricExpr": "INST_RETIRED.ANY / (BR_MISP_RETIRED.COND - BR_MISP_RETIRED.COND_TAKEN)", - "MetricName": "tma_info_ipmisp_cond_ntaken", + "MetricExpr": "INST_RETIRED.ANY / (cpu_atom@BR_MISP_RETIRED.COND@ - cpu_atom@BR_MISP_RETIRED.COND_TAKEN@)", + "MetricName": "tma_info_inst_mix_ipmisp_cond_ntaken", "Unit": "cpu_atom" }, { "BriefDescription": "Instructions per retired conditional Branch Misprediction where the branch was taken", "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_TAKEN", - "MetricName": "tma_info_ipmisp_cond_taken", + "MetricName": "tma_info_inst_mix_ipmisp_cond_taken", "Unit": "cpu_atom" }, { "BriefDescription": "Instructions per retired indirect call or jump Branch Misprediction", "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.INDIRECT", - "MetricName": "tma_info_ipmisp_indirect", + "MetricName": "tma_info_inst_mix_ipmisp_indirect", "Unit": "cpu_atom" }, { "BriefDescription": "Instructions per retired return Branch Misprediction", "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.RETURN", - "MetricName": "tma_info_ipmisp_ret", + "MetricName": "tma_info_inst_mix_ipmisp_ret", "Unit": "cpu_atom" }, { "BriefDescription": "Instructions per retired Branch Misprediction", "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": " ", - "MetricName": "tma_info_ipmispredict", + "MetricName": "tma_info_inst_mix_ipmispredict", "Unit": "cpu_atom" }, { "BriefDescription": "Instructions per Store", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_STORES", - "MetricGroup": " ", - "MetricName": "tma_info_ipstore", + "MetricName": "tma_info_inst_mix_ipstore", "Unit": "cpu_atom" }, { - "BriefDescription": "Fraction of cycles spent in Kernel mode", - "MetricExpr": "cpu_atom@CPU_CLK_UNHALTED.CORE@k / CPU_CLK_UNHALTED.CORE", - "MetricGroup": " ", - "MetricName": "tma_info_kernel_utilization", + "BriefDescription": "Percentage of all uops which are ucode ops", + "MetricExpr": "100 * cpu_atom@UOPS_RETIRED.MS@ / UOPS_RETIRED.ALL", + "MetricName": "tma_info_inst_mix_microcode_uop_ratio", + "Unit": "cpu_atom" + }, + { + "BriefDescription": "Percentage of all uops which are x87 uops", + "MetricExpr": "100 * cpu_atom@UOPS_RETIRED.X87@ / UOPS_RETIRED.ALL", + "MetricName": "tma_info_inst_mix_x87_uop_ratio", + "Unit": "cpu_atom" + }, + { + "BriefDescription": "Percentage of total non-speculative loads with a address aliasing block", + "MetricExpr": "100 * cpu_atom@LD_BLOCKS.4K_ALIAS@ / MEM_UOPS_RETIRED.ALL_LOADS", + "MetricName": "tma_info_l1_bound_address_alias_blocks", "Unit": "cpu_atom" }, { "BriefDescription": "Percentage of total non-speculative loads that are splits", - "MetricExpr": "100 * MEM_UOPS_RETIRED.SPLIT_LOADS / MEM_UOPS_RETIRED.ALL_LOADS", - "MetricName": "tma_info_load_splits", + "MetricExpr": "100 * cpu_atom@MEM_UOPS_RETIRED.SPLIT_LOADS@ / MEM_UOPS_RETIRED.ALL_LOADS", + "MetricName": "tma_info_l1_bound_load_splits", "Unit": "cpu_atom" }, { - "BriefDescription": "load ops retired per 1000 instruction", - "MetricExpr": "1e3 * MEM_UOPS_RETIRED.ALL_LOADS / INST_RETIRED.ANY", - "MetricGroup": " ", - "MetricName": "tma_info_memloadpki", + "BriefDescription": "Percentage of total non-speculative loads with a store forward or unknown store address block", + "MetricExpr": "100 * cpu_atom@LD_BLOCKS.DATA_UNKNOWN@ / MEM_UOPS_RETIRED.ALL_LOADS", + "MetricName": "tma_info_l1_bound_store_fwd_blocks", "Unit": "cpu_atom" }, { - "BriefDescription": "Percentage of all uops which are ucode ops", - "MetricExpr": "100 * UOPS_RETIRED.MS / UOPS_RETIRED.ALL", - "MetricGroup": " ", - "MetricName": "tma_info_microcode_uop_ratio", + "BriefDescription": "Cycle cost per DRAM hit", + "MetricExpr": "MEM_BOUND_STALLS.LOAD_DRAM_HIT / MEM_LOAD_UOPS_RETIRED.DRAM_HIT", + "MetricName": "tma_info_memory_cycles_per_demand_load_dram_hit", "Unit": "cpu_atom" }, { - "BriefDescription": "", - "MetricExpr": "5 * tma_info_clks", - "MetricGroup": " ", - "MetricName": "tma_info_slots", + "BriefDescription": "Cycle cost per L2 hit", + "MetricExpr": "MEM_BOUND_STALLS.LOAD_L2_HIT / MEM_LOAD_UOPS_RETIRED.L2_HIT", + "MetricName": "tma_info_memory_cycles_per_demand_load_l2_hit", "Unit": "cpu_atom" }, { - "BriefDescription": "Percentage of total non-speculative loads with a store forward or unknown store address block", - "MetricExpr": "100 * LD_BLOCKS.DATA_UNKNOWN / MEM_UOPS_RETIRED.ALL_LOADS", - "MetricName": "tma_info_store_fwd_blocks", + "BriefDescription": "Cycle cost per LLC hit", + "MetricExpr": "MEM_BOUND_STALLS.LOAD_LLC_HIT / MEM_LOAD_UOPS_RETIRED.L3_HIT", + "MetricName": "tma_info_memory_cycles_per_demand_load_l3_hit", "Unit": "cpu_atom" }, { - "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", - "MetricGroup": " ", - "MetricName": "tma_info_turbo_utilization", + "BriefDescription": "load ops retired per 1000 instruction", + "MetricExpr": "1e3 * cpu_atom@MEM_UOPS_RETIRED.ALL_LOADS@ / INST_RETIRED.ANY", + "MetricName": "tma_info_memory_memloadpki", "Unit": "cpu_atom" }, { - "BriefDescription": "Uops Per Instruction", - "MetricExpr": "UOPS_RETIRED.ALL / INST_RETIRED.ANY", - "MetricGroup": " ", - "MetricName": "tma_info_upi", + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricName": "tma_info_system_cpu_utilization", "Unit": "cpu_atom" }, { - "BriefDescription": "Percentage of all uops which are x87 uops", - "MetricExpr": "100 * UOPS_RETIRED.X87 / UOPS_RETIRED.ALL", - "MetricGroup": " ", - "MetricName": "tma_info_x87_uop_ratio", + "BriefDescription": "Fraction of cycles spent in Kernel mode", + "MetricExpr": "cpu_atom@CPU_CLK_UNHALTED.CORE@k / CPU_CLK_UNHALTED.CORE", + "MetricGroup": "Summary", + "MetricName": "tma_info_system_kernel_utilization", + "Unit": "cpu_atom" + }, + { + "BriefDescription": "Average Frequency Utilization relative nominal frequency", + "MetricExpr": "tma_info_core_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricGroup": "Power", + "MetricName": "tma_info_system_turbo_utilization", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to Instruction Table Lookaside Buffer (ITLB) misses.", - "MetricExpr": "TOPDOWN_FE_BOUND.ITLB / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.ITLB / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05", @@ -552,7 +539,7 @@ }, { "BriefDescription": "Counts the number of cycles that the oldest load of the load buffer is stalled at retirement due to a load block.", - "MetricExpr": "LD_HEAD.L1_BOUND_AT_RET / tma_info_clks", + "MetricExpr": "LD_HEAD.L1_BOUND_AT_RET / tma_info_core_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l1_bound", "MetricThreshold": "tma_l1_bound > 0.1", @@ -562,7 +549,7 @@ { "BriefDescription": "Counts the number of cycles a core is stalled due to a demand load which hit in the L2 Cache.", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEM_BOUND_STALLS.LOAD_L2_HIT / tma_info_clks - MEM_BOUND_STALLS_AT_RET_CORRECTION * MEM_BOUND_STALLS.LOAD_L2_HIT / MEM_BOUND_STALLS.LOAD", + "MetricExpr": "cpu_atom@MEM_BOUND_STALLS.LOAD_L2_HIT@ / tma_info_core_clks - max((cpu_atom@MEM_BOUND_STALLS.LOAD@ - cpu_atom@LD_HEAD.L1_MISS_AT_RET@) / tma_info_core_clks, 0) * cpu_atom@MEM_BOUND_STALLS.LOAD_L2_HIT@ / cpu_atom@MEM_BOUND_STALLS.LOAD@", "MetricGroup": "TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l2_bound", "MetricThreshold": "tma_l2_bound > 0.1", @@ -571,7 +558,7 @@ }, { "BriefDescription": "Counts the number of cycles a core is stalled due to a demand load which hit in the Last Level Cache (LLC) or other core with HITE/F/M.", - "MetricExpr": "MEM_BOUND_STALLS.LOAD_LLC_HIT / tma_info_clks - MEM_BOUND_STALLS_AT_RET_CORRECTION * MEM_BOUND_STALLS.LOAD_LLC_HIT / MEM_BOUND_STALLS.LOAD", + "MetricExpr": "cpu_atom@MEM_BOUND_STALLS.LOAD_LLC_HIT@ / tma_info_core_clks - max((cpu_atom@MEM_BOUND_STALLS.LOAD@ - cpu_atom@LD_HEAD.L1_MISS_AT_RET@) / tma_info_core_clks, 0) * cpu_atom@MEM_BOUND_STALLS.LOAD_LLC_HIT@ / cpu_atom@MEM_BOUND_STALLS.LOAD@", "MetricGroup": "TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.1", @@ -580,7 +567,7 @@ }, { "BriefDescription": "Counts the number of cycles, relative to the number of mem_scheduler slots, in which uops are blocked due to load buffer full", - "MetricExpr": "tma_mem_scheduler * MEM_SCHEDULER_BLOCK.LD_BUF / MEM_SCHEDULER_BLOCK.ALL", + "MetricExpr": "tma_mem_scheduler * cpu_atom@MEM_SCHEDULER_BLOCK.LD_BUF@ / MEM_SCHEDULER_BLOCK.ALL", "MetricGroup": "TopdownL4;tma_L4_group;tma_mem_scheduler_group", "MetricName": "tma_ld_buffer", "MetricThreshold": "tma_ld_buffer > 0.05", @@ -589,16 +576,17 @@ }, { "BriefDescription": "Counts the total number of issue slots that were not consumed by the backend because allocation is stalled due to a machine clear (nuke) of any kind including memory ordering and memory disambiguation.", - "MetricExpr": "TOPDOWN_BAD_SPECULATION.MACHINE_CLEARS / tma_info_slots", + "MetricExpr": "TOPDOWN_BAD_SPECULATION.MACHINE_CLEARS / tma_info_core_slots", "MetricGroup": "TopdownL2;tma_L2_group;tma_bad_speculation_group", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.05", + "MetricgroupNoGroup": "TopdownL2", "ScaleUnit": "100%", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to memory reservation stalls in which a scheduler is not able to accept uops.", - "MetricExpr": "TOPDOWN_BE_BOUND.MEM_SCHEDULER / tma_info_slots", + "MetricExpr": "TOPDOWN_BE_BOUND.MEM_SCHEDULER / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_resource_bound_group", "MetricName": "tma_mem_scheduler", "MetricThreshold": "tma_mem_scheduler > 0.1", @@ -607,16 +595,17 @@ }, { "BriefDescription": "Counts the number of cycles the core is stalled due to stores or loads.", - "MetricExpr": "min(tma_backend_bound, LD_HEAD.ANY_AT_RET / tma_info_clks + tma_store_bound)", + "MetricExpr": "min(cpu_atom@TOPDOWN_BE_BOUND.ALL@ / tma_info_core_slots, cpu_atom@LD_HEAD.ANY_AT_RET@ / tma_info_core_clks + tma_store_bound)", "MetricGroup": "TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "ScaleUnit": "100%", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of machine clears relative to the number of nuke slots due to memory ordering.", - "MetricExpr": "tma_nuke * (MACHINE_CLEARS.MEMORY_ORDERING / MACHINE_CLEARS.SLOW)", + "MetricExpr": "tma_nuke * (cpu_atom@MACHINE_CLEARS.MEMORY_ORDERING@ / cpu_atom@MACHINE_CLEARS.SLOW@)", "MetricGroup": "TopdownL4;tma_L4_group;tma_nuke_group", "MetricName": "tma_memory_ordering", "MetricThreshold": "tma_memory_ordering > 0.02", @@ -625,17 +614,18 @@ }, { "BriefDescription": "Counts the number of uops that are from the complex flows issued by the micro-sequencer (MS)", - "MetricExpr": "tma_microcode_sequencer", + "MetricExpr": "UOPS_RETIRED.MS / tma_info_core_slots", "MetricGroup": "TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_ms_uops", "MetricThreshold": "tma_ms_uops > 0.05", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "Counts the number of uops that are from the complex flows issued by the micro-sequencer (MS). This includes uops from flows due to complex instructions, faults, assists, and inserted flows.", "ScaleUnit": "100%", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to IEC or FPC RAT stalls, which can be due to FIQ or IEC reservation stalls in which the integer, floating point or SIMD scheduler is not able to accept uops.", - "MetricExpr": "TOPDOWN_BE_BOUND.NON_MEM_SCHEDULER / tma_info_slots", + "MetricExpr": "TOPDOWN_BE_BOUND.NON_MEM_SCHEDULER / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_resource_bound_group", "MetricName": "tma_non_mem_scheduler", "MetricThreshold": "tma_non_mem_scheduler > 0.1", @@ -644,7 +634,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to a machine clear (slow nuke).", - "MetricExpr": "TOPDOWN_BAD_SPECULATION.NUKE / tma_info_slots", + "MetricExpr": "TOPDOWN_BAD_SPECULATION.NUKE / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_machine_clears_group", "MetricName": "tma_nuke", "MetricThreshold": "tma_nuke > 0.05", @@ -653,7 +643,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to other common frontend stalls not categorized.", - "MetricExpr": "TOPDOWN_FE_BOUND.OTHER / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.OTHER / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_other_fb", "MetricThreshold": "tma_other_fb > 0.05", @@ -662,7 +652,7 @@ }, { "BriefDescription": "Counts the number of cycles that the oldest load of the load buffer is stalled at retirement due to a number of other load blocks.", - "MetricExpr": "LD_HEAD.OTHER_AT_RET / tma_info_clks", + "MetricExpr": "LD_HEAD.OTHER_AT_RET / tma_info_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_other_l1", "MetricThreshold": "tma_other_l1 > 0.05", @@ -680,7 +670,7 @@ }, { "BriefDescription": "Counts the number of uops retired excluding ms and fp div uops.", - "MetricExpr": "(TOPDOWN_RETIRING.ALL - UOPS_RETIRED.MS - UOPS_RETIRED.FPDIV) / tma_info_slots", + "MetricExpr": "(cpu_atom@TOPDOWN_RETIRING.ALL@ - cpu_atom@UOPS_RETIRED.MS@ - cpu_atom@UOPS_RETIRED.FPDIV@) / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_base_group", "MetricName": "tma_other_ret", "MetricThreshold": "tma_other_ret > 0.3", @@ -689,7 +679,7 @@ }, { "BriefDescription": "Counts the number of machine clears relative to the number of nuke slots due to page faults.", - "MetricExpr": "tma_nuke * (MACHINE_CLEARS.PAGE_FAULT / MACHINE_CLEARS.SLOW)", + "MetricExpr": "tma_nuke * (cpu_atom@MACHINE_CLEARS.PAGE_FAULT@ / cpu_atom@MACHINE_CLEARS.SLOW@)", "MetricGroup": "TopdownL4;tma_L4_group;tma_nuke_group", "MetricName": "tma_page_fault", "MetricThreshold": "tma_page_fault > 0.02", @@ -698,7 +688,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to wrong predecodes.", - "MetricExpr": "TOPDOWN_FE_BOUND.PREDECODE / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.PREDECODE / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_predecode", "MetricThreshold": "tma_predecode > 0.05", @@ -707,7 +697,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to the physical register file unable to accept an entry (marble stalls).", - "MetricExpr": "TOPDOWN_BE_BOUND.REGISTER / tma_info_slots", + "MetricExpr": "TOPDOWN_BE_BOUND.REGISTER / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_resource_bound_group", "MetricName": "tma_register", "MetricThreshold": "tma_register > 0.1", @@ -716,7 +706,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to the reorder buffer being full (ROB stalls).", - "MetricExpr": "TOPDOWN_BE_BOUND.REORDER_BUFFER / tma_info_slots", + "MetricExpr": "TOPDOWN_BE_BOUND.REORDER_BUFFER / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_resource_bound_group", "MetricName": "tma_reorder_buffer", "MetricThreshold": "tma_reorder_buffer > 0.1", @@ -729,22 +719,25 @@ "MetricGroup": "TopdownL2;tma_L2_group;tma_backend_bound_aux_group", "MetricName": "tma_resource_bound", "MetricThreshold": "tma_resource_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "Counts the total number of issue slots that were not consumed by the backend due to backend stalls. Note that uops must be available for consumption in order for this event to count. If a uop is not available (IQ is empty), this event will not count.", "ScaleUnit": "100%", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the numer of issue slots that result in retirement slots.", - "MetricExpr": "TOPDOWN_RETIRING.ALL / tma_info_slots", - "MetricGroup": "TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "TOPDOWN_RETIRING.ALL / tma_info_core_slots", + "MetricGroup": "Default;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.75", + "MetricgroupNoGroup": "TopdownL1;Default", "ScaleUnit": "100%", "Unit": "cpu_atom" }, { "BriefDescription": "Counts the number of cycles, relative to the number of mem_scheduler slots, in which uops are blocked due to RSV full relative", - "MetricExpr": "tma_mem_scheduler * MEM_SCHEDULER_BLOCK.RSV / MEM_SCHEDULER_BLOCK.ALL", + "MetricExpr": "tma_mem_scheduler * cpu_atom@MEM_SCHEDULER_BLOCK.RSV@ / MEM_SCHEDULER_BLOCK.ALL", "MetricGroup": "TopdownL4;tma_L4_group;tma_mem_scheduler_group", "MetricName": "tma_rsv", "MetricThreshold": "tma_rsv > 0.05", @@ -753,7 +746,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to scoreboards from the instruction queue (IQ), jump execution unit (JEU), or microcode sequencer (MS).", - "MetricExpr": "TOPDOWN_BE_BOUND.SERIALIZATION / tma_info_slots", + "MetricExpr": "TOPDOWN_BE_BOUND.SERIALIZATION / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_resource_bound_group", "MetricName": "tma_serialization", "MetricThreshold": "tma_serialization > 0.1", @@ -762,7 +755,7 @@ }, { "BriefDescription": "Counts the number of machine clears relative to the number of nuke slots due to SMC.", - "MetricExpr": "tma_nuke * (MACHINE_CLEARS.SMC / MACHINE_CLEARS.SLOW)", + "MetricExpr": "tma_nuke * (cpu_atom@MACHINE_CLEARS.SMC@ / cpu_atom@MACHINE_CLEARS.SLOW@)", "MetricGroup": "TopdownL4;tma_L4_group;tma_nuke_group", "MetricName": "tma_smc", "MetricThreshold": "tma_smc > 0.02", @@ -780,7 +773,7 @@ }, { "BriefDescription": "Counts the number of cycles that the oldest load of the load buffer is stalled at retirement due to a first level TLB miss.", - "MetricExpr": "LD_HEAD.DTLB_MISS_AT_RET / tma_info_clks", + "MetricExpr": "LD_HEAD.DTLB_MISS_AT_RET / tma_info_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_stlb_hit", "MetricThreshold": "tma_stlb_hit > 0.05", @@ -789,7 +782,7 @@ }, { "BriefDescription": "Counts the number of cycles that the oldest load of the load buffer is stalled at retirement due to a second level TLB miss requiring a page walk.", - "MetricExpr": "LD_HEAD.PGWALK_AT_RET / tma_info_clks", + "MetricExpr": "LD_HEAD.PGWALK_AT_RET / tma_info_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_stlb_miss", "MetricThreshold": "tma_stlb_miss > 0.05", @@ -798,7 +791,7 @@ }, { "BriefDescription": "Counts the number of cycles the core is stalled due to store buffer full.", - "MetricExpr": "tma_mem_scheduler * (MEM_SCHEDULER_BLOCK.ST_BUF / MEM_SCHEDULER_BLOCK.ALL)", + "MetricExpr": "tma_mem_scheduler * (cpu_atom@MEM_SCHEDULER_BLOCK.ST_BUF@ / cpu_atom@MEM_SCHEDULER_BLOCK.ALL@)", "MetricGroup": "TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.1", @@ -807,7 +800,7 @@ }, { "BriefDescription": "Counts the number of cycles that the oldest load of the load buffer is stalled at retirement due to a store forward block.", - "MetricExpr": "LD_HEAD.ST_ADDR_AT_RET / tma_info_clks", + "MetricExpr": "LD_HEAD.ST_ADDR_AT_RET / tma_info_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_store_fwd_blk", "MetricThreshold": "tma_store_fwd_blk > 0.05", @@ -816,7 +809,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution ports for ALU operations.", - "MetricExpr": "(UOPS_DISPATCHED.PORT_0 + UOPS_DISPATCHED.PORT_1 + UOPS_DISPATCHED.PORT_5_11 + UOPS_DISPATCHED.PORT_6) / (5 * tma_info_core_clks)", + "MetricExpr": "(cpu_core@UOPS_DISPATCHED.PORT_0@ + cpu_core@UOPS_DISPATCHED.PORT_1@ + cpu_core@UOPS_DISPATCHED.PORT_5_11@ + cpu_core@UOPS_DISPATCHED.PORT_6@) / (5 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_alu_op_utilization", "MetricThreshold": "tma_alu_op_utilization > 0.6", @@ -825,7 +818,7 @@ }, { "BriefDescription": "This metric estimates fraction of slots the CPU retired uops delivered by the Microcode_Sequencer as a result of Assists", - "MetricExpr": "100 * cpu_core@ASSISTS.ANY\\,umask\\=0x1B@ / tma_info_slots", + "MetricExpr": "100 * cpu_core@ASSISTS.ANY\\,umask\\=0x1B@ / tma_info_thread_slots", "MetricGroup": "TopdownL4;tma_L4_group;tma_microcode_sequencer_group", "MetricName": "tma_assists", "MetricThreshold": "tma_assists > 0.1 & (tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1)", @@ -835,7 +828,7 @@ }, { "BriefDescription": "This metric estimates fraction of slots the CPU retired uops as a result of handing SSE to AVX* or AVX* to SSE transition Assists.", - "MetricExpr": "63 * ASSISTS.SSE_AVX_MIX / tma_info_slots", + "MetricExpr": "63 * cpu_core@ASSISTS.SSE_AVX_MIX@ / tma_info_thread_slots", "MetricGroup": "HPC;TopdownL5;tma_L5_group;tma_assists_group", "MetricName": "tma_avx_assists", "MetricThreshold": "tma_avx_assists > 0.1", @@ -844,37 +837,42 @@ }, { "BriefDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend", - "MetricExpr": "topdown\\-be\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_slots", - "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "cpu_core@topdown\\-be\\-bound@ / (cpu_core@topdown\\-fe\\-bound@ + cpu_core@topdown\\-bad\\-spec@ + cpu_core@topdown\\-retiring@ + cpu_core@topdown\\-be\\-bound@) + 0 * tma_info_thread_slots", + "MetricGroup": "Default;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend. Backend is the portion of the processor core where the out-of-order scheduler dispatches ready uops into their respective execution units; and once completed these uops get retired according to program order. For example; stalls due to data-cache misses or stalls due to the divider unit being overloaded are both categorized under Backend Bound. Backend Bound is further divided into two main categories: Memory Bound and Core Bound. Sample with: TOPDOWN.BACKEND_BOUND_SLOTS", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations", + "DefaultMetricgroupName": "TopdownL1", "MetricExpr": "max(1 - (tma_frontend_bound + tma_backend_bound + tma_retiring), 0)", - "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", + "MetricGroup": "Default;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots wasted due to incorrect speculations. This include slots used to issue uops that do not eventually get retired and slots for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work due to miss-predicted branches are categorized under Bad Speculation category. Incorrect data speculation followed by Memory Ordering Nukes is another example.", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction", - "MetricExpr": "topdown\\-br\\-mispredict / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_slots", + "MetricExpr": "cpu_core@topdown\\-br\\-mispredict@ / (cpu_core@topdown\\-fe\\-bound@ + cpu_core@topdown\\-bad\\-spec@ + cpu_core@topdown\\-retiring@ + cpu_core@topdown\\-be\\-bound@) + 0 * tma_info_thread_slots", "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15", - "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: TOPDOWN.BR_MISPREDICT_SLOTS. Related metrics: tma_info_branch_misprediction_cost, tma_info_mispredictions, tma_mispredicts_resteers", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: TOPDOWN.BR_MISPREDICT_SLOTS. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_info_bottleneck_mispredictions, tma_mispredicts_resteers", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers", - "MetricExpr": "INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks + tma_unknown_branches", + "MetricExpr": "INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks + tma_unknown_branches", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteers", "MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -894,7 +892,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Machine Clears", - "MetricExpr": "(1 - tma_branch_mispredicts / tma_bad_speculation) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks", + "MetricExpr": "(1 - tma_branch_mispredicts / tma_bad_speculation) * cpu_core@INT_MISC.CLEAR_RESTEER_CYCLES@ / tma_info_thread_clks", "MetricGroup": "BadSpec;MachineClears;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueMC", "MetricName": "tma_clears_resteers", "MetricThreshold": "tma_clears_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", @@ -905,7 +903,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to contested accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(25 * tma_info_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_FWD * (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM / (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM + OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) + 24 * tma_info_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "(25 * tma_info_system_average_frequency * (cpu_core@MEM_LOAD_L3_HIT_RETIRED.XSNP_FWD@ * (cpu_core@OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM@ / (cpu_core@OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM@ + cpu_core@OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD@))) + 24 * tma_info_system_average_frequency * cpu_core@MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS@) * (1 + cpu_core@MEM_LOAD_RETIRED.FB_HIT@ / cpu_core@MEM_LOAD_RETIRED.L1_MISS@ / 2) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_contested_accesses", "MetricThreshold": "tma_contested_accesses > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -919,6 +917,7 @@ "MetricGroup": "Backend;Compute;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck. Shortage in hardware compute resources; or dependencies in software's instructions are both categorized under Core Bound. Hence it may indicate the machine ran out of an out-of-order resource; certain execution units are overloaded or dependencies in program's data- or instruction-flow are limiting the performance (e.g. FP-chained long-latency arithmetic operations).", "ScaleUnit": "100%", "Unit": "cpu_core" @@ -926,7 +925,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to data-sharing accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "24 * tma_info_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_NO_FWD + MEM_LOAD_L3_HIT_RETIRED.XSNP_FWD * (1 - OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM / (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM + OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "24 * tma_info_system_average_frequency * (cpu_core@MEM_LOAD_L3_HIT_RETIRED.XSNP_NO_FWD@ + cpu_core@MEM_LOAD_L3_HIT_RETIRED.XSNP_FWD@ * (1 - cpu_core@OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM@ / (cpu_core@OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM@ + cpu_core@OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD@))) * (1 + cpu_core@MEM_LOAD_RETIRED.FB_HIT@ / cpu_core@MEM_LOAD_RETIRED.L1_MISS@ / 2) / tma_info_thread_clks", "MetricGroup": "Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_data_sharing", "MetricThreshold": "tma_data_sharing > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -936,17 +935,17 @@ }, { "BriefDescription": "This metric represents fraction of cycles where decoder-0 was the only active decoder", - "MetricExpr": "(cpu_core@INST_DECODED.DECODERS\\,cmask\\=1@ - cpu_core@INST_DECODED.DECODERS\\,cmask\\=2@) / tma_info_core_clks / 2", + "MetricExpr": "(cpu_core@INST_DECODED.DECODERS\\,cmask\\=1@ - cpu_core@INST_DECODED.DECODERS\\,cmask\\=2@) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL4;tma_L4_group;tma_issueD0;tma_mite_group", "MetricName": "tma_decoder0_alone", - "MetricThreshold": "tma_decoder0_alone > 0.1 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 6 > 0.35))", + "MetricThreshold": "tma_decoder0_alone > 0.1 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 6 > 0.35))", "PublicDescription": "This metric represents fraction of cycles where decoder-0 was the only active decoder. Related metrics: tma_few_uops_instructions", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric represents fraction of cycles where the Divider unit was active", - "MetricExpr": "ARITH.DIV_ACTIVE / tma_info_clks", + "MetricExpr": "ARITH.DIV_ACTIVE / tma_info_thread_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_divider", "MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -957,7 +956,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEMORY_ACTIVITY.STALLS_L3_MISS / tma_info_clks", + "MetricExpr": "cpu_core@MEMORY_ACTIVITY.STALLS_L3_MISS@ / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -967,47 +966,47 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline", - "MetricExpr": "(IDQ.DSB_CYCLES_ANY - IDQ.DSB_CYCLES_OK) / tma_info_core_clks / 2", + "MetricExpr": "(cpu_core@IDQ.DSB_CYCLES_ANY@ - cpu_core@IDQ.DSB_CYCLES_OK@) / tma_info_core_core_clks / 2", "MetricGroup": "DSB;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_dsb", - "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 6 > 0.35)", + "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 6 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline. For example; inefficient utilization of the DSB cache structure or bank conflict when reading from it; are categorized here.", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_dsb_switches", "MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb, tma_lcp", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses", - "MetricExpr": "min(7 * cpu_core@DTLB_LOAD_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_LOAD_MISSES.WALK_ACTIVE, max(CYCLE_ACTIVITY.CYCLES_MEM_ANY - MEMORY_ACTIVITY.CYCLES_L1D_MISS, 0)) / tma_info_clks", + "MetricExpr": "min(7 * cpu_core@DTLB_LOAD_MISSES.STLB_HIT\\,cmask\\=1@ + cpu_core@DTLB_LOAD_MISSES.WALK_ACTIVE@, max(cpu_core@CYCLE_ACTIVITY.CYCLES_MEM_ANY@ - cpu_core@MEMORY_ACTIVITY.CYCLES_L1D_MISS@, 0)) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group", "MetricName": "tma_dtlb_load", "MetricThreshold": "tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses. TLBs (Translation Look-aside Buffers) are processor caches for recently used entries out of the Page Tables that are used to map virtual- to physical-addresses by the operating system. This metric approximates the potential delay of demand loads missing the first-level data TLB (assuming worst case scenario with back to back misses to different pages). This includes hitting in the second-level TLB (STLB) as well as performing a hardware page walk on an STLB miss. Sample with: MEM_INST_RETIRED.STLB_MISS_LOADS_PS. Related metrics: tma_dtlb_store, tma_info_memory_data_tlbs", + "PublicDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses. TLBs (Translation Look-aside Buffers) are processor caches for recently used entries out of the Page Tables that are used to map virtual- to physical-addresses by the operating system. This metric approximates the potential delay of demand loads missing the first-level data TLB (assuming worst case scenario with back to back misses to different pages). This includes hitting in the second-level TLB (STLB) as well as performing a hardware page walk on an STLB miss. Sample with: MEM_INST_RETIRED.STLB_MISS_LOADS_PS. Related metrics: tma_dtlb_store, tma_info_bottleneck_memory_data_tlbs", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses", - "MetricExpr": "(7 * cpu_core@DTLB_STORE_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_STORE_MISSES.WALK_ACTIVE) / tma_info_core_clks", + "MetricExpr": "(7 * cpu_core@DTLB_STORE_MISSES.STLB_HIT\\,cmask\\=1@ + cpu_core@DTLB_STORE_MISSES.WALK_ACTIVE@) / tma_info_core_core_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_store_bound_group", "MetricName": "tma_dtlb_store", "MetricThreshold": "tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses. As with ordinary data caching; focus on improving data locality and reducing working-set size to reduce DTLB overhead. Additionally; consider using profile-guided optimization (PGO) to collocate frequently-used data on the same page. Try using larger page sizes for large amounts of frequently-used data. Sample with: MEM_INST_RETIRED.STLB_MISS_STORES_PS. Related metrics: tma_dtlb_load, tma_info_memory_data_tlbs", + "PublicDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses. As with ordinary data caching; focus on improving data locality and reducing working-set size to reduce DTLB overhead. Additionally; consider using profile-guided optimization (PGO) to collocate frequently-used data on the same page. Try using larger page sizes for large amounts of frequently-used data. Sample with: MEM_INST_RETIRED.STLB_MISS_STORES_PS. Related metrics: tma_dtlb_load, tma_info_bottleneck_memory_data_tlbs", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric roughly estimates how often CPU was handling synchronizations due to False Sharing", - "MetricExpr": "28 * tma_info_average_frequency * OCR.DEMAND_RFO.L3_HIT.SNOOP_HITM / tma_info_clks", + "MetricExpr": "28 * tma_info_system_average_frequency * cpu_core@OCR.DEMAND_RFO.L3_HIT.SNOOP_HITM@ / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_store_bound_group", "MetricName": "tma_false_sharing", "MetricThreshold": "tma_false_sharing > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1017,11 +1016,11 @@ }, { "BriefDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed", - "MetricExpr": "L1D_PEND_MISS.FB_FULL / tma_info_clks", + "MetricExpr": "L1D_PEND_MISS.FB_FULL / tma_info_thread_clks", "MetricGroup": "MemoryBW;TopdownL4;tma_L4_group;tma_issueBW;tma_issueSL;tma_issueSmSt;tma_l1_bound_group", "MetricName": "tma_fb_full", "MetricThreshold": "tma_fb_full > 0.3", - "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", + "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", "ScaleUnit": "100%", "Unit": "cpu_core" }, @@ -1030,17 +1029,19 @@ "MetricExpr": "max(0, tma_frontend_bound - tma_fetch_latency)", "MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", "MetricName": "tma_fetch_bandwidth", - "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 6 > 0.35", - "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb, tma_lcp", + "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 6 > 0.35", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues", - "MetricExpr": "topdown\\-fetch\\-lat / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) - INT_MISC.UOP_DROPPING / tma_info_slots", + "MetricExpr": "cpu_core@topdown\\-fetch\\-lat@ / (cpu_core@topdown\\-fe\\-bound@ + cpu_core@topdown\\-bad\\-spec@ + cpu_core@topdown\\-retiring@ + cpu_core@topdown\\-be\\-bound@) - cpu_core@INT_MISC.UOP_DROPPING@ / tma_info_thread_slots", "MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: FRONTEND_RETIRED.LATENCY_GE_16_PS;FRONTEND_RETIRED.LATENCY_GE_8_PS", "ScaleUnit": "100%", "Unit": "cpu_core" @@ -1057,6 +1058,7 @@ }, { "BriefDescription": "This metric represents overall arithmetic floating-point (FP) operations fraction the CPU has executed (retired)", + "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "tma_x87_use + tma_fp_scalar + tma_fp_vector", "MetricGroup": "HPC;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_fp_arith", @@ -1067,7 +1069,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of slots the CPU retired uops as a result of handing Floating Point (FP) Assists", - "MetricExpr": "30 * ASSISTS.FP / tma_info_slots", + "MetricExpr": "30 * cpu_core@ASSISTS.FP@ / tma_info_thread_slots", "MetricGroup": "HPC;TopdownL5;tma_L5_group;tma_assists_group", "MetricName": "tma_fp_assists", "MetricThreshold": "tma_fp_assists > 0.1", @@ -1077,7 +1079,7 @@ }, { "BriefDescription": "This metric approximates arithmetic floating-point (FP) scalar uops fraction the CPU has retired", - "MetricExpr": "cpu_core@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ / (tma_retiring * tma_info_slots)", + "MetricExpr": "cpu_core@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL4;tma_L4_group;tma_fp_arith_group;tma_issue2P", "MetricName": "tma_fp_scalar", "MetricThreshold": "tma_fp_scalar > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6)", @@ -1087,7 +1089,7 @@ }, { "BriefDescription": "This metric approximates arithmetic floating-point (FP) vector uops fraction the CPU has retired aggregated across all vector widths", - "MetricExpr": "cpu_core@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@ / (tma_retiring * tma_info_slots)", + "MetricExpr": "cpu_core@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@ / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL4;tma_L4_group;tma_fp_arith_group;tma_issue2P", "MetricName": "tma_fp_vector", "MetricThreshold": "tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6)", @@ -1097,7 +1099,7 @@ }, { "BriefDescription": "This metric approximates arithmetic FP vector uops fraction the CPU has retired for 128-bit wide vectors", - "MetricExpr": "(FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(cpu_core@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE@ + cpu_core@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE@) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL5;tma_L5_group;tma_fp_vector_group;tma_issue2P", "MetricName": "tma_fp_vector_128b", "MetricThreshold": "tma_fp_vector_128b > 0.1 & (tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6))", @@ -1107,7 +1109,7 @@ }, { "BriefDescription": "This metric approximates arithmetic FP vector uops fraction the CPU has retired for 256-bit wide vectors", - "MetricExpr": "(FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(cpu_core@FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE@ + cpu_core@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE@) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL5;tma_L5_group;tma_fp_vector_group;tma_issue2P", "MetricName": "tma_fp_vector_256b", "MetricThreshold": "tma_fp_vector_256b > 0.1 & (tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6))", @@ -1117,17 +1119,19 @@ }, { "BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend", - "MetricExpr": "topdown\\-fe\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) - INT_MISC.UOP_DROPPING / tma_info_slots", - "MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "cpu_core@topdown\\-fe\\-bound@ / (cpu_core@topdown\\-fe\\-bound@ + cpu_core@topdown\\-bad\\-spec@ + cpu_core@topdown\\-retiring@ + cpu_core@topdown\\-be\\-bound@) - cpu_core@INT_MISC.UOP_DROPPING@ / tma_info_thread_slots", + "MetricGroup": "Default;PGO;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound. Sample with: FRONTEND_RETIRED.LATENCY_GE_4_PS", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring fused instructions -- where one uop can represent multiple contiguous instructions", - "MetricExpr": "tma_light_operations * INST_RETIRED.MACRO_FUSED / (tma_retiring * tma_info_slots)", + "MetricExpr": "tma_light_operations * cpu_core@INST_RETIRED.MACRO_FUSED@ / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_fused_instructions", "MetricThreshold": "tma_fused_instructions > 0.1 & tma_light_operations > 0.6", @@ -1137,17 +1141,18 @@ }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences", - "MetricExpr": "topdown\\-heavy\\-ops / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_slots", + "MetricExpr": "cpu_core@topdown\\-heavy\\-ops@ / (cpu_core@topdown\\-fe\\-bound@ + cpu_core@topdown\\-bad\\-spec@ + cpu_core@topdown\\-retiring@ + cpu_core@topdown\\-be\\-bound@) + 0 * tma_info_thread_slots", "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_heavy_operations", "MetricThreshold": "tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences. Sample with: UOPS_RETIRED.HEAVY", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to instruction cache misses", - "MetricExpr": "ICACHE_DATA.STALLS / tma_info_clks", + "MetricExpr": "ICACHE_DATA.STALLS / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;IcMiss;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_icache_misses", "MetricThreshold": "tma_icache_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -1156,247 +1161,300 @@ "Unit": "cpu_core" }, { - "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", - "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time", - "MetricGroup": "Power;Summary", - "MetricName": "tma_info_average_frequency", - "Unit": "cpu_core" - }, - { - "BriefDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses)", + "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * tma_fetch_latency * (tma_itlb_misses + tma_icache_misses + tma_unknown_branches) / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)", - "MetricGroup": "BigFoot;Fed;Frontend;IcMiss;MemoryTLB;tma_issueBC", - "MetricName": "tma_info_big_code", - "MetricThreshold": "tma_info_big_code > 20", - "PublicDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses). Related metrics: tma_info_branching_overhead", + "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_thread_slots / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BrMispredicts;tma_issueBM", + "MetricName": "tma_info_bad_spec_branch_misprediction_cost", + "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_info_bottleneck_mispredictions, tma_mispredicts_resteers", "Unit": "cpu_core" }, { - "BriefDescription": "Branch instructions per taken branch.", - "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", - "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_bptkbranch", + "BriefDescription": "Instructions per retired mispredicts for conditional non-taken branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_NTAKEN", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_cond_ntaken", + "MetricThreshold": "tma_info_bad_spec_ipmisp_cond_ntaken < 200", "Unit": "cpu_core" }, { - "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", - "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_slots / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_branch_misprediction_cost", - "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_info_mispredictions, tma_mispredicts_resteers", + "BriefDescription": "Instructions per retired mispredicts for conditional taken branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_TAKEN", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_cond_taken", + "MetricThreshold": "tma_info_bad_spec_ipmisp_cond_taken < 200", "Unit": "cpu_core" }, { - "BriefDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls)", - "MetricExpr": "100 * ((BR_INST_RETIRED.COND + 3 * BR_INST_RETIRED.NEAR_CALL + (BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL)) / tma_info_slots)", - "MetricGroup": "Ret;tma_issueBC", - "MetricName": "tma_info_branching_overhead", - "MetricThreshold": "tma_info_branching_overhead > 10", - "PublicDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls). Related metrics: tma_info_big_code", + "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", + "MetricExpr": "cpu_core@BR_MISP_RETIRED.INDIRECT_CALL\\,umask\\=0x80@ / BR_MISP_RETIRED.INDIRECT", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_indirect", + "MetricThreshold": "tma_info_bad_spec_ipmisp_indirect < 1e3", "Unit": "cpu_core" }, { - "BriefDescription": "Fraction of branches that are CALL or RET", - "MetricExpr": "(BR_INST_RETIRED.NEAR_CALL + BR_INST_RETIRED.NEAR_RETURN) / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_callret", + "BriefDescription": "Instructions per retired mispredicts for return branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.RET", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_ret", + "MetricThreshold": "tma_info_bad_spec_ipmisp_ret < 500", "Unit": "cpu_core" }, { - "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "Pipeline", - "MetricName": "tma_info_clks", + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BadSpec;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmispredict", + "MetricThreshold": "tma_info_bad_spec_ipmispredict < 200", "Unit": "cpu_core" }, { - "BriefDescription": "STLB (2nd level TLB) code speculative misses per kilo instruction (misses of any page-size that complete the page walk)", - "MetricExpr": "1e3 * ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", - "MetricGroup": "Fed;MemoryTLB", - "MetricName": "tma_info_code_stlb_mpki", + "BriefDescription": "Probability of Core Bound bottleneck hidden by SMT-profiling artifacts", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(100 * (1 - tma_core_bound / tma_ports_utilization if tma_core_bound < tma_ports_utilization else 1) if tma_info_system_smt_2t_utilization > 0.5 else 0)", + "MetricGroup": "Cor;SMT", + "MetricName": "tma_info_botlnk_l0_core_bound_likely", + "MetricThreshold": "tma_info_botlnk_l0_core_bound_likely > 0.5", "Unit": "cpu_core" }, { - "BriefDescription": "Fraction of branches that are non-taken conditionals", - "MetricExpr": "BR_INST_RETIRED.COND_NTAKEN / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches;CodeGen;PGO", - "MetricName": "tma_info_cond_nt", + "BriefDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_fetch_latency * tma_dsb_switches / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches) + tma_fetch_bandwidth * tma_mite / (tma_dsb + tma_lsd + tma_mite))", + "MetricGroup": "DSBmiss;Fed;tma_issueFB", + "MetricName": "tma_info_botlnk_l2_dsb_misses", + "MetricThreshold": "tma_info_botlnk_l2_dsb_misses > 10", + "PublicDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "Unit": "cpu_core" }, { - "BriefDescription": "Fraction of branches that are taken conditionals", - "MetricExpr": "BR_INST_RETIRED.COND_TAKEN / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches;CodeGen;PGO", - "MetricName": "tma_info_cond_tk", + "BriefDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck", + "MetricExpr": "100 * (tma_fetch_latency * tma_icache_misses / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", + "MetricGroup": "Fed;FetchLat;IcMiss;tma_issueFL", + "MetricName": "tma_info_botlnk_l2_ic_misses", + "MetricThreshold": "tma_info_botlnk_l2_ic_misses > 5", + "PublicDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck. Related metrics: ", "Unit": "cpu_core" }, { - "BriefDescription": "Probability of Core Bound bottleneck hidden by SMT-profiling artifacts", - "MetricExpr": "(100 * (1 - tma_core_bound / tma_ports_utilization if tma_core_bound < tma_ports_utilization else 1) if tma_info_smt_2t_utilization > 0.5 else 0)", - "MetricGroup": "Cor;SMT", - "MetricName": "tma_info_core_bound_likely", - "MetricThreshold": "tma_info_core_bound_likely > 0.5", + "BriefDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_fetch_latency * (tma_itlb_misses + tma_icache_misses + tma_unknown_branches) / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)", + "MetricGroup": "BigFoot;Fed;Frontend;IcMiss;MemoryTLB;tma_issueBC", + "MetricName": "tma_info_bottleneck_big_code", + "MetricThreshold": "tma_info_bottleneck_big_code > 20", + "PublicDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses). Related metrics: tma_info_bottleneck_branching_overhead", "Unit": "cpu_core" }, { - "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", - "MetricExpr": "CPU_CLK_UNHALTED.DISTRIBUTED", - "MetricGroup": "SMT", - "MetricName": "tma_info_core_clks", + "BriefDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls)", + "MetricExpr": "100 * ((cpu_core@BR_INST_RETIRED.COND@ + 3 * cpu_core@BR_INST_RETIRED.NEAR_CALL@ + (cpu_core@BR_INST_RETIRED.NEAR_TAKEN@ - cpu_core@BR_INST_RETIRED.COND_TAKEN@ - 2 * cpu_core@BR_INST_RETIRED.NEAR_CALL@)) / tma_info_thread_slots)", + "MetricGroup": "Ret;tma_issueBC", + "MetricName": "tma_info_bottleneck_branching_overhead", + "MetricThreshold": "tma_info_bottleneck_branching_overhead > 10", + "PublicDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls). Related metrics: tma_info_bottleneck_big_code", "Unit": "cpu_core" }, { - "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", - "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_coreipc", + "BriefDescription": "Total pipeline cost of instruction fetch bandwidth related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_frontend_bound - tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) - tma_info_bottleneck_big_code", + "MetricGroup": "Fed;FetchBW;Frontend", + "MetricName": "tma_info_bottleneck_instruction_fetch_bw", + "MetricThreshold": "tma_info_bottleneck_instruction_fetch_bw > 20", "Unit": "cpu_core" }, { - "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricExpr": "1 / tma_info_ipc", - "MetricGroup": "Mem;Pipeline", - "MetricName": "tma_info_cpi", + "BriefDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks", + "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_bandwidth / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_sq_full / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full))) + tma_l1_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_fb_full / (tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk))", + "MetricGroup": "Mem;MemoryBW;Offcore;tma_issueBW", + "MetricName": "tma_info_bottleneck_memory_bandwidth", + "MetricThreshold": "tma_info_bottleneck_memory_bandwidth > 20", + "PublicDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks. Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full", "Unit": "cpu_core" }, { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": "HPC;Summary", - "MetricName": "tma_info_cpu_utilization", + "BriefDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_l1_bound / max(tma_memory_bound, tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_load / max(tma_l1_bound, tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk)) + tma_store_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_store / (tma_dtlb_store + tma_false_sharing + tma_split_stores + tma_store_latency + tma_streaming_stores)))", + "MetricGroup": "Mem;MemoryTLB;Offcore;tma_issueTLB", + "MetricName": "tma_info_bottleneck_memory_data_tlbs", + "MetricThreshold": "tma_info_bottleneck_memory_data_tlbs > 20", + "PublicDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs). Related metrics: tma_dtlb_load, tma_dtlb_store", "Unit": "cpu_core" }, { - "BriefDescription": "Average Parallel L2 cache miss data reads", - "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", - "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_data_l2_mlp", + "BriefDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_latency / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_l3_hit_latency / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full)) + tma_l2_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound))", + "MetricGroup": "Mem;MemoryLat;Offcore;tma_issueLat", + "MetricName": "tma_info_bottleneck_memory_latency", + "MetricThreshold": "tma_info_bottleneck_memory_latency > 20", + "PublicDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches). Related metrics: tma_l3_hit_latency, tma_mem_latency", "Unit": "cpu_core" }, { - "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricExpr": "64 * (UNC_ARB_TRK_REQUESTS.ALL + UNC_ARB_COH_TRK_REQUESTS.ALL) / 1e6 / duration_time / 1e3", - "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", - "MetricName": "tma_info_dram_bw_use", - "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_info_memory_bandwidth, tma_mem_bandwidth, tma_sq_full", + "BriefDescription": "Total pipeline cost of Branch Misprediction related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", + "MetricGroup": "Bad;BadSpec;BrMispredicts;tma_issueBM", + "MetricName": "tma_info_bottleneck_mispredictions", + "MetricThreshold": "tma_info_bottleneck_mispredictions > 20", + "PublicDescription": "Total pipeline cost of Branch Misprediction related bottlenecks. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers", "Unit": "cpu_core" }, { - "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", - "MetricExpr": "IDQ.DSB_UOPS / UOPS_ISSUED.ANY", - "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", - "MetricName": "tma_info_dsb_coverage", - "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 6 > 0.35", - "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_misses, tma_info_iptb, tma_lcp", + "BriefDescription": "Fraction of branches that are CALL or RET", + "MetricExpr": "(cpu_core@BR_INST_RETIRED.NEAR_CALL@ + cpu_core@BR_INST_RETIRED.NEAR_RETURN@) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_callret", "Unit": "cpu_core" }, { - "BriefDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck", - "MetricExpr": "100 * (tma_fetch_latency * tma_dsb_switches / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches) + tma_fetch_bandwidth * tma_mite / (tma_dsb + tma_lsd + tma_mite))", - "MetricGroup": "DSBmiss;Fed;tma_issueFB", - "MetricName": "tma_info_dsb_misses", - "MetricThreshold": "tma_info_dsb_misses > 10", - "PublicDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb, tma_lcp", + "BriefDescription": "Fraction of branches that are non-taken conditionals", + "MetricExpr": "BR_INST_RETIRED.COND_NTAKEN / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches;CodeGen;PGO", + "MetricName": "tma_info_branches_cond_nt", "Unit": "cpu_core" }, { - "BriefDescription": "Average number of cycles of a switch from the DSB fetch-unit to MITE fetch unit - see DSB_Switches tree node for details.", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / cpu_core@DSB2MITE_SWITCHES.PENALTY_CYCLES\\,cmask\\=1\\,edge@", - "MetricGroup": "DSBmiss", - "MetricName": "tma_info_dsb_switch_cost", + "BriefDescription": "Fraction of branches that are taken conditionals", + "MetricExpr": "BR_INST_RETIRED.COND_TAKEN / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches;CodeGen;PGO", + "MetricName": "tma_info_branches_cond_tk", "Unit": "cpu_core" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", - "MetricExpr": "UOPS_EXECUTED.THREAD / cpu_core@UOPS_EXECUTED.THREAD\\,cmask\\=1@", - "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", - "MetricName": "tma_info_execute", + "BriefDescription": "Fraction of branches that are unconditional (direct or indirect) jumps", + "MetricExpr": "(cpu_core@BR_INST_RETIRED.NEAR_TAKEN@ - cpu_core@BR_INST_RETIRED.COND_TAKEN@ - 2 * cpu_core@BR_INST_RETIRED.NEAR_CALL@) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_jump", "Unit": "cpu_core" }, { - "BriefDescription": "The ratio of Executed- by Issued-Uops", - "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", - "MetricGroup": "Cor;Pipeline", - "MetricName": "tma_info_execute_per_issue", - "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage.", + "BriefDescription": "Fraction of branches of other types (not individually covered by other metrics in Info.Branches group)", + "MetricExpr": "1 - (tma_info_branches_cond_nt + tma_info_branches_cond_tk + tma_info_branches_callret + tma_info_branches_jump)", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_other_branches", "Unit": "cpu_core" }, { - "BriefDescription": "Fill Buffer (FB) hits per kilo instructions for retired demand loads (L1D misses that merge into ongoing miss-handling entries)", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.FB_HIT / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_fb_hpki", + "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", + "MetricExpr": "cpu_core@CPU_CLK_UNHALTED.DISTRIBUTED@", + "MetricGroup": "SMT", + "MetricName": "tma_info_core_core_clks", "Unit": "cpu_core" }, { - "BriefDescription": "Average number of Uops issued by front-end when it issued something", - "MetricExpr": "UOPS_ISSUED.ANY / cpu_core@UOPS_ISSUED.ANY\\,cmask\\=1@", - "MetricGroup": "Fed;FetchBW", - "MetricName": "tma_info_fetch_upc", + "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks", + "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", + "MetricName": "tma_info_core_coreipc", "Unit": "cpu_core" }, { "BriefDescription": "Floating Point Operations Per Cycle", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / tma_info_core_clks", + "MetricExpr": "(cpu_core@FP_ARITH_INST_RETIRED.SCALAR_SINGLE@ + cpu_core@FP_ARITH_INST_RETIRED.SCALAR_DOUBLE@ + 2 * cpu_core@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE@ + 4 * (cpu_core@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE@ + cpu_core@FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE@) + 8 * cpu_core@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE@) / tma_info_core_core_clks", "MetricGroup": "Flops;Ret", - "MetricName": "tma_info_flopc", + "MetricName": "tma_info_core_flopc", "Unit": "cpu_core" }, { "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(FP_ARITH_DISPATCHED.PORT_0 + FP_ARITH_DISPATCHED.PORT_1 + FP_ARITH_DISPATCHED.PORT_5) / (2 * tma_info_core_clks)", + "MetricExpr": "(cpu_core@FP_ARITH_DISPATCHED.PORT_0@ + cpu_core@FP_ARITH_DISPATCHED.PORT_1@ + cpu_core@FP_ARITH_DISPATCHED.PORT_5@) / (2 * tma_info_core_core_clks)", "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_fp_arith_utilization", + "MetricName": "tma_info_core_fp_arith_utilization", "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common).", "Unit": "cpu_core" }, { - "BriefDescription": "Giga Floating Point Operations Per Second", - "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / 1e9 / duration_time", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_gflops", - "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine.", + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", + "MetricExpr": "UOPS_EXECUTED.THREAD / (cpu_core@UOPS_EXECUTED.CORE_CYCLES_GE_1@ / 2 if #SMT_on else cpu_core@UOPS_EXECUTED.CORE_CYCLES_GE_1@)", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", + "MetricName": "tma_info_core_ilp", "Unit": "cpu_core" }, { - "BriefDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck", - "MetricExpr": "100 * (tma_fetch_latency * tma_icache_misses / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", - "MetricGroup": "Fed;FetchLat;IcMiss;tma_issueFL", - "MetricName": "tma_info_ic_misses", - "MetricThreshold": "tma_info_ic_misses > 5", - "PublicDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck. Related metrics: ", + "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", + "MetricExpr": "IDQ.DSB_UOPS / cpu_core@UOPS_ISSUED.ANY@", + "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", + "MetricName": "tma_info_frontend_dsb_coverage", + "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 6 > 0.35", + "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_inst_mix_iptb, tma_lcp", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Average number of cycles of a switch from the DSB fetch-unit to MITE fetch unit - see DSB_Switches tree node for details.", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / cpu_core@DSB2MITE_SWITCHES.PENALTY_CYCLES\\,cmask\\=1\\,edge@", + "MetricGroup": "DSBmiss", + "MetricName": "tma_info_frontend_dsb_switch_cost", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Average number of Uops issued by front-end when it issued something", + "MetricExpr": "UOPS_ISSUED.ANY / cpu_core@UOPS_ISSUED.ANY\\,cmask\\=1@", + "MetricGroup": "Fed;FetchBW", + "MetricName": "tma_info_frontend_fetch_upc", "Unit": "cpu_core" }, { "BriefDescription": "Average Latency for L1 instruction cache misses", "MetricExpr": "ICACHE_DATA.STALLS / cpu_core@ICACHE_DATA.STALLS\\,cmask\\=1\\,edge@", "MetricGroup": "Fed;FetchLat;IcMiss", - "MetricName": "tma_info_icache_miss_latency", + "MetricName": "tma_info_frontend_icache_miss_latency", "Unit": "cpu_core" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", - "MetricExpr": "UOPS_EXECUTED.THREAD / (UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1)", - "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", - "MetricName": "tma_info_ilp", + "BriefDescription": "Instructions per non-speculative DSB miss (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / FRONTEND_RETIRED.ANY_DSB_MISS", + "MetricGroup": "DSBmiss;Fed", + "MetricName": "tma_info_frontend_ipdsb_miss_ret", + "MetricThreshold": "tma_info_frontend_ipdsb_miss_ret < 50", "Unit": "cpu_core" }, { - "BriefDescription": "Total pipeline cost of instruction fetch bandwidth related bottlenecks", - "MetricExpr": "100 * (tma_frontend_bound - tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) - tma_info_big_code", - "MetricGroup": "Fed;FetchBW;Frontend", - "MetricName": "tma_info_instruction_fetch_bw", - "MetricThreshold": "tma_info_instruction_fetch_bw > 20", + "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_inst_mix_instructions / BACLEARS.ANY", + "MetricGroup": "Fed", + "MetricName": "tma_info_frontend_ipunknown_branch", + "Unit": "cpu_core" + }, + { + "BriefDescription": "L2 cache true code cacheline misses per kilo instruction", + "MetricExpr": "1e3 * cpu_core@FRONTEND_RETIRED.L2_MISS@ / INST_RETIRED.ANY", + "MetricGroup": "IcMiss", + "MetricName": "tma_info_frontend_l2mpki_code", + "Unit": "cpu_core" + }, + { + "BriefDescription": "L2 cache speculative code cacheline misses per kilo instruction", + "MetricExpr": "1e3 * cpu_core@L2_RQSTS.CODE_RD_MISS@ / INST_RETIRED.ANY", + "MetricGroup": "IcMiss", + "MetricName": "tma_info_frontend_l2mpki_code_all", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Fraction of Uops delivered by the LSD (Loop Stream Detector; aka Loop Cache)", + "MetricExpr": "LSD.UOPS / cpu_core@UOPS_ISSUED.ANY@", + "MetricGroup": "Fed;LSD", + "MetricName": "tma_info_frontend_lsd_coverage", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Branch instructions per taken branch.", + "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;PGO", + "MetricName": "tma_info_inst_mix_bptkbranch", "Unit": "cpu_core" }, { "BriefDescription": "Total number of retired Instructions", - "MetricExpr": "INST_RETIRED.ANY", + "MetricExpr": "cpu_core@INST_RETIRED.ANY@", "MetricGroup": "Summary;TmaL1;tma_L1_group", - "MetricName": "tma_info_instructions", + "MetricName": "tma_info_inst_mix_instructions", "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST", "Unit": "cpu_core" }, @@ -1404,26 +1462,26 @@ "BriefDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (cpu_core@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu_core@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_iparith", - "MetricThreshold": "tma_info_iparith < 10", + "MetricName": "tma_info_inst_mix_iparith", + "MetricThreshold": "tma_info_inst_mix_iparith < 10", "PublicDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate). May undercount due to FMA double counting. Approximated prior to BDW.", "Unit": "cpu_core" }, { "BriefDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE)", + "MetricExpr": "INST_RETIRED.ANY / (cpu_core@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE@ + cpu_core@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE@)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx128", - "MetricThreshold": "tma_info_iparith_avx128 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx128", + "MetricThreshold": "tma_info_inst_mix_iparith_avx128 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting.", "Unit": "cpu_core" }, { "BriefDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE)", + "MetricExpr": "INST_RETIRED.ANY / (cpu_core@FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE@ + cpu_core@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE@)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx256", - "MetricThreshold": "tma_info_iparith_avx256 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx256", + "MetricThreshold": "tma_info_inst_mix_iparith_avx256 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting.", "Unit": "cpu_core" }, @@ -1431,8 +1489,8 @@ "BriefDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_dp", - "MetricThreshold": "tma_info_iparith_scalar_dp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_dp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_dp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting.", "Unit": "cpu_core" }, @@ -1440,490 +1498,445 @@ "BriefDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_SINGLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_sp", - "MetricThreshold": "tma_info_iparith_scalar_sp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_sp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_sp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting.", "Unit": "cpu_core" }, { - "BriefDescription": "Instructions per a microcode Assist invocation", - "MetricExpr": "INST_RETIRED.ANY / cpu_core@ASSISTS.ANY\\,umask\\=0x1B@", - "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_ipassist", - "MetricThreshold": "tma_info_ipassist < 100e3", - "PublicDescription": "Instructions per a microcode Assist invocation. See Assists tree node for details (lower number means higher occurrence rate)", - "Unit": "cpu_core" - }, - { "BriefDescription": "Instructions per Branch (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Branches;Fed;InsType", - "MetricName": "tma_info_ipbranch", - "MetricThreshold": "tma_info_ipbranch < 8", - "Unit": "cpu_core" - }, - { - "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": "Ret;Summary", - "MetricName": "tma_info_ipc", + "MetricName": "tma_info_inst_mix_ipbranch", + "MetricThreshold": "tma_info_inst_mix_ipbranch < 8", "Unit": "cpu_core" }, { "BriefDescription": "Instructions per (near) call (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_CALL", "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_ipcall", - "MetricThreshold": "tma_info_ipcall < 200", - "Unit": "cpu_core" - }, - { - "BriefDescription": "Instructions per non-speculative DSB miss (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / FRONTEND_RETIRED.ANY_DSB_MISS", - "MetricGroup": "DSBmiss;Fed", - "MetricName": "tma_info_ipdsb_miss_ret", - "MetricThreshold": "tma_info_ipdsb_miss_ret < 50", - "Unit": "cpu_core" - }, - { - "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", - "MetricGroup": "Branches;OS", - "MetricName": "tma_info_ipfarbranch", - "MetricThreshold": "tma_info_ipfarbranch < 1e6", + "MetricName": "tma_info_inst_mix_ipcall", + "MetricThreshold": "tma_info_inst_mix_ipcall < 200", "Unit": "cpu_core" }, { "BriefDescription": "Instructions per Floating Point (FP) Operation (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE)", + "MetricExpr": "INST_RETIRED.ANY / (cpu_core@FP_ARITH_INST_RETIRED.SCALAR_SINGLE@ + cpu_core@FP_ARITH_INST_RETIRED.SCALAR_DOUBLE@ + 2 * cpu_core@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE@ + 4 * (cpu_core@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE@ + cpu_core@FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE@) + 8 * cpu_core@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE@)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_ipflop", - "MetricThreshold": "tma_info_ipflop < 10", + "MetricName": "tma_info_inst_mix_ipflop", + "MetricThreshold": "tma_info_inst_mix_ipflop < 10", "Unit": "cpu_core" }, { "BriefDescription": "Instructions per Load (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_LOADS", "MetricGroup": "InsType", - "MetricName": "tma_info_ipload", - "MetricThreshold": "tma_info_ipload < 3", - "Unit": "cpu_core" - }, - { - "BriefDescription": "Instructions per retired mispredicts for conditional non-taken branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_NTAKEN", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_cond_ntaken", - "MetricThreshold": "tma_info_ipmisp_cond_ntaken < 200", - "Unit": "cpu_core" - }, - { - "BriefDescription": "Instructions per retired mispredicts for conditional taken branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_TAKEN", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_cond_taken", - "MetricThreshold": "tma_info_ipmisp_cond_taken < 200", - "Unit": "cpu_core" - }, - { - "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", - "MetricExpr": "cpu_core@BR_MISP_RETIRED.INDIRECT_CALL\\,umask\\=0x80@ / BR_MISP_RETIRED.INDIRECT", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_indirect", - "MetricThreshold": "tma_info_ipmisp_indirect < 1e3", - "Unit": "cpu_core" - }, - { - "BriefDescription": "Instructions per retired mispredicts for return branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.RET", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_ret", - "MetricThreshold": "tma_info_ipmisp_ret < 500", - "Unit": "cpu_core" - }, - { - "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BadSpec;BrMispredicts", - "MetricName": "tma_info_ipmispredict", - "MetricThreshold": "tma_info_ipmispredict < 200", + "MetricName": "tma_info_inst_mix_ipload", + "MetricThreshold": "tma_info_inst_mix_ipload < 3", "Unit": "cpu_core" }, { "BriefDescription": "Instructions per Store (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_STORES", "MetricGroup": "InsType", - "MetricName": "tma_info_ipstore", - "MetricThreshold": "tma_info_ipstore < 8", + "MetricName": "tma_info_inst_mix_ipstore", + "MetricThreshold": "tma_info_inst_mix_ipstore < 8", "Unit": "cpu_core" }, { "BriefDescription": "Instructions per Software prefetch instruction (of any type: NTA/T0/T1/T2/Prefetch) (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / cpu_core@SW_PREFETCH_ACCESS.T0\\,umask\\=0xF@", "MetricGroup": "Prefetches", - "MetricName": "tma_info_ipswpf", - "MetricThreshold": "tma_info_ipswpf < 100", + "MetricName": "tma_info_inst_mix_ipswpf", + "MetricThreshold": "tma_info_inst_mix_ipswpf < 100", "Unit": "cpu_core" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW;Frontend;PGO;tma_issueFB", - "MetricName": "tma_info_iptb", - "MetricThreshold": "tma_info_iptb < 13", - "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_lcp", - "Unit": "cpu_core" - }, - { - "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", - "MetricExpr": "tma_info_instructions / BACLEARS.ANY", - "MetricGroup": "Fed", - "MetricName": "tma_info_ipunknown_branch", + "MetricName": "tma_info_inst_mix_iptb", + "MetricThreshold": "tma_info_inst_mix_iptb < 13", + "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_lcp", "Unit": "cpu_core" }, { - "BriefDescription": "Fraction of branches that are unconditional (direct or indirect) jumps", - "MetricExpr": "(BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL) / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_jump", + "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "64 * cpu_core@L1D.REPLACEMENT@ / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_core_l1d_cache_fill_bw", "Unit": "cpu_core" }, { - "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_cpi", + "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "64 * cpu_core@L2_LINES_IN.ALL@ / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_core_l2_cache_fill_bw", "Unit": "cpu_core" }, { - "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_utilization", - "MetricThreshold": "tma_info_kernel_utilization > 0.05", + "BriefDescription": "Average per-core data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * cpu_core@OFFCORE_REQUESTS.ALL_REQUESTS@ / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_core_l3_cache_access_bw", "Unit": "cpu_core" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", + "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * cpu_core@LONGEST_LAT_CACHE.MISS@ / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw", + "MetricName": "tma_info_memory_core_l3_cache_fill_bw", "Unit": "cpu_core" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "tma_info_l1d_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw_1t", + "BriefDescription": "Fill Buffer (FB) hits per kilo instructions for retired demand loads (L1D misses that merge into ongoing miss-handling entries)", + "MetricExpr": "1e3 * cpu_core@MEM_LOAD_RETIRED.FB_HIT@ / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_fb_hpki", "Unit": "cpu_core" }, { "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L1_MISS / INST_RETIRED.ANY", + "MetricExpr": "1e3 * cpu_core@MEM_LOAD_RETIRED.L1_MISS@ / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki", + "MetricName": "tma_info_memory_l1mpki", "Unit": "cpu_core" }, { "BriefDescription": "L1 cache true misses per kilo instruction for all demand loads (including speculative)", - "MetricExpr": "1e3 * L2_RQSTS.ALL_DEMAND_DATA_RD / INST_RETIRED.ANY", + "MetricExpr": "1e3 * cpu_core@L2_RQSTS.ALL_DEMAND_DATA_RD@ / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki_load", - "Unit": "cpu_core" - }, - { - "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw", - "Unit": "cpu_core" - }, - { - "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "tma_info_l2_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw_1t", + "MetricName": "tma_info_memory_l1mpki_load", "Unit": "cpu_core" }, { "BriefDescription": "L2 cache hits per kilo instruction for all request types (including speculative)", - "MetricExpr": "1e3 * (L2_RQSTS.REFERENCES - L2_RQSTS.MISS) / INST_RETIRED.ANY", + "MetricExpr": "1e3 * (cpu_core@L2_RQSTS.REFERENCES@ - cpu_core@L2_RQSTS.MISS@) / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_all", + "MetricName": "tma_info_memory_l2hpki_all", "Unit": "cpu_core" }, { "BriefDescription": "L2 cache hits per kilo instruction for all demand loads (including speculative)", - "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_HIT / INST_RETIRED.ANY", + "MetricExpr": "1e3 * cpu_core@L2_RQSTS.DEMAND_DATA_RD_HIT@ / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_load", + "MetricName": "tma_info_memory_l2hpki_load", "Unit": "cpu_core" }, { "BriefDescription": "L2 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L2_MISS / INST_RETIRED.ANY", + "MetricExpr": "1e3 * cpu_core@MEM_LOAD_RETIRED.L2_MISS@ / INST_RETIRED.ANY", "MetricGroup": "Backend;CacheMisses;Mem", - "MetricName": "tma_info_l2mpki", + "MetricName": "tma_info_memory_l2mpki", "Unit": "cpu_core" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all request types (including speculative)", - "MetricExpr": "1e3 * L2_RQSTS.MISS / INST_RETIRED.ANY", + "MetricExpr": "1e3 * cpu_core@L2_RQSTS.MISS@ / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem;Offcore", - "MetricName": "tma_info_l2mpki_all", - "Unit": "cpu_core" - }, - { - "BriefDescription": "L2 cache true code cacheline misses per kilo instruction", - "MetricExpr": "1e3 * FRONTEND_RETIRED.L2_MISS / INST_RETIRED.ANY", - "MetricGroup": "IcMiss", - "MetricName": "tma_info_l2mpki_code", - "Unit": "cpu_core" - }, - { - "BriefDescription": "L2 cache speculative code cacheline misses per kilo instruction", - "MetricExpr": "1e3 * L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", - "MetricGroup": "IcMiss", - "MetricName": "tma_info_l2mpki_code_all", + "MetricName": "tma_info_memory_l2mpki_all", "Unit": "cpu_core" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all demand loads (including speculative)", - "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_MISS / INST_RETIRED.ANY", + "MetricExpr": "1e3 * cpu_core@L2_RQSTS.DEMAND_DATA_RD_MISS@ / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2mpki_load", + "MetricName": "tma_info_memory_l2mpki_load", "Unit": "cpu_core" }, { - "BriefDescription": "Average per-core data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * OFFCORE_REQUESTS.ALL_REQUESTS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw", - "Unit": "cpu_core" - }, - { - "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_access_bw", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw_1t", + "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * cpu_core@MEM_LOAD_RETIRED.L3_MISS@ / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l3mpki", "Unit": "cpu_core" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw", + "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", + "MetricExpr": "L1D_PEND_MISS.PENDING / MEM_LOAD_COMPLETED.L1_MISS_ANY", + "MetricGroup": "Mem;MemoryBound;MemoryLat", + "MetricName": "tma_info_memory_load_miss_real_latency", "Unit": "cpu_core" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw_1t", + "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", + "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", + "MetricGroup": "Mem;MemoryBW;MemoryBound", + "MetricName": "tma_info_memory_mlp", + "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)", "Unit": "cpu_core" }, { - "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L3_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l3mpki", + "BriefDescription": "Average Parallel L2 cache miss data reads", + "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", + "MetricGroup": "Memory_BW;Offcore", + "MetricName": "tma_info_memory_oro_data_l2_mlp", "Unit": "cpu_core" }, { "BriefDescription": "Average Latency for L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS.DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l2_miss_latency", + "MetricName": "tma_info_memory_oro_load_l2_miss_latency", "Unit": "cpu_core" }, { "BriefDescription": "Average Parallel L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / cpu_core@OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD\\,cmask\\=1@", "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_load_l2_mlp", + "MetricName": "tma_info_memory_oro_load_l2_mlp", "Unit": "cpu_core" }, { "BriefDescription": "Average Latency for L3 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.L3_MISS_DEMAND_DATA_RD / OFFCORE_REQUESTS.L3_MISS_DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l3_miss_latency", + "MetricName": "tma_info_memory_oro_load_l3_miss_latency", "Unit": "cpu_core" }, { - "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", - "MetricExpr": "L1D_PEND_MISS.PENDING / MEM_LOAD_COMPLETED.L1_MISS_ANY", - "MetricGroup": "Mem;MemoryBound;MemoryLat", - "MetricName": "tma_info_load_miss_real_latency", + "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l1d_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l1d_cache_fill_bw_1t", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l2_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l2_cache_fill_bw_1t", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_access_bw", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_thread_l3_cache_access_bw_1t", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l3_cache_fill_bw_1t", + "Unit": "cpu_core" + }, + { + "BriefDescription": "STLB (2nd level TLB) code speculative misses per kilo instruction (misses of any page-size that complete the page walk)", + "MetricExpr": "1e3 * cpu_core@ITLB_MISSES.WALK_COMPLETED@ / INST_RETIRED.ANY", + "MetricGroup": "Fed;MemoryTLB", + "MetricName": "tma_info_memory_tlb_code_stlb_mpki", "Unit": "cpu_core" }, { "BriefDescription": "STLB (2nd level TLB) data load speculative misses per kilo instruction (misses of any page-size that complete the page walk)", - "MetricExpr": "1e3 * DTLB_LOAD_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricExpr": "1e3 * cpu_core@DTLB_LOAD_MISSES.WALK_COMPLETED@ / INST_RETIRED.ANY", "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_load_stlb_mpki", + "MetricName": "tma_info_memory_tlb_load_stlb_mpki", "Unit": "cpu_core" }, { - "BriefDescription": "Fraction of Uops delivered by the LSD (Loop Stream Detector; aka Loop Cache)", - "MetricExpr": "LSD.UOPS / UOPS_ISSUED.ANY", - "MetricGroup": "Fed;LSD", - "MetricName": "tma_info_lsd_coverage", + "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", + "MetricExpr": "(cpu_core@ITLB_MISSES.WALK_PENDING@ + cpu_core@DTLB_LOAD_MISSES.WALK_PENDING@ + cpu_core@DTLB_STORE_MISSES.WALK_PENDING@) / (4 * tma_info_core_core_clks)", + "MetricGroup": "Mem;MemoryTLB", + "MetricName": "tma_info_memory_tlb_page_walks_utilization", + "MetricThreshold": "tma_info_memory_tlb_page_walks_utilization > 0.5", "Unit": "cpu_core" }, { - "BriefDescription": "Average number of parallel data read requests to external memory", - "MetricExpr": "UNC_ARB_DAT_OCCUPANCY.RD / cpu_core@UNC_ARB_DAT_OCCUPANCY.RD\\,cmask\\=1@", - "MetricGroup": "Mem;MemoryBW;SoC", - "MetricName": "tma_info_mem_parallel_reads", - "PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches", + "BriefDescription": "STLB (2nd level TLB) data store speculative misses per kilo instruction (misses of any page-size that complete the page walk)", + "MetricExpr": "1e3 * cpu_core@DTLB_STORE_MISSES.WALK_COMPLETED@ / INST_RETIRED.ANY", + "MetricGroup": "Mem;MemoryTLB", + "MetricName": "tma_info_memory_tlb_store_stlb_mpki", "Unit": "cpu_core" }, { - "BriefDescription": "Average latency of data read request to external memory (in nanoseconds)", - "MetricExpr": "(UNC_ARB_TRK_OCCUPANCY.RD + UNC_ARB_DAT_OCCUPANCY.RD) / UNC_ARB_TRK_REQUESTS.RD", - "MetricGroup": "Mem;MemoryLat;SoC", - "MetricName": "tma_info_mem_read_latency", - "PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)", + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", + "MetricExpr": "UOPS_EXECUTED.THREAD / cpu_core@UOPS_EXECUTED.THREAD\\,cmask\\=1@", + "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", + "MetricName": "tma_info_pipeline_execute", "Unit": "cpu_core" }, { - "BriefDescription": "Average latency of all requests to external memory (in Uncore cycles)", - "MetricExpr": "(UNC_ARB_TRK_OCCUPANCY.ALL + UNC_ARB_DAT_OCCUPANCY.RD) / UNC_ARB_TRK_REQUESTS.ALL", - "MetricGroup": "Mem;SoC", - "MetricName": "tma_info_mem_request_latency", + "BriefDescription": "Instructions per a microcode Assist invocation", + "MetricExpr": "INST_RETIRED.ANY / cpu_core@ASSISTS.ANY\\,umask\\=0x1B@", + "MetricGroup": "Pipeline;Ret;Retire", + "MetricName": "tma_info_pipeline_ipassist", + "MetricThreshold": "tma_info_pipeline_ipassist < 100e3", + "PublicDescription": "Instructions per a microcode Assist invocation. See Assists tree node for details (lower number means higher occurrence rate)", "Unit": "cpu_core" }, { - "BriefDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks", - "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_bandwidth / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_sq_full / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full))) + tma_l1_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_fb_full / (tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk))", - "MetricGroup": "Mem;MemoryBW;Offcore;tma_issueBW", - "MetricName": "tma_info_memory_bandwidth", - "MetricThreshold": "tma_info_memory_bandwidth > 20", - "PublicDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks. Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_mem_bandwidth, tma_sq_full", + "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "tma_retiring * tma_info_thread_slots / cpu_core@UOPS_RETIRED.SLOTS\\,cmask\\=1@", + "MetricGroup": "Pipeline;Ret", + "MetricName": "tma_info_pipeline_retire", "Unit": "cpu_core" }, { - "BriefDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs)", - "MetricExpr": "100 * tma_memory_bound * (tma_l1_bound / max(tma_memory_bound, tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_load / max(tma_l1_bound, tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk)) + tma_store_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_store / (tma_dtlb_store + tma_false_sharing + tma_split_stores + tma_store_latency + tma_streaming_stores)))", - "MetricGroup": "Mem;MemoryTLB;Offcore;tma_issueTLB", - "MetricName": "tma_info_memory_data_tlbs", - "MetricThreshold": "tma_info_memory_data_tlbs > 20", - "PublicDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs). Related metrics: tma_dtlb_load, tma_dtlb_store", + "BriefDescription": "Estimated fraction of retirement-cycles dealing with repeat instructions", + "MetricExpr": "INST_RETIRED.REP_ITERATION / cpu_core@UOPS_RETIRED.SLOTS\\,cmask\\=1@", + "MetricGroup": "Pipeline;Ret", + "MetricName": "tma_info_pipeline_strings_cycles", + "MetricThreshold": "tma_info_pipeline_strings_cycles > 0.1", "Unit": "cpu_core" }, { - "BriefDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches)", - "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_latency / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_l3_hit_latency / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full)) + tma_l2_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound))", - "MetricGroup": "Mem;MemoryLat;Offcore;tma_issueLat", - "MetricName": "tma_info_memory_latency", - "MetricThreshold": "tma_info_memory_latency > 20", - "PublicDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches). Related metrics: tma_l3_hit_latency, tma_mem_latency", + "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time", + "MetricGroup": "Power;Summary", + "MetricName": "tma_info_system_average_frequency", "Unit": "cpu_core" }, { - "BriefDescription": "Total pipeline cost of Branch Misprediction related bottlenecks", - "MetricExpr": "100 * (tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", - "MetricGroup": "Bad;BadSpec;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_mispredictions", - "MetricThreshold": "tma_info_mispredictions > 20", - "PublicDescription": "Total pipeline cost of Branch Misprediction related bottlenecks. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost, tma_mispredicts_resteers", + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricGroup": "HPC;Summary", + "MetricName": "tma_info_system_cpu_utilization", "Unit": "cpu_core" }, { - "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", - "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", - "MetricGroup": "Mem;MemoryBW;MemoryBound", - "MetricName": "tma_info_mlp", - "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)", + "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", + "MetricExpr": "64 * (UNC_ARB_TRK_REQUESTS.ALL + UNC_ARB_COH_TRK_REQUESTS.ALL) / 1e6 / duration_time / 1e3", + "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", + "MetricName": "tma_info_system_dram_bw_use", + "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_mem_bandwidth, tma_sq_full", "Unit": "cpu_core" }, { - "BriefDescription": "Fraction of branches of other types (not individually covered by other metrics in Info.Branches group)", - "MetricExpr": "1 - (tma_info_cond_nt + tma_info_cond_tk + tma_info_callret + tma_info_jump)", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_other_branches", + "BriefDescription": "Giga Floating Point Operations Per Second", + "MetricExpr": "(cpu_core@FP_ARITH_INST_RETIRED.SCALAR_SINGLE@ + cpu_core@FP_ARITH_INST_RETIRED.SCALAR_DOUBLE@ + 2 * cpu_core@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE@ + 4 * (cpu_core@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE@ + cpu_core@FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE@) + 8 * cpu_core@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE@) / 1e9 / duration_time", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_system_gflops", + "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine.", "Unit": "cpu_core" }, { - "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", - "MetricExpr": "(ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING) / (4 * tma_info_core_clks)", - "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_page_walks_utilization", - "MetricThreshold": "tma_info_page_walks_utilization > 0.5", + "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", + "MetricExpr": "INST_RETIRED.ANY / cpu_core@BR_INST_RETIRED.FAR_BRANCH@u", + "MetricGroup": "Branches;OS", + "MetricName": "tma_info_system_ipfarbranch", + "MetricThreshold": "tma_info_system_ipfarbranch < 1e6", "Unit": "cpu_core" }, { - "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", - "MetricExpr": "tma_retiring * tma_info_slots / cpu_core@UOPS_RETIRED.SLOTS\\,cmask\\=1@", - "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_retire", + "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / cpu_core@INST_RETIRED.ANY_P@k", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_cpi", "Unit": "cpu_core" }, { - "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", - "MetricExpr": "TOPDOWN.SLOTS", - "MetricGroup": "TmaL1;tma_L1_group", - "MetricName": "tma_info_slots", + "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_utilization", + "MetricThreshold": "tma_info_system_kernel_utilization > 0.05", "Unit": "cpu_core" }, { - "BriefDescription": "Fraction of Physical Core issue-slots utilized by this Logical Processor", - "MetricExpr": "(tma_info_slots / (TOPDOWN.SLOTS / 2) if #SMT_on else 1)", - "MetricGroup": "SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_slots_utilization", + "BriefDescription": "Average number of parallel data read requests to external memory", + "MetricExpr": "UNC_ARB_DAT_OCCUPANCY.RD / cpu_core@UNC_ARB_DAT_OCCUPANCY.RD\\,cmask\\=1@", + "MetricGroup": "Mem;MemoryBW;SoC", + "MetricName": "tma_info_system_mem_parallel_reads", + "PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Average latency of data read request to external memory (in nanoseconds)", + "MetricExpr": "(UNC_ARB_TRK_OCCUPANCY.RD + UNC_ARB_DAT_OCCUPANCY.RD) / UNC_ARB_TRK_REQUESTS.RD", + "MetricGroup": "Mem;MemoryLat;SoC", + "MetricName": "tma_info_system_mem_read_latency", + "PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Average latency of all requests to external memory (in Uncore cycles)", + "MetricExpr": "(UNC_ARB_TRK_OCCUPANCY.ALL + UNC_ARB_DAT_OCCUPANCY.RD) / UNC_ARB_TRK_REQUESTS.ALL", + "MetricGroup": "Mem;SoC", + "MetricName": "tma_info_system_mem_request_latency", "Unit": "cpu_core" }, { "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", - "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_DISTRIBUTED if #SMT_on else 0)", + "MetricExpr": "(1 - cpu_core@CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE@ / cpu_core@CPU_CLK_UNHALTED.REF_DISTRIBUTED@ if #SMT_on else 0)", "MetricGroup": "SMT", - "MetricName": "tma_info_smt_2t_utilization", + "MetricName": "tma_info_system_smt_2t_utilization", "Unit": "cpu_core" }, { "BriefDescription": "Socket actual clocks when any core is active on that socket", "MetricExpr": "UNC_CLOCK.SOCKET", "MetricGroup": "SoC", - "MetricName": "tma_info_socket_clks", + "MetricName": "tma_info_system_socket_clks", "Unit": "cpu_core" }, { - "BriefDescription": "STLB (2nd level TLB) data store speculative misses per kilo instruction (misses of any page-size that complete the page walk)", - "MetricExpr": "1e3 * DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", - "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_store_stlb_mpki", + "BriefDescription": "Average Frequency Utilization relative nominal frequency", + "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricGroup": "Power", + "MetricName": "tma_info_system_turbo_utilization", "Unit": "cpu_core" }, { - "BriefDescription": "Estimated fraction of retirement-cycles dealing with repeat instructions", - "MetricExpr": "INST_RETIRED.REP_ITERATION / cpu_core@UOPS_RETIRED.SLOTS\\,cmask\\=1@", - "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_strings_cycles", - "MetricThreshold": "tma_info_strings_cycles > 0.1", + "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "cpu_core@CPU_CLK_UNHALTED.THREAD@", + "MetricGroup": "Pipeline", + "MetricName": "tma_info_thread_clks", "Unit": "cpu_core" }, { - "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", - "MetricGroup": "Power", - "MetricName": "tma_info_turbo_utilization", + "BriefDescription": "Cycles Per Instruction (per Logical Processor)", + "MetricExpr": "1 / tma_info_thread_ipc", + "MetricGroup": "Mem;Pipeline", + "MetricName": "tma_info_thread_cpi", + "Unit": "cpu_core" + }, + { + "BriefDescription": "The ratio of Executed- by Issued-Uops", + "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", + "MetricGroup": "Cor;Pipeline", + "MetricName": "tma_info_thread_execute_per_issue", + "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage.", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Instructions Per Cycle (per Logical Processor)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks", + "MetricGroup": "Ret;Summary", + "MetricName": "tma_info_thread_ipc", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", + "MetricExpr": "cpu_core@TOPDOWN.SLOTS@", + "MetricGroup": "TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots", + "Unit": "cpu_core" + }, + { + "BriefDescription": "Fraction of Physical Core issue-slots utilized by this Logical Processor", + "MetricExpr": "(tma_info_thread_slots / (cpu_core@TOPDOWN.SLOTS@ / 2) if #SMT_on else 1)", + "MetricGroup": "SMT;TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots_utilization", "Unit": "cpu_core" }, { "BriefDescription": "Uops Per Instruction", - "MetricExpr": "tma_retiring * tma_info_slots / INST_RETIRED.ANY", + "MetricExpr": "tma_retiring * tma_info_thread_slots / INST_RETIRED.ANY", "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_uoppi", - "MetricThreshold": "tma_info_uoppi > 1.05", + "MetricName": "tma_info_thread_uoppi", + "MetricThreshold": "tma_info_thread_uoppi > 1.05", "Unit": "cpu_core" }, { "BriefDescription": "Instruction per taken branch", - "MetricExpr": "tma_retiring * tma_info_slots / BR_INST_RETIRED.NEAR_TAKEN", + "MetricExpr": "tma_retiring * tma_info_thread_slots / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW", - "MetricName": "tma_info_uptb", - "MetricThreshold": "tma_info_uptb < 9", + "MetricName": "tma_info_thread_uptb", + "MetricThreshold": "tma_info_thread_uptb < 9", "Unit": "cpu_core" }, { @@ -1938,7 +1951,7 @@ }, { "BriefDescription": "This metric represents 128-bit vector Integer ADD/SUB/SAD or VNNI (Vector Neural Network Instructions) uops fraction the CPU has retired", - "MetricExpr": "(INT_VEC_RETIRED.ADD_128 + INT_VEC_RETIRED.VNNI_128) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(cpu_core@INT_VEC_RETIRED.ADD_128@ + cpu_core@INT_VEC_RETIRED.VNNI_128@) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;IntVector;Pipeline;TopdownL4;tma_L4_group;tma_int_operations_group;tma_issue2P", "MetricName": "tma_int_vector_128b", "MetricThreshold": "tma_int_vector_128b > 0.1 & (tma_int_operations > 0.1 & tma_light_operations > 0.6)", @@ -1948,7 +1961,7 @@ }, { "BriefDescription": "This metric represents 256-bit vector Integer ADD/SUB/SAD or VNNI (Vector Neural Network Instructions) uops fraction the CPU has retired", - "MetricExpr": "(INT_VEC_RETIRED.ADD_256 + INT_VEC_RETIRED.MUL_256 + INT_VEC_RETIRED.VNNI_256) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(cpu_core@INT_VEC_RETIRED.ADD_256@ + cpu_core@INT_VEC_RETIRED.MUL_256@ + cpu_core@INT_VEC_RETIRED.VNNI_256@) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;IntVector;Pipeline;TopdownL4;tma_L4_group;tma_int_operations_group;tma_issue2P", "MetricName": "tma_int_vector_256b", "MetricThreshold": "tma_int_vector_256b > 0.1 & (tma_int_operations > 0.1 & tma_light_operations > 0.6)", @@ -1958,7 +1971,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses", - "MetricExpr": "ICACHE_TAG.STALLS / tma_info_clks", + "MetricExpr": "ICACHE_TAG.STALLS / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -1968,7 +1981,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled without loads missing the L1 data cache", - "MetricExpr": "max((EXE_ACTIVITY.BOUND_ON_LOADS - MEMORY_ACTIVITY.STALLS_L1D_MISS) / tma_info_clks, 0)", + "MetricExpr": "max((cpu_core@EXE_ACTIVITY.BOUND_ON_LOADS@ - cpu_core@MEMORY_ACTIVITY.STALLS_L1D_MISS@) / tma_info_thread_clks, 0)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_issueL1;tma_issueMC;tma_memory_bound_group", "MetricName": "tma_l1_bound", "MetricThreshold": "tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1979,7 +1992,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled due to L2 cache accesses by loads", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(MEMORY_ACTIVITY.STALLS_L1D_MISS - MEMORY_ACTIVITY.STALLS_L2_MISS) / tma_info_clks", + "MetricExpr": "(cpu_core@MEMORY_ACTIVITY.STALLS_L1D_MISS@ - cpu_core@MEMORY_ACTIVITY.STALLS_L2_MISS@) / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l2_bound", "MetricThreshold": "tma_l2_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1989,7 +2002,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core", - "MetricExpr": "(MEMORY_ACTIVITY.STALLS_L2_MISS - MEMORY_ACTIVITY.STALLS_L3_MISS) / tma_info_clks", + "MetricExpr": "(cpu_core@MEMORY_ACTIVITY.STALLS_L2_MISS@ - cpu_core@MEMORY_ACTIVITY.STALLS_L3_MISS@) / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1999,21 +2012,21 @@ }, { "BriefDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited)", - "MetricExpr": "9 * tma_info_average_frequency * MEM_LOAD_RETIRED.L3_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "9 * tma_info_system_average_frequency * cpu_core@MEM_LOAD_RETIRED.L3_HIT@ * (1 + cpu_core@MEM_LOAD_RETIRED.FB_HIT@ / cpu_core@MEM_LOAD_RETIRED.L1_MISS@ / 2) / tma_info_thread_clks", "MetricGroup": "MemoryLat;TopdownL4;tma_L4_group;tma_issueLat;tma_l3_bound_group", "MetricName": "tma_l3_hit_latency", "MetricThreshold": "tma_l3_hit_latency > 0.1 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited). Avoiding private cache misses (i.e. L2 misses/L3 hits) will improve the latency; reduce contention with sibling physical cores and increase performance. Note the value of this node may overlap with its siblings. Sample with: MEM_LOAD_RETIRED.L3_HIT_PS. Related metrics: tma_info_memory_latency, tma_mem_latency", + "PublicDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited). Avoiding private cache misses (i.e. L2 misses/L3 hits) will improve the latency; reduce contention with sibling physical cores and increase performance. Note the value of this node may overlap with its siblings. Sample with: MEM_LOAD_RETIRED.L3_HIT_PS. Related metrics: tma_info_bottleneck_memory_latency, tma_mem_latency", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)", - "MetricExpr": "DECODE.LCP / tma_info_clks", + "MetricExpr": "DECODE.LCP / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_lcp", "MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb", + "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb", "ScaleUnit": "100%", "Unit": "cpu_core" }, @@ -2023,13 +2036,14 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_light_operations", "MetricThreshold": "tma_light_operations > 0.6", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation). This correlates with total number of instructions used by the program. A uops-per-instruction (see UopPI metric) ratio of 1 or less should be expected for decently optimized software running on Intel Core/Xeon products. While this often indicates efficient X86 instructions were executed; high value does not necessarily mean better performance cannot be achieved. Sample with: INST_RETIRED.PREC_DIST", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Load operations", - "MetricExpr": "UOPS_DISPATCHED.PORT_2_3_10 / (3 * tma_info_core_clks)", + "MetricExpr": "UOPS_DISPATCHED.PORT_2_3_10 / (3 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_load_op_utilization", "MetricThreshold": "tma_load_op_utilization > 0.6", @@ -2048,7 +2062,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles where the Second-level TLB (STLB) was missed by load accesses, performing a hardware page walk", - "MetricExpr": "DTLB_LOAD_MISSES.WALK_ACTIVE / tma_info_clks", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_ACTIVE / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL5;tma_L5_group;tma_dtlb_load_group", "MetricName": "tma_load_stlb_miss", "MetricThreshold": "tma_load_stlb_miss > 0.05 & (tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -2058,7 +2072,7 @@ { "BriefDescription": "This metric represents fraction of cycles the CPU spent handling cache misses due to lock operations", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(16 * max(0, MEM_INST_RETIRED.LOCK_LOADS - L2_RQSTS.ALL_RFO) + MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES * (10 * L2_RQSTS.RFO_HIT + min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO))) / tma_info_clks", + "MetricExpr": "(16 * max(0, cpu_core@MEM_INST_RETIRED.LOCK_LOADS@ - cpu_core@L2_RQSTS.ALL_RFO@) + cpu_core@MEM_INST_RETIRED.LOCK_LOADS@ / cpu_core@MEM_INST_RETIRED.ALL_STORES@ * (10 * cpu_core@L2_RQSTS.RFO_HIT@ + min(cpu_core@CPU_CLK_UNHALTED.THREAD@, cpu_core@OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO@))) / tma_info_thread_clks", "MetricGroup": "Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_l1_bound_group", "MetricName": "tma_lock_latency", "MetricThreshold": "tma_lock_latency > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -2068,10 +2082,10 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to LSD (Loop Stream Detector) unit", - "MetricExpr": "(LSD.CYCLES_ACTIVE - LSD.CYCLES_OK) / tma_info_core_clks / 2", + "MetricExpr": "(cpu_core@LSD.CYCLES_ACTIVE@ - cpu_core@LSD.CYCLES_OK@) / tma_info_core_core_clks / 2", "MetricGroup": "FetchBW;LSD;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_lsd", - "MetricThreshold": "tma_lsd > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 6 > 0.35)", + "MetricThreshold": "tma_lsd > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 6 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to LSD (Loop Stream Detector) unit. LSD typically does well sustaining Uop supply. However; in some rare cases; optimal uop-delivery could not be reached for small loops whose size (in terms of number of uops) does not suit well the LSD structure.", "ScaleUnit": "100%", "Unit": "cpu_core" @@ -2082,43 +2096,45 @@ "MetricGroup": "BadSpec;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.1 & tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears. These slots are either wasted by uops fetched prior to the clear; or stalls the out-of-order portion of the machine needs to recover its state after the clear. For example; this can happen due to memory ordering Nukes (e.g. Memory Disambiguation) or Self-Modifying-Code (SMC) nukes. Sample with: MACHINE_CLEARS.COUNT. Related metrics: tma_clears_resteers, tma_contested_accesses, tma_data_sharing, tma_false_sharing, tma_l1_bound, tma_microcode_sequencer, tma_ms_switches, tma_remote_cache", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu_core@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_clks", + "MetricExpr": "min(cpu_core@CPU_CLK_UNHALTED.THREAD@, cpu_core@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW", "MetricName": "tma_mem_bandwidth", "MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_sq_full", + "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_sq_full", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth", + "MetricExpr": "min(cpu_core@CPU_CLK_UNHALTED.THREAD@, cpu_core@OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD@) / tma_info_thread_clks - tma_mem_bandwidth", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat", "MetricName": "tma_mem_latency", "MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM). This metric does not aggregate requests from other Logical Processors/Physical Cores/sockets (see Uncore counters for that). Related metrics: tma_info_memory_latency, tma_l3_hit_latency", + "PublicDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM). This metric does not aggregate requests from other Logical Processors/Physical Cores/sockets (see Uncore counters for that). Related metrics: tma_info_bottleneck_memory_latency, tma_l3_hit_latency", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck", - "MetricExpr": "min(tma_backend_bound, LD_HEAD.ANY_AT_RET / tma_info_clks + tma_store_bound)", + "MetricExpr": "cpu_core@topdown\\-mem\\-bound@ / (cpu_core@topdown\\-fe\\-bound@ + cpu_core@topdown\\-bad\\-spec@ + cpu_core@topdown\\-retiring@ + cpu_core@topdown\\-be\\-bound@) + 0 * tma_info_thread_slots", "MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck. Memory Bound estimates fraction of slots where pipeline is likely stalled due to demand load or store instructions. This accounts mainly for (1) non-completed in-flight memory demand loads which coincides with execution units starvation; in addition to (2) cases where stores could impose backpressure on the pipeline when many of them get buffered at the same time (less common out of the two).", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to LFENCE Instructions.", - "MetricExpr": "13 * MISC2_RETIRED.LFENCE / tma_info_clks", + "MetricExpr": "13 * cpu_core@MISC2_RETIRED.LFENCE@ / tma_info_thread_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_serializing_operation_group", "MetricName": "tma_memory_fence", "MetricThreshold": "tma_memory_fence > 0.05 & (tma_serializing_operation > 0.1 & (tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))))", @@ -2127,7 +2143,8 @@ }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring memory operations -- uops for memory load or store accesses.", - "MetricExpr": "tma_light_operations * MEM_UOP_RETIRED.ANY / (tma_retiring * tma_info_slots)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "tma_light_operations * cpu_core@MEM_UOP_RETIRED.ANY@ / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_memory_operations", "MetricThreshold": "tma_memory_operations > 0.1 & tma_light_operations > 0.6", @@ -2136,7 +2153,7 @@ }, { "BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit", - "MetricExpr": "UOPS_RETIRED.MS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.MS / tma_info_thread_slots", "MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS", "MetricName": "tma_microcode_sequencer", "MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1", @@ -2146,27 +2163,27 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage", - "MetricExpr": "tma_branch_mispredicts / tma_bad_speculation * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks", + "MetricExpr": "tma_branch_mispredicts / tma_bad_speculation * cpu_core@INT_MISC.CLEAR_RESTEER_CYCLES@ / tma_info_thread_clks", "MetricGroup": "BadSpec;BrMispredicts;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueBM", "MetricName": "tma_mispredicts_resteers", "MetricThreshold": "tma_mispredicts_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost, tma_info_mispredictions", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost, tma_info_bottleneck_mispredictions", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline)", - "MetricExpr": "(IDQ.MITE_CYCLES_ANY - IDQ.MITE_CYCLES_OK) / tma_info_core_clks / 2", + "MetricExpr": "(cpu_core@IDQ.MITE_CYCLES_ANY@ - cpu_core@IDQ.MITE_CYCLES_OK@) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_mite", - "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 6 > 0.35)", + "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 6 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline). This pipeline is used for code that was not pre-cached in the DSB or LSD. For example; inefficiencies due to asymmetric decoders; use of long immediate or LCP can manifest as MITE fetch bandwidth bottleneck. Sample with: FRONTEND_RETIRED.ANY_DSB_MISS", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "The Mixing_Vectors metric gives the percentage of injected blend uops out of all uops issued", - "MetricExpr": "160 * ASSISTS.SSE_AVX_MIX / tma_info_clks", + "MetricExpr": "160 * cpu_core@ASSISTS.SSE_AVX_MIX@ / tma_info_thread_clks", "MetricGroup": "TopdownL5;tma_L5_group;tma_issueMV;tma_ports_utilized_0_group", "MetricName": "tma_mixing_vectors", "MetricThreshold": "tma_mixing_vectors > 0.05", @@ -2176,7 +2193,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)", - "MetricExpr": "3 * cpu_core@UOPS_RETIRED.MS\\,cmask\\=1\\,edge@ / (tma_retiring * tma_info_slots / UOPS_ISSUED.ANY) / tma_info_clks", + "MetricExpr": "3 * cpu_core@UOPS_RETIRED.MS\\,cmask\\=1\\,edge@ / (tma_retiring * tma_info_thread_slots / cpu_core@UOPS_ISSUED.ANY@) / tma_info_thread_clks", "MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO", "MetricName": "tma_ms_switches", "MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -2186,7 +2203,7 @@ }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring branch instructions that were not fused", - "MetricExpr": "tma_light_operations * (BR_INST_RETIRED.ALL_BRANCHES - INST_RETIRED.MACRO_FUSED) / (tma_retiring * tma_info_slots)", + "MetricExpr": "tma_light_operations * (cpu_core@BR_INST_RETIRED.ALL_BRANCHES@ - cpu_core@INST_RETIRED.MACRO_FUSED@) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_non_fused_branches", "MetricThreshold": "tma_non_fused_branches > 0.1 & tma_light_operations > 0.6", @@ -2196,7 +2213,7 @@ }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring NOP (no op) instructions", - "MetricExpr": "tma_light_operations * INST_RETIRED.NOP / (tma_retiring * tma_info_slots)", + "MetricExpr": "tma_light_operations * cpu_core@INST_RETIRED.NOP@ / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_nop_instructions", "MetricThreshold": "tma_nop_instructions > 0.1 & tma_light_operations > 0.6", @@ -2206,6 +2223,7 @@ }, { "BriefDescription": "This metric represents the remaining light uops fraction the CPU has executed - remaining means not covered by other sibling nodes", + "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "max(0, tma_light_operations - (tma_fp_arith + tma_int_operations + tma_memory_operations + tma_fused_instructions + tma_non_fused_branches + tma_nop_instructions))", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_other_light_ops", @@ -2216,7 +2234,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of slots the CPU retired uops as a result of handing Page Faults", - "MetricExpr": "99 * ASSISTS.PAGE_FAULT / tma_info_slots", + "MetricExpr": "99 * cpu_core@ASSISTS.PAGE_FAULT@ / tma_info_thread_slots", "MetricGroup": "TopdownL5;tma_L5_group;tma_assists_group", "MetricName": "tma_page_faults", "MetricThreshold": "tma_page_faults > 0.05", @@ -2226,7 +2244,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 0 ([SNB+] ALU; [HSW+] ALU and 2nd branch)", - "MetricExpr": "UOPS_DISPATCHED.PORT_0 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_0 / tma_info_core_core_clks", "MetricGroup": "Compute;TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_0", "MetricThreshold": "tma_port_0 > 0.6", @@ -2236,7 +2254,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 1 (ALU)", - "MetricExpr": "UOPS_DISPATCHED.PORT_1 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_1 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_1", "MetricThreshold": "tma_port_1 > 0.6", @@ -2246,7 +2264,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 6 ([HSW+]Primary Branch and simple ALU)", - "MetricExpr": "UOPS_DISPATCHED.PORT_6 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_6 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_6", "MetricThreshold": "tma_port_6 > 0.6", @@ -2256,7 +2274,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)", - "MetricExpr": "((cpu_core@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - EXE_ACTIVITY.BOUND_ON_LOADS) + (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * cpu_core@EXE_ACTIVITY.2_PORTS_UTIL\\,umask\\=0xc@)) / tma_info_clks if ARITH.DIV_ACTIVE < CYCLE_ACTIVITY.STALLS_TOTAL - EXE_ACTIVITY.BOUND_ON_LOADS else (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * cpu_core@EXE_ACTIVITY.2_PORTS_UTIL\\,umask\\=0xc@) / tma_info_clks)", + "MetricExpr": "((cpu_core@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ + tma_serializing_operation * (cpu_core@CYCLE_ACTIVITY.STALLS_TOTAL@ - cpu_core@EXE_ACTIVITY.BOUND_ON_LOADS@) + (cpu_core@EXE_ACTIVITY.1_PORTS_UTIL@ + tma_retiring * cpu_core@EXE_ACTIVITY.2_PORTS_UTIL\\,umask\\=0xc@)) / tma_info_thread_clks if cpu_core@ARITH.DIV_ACTIVE@ < cpu_core@CYCLE_ACTIVITY.STALLS_TOTAL@ - cpu_core@EXE_ACTIVITY.BOUND_ON_LOADS@ else (cpu_core@EXE_ACTIVITY.1_PORTS_UTIL@ + tma_retiring * cpu_core@EXE_ACTIVITY.2_PORTS_UTIL\\,umask\\=0xc@) / tma_info_thread_clks)", "MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_ports_utilization", "MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -2266,7 +2284,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed no uops on any execution port (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "cpu_core@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ / tma_info_clks + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - EXE_ACTIVITY.BOUND_ON_LOADS) / tma_info_clks", + "MetricExpr": "cpu_core@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ / tma_info_thread_clks + tma_serializing_operation * (cpu_core@CYCLE_ACTIVITY.STALLS_TOTAL@ - cpu_core@EXE_ACTIVITY.BOUND_ON_LOADS@) / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_0", "MetricThreshold": "tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -2276,7 +2294,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the CPU executed total of 1 uop per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "EXE_ACTIVITY.1_PORTS_UTIL / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.1_PORTS_UTIL / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issueL1;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_1", "MetricThreshold": "tma_ports_utilized_1 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -2286,7 +2304,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 2 uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "EXE_ACTIVITY.2_PORTS_UTIL / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.2_PORTS_UTIL / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issue2P;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_2", "MetricThreshold": "tma_ports_utilized_2 > 0.15 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -2296,7 +2314,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 3 or more uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "UOPS_EXECUTED.CYCLES_GE_3 / tma_info_clks", + "MetricExpr": "UOPS_EXECUTED.CYCLES_GE_3 / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_3m", "MetricThreshold": "tma_ports_utilized_3m > 0.7 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -2306,17 +2324,19 @@ }, { "BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired", - "MetricExpr": "topdown\\-retiring / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_slots", - "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "cpu_core@topdown\\-retiring@ / (cpu_core@topdown\\-fe\\-bound@ + cpu_core@topdown\\-bad\\-spec@ + cpu_core@topdown\\-retiring@ + cpu_core@topdown\\-be\\-bound@) + 0 * tma_info_thread_slots", + "MetricGroup": "Default;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired. Ideally; all pipeline slots would be attributed to the Retiring category. Retiring of 100% would indicate the maximum Pipeline_Width throughput was achieved. Maximizing Retiring typically increases the Instructions-per-cycle (see IPC metric). Note that a high Retiring value does not necessary mean there is no room for more performance. For example; Heavy-operations or Microcode Assists are categorized under Retiring. They often indicate suboptimal performance and can often be optimized or avoided. Sample with: UOPS_RETIRED.SLOTS", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric represents fraction of cycles the CPU issue-pipeline was stalled due to serializing operations", - "MetricExpr": "RESOURCE_STALLS.SCOREBOARD / tma_info_clks", + "MetricExpr": "RESOURCE_STALLS.SCOREBOARD / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL5;tma_L5_group;tma_issueSO;tma_ports_utilized_0_group", "MetricName": "tma_serializing_operation", "MetricThreshold": "tma_serializing_operation > 0.1 & (tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)))", @@ -2326,7 +2346,7 @@ }, { "BriefDescription": "This metric represents Shuffle (cross \"vector lane\" data transfers) uops fraction the CPU has retired.", - "MetricExpr": "INT_VEC_RETIRED.SHUFFLES / (tma_retiring * tma_info_slots)", + "MetricExpr": "INT_VEC_RETIRED.SHUFFLES / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "HPC;Pipeline;TopdownL4;tma_L4_group;tma_int_operations_group", "MetricName": "tma_shuffles", "MetricThreshold": "tma_shuffles > 0.1 & (tma_int_operations > 0.1 & tma_light_operations > 0.6)", @@ -2335,7 +2355,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to PAUSE Instructions", - "MetricExpr": "CPU_CLK_UNHALTED.PAUSE / tma_info_clks", + "MetricExpr": "CPU_CLK_UNHALTED.PAUSE / tma_info_thread_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_serializing_operation_group", "MetricName": "tma_slow_pause", "MetricThreshold": "tma_slow_pause > 0.05 & (tma_serializing_operation > 0.1 & (tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))))", @@ -2345,7 +2365,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles handling memory load split accesses - load that cross 64-byte cache line boundary", - "MetricExpr": "tma_info_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * cpu_core@LD_BLOCKS.NO_SR@ / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_split_loads", "MetricThreshold": "tma_split_loads > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -2355,7 +2375,7 @@ }, { "BriefDescription": "This metric represents rate of split store accesses", - "MetricExpr": "MEM_INST_RETIRED.SPLIT_STORES / tma_info_core_clks", + "MetricExpr": "MEM_INST_RETIRED.SPLIT_STORES / tma_info_core_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_issueSpSt;tma_store_bound_group", "MetricName": "tma_split_stores", "MetricThreshold": "tma_split_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -2365,17 +2385,17 @@ }, { "BriefDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors)", - "MetricExpr": "(XQ.FULL_CYCLES + L1D_PEND_MISS.L2_STALLS) / tma_info_clks", + "MetricExpr": "(cpu_core@XQ.FULL_CYCLES@ + cpu_core@L1D_PEND_MISS.L2_STALLS@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueBW;tma_l3_bound_group", "MetricName": "tma_sq_full", "MetricThreshold": "tma_sq_full > 0.3 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_mem_bandwidth", + "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_mem_bandwidth", "ScaleUnit": "100%", "Unit": "cpu_core" }, { "BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write", - "MetricExpr": "EXE_ACTIVITY.BOUND_ON_STORES / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.BOUND_ON_STORES / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -2385,7 +2405,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of cycles when the memory subsystem had loads blocked since they could not forward data from earlier (in program order) overlapping stores", - "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_clks", + "MetricExpr": "13 * cpu_core@LD_BLOCKS.STORE_FORWARD@ / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_store_fwd_blk", "MetricThreshold": "tma_store_fwd_blk > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -2395,7 +2415,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles the CPU spent handling L1D store misses", - "MetricExpr": "(MEM_STORE_RETIRED.L2_HIT * 10 * (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) + (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_clks", + "MetricExpr": "(cpu_core@MEM_STORE_RETIRED.L2_HIT@ * 10 * (1 - cpu_core@MEM_INST_RETIRED.LOCK_LOADS@ / cpu_core@MEM_INST_RETIRED.ALL_STORES@) + (1 - cpu_core@MEM_INST_RETIRED.LOCK_LOADS@ / cpu_core@MEM_INST_RETIRED.ALL_STORES@) * min(cpu_core@CPU_CLK_UNHALTED.THREAD@, cpu_core@OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO@)) / tma_info_thread_clks", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_issueSL;tma_store_bound_group", "MetricName": "tma_store_latency", "MetricThreshold": "tma_store_latency > 0.1 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -2405,7 +2425,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Store operations", - "MetricExpr": "(UOPS_DISPATCHED.PORT_4_9 + UOPS_DISPATCHED.PORT_7_8) / (4 * tma_info_core_clks)", + "MetricExpr": "(cpu_core@UOPS_DISPATCHED.PORT_4_9@ + cpu_core@UOPS_DISPATCHED.PORT_7_8@) / (4 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_store_op_utilization", "MetricThreshold": "tma_store_op_utilization > 0.6", @@ -2424,7 +2444,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles where the STLB was missed by store accesses, performing a hardware page walk", - "MetricExpr": "DTLB_STORE_MISSES.WALK_ACTIVE / tma_info_core_clks", + "MetricExpr": "DTLB_STORE_MISSES.WALK_ACTIVE / tma_info_core_core_clks", "MetricGroup": "MemoryTLB;TopdownL5;tma_L5_group;tma_dtlb_store_group", "MetricName": "tma_store_stlb_miss", "MetricThreshold": "tma_store_stlb_miss > 0.05 & (tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -2433,7 +2453,7 @@ }, { "BriefDescription": "This metric estimates how often CPU was stalled due to Streaming store memory accesses; Streaming store optimize out a read request required by RFO stores", - "MetricExpr": "9 * OCR.STREAMING_WR.ANY_RESPONSE / tma_info_clks", + "MetricExpr": "9 * cpu_core@OCR.STREAMING_WR.ANY_RESPONSE@ / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueSmSt;tma_store_bound_group", "MetricName": "tma_streaming_stores", "MetricThreshold": "tma_streaming_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -2443,7 +2463,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to new branch address clears", - "MetricExpr": "INT_MISC.UNKNOWN_BRANCH_CYCLES / tma_info_clks", + "MetricExpr": "INT_MISC.UNKNOWN_BRANCH_CYCLES / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;TopdownL4;tma_L4_group;tma_branch_resteers_group", "MetricName": "tma_unknown_branches", "MetricThreshold": "tma_unknown_branches > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", @@ -2453,7 +2473,7 @@ }, { "BriefDescription": "This metric serves as an approximation of legacy x87 usage", - "MetricExpr": "tma_retiring * UOPS_EXECUTED.X87 / UOPS_EXECUTED.THREAD", + "MetricExpr": "tma_retiring * cpu_core@UOPS_EXECUTED.X87@ / UOPS_EXECUTED.THREAD", "MetricGroup": "Compute;TopdownL4;tma_L4_group;tma_fp_arith_group", "MetricName": "tma_x87_use", "MetricThreshold": "tma_x87_use > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6)", diff --git a/tools/perf/pmu-events/arch/x86/alderlake/cache.json b/tools/perf/pmu-events/arch/x86/alderlake/cache.json index 51770416bcc2..b3d7f8fb50df 100644 --- a/tools/perf/pmu-events/arch/x86/alderlake/cache.json +++ b/tools/perf/pmu-events/arch/x86/alderlake/cache.json @@ -1018,6 +1018,15 @@ "Unit": "cpu_core" }, { + "BriefDescription": "Counts bus locks, accounts for cache line split locks and UC locks.", + "EventCode": "0x2c", + "EventName": "SQ_MISC.BUS_LOCK", + "PublicDescription": "Counts the more expensive bus lock needed to enforce cache coherency for certain memory accesses that need to be done atomically. Can be created by issuing an atomic instruction (via the LOCK prefix) which causes a cache line split or accesses uncacheable memory.", + "SampleAfterValue": "100003", + "UMask": "0x10", + "Unit": "cpu_core" + }, + { "BriefDescription": "Number of PREFETCHNTA instructions executed.", "EventCode": "0x40", "EventName": "SW_PREFETCH_ACCESS.NTA", diff --git a/tools/perf/pmu-events/arch/x86/alderlake/memory.json b/tools/perf/pmu-events/arch/x86/alderlake/memory.json index 55827b276e6e..73d92d5c9f9d 100644 --- a/tools/perf/pmu-events/arch/x86/alderlake/memory.json +++ b/tools/perf/pmu-events/arch/x86/alderlake/memory.json @@ -93,19 +93,21 @@ "Unit": "cpu_core" }, { - "BriefDescription": "MEMORY_ACTIVITY.STALLS_L2_MISS", + "BriefDescription": "Execution stalls while L2 cache miss demand cacheable load request is outstanding.", "CounterMask": "5", "EventCode": "0x47", "EventName": "MEMORY_ACTIVITY.STALLS_L2_MISS", + "PublicDescription": "Execution stalls while L2 cache miss demand cacheable load request is outstanding (will not count for uncacheable demand requests e.g. bus lock).", "SampleAfterValue": "1000003", "UMask": "0x5", "Unit": "cpu_core" }, { - "BriefDescription": "MEMORY_ACTIVITY.STALLS_L3_MISS", + "BriefDescription": "Execution stalls while L3 cache miss demand cacheable load request is outstanding.", "CounterMask": "9", "EventCode": "0x47", "EventName": "MEMORY_ACTIVITY.STALLS_L3_MISS", + "PublicDescription": "Execution stalls while L3 cache miss demand cacheable load request is outstanding (will not count for uncacheable demand requests e.g. bus lock).", "SampleAfterValue": "1000003", "UMask": "0x9", "Unit": "cpu_core" diff --git a/tools/perf/pmu-events/arch/x86/alderlake/metricgroups.json b/tools/perf/pmu-events/arch/x86/alderlake/metricgroups.json new file mode 100644 index 000000000000..516eb0f93f02 --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/alderlake/metricgroups.json @@ -0,0 +1,122 @@ +{ + "Backend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Bad": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BadSpec": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BigFoot": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BrMispredicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Branches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CacheMisses": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CodeGen": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Compute": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Cor": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSBmiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DataSharing": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Fed": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Flops": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpScalar": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Frontend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "HPC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IcMiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "InsType": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IntVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "L2Evicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "LSD": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MachineClears": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBound": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryTLB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_BW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_Lat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MicroSeq": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "OS": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Offcore": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PGO": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Pipeline": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PortsUtil": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Prefetches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Ret": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Retire": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SMT": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Server": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Snoop": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SoC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL1": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL2": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL3mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "TopdownL5": "Metrics for top-down breakdown at level 5", + "TopdownL6": "Metrics for top-down breakdown at level 6", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_L5_group": "Metrics for top-down breakdown at level 5", + "tma_L6_group": "Metrics for top-down breakdown at level 6", + "tma_alu_op_utilization_group": "Metrics contributing to tma_alu_op_utilization category", + "tma_assists_group": "Metrics contributing to tma_assists category", + "tma_backend_bound_aux_group": "Metrics contributing to tma_backend_bound_aux category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_base_group": "Metrics contributing to tma_base category", + "tma_branch_resteers_group": "Metrics contributing to tma_branch_resteers category", + "tma_core_bound_group": "Metrics contributing to tma_core_bound category", + "tma_dram_bound_group": "Metrics contributing to tma_dram_bound category", + "tma_dtlb_load_group": "Metrics contributing to tma_dtlb_load category", + "tma_dtlb_store_group": "Metrics contributing to tma_dtlb_store category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_fp_arith_group": "Metrics contributing to tma_fp_arith category", + "tma_fp_vector_group": "Metrics contributing to tma_fp_vector category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_heavy_operations_group": "Metrics contributing to tma_heavy_operations category", + "tma_int_operations_group": "Metrics contributing to tma_int_operations category", + "tma_issue2P": "Metrics related by the issue $issue2P", + "tma_issueBC": "Metrics related by the issue $issueBC", + "tma_issueBM": "Metrics related by the issue $issueBM", + "tma_issueBW": "Metrics related by the issue $issueBW", + "tma_issueD0": "Metrics related by the issue $issueD0", + "tma_issueFB": "Metrics related by the issue $issueFB", + "tma_issueFL": "Metrics related by the issue $issueFL", + "tma_issueL1": "Metrics related by the issue $issueL1", + "tma_issueLat": "Metrics related by the issue $issueLat", + "tma_issueMC": "Metrics related by the issue $issueMC", + "tma_issueMS": "Metrics related by the issue $issueMS", + "tma_issueMV": "Metrics related by the issue $issueMV", + "tma_issueRFO": "Metrics related by the issue $issueRFO", + "tma_issueSL": "Metrics related by the issue $issueSL", + "tma_issueSO": "Metrics related by the issue $issueSO", + "tma_issueSmSt": "Metrics related by the issue $issueSmSt", + "tma_issueSpSt": "Metrics related by the issue $issueSpSt", + "tma_issueSyncxn": "Metrics related by the issue $issueSyncxn", + "tma_issueTLB": "Metrics related by the issue $issueTLB", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_l3_bound_group": "Metrics contributing to tma_l3_bound category", + "tma_light_operations_group": "Metrics contributing to tma_light_operations category", + "tma_load_op_utilization_group": "Metrics contributing to tma_load_op_utilization category", + "tma_machine_clears_group": "Metrics contributing to tma_machine_clears category", + "tma_mem_latency_group": "Metrics contributing to tma_mem_latency category", + "tma_mem_scheduler_group": "Metrics contributing to tma_mem_scheduler category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_microcode_sequencer_group": "Metrics contributing to tma_microcode_sequencer category", + "tma_mite_group": "Metrics contributing to tma_mite category", + "tma_nuke_group": "Metrics contributing to tma_nuke category", + "tma_ports_utilization_group": "Metrics contributing to tma_ports_utilization category", + "tma_ports_utilized_0_group": "Metrics contributing to tma_ports_utilized_0 category", + "tma_ports_utilized_3m_group": "Metrics contributing to tma_ports_utilized_3m category", + "tma_resource_bound_group": "Metrics contributing to tma_resource_bound category", + "tma_retiring_group": "Metrics contributing to tma_retiring category", + "tma_serializing_operation_group": "Metrics contributing to tma_serializing_operation category", + "tma_store_bound_group": "Metrics contributing to tma_store_bound category", + "tma_store_op_utilization_group": "Metrics contributing to tma_store_op_utilization category" +} diff --git a/tools/perf/pmu-events/arch/x86/alderlaken/adln-metrics.json b/tools/perf/pmu-events/arch/x86/alderlaken/adln-metrics.json index 1a85d935c733..0f1628d698da 100644 --- a/tools/perf/pmu-events/arch/x86/alderlaken/adln-metrics.json +++ b/tools/perf/pmu-events/arch/x86/alderlaken/adln-metrics.json @@ -86,7 +86,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to certain allocation restrictions.", - "MetricExpr": "TOPDOWN_BE_BOUND.ALLOC_RESTRICTIONS / tma_info_slots", + "MetricExpr": "TOPDOWN_BE_BOUND.ALLOC_RESTRICTIONS / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_resource_bound_group", "MetricName": "tma_alloc_restriction", "MetricThreshold": "tma_alloc_restriction > 0.1", @@ -94,42 +94,49 @@ }, { "BriefDescription": "Counts the total number of issue slots that were not consumed by the backend due to backend stalls", - "MetricExpr": "TOPDOWN_BE_BOUND.ALL / tma_info_slots", - "MetricGroup": "TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "TOPDOWN_BE_BOUND.ALL / tma_info_core_slots", + "MetricGroup": "Default;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.1", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "Counts the total number of issue slots that were not consumed by the backend due to backend stalls. Note that uops must be available for consumption in order for this event to count. If a uop is not available (IQ is empty), this event will not count. The rest of these subevents count backend stalls, in cycles, due to an outstanding request which is memory bound vs core bound. The subevents are not slot based events and therefore can not be precisely added or subtracted from the Backend_Bound_Aux subevents which are slot based.", "ScaleUnit": "100%" }, { "BriefDescription": "Counts the total number of issue slots that were not consumed by the backend due to backend stalls", + "DefaultMetricgroupName": "TopdownL1", "MetricExpr": "tma_backend_bound", - "MetricGroup": "TopdownL1;tma_L1_group", + "MetricGroup": "Default;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound_aux", "MetricThreshold": "tma_backend_bound_aux > 0.2", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "Counts the total number of issue slots that were not consumed by the backend due to backend stalls. Note that UOPS must be available for consumption in order for this event to count. If a uop is not available (IQ is empty), this event will not count. All of these subevents count backend stalls, in slots, due to a resource limitation. These are not cycle based events and therefore can not be precisely added or subtracted from the Backend_Bound subevents which are cycle based. These subevents are supplementary to Backend_Bound and can be used to analyze results from a resource perspective at allocation.", "ScaleUnit": "100%" }, { "BriefDescription": "Counts the total number of issue slots that were not consumed by the backend because allocation is stalled due to a mispredicted jump or a machine clear", - "MetricExpr": "(tma_info_slots - (TOPDOWN_FE_BOUND.ALL + TOPDOWN_BE_BOUND.ALL + TOPDOWN_RETIRING.ALL)) / tma_info_slots", - "MetricGroup": "TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "(tma_info_core_slots - (TOPDOWN_FE_BOUND.ALL + TOPDOWN_BE_BOUND.ALL + TOPDOWN_RETIRING.ALL)) / tma_info_core_slots", + "MetricGroup": "Default;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "Counts the total number of issue slots that were not consumed by the backend because allocation is stalled due to a mispredicted jump or a machine clear. Only issue slots wasted due to fast nukes such as memory ordering nukes are counted. Other nukes are not accounted for. Counts all issue slots blocked during this recovery window including relevant microcode flows and while uops are not yet available in the instruction queue (IQ). Also includes the issue slots that were consumed by the backend but were thrown away because they were younger than the mispredict or machine clear.", "ScaleUnit": "100%" }, { "BriefDescription": "Counts the number of uops that are not from the microsequencer.", - "MetricExpr": "(TOPDOWN_RETIRING.ALL - UOPS_RETIRED.MS) / tma_info_slots", + "MetricExpr": "(TOPDOWN_RETIRING.ALL - UOPS_RETIRED.MS) / tma_info_core_slots", "MetricGroup": "TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_base", "MetricThreshold": "tma_base > 0.6", + "MetricgroupNoGroup": "TopdownL2", "ScaleUnit": "100%" }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to BACLEARS, which occurs when the Branch Target Buffer (BTB) prediction or lack thereof, was corrected by a later branch predictor in the frontend", - "MetricExpr": "TOPDOWN_FE_BOUND.BRANCH_DETECT / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.BRANCH_DETECT / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_detect", "MetricThreshold": "tma_branch_detect > 0.05", @@ -138,15 +145,16 @@ }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to branch mispredicts.", - "MetricExpr": "TOPDOWN_BAD_SPECULATION.MISPREDICT / tma_info_slots", + "MetricExpr": "TOPDOWN_BAD_SPECULATION.MISPREDICT / tma_info_core_slots", "MetricGroup": "TopdownL2;tma_L2_group;tma_bad_speculation_group", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.05", + "MetricgroupNoGroup": "TopdownL2", "ScaleUnit": "100%" }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to BTCLEARS, which occurs when the Branch Target Buffer (BTB) predicts a taken branch.", - "MetricExpr": "TOPDOWN_FE_BOUND.BRANCH_RESTEER / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.BRANCH_RESTEER / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteer", "MetricThreshold": "tma_branch_resteer > 0.05", @@ -154,7 +162,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to the microcode sequencer (MS).", - "MetricExpr": "TOPDOWN_FE_BOUND.CISC / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.CISC / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_cisc", "MetricThreshold": "tma_cisc > 0.05", @@ -166,11 +174,12 @@ "MetricGroup": "TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1", + "MetricgroupNoGroup": "TopdownL2", "ScaleUnit": "100%" }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to decode stalls.", - "MetricExpr": "TOPDOWN_FE_BOUND.DECODE / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.DECODE / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_decode", "MetricThreshold": "tma_decode > 0.05", @@ -187,7 +196,7 @@ { "BriefDescription": "Counts the number of cycles the core is stalled due to a demand load miss which hit in DRAM or MMIO (Non-DRAM).", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEM_BOUND_STALLS.LOAD_DRAM_HIT / tma_info_clks - MEM_BOUND_STALLS_AT_RET_CORRECTION * MEM_BOUND_STALLS.LOAD_DRAM_HIT / MEM_BOUND_STALLS.LOAD", + "MetricExpr": "MEM_BOUND_STALLS.LOAD_DRAM_HIT / tma_info_core_clks - max((MEM_BOUND_STALLS.LOAD - LD_HEAD.L1_MISS_AT_RET) / tma_info_core_clks, 0) * MEM_BOUND_STALLS.LOAD_DRAM_HIT / MEM_BOUND_STALLS.LOAD", "MetricGroup": "TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1", @@ -195,7 +204,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to a machine clear classified as a fast nuke due to memory ordering, memory disambiguation and memory renaming.", - "MetricExpr": "TOPDOWN_BAD_SPECULATION.FASTNUKE / tma_info_slots", + "MetricExpr": "TOPDOWN_BAD_SPECULATION.FASTNUKE / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_machine_clears_group", "MetricName": "tma_fast_nuke", "MetricThreshold": "tma_fast_nuke > 0.05", @@ -203,18 +212,20 @@ }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to frontend bandwidth restrictions due to decode, predecode, cisc, and other limitations.", - "MetricExpr": "TOPDOWN_FE_BOUND.FRONTEND_BANDWIDTH / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.FRONTEND_BANDWIDTH / tma_info_core_slots", "MetricGroup": "TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_bandwidth", "MetricThreshold": "tma_fetch_bandwidth > 0.1", + "MetricgroupNoGroup": "TopdownL2", "ScaleUnit": "100%" }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to frontend bandwidth restrictions due to decode, predecode, cisc, and other limitations.", - "MetricExpr": "TOPDOWN_FE_BOUND.FRONTEND_LATENCY / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.FRONTEND_LATENCY / tma_info_core_slots", "MetricGroup": "TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.15", + "MetricgroupNoGroup": "TopdownL2", "ScaleUnit": "100%" }, { @@ -227,7 +238,7 @@ }, { "BriefDescription": "Counts the number of floating point divide operations per uop.", - "MetricExpr": "UOPS_RETIRED.FPDIV / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.FPDIV / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_base_group", "MetricName": "tma_fpdiv_uops", "MetricThreshold": "tma_fpdiv_uops > 0.2", @@ -235,226 +246,202 @@ }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to frontend stalls.", - "MetricExpr": "TOPDOWN_FE_BOUND.ALL / tma_info_slots", - "MetricGroup": "TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "TOPDOWN_FE_BOUND.ALL / tma_info_core_slots", + "MetricGroup": "Default;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1;Default", "ScaleUnit": "100%" }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to instruction cache misses.", - "MetricExpr": "TOPDOWN_FE_BOUND.ICACHE / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.ICACHE / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_icache_misses", "MetricThreshold": "tma_icache_misses > 0.05", "ScaleUnit": "100%" }, { - "BriefDescription": "Percentage of total non-speculative loads with a address aliasing block", - "MetricExpr": "100 * LD_BLOCKS.4K_ALIAS / MEM_UOPS_RETIRED.ALL_LOADS", - "MetricName": "tma_info_address_alias_blocks" - }, - { - "BriefDescription": "Ratio of all branches which mispredict", - "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": " ", - "MetricName": "tma_info_branch_mispredict_ratio" - }, - { - "BriefDescription": "Ratio between Mispredicted branches and unknown branches", - "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / BACLEARS.ANY", - "MetricGroup": " ", - "MetricName": "tma_info_branch_mispredict_to_unknown_branch_ratio" - }, - { "BriefDescription": "", "MetricExpr": "CPU_CLK_UNHALTED.CORE", - "MetricGroup": " ", - "MetricName": "tma_info_clks" + "MetricName": "tma_info_core_clks" }, { "BriefDescription": "", "MetricExpr": "CPU_CLK_UNHALTED.CORE_P", - "MetricGroup": " ", - "MetricName": "tma_info_clks_p" + "MetricName": "tma_info_core_clks_p" }, { "BriefDescription": "Cycles Per Instruction", - "MetricExpr": "tma_info_clks / INST_RETIRED.ANY", - "MetricGroup": " ", - "MetricName": "tma_info_cpi" - }, - { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": " ", - "MetricName": "tma_info_cpu_utilization" - }, - { - "BriefDescription": "Cycle cost per DRAM hit", - "MetricExpr": "MEM_BOUND_STALLS.LOAD_DRAM_HIT / MEM_LOAD_UOPS_RETIRED.DRAM_HIT", - "MetricGroup": " ", - "MetricName": "tma_info_cycles_per_demand_load_dram_hit" - }, - { - "BriefDescription": "Cycle cost per L2 hit", - "MetricExpr": "MEM_BOUND_STALLS.LOAD_L2_HIT / MEM_LOAD_UOPS_RETIRED.L2_HIT", - "MetricGroup": " ", - "MetricName": "tma_info_cycles_per_demand_load_l2_hit" + "MetricExpr": "tma_info_core_clks / INST_RETIRED.ANY", + "MetricName": "tma_info_core_cpi" }, { - "BriefDescription": "Cycle cost per LLC hit", - "MetricExpr": "MEM_BOUND_STALLS.LOAD_LLC_HIT / MEM_LOAD_UOPS_RETIRED.L3_HIT", - "MetricGroup": " ", - "MetricName": "tma_info_cycles_per_demand_load_l3_hit" + "BriefDescription": "Instructions Per Cycle", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricName": "tma_info_core_ipc" }, { - "BriefDescription": "Percentage of all uops which are FPDiv uops", - "MetricExpr": "100 * UOPS_RETIRED.FPDIV / UOPS_RETIRED.ALL", - "MetricGroup": " ", - "MetricName": "tma_info_fpdiv_uop_ratio" + "BriefDescription": "", + "MetricExpr": "5 * tma_info_core_clks", + "MetricName": "tma_info_core_slots" }, { - "BriefDescription": "Percentage of all uops which are IDiv uops", - "MetricExpr": "100 * UOPS_RETIRED.IDIV / UOPS_RETIRED.ALL", - "MetricGroup": " ", - "MetricName": "tma_info_idiv_uop_ratio" + "BriefDescription": "Uops Per Instruction", + "MetricExpr": "UOPS_RETIRED.ALL / INST_RETIRED.ANY", + "MetricName": "tma_info_core_upi" }, { "BriefDescription": "Percent of instruction miss cost that hit in DRAM", "MetricExpr": "100 * MEM_BOUND_STALLS.IFETCH_DRAM_HIT / MEM_BOUND_STALLS.IFETCH", - "MetricGroup": " ", - "MetricName": "tma_info_inst_miss_cost_dramhit_percent" + "MetricName": "tma_info_frontend_inst_miss_cost_dramhit_percent" }, { "BriefDescription": "Percent of instruction miss cost that hit in the L2", "MetricExpr": "100 * MEM_BOUND_STALLS.IFETCH_L2_HIT / MEM_BOUND_STALLS.IFETCH", - "MetricGroup": " ", - "MetricName": "tma_info_inst_miss_cost_l2hit_percent" + "MetricName": "tma_info_frontend_inst_miss_cost_l2hit_percent" }, { "BriefDescription": "Percent of instruction miss cost that hit in the L3", "MetricExpr": "100 * MEM_BOUND_STALLS.IFETCH_LLC_HIT / MEM_BOUND_STALLS.IFETCH", - "MetricGroup": " ", - "MetricName": "tma_info_inst_miss_cost_l3hit_percent" + "MetricName": "tma_info_frontend_inst_miss_cost_l3hit_percent" }, { - "BriefDescription": "Instructions per Branch (lower number means higher occurance rate)", - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": " ", - "MetricName": "tma_info_ipbranch" + "BriefDescription": "Ratio of all branches which mispredict", + "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.ALL_BRANCHES", + "MetricName": "tma_info_inst_mix_branch_mispredict_ratio" }, { - "BriefDescription": "Instructions Per Cycle", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": " ", - "MetricName": "tma_info_ipc" + "BriefDescription": "Ratio between Mispredicted branches and unknown branches", + "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / BACLEARS.ANY", + "MetricName": "tma_info_inst_mix_branch_mispredict_to_unknown_branch_ratio" + }, + { + "BriefDescription": "Percentage of all uops which are FPDiv uops", + "MetricExpr": "100 * UOPS_RETIRED.FPDIV / UOPS_RETIRED.ALL", + "MetricName": "tma_info_inst_mix_fpdiv_uop_ratio" + }, + { + "BriefDescription": "Percentage of all uops which are IDiv uops", + "MetricExpr": "100 * UOPS_RETIRED.IDIV / UOPS_RETIRED.ALL", + "MetricName": "tma_info_inst_mix_idiv_uop_ratio" + }, + { + "BriefDescription": "Instructions per Branch (lower number means higher occurance rate)", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", + "MetricName": "tma_info_inst_mix_ipbranch" }, { "BriefDescription": "Instruction per (near) call (lower number means higher occurance rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.CALL", - "MetricGroup": " ", - "MetricName": "tma_info_ipcall" + "MetricName": "tma_info_inst_mix_ipcall" }, { "BriefDescription": "Instructions per Far Branch", "MetricExpr": "INST_RETIRED.ANY / (BR_INST_RETIRED.FAR_BRANCH / 2)", - "MetricGroup": " ", - "MetricName": "tma_info_ipfarbranch" + "MetricName": "tma_info_inst_mix_ipfarbranch" }, { "BriefDescription": "Instructions per Load", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_LOADS", - "MetricGroup": " ", - "MetricName": "tma_info_ipload" + "MetricName": "tma_info_inst_mix_ipload" }, { "BriefDescription": "Instructions per retired conditional Branch Misprediction where the branch was not taken", "MetricExpr": "INST_RETIRED.ANY / (BR_MISP_RETIRED.COND - BR_MISP_RETIRED.COND_TAKEN)", - "MetricName": "tma_info_ipmisp_cond_ntaken" + "MetricName": "tma_info_inst_mix_ipmisp_cond_ntaken" }, { "BriefDescription": "Instructions per retired conditional Branch Misprediction where the branch was taken", "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_TAKEN", - "MetricName": "tma_info_ipmisp_cond_taken" + "MetricName": "tma_info_inst_mix_ipmisp_cond_taken" }, { "BriefDescription": "Instructions per retired indirect call or jump Branch Misprediction", "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.INDIRECT", - "MetricName": "tma_info_ipmisp_indirect" + "MetricName": "tma_info_inst_mix_ipmisp_indirect" }, { "BriefDescription": "Instructions per retired return Branch Misprediction", "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.RETURN", - "MetricName": "tma_info_ipmisp_ret" + "MetricName": "tma_info_inst_mix_ipmisp_ret" }, { "BriefDescription": "Instructions per retired Branch Misprediction", "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": " ", - "MetricName": "tma_info_ipmispredict" + "MetricName": "tma_info_inst_mix_ipmispredict" }, { "BriefDescription": "Instructions per Store", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_STORES", - "MetricGroup": " ", - "MetricName": "tma_info_ipstore" + "MetricName": "tma_info_inst_mix_ipstore" }, { - "BriefDescription": "Fraction of cycles spent in Kernel mode", - "MetricExpr": "cpu@CPU_CLK_UNHALTED.CORE@k / CPU_CLK_UNHALTED.CORE", - "MetricGroup": " ", - "MetricName": "tma_info_kernel_utilization" + "BriefDescription": "Percentage of all uops which are ucode ops", + "MetricExpr": "100 * UOPS_RETIRED.MS / UOPS_RETIRED.ALL", + "MetricName": "tma_info_inst_mix_microcode_uop_ratio" + }, + { + "BriefDescription": "Percentage of all uops which are x87 uops", + "MetricExpr": "100 * UOPS_RETIRED.X87 / UOPS_RETIRED.ALL", + "MetricName": "tma_info_inst_mix_x87_uop_ratio" + }, + { + "BriefDescription": "Percentage of total non-speculative loads with a address aliasing block", + "MetricExpr": "100 * LD_BLOCKS.4K_ALIAS / MEM_UOPS_RETIRED.ALL_LOADS", + "MetricName": "tma_info_l1_bound_address_alias_blocks" }, { "BriefDescription": "Percentage of total non-speculative loads that are splits", "MetricExpr": "100 * MEM_UOPS_RETIRED.SPLIT_LOADS / MEM_UOPS_RETIRED.ALL_LOADS", - "MetricName": "tma_info_load_splits" + "MetricName": "tma_info_l1_bound_load_splits" }, { - "BriefDescription": "load ops retired per 1000 instruction", - "MetricExpr": "1e3 * MEM_UOPS_RETIRED.ALL_LOADS / INST_RETIRED.ANY", - "MetricGroup": " ", - "MetricName": "tma_info_memloadpki" + "BriefDescription": "Percentage of total non-speculative loads with a store forward or unknown store address block", + "MetricExpr": "100 * LD_BLOCKS.DATA_UNKNOWN / MEM_UOPS_RETIRED.ALL_LOADS", + "MetricName": "tma_info_l1_bound_store_fwd_blocks" }, { - "BriefDescription": "Percentage of all uops which are ucode ops", - "MetricExpr": "100 * UOPS_RETIRED.MS / UOPS_RETIRED.ALL", - "MetricGroup": " ", - "MetricName": "tma_info_microcode_uop_ratio" + "BriefDescription": "Cycle cost per DRAM hit", + "MetricExpr": "MEM_BOUND_STALLS.LOAD_DRAM_HIT / MEM_LOAD_UOPS_RETIRED.DRAM_HIT", + "MetricName": "tma_info_memory_cycles_per_demand_load_dram_hit" }, { - "BriefDescription": "", - "MetricExpr": "5 * tma_info_clks", - "MetricGroup": " ", - "MetricName": "tma_info_slots" + "BriefDescription": "Cycle cost per L2 hit", + "MetricExpr": "MEM_BOUND_STALLS.LOAD_L2_HIT / MEM_LOAD_UOPS_RETIRED.L2_HIT", + "MetricName": "tma_info_memory_cycles_per_demand_load_l2_hit" }, { - "BriefDescription": "Percentage of total non-speculative loads with a store forward or unknown store address block", - "MetricExpr": "100 * LD_BLOCKS.DATA_UNKNOWN / MEM_UOPS_RETIRED.ALL_LOADS", - "MetricName": "tma_info_store_fwd_blocks" + "BriefDescription": "Cycle cost per LLC hit", + "MetricExpr": "MEM_BOUND_STALLS.LOAD_LLC_HIT / MEM_LOAD_UOPS_RETIRED.L3_HIT", + "MetricName": "tma_info_memory_cycles_per_demand_load_l3_hit" }, { - "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", - "MetricGroup": " ", - "MetricName": "tma_info_turbo_utilization" + "BriefDescription": "load ops retired per 1000 instruction", + "MetricExpr": "1e3 * MEM_UOPS_RETIRED.ALL_LOADS / INST_RETIRED.ANY", + "MetricName": "tma_info_memory_memloadpki" }, { - "BriefDescription": "Uops Per Instruction", - "MetricExpr": "UOPS_RETIRED.ALL / INST_RETIRED.ANY", - "MetricGroup": " ", - "MetricName": "tma_info_upi" + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricName": "tma_info_system_cpu_utilization" }, { - "BriefDescription": "Percentage of all uops which are x87 uops", - "MetricExpr": "100 * UOPS_RETIRED.X87 / UOPS_RETIRED.ALL", - "MetricGroup": " ", - "MetricName": "tma_info_x87_uop_ratio" + "BriefDescription": "Fraction of cycles spent in Kernel mode", + "MetricExpr": "cpu@CPU_CLK_UNHALTED.CORE@k / CPU_CLK_UNHALTED.CORE", + "MetricGroup": "Summary", + "MetricName": "tma_info_system_kernel_utilization" + }, + { + "BriefDescription": "Average Frequency Utilization relative nominal frequency", + "MetricExpr": "tma_info_core_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricGroup": "Power", + "MetricName": "tma_info_system_turbo_utilization" }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to Instruction Table Lookaside Buffer (ITLB) misses.", - "MetricExpr": "TOPDOWN_FE_BOUND.ITLB / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.ITLB / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05", @@ -462,7 +449,7 @@ }, { "BriefDescription": "Counts the number of cycles that the oldest load of the load buffer is stalled at retirement due to a load block.", - "MetricExpr": "LD_HEAD.L1_BOUND_AT_RET / tma_info_clks", + "MetricExpr": "LD_HEAD.L1_BOUND_AT_RET / tma_info_core_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l1_bound", "MetricThreshold": "tma_l1_bound > 0.1", @@ -471,7 +458,7 @@ { "BriefDescription": "Counts the number of cycles a core is stalled due to a demand load which hit in the L2 Cache.", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEM_BOUND_STALLS.LOAD_L2_HIT / tma_info_clks - MEM_BOUND_STALLS_AT_RET_CORRECTION * MEM_BOUND_STALLS.LOAD_L2_HIT / MEM_BOUND_STALLS.LOAD", + "MetricExpr": "MEM_BOUND_STALLS.LOAD_L2_HIT / tma_info_core_clks - max((MEM_BOUND_STALLS.LOAD - LD_HEAD.L1_MISS_AT_RET) / tma_info_core_clks, 0) * MEM_BOUND_STALLS.LOAD_L2_HIT / MEM_BOUND_STALLS.LOAD", "MetricGroup": "TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l2_bound", "MetricThreshold": "tma_l2_bound > 0.1", @@ -479,7 +466,7 @@ }, { "BriefDescription": "Counts the number of cycles a core is stalled due to a demand load which hit in the Last Level Cache (LLC) or other core with HITE/F/M.", - "MetricExpr": "MEM_BOUND_STALLS.LOAD_LLC_HIT / tma_info_clks - MEM_BOUND_STALLS_AT_RET_CORRECTION * MEM_BOUND_STALLS.LOAD_LLC_HIT / MEM_BOUND_STALLS.LOAD", + "MetricExpr": "MEM_BOUND_STALLS.LOAD_LLC_HIT / tma_info_core_clks - max((MEM_BOUND_STALLS.LOAD - LD_HEAD.L1_MISS_AT_RET) / tma_info_core_clks, 0) * MEM_BOUND_STALLS.LOAD_LLC_HIT / MEM_BOUND_STALLS.LOAD", "MetricGroup": "TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.1", @@ -495,15 +482,16 @@ }, { "BriefDescription": "Counts the total number of issue slots that were not consumed by the backend because allocation is stalled due to a machine clear (nuke) of any kind including memory ordering and memory disambiguation.", - "MetricExpr": "TOPDOWN_BAD_SPECULATION.MACHINE_CLEARS / tma_info_slots", + "MetricExpr": "TOPDOWN_BAD_SPECULATION.MACHINE_CLEARS / tma_info_core_slots", "MetricGroup": "TopdownL2;tma_L2_group;tma_bad_speculation_group", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.05", + "MetricgroupNoGroup": "TopdownL2", "ScaleUnit": "100%" }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to memory reservation stalls in which a scheduler is not able to accept uops.", - "MetricExpr": "TOPDOWN_BE_BOUND.MEM_SCHEDULER / tma_info_slots", + "MetricExpr": "TOPDOWN_BE_BOUND.MEM_SCHEDULER / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_resource_bound_group", "MetricName": "tma_mem_scheduler", "MetricThreshold": "tma_mem_scheduler > 0.1", @@ -511,10 +499,11 @@ }, { "BriefDescription": "Counts the number of cycles the core is stalled due to stores or loads.", - "MetricExpr": "min(tma_backend_bound, LD_HEAD.ANY_AT_RET / tma_info_clks + tma_store_bound)", + "MetricExpr": "min(tma_backend_bound, LD_HEAD.ANY_AT_RET / tma_info_core_clks + tma_store_bound)", "MetricGroup": "TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "ScaleUnit": "100%" }, { @@ -527,16 +516,17 @@ }, { "BriefDescription": "Counts the number of uops that are from the complex flows issued by the micro-sequencer (MS)", - "MetricExpr": "UOPS_RETIRED.MS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.MS / tma_info_core_slots", "MetricGroup": "TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_ms_uops", "MetricThreshold": "tma_ms_uops > 0.05", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "Counts the number of uops that are from the complex flows issued by the micro-sequencer (MS). This includes uops from flows due to complex instructions, faults, assists, and inserted flows.", "ScaleUnit": "100%" }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to IEC or FPC RAT stalls, which can be due to FIQ or IEC reservation stalls in which the integer, floating point or SIMD scheduler is not able to accept uops.", - "MetricExpr": "TOPDOWN_BE_BOUND.NON_MEM_SCHEDULER / tma_info_slots", + "MetricExpr": "TOPDOWN_BE_BOUND.NON_MEM_SCHEDULER / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_resource_bound_group", "MetricName": "tma_non_mem_scheduler", "MetricThreshold": "tma_non_mem_scheduler > 0.1", @@ -544,7 +534,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to a machine clear (slow nuke).", - "MetricExpr": "TOPDOWN_BAD_SPECULATION.NUKE / tma_info_slots", + "MetricExpr": "TOPDOWN_BAD_SPECULATION.NUKE / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_machine_clears_group", "MetricName": "tma_nuke", "MetricThreshold": "tma_nuke > 0.05", @@ -552,7 +542,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to other common frontend stalls not categorized.", - "MetricExpr": "TOPDOWN_FE_BOUND.OTHER / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.OTHER / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_other_fb", "MetricThreshold": "tma_other_fb > 0.05", @@ -560,7 +550,7 @@ }, { "BriefDescription": "Counts the number of cycles that the oldest load of the load buffer is stalled at retirement due to a number of other load blocks.", - "MetricExpr": "LD_HEAD.OTHER_AT_RET / tma_info_clks", + "MetricExpr": "LD_HEAD.OTHER_AT_RET / tma_info_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_other_l1", "MetricThreshold": "tma_other_l1 > 0.05", @@ -576,7 +566,7 @@ }, { "BriefDescription": "Counts the number of uops retired excluding ms and fp div uops.", - "MetricExpr": "(TOPDOWN_RETIRING.ALL - UOPS_RETIRED.MS - UOPS_RETIRED.FPDIV) / tma_info_slots", + "MetricExpr": "(TOPDOWN_RETIRING.ALL - UOPS_RETIRED.MS - UOPS_RETIRED.FPDIV) / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_base_group", "MetricName": "tma_other_ret", "MetricThreshold": "tma_other_ret > 0.3", @@ -592,7 +582,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not delivered by the frontend due to wrong predecodes.", - "MetricExpr": "TOPDOWN_FE_BOUND.PREDECODE / tma_info_slots", + "MetricExpr": "TOPDOWN_FE_BOUND.PREDECODE / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_predecode", "MetricThreshold": "tma_predecode > 0.05", @@ -600,7 +590,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to the physical register file unable to accept an entry (marble stalls).", - "MetricExpr": "TOPDOWN_BE_BOUND.REGISTER / tma_info_slots", + "MetricExpr": "TOPDOWN_BE_BOUND.REGISTER / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_resource_bound_group", "MetricName": "tma_register", "MetricThreshold": "tma_register > 0.1", @@ -608,7 +598,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to the reorder buffer being full (ROB stalls).", - "MetricExpr": "TOPDOWN_BE_BOUND.REORDER_BUFFER / tma_info_slots", + "MetricExpr": "TOPDOWN_BE_BOUND.REORDER_BUFFER / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_resource_bound_group", "MetricName": "tma_reorder_buffer", "MetricThreshold": "tma_reorder_buffer > 0.1", @@ -620,15 +610,18 @@ "MetricGroup": "TopdownL2;tma_L2_group;tma_backend_bound_aux_group", "MetricName": "tma_resource_bound", "MetricThreshold": "tma_resource_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "Counts the total number of issue slots that were not consumed by the backend due to backend stalls. Note that uops must be available for consumption in order for this event to count. If a uop is not available (IQ is empty), this event will not count.", "ScaleUnit": "100%" }, { "BriefDescription": "Counts the numer of issue slots that result in retirement slots.", - "MetricExpr": "TOPDOWN_RETIRING.ALL / tma_info_slots", - "MetricGroup": "TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "TOPDOWN_RETIRING.ALL / tma_info_core_slots", + "MetricGroup": "Default;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.75", + "MetricgroupNoGroup": "TopdownL1;Default", "ScaleUnit": "100%" }, { @@ -641,7 +634,7 @@ }, { "BriefDescription": "Counts the number of issue slots that were not consumed by the backend due to scoreboards from the instruction queue (IQ), jump execution unit (JEU), or microcode sequencer (MS).", - "MetricExpr": "TOPDOWN_BE_BOUND.SERIALIZATION / tma_info_slots", + "MetricExpr": "TOPDOWN_BE_BOUND.SERIALIZATION / tma_info_core_slots", "MetricGroup": "TopdownL3;tma_L3_group;tma_resource_bound_group", "MetricName": "tma_serialization", "MetricThreshold": "tma_serialization > 0.1", @@ -665,7 +658,7 @@ }, { "BriefDescription": "Counts the number of cycles that the oldest load of the load buffer is stalled at retirement due to a first level TLB miss.", - "MetricExpr": "LD_HEAD.DTLB_MISS_AT_RET / tma_info_clks", + "MetricExpr": "LD_HEAD.DTLB_MISS_AT_RET / tma_info_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_stlb_hit", "MetricThreshold": "tma_stlb_hit > 0.05", @@ -673,7 +666,7 @@ }, { "BriefDescription": "Counts the number of cycles that the oldest load of the load buffer is stalled at retirement due to a second level TLB miss requiring a page walk.", - "MetricExpr": "LD_HEAD.PGWALK_AT_RET / tma_info_clks", + "MetricExpr": "LD_HEAD.PGWALK_AT_RET / tma_info_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_stlb_miss", "MetricThreshold": "tma_stlb_miss > 0.05", @@ -689,7 +682,7 @@ }, { "BriefDescription": "Counts the number of cycles that the oldest load of the load buffer is stalled at retirement due to a store forward block.", - "MetricExpr": "LD_HEAD.ST_ADDR_AT_RET / tma_info_clks", + "MetricExpr": "LD_HEAD.ST_ADDR_AT_RET / tma_info_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_store_fwd_blk", "MetricThreshold": "tma_store_fwd_blk > 0.05", diff --git a/tools/perf/pmu-events/arch/x86/alderlaken/metricgroups.json b/tools/perf/pmu-events/arch/x86/alderlaken/metricgroups.json new file mode 100644 index 000000000000..7b2049cd2694 --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/alderlaken/metricgroups.json @@ -0,0 +1,26 @@ +{ + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_backend_bound_aux_group": "Metrics contributing to tma_backend_bound_aux category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_base_group": "Metrics contributing to tma_base category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_machine_clears_group": "Metrics contributing to tma_machine_clears category", + "tma_mem_scheduler_group": "Metrics contributing to tma_mem_scheduler category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_nuke_group": "Metrics contributing to tma_nuke category", + "tma_resource_bound_group": "Metrics contributing to tma_resource_bound category", + "tma_retiring_group": "Metrics contributing to tma_retiring category" +} diff --git a/tools/perf/pmu-events/arch/x86/broadwell/bdw-metrics.json b/tools/perf/pmu-events/arch/x86/broadwell/bdw-metrics.json index 51cf8560a8d3..55a10b0bf36f 100644 --- a/tools/perf/pmu-events/arch/x86/broadwell/bdw-metrics.json +++ b/tools/perf/pmu-events/arch/x86/broadwell/bdw-metrics.json @@ -50,7 +50,7 @@ }, { "BriefDescription": "Uncore frequency per die [GHZ]", - "MetricExpr": "tma_info_socket_clks / #num_dies / duration_time / 1e9", + "MetricExpr": "tma_info_system_socket_clks / #num_dies / duration_time / 1e9", "MetricGroup": "SoC", "MetricName": "UNCORE_FREQ" }, @@ -71,7 +71,7 @@ }, { "BriefDescription": "This metric estimates how often memory load accesses were aliased by preceding stores (in program order) with a 4K address offset", - "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_clks", + "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_4k_aliasing", "MetricThreshold": "tma_4k_aliasing > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -81,7 +81,7 @@ { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution ports for ALU operations.", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5 + UOPS_DISPATCHED_PORT.PORT_6) / tma_info_slots", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5 + UOPS_DISPATCHED_PORT.PORT_6) / tma_info_thread_slots", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_alu_op_utilization", "MetricThreshold": "tma_alu_op_utilization > 0.6", @@ -89,7 +89,7 @@ }, { "BriefDescription": "This metric estimates fraction of slots the CPU retired uops delivered by the Microcode_Sequencer as a result of Assists", - "MetricExpr": "100 * OTHER_ASSISTS.ANY_WB_ASSIST / tma_info_slots", + "MetricExpr": "100 * OTHER_ASSISTS.ANY_WB_ASSIST / tma_info_thread_slots", "MetricGroup": "TopdownL4;tma_L4_group;tma_microcode_sequencer_group", "MetricName": "tma_assists", "MetricThreshold": "tma_assists > 0.1 & (tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1)", @@ -103,15 +103,17 @@ "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend. Backend is the portion of the processor core where the out-of-order scheduler dispatches ready uops into their respective execution units; and once completed these uops get retired according to program order. For example; stalls due to data-cache misses or stalls due to the divider unit being overloaded are both categorized under Backend Bound. Backend Bound is further divided into two main categories: Memory Bound and Core Bound.", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations", - "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_slots", + "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots wasted due to incorrect speculations. This include slots used to issue uops that do not eventually get retired and slots for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work due to miss-predicted branches are categorized under Bad Speculation category. Incorrect data speculation followed by Memory Ordering Nukes is another example.", "ScaleUnit": "100%" }, @@ -122,12 +124,13 @@ "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15", - "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_branch_misprediction_cost, tma_mispredicts_resteers", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers", - "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_clks", + "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteers", "MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -156,7 +159,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to contested accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(60 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) + 43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS)))) / tma_info_clks", + "MetricExpr": "(60 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) + 43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS)))) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_contested_accesses", "MetricThreshold": "tma_contested_accesses > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -170,13 +173,14 @@ "MetricGroup": "Backend;Compute;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck. Shortage in hardware compute resources; or dependencies in software's instructions are both categorized under Core Bound. Hence it may indicate the machine ran out of an out-of-order resource; certain execution units are overloaded or dependencies in program's data- or instruction-flow are limiting the performance (e.g. FP-chained long-latency arithmetic operations).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to data-sharing accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) / tma_info_clks", + "MetricExpr": "43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) / tma_info_thread_clks", "MetricGroup": "Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_data_sharing", "MetricThreshold": "tma_data_sharing > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -185,7 +189,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the Divider unit was active", - "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_clks", + "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_core_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_divider", "MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -195,7 +199,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS)) * CYCLE_ACTIVITY.STALLS_L2_MISS / tma_info_clks", + "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS)) * CYCLE_ACTIVITY.STALLS_L2_MISS / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -204,25 +208,25 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline", - "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSB;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_dsb", - "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline. For example; inefficient utilization of the DSB cache structure or bank conflict when reading from it; are categorized here.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_dsb_switches", "MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb, tma_lcp", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses", - "MetricExpr": "(8 * DTLB_LOAD_MISSES.STLB_HIT + cpu@DTLB_LOAD_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * DTLB_LOAD_MISSES.WALK_COMPLETED) / tma_info_clks", + "MetricExpr": "(8 * DTLB_LOAD_MISSES.STLB_HIT + cpu@DTLB_LOAD_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * DTLB_LOAD_MISSES.WALK_COMPLETED) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group", "MetricName": "tma_dtlb_load", "MetricThreshold": "tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -231,7 +235,7 @@ }, { "BriefDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses", - "MetricExpr": "(8 * DTLB_STORE_MISSES.STLB_HIT + cpu@DTLB_STORE_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * DTLB_STORE_MISSES.WALK_COMPLETED) / tma_info_clks", + "MetricExpr": "(8 * DTLB_STORE_MISSES.STLB_HIT + cpu@DTLB_STORE_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * DTLB_STORE_MISSES.WALK_COMPLETED) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_store_bound_group", "MetricName": "tma_dtlb_store", "MetricThreshold": "tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -240,7 +244,7 @@ }, { "BriefDescription": "This metric roughly estimates how often CPU was handling synchronizations due to False Sharing", - "MetricExpr": "60 * OFFCORE_RESPONSE.DEMAND_RFO.L3_HIT.SNOOP_HITM / tma_info_clks", + "MetricExpr": "60 * OFFCORE_RESPONSE.DEMAND_RFO.L3_HIT.SNOOP_HITM / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_store_bound_group", "MetricName": "tma_false_sharing", "MetricThreshold": "tma_false_sharing > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -250,11 +254,11 @@ { "BriefDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "tma_info_load_miss_real_latency * cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@ / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@ / tma_info_thread_clks", "MetricGroup": "MemoryBW;TopdownL4;tma_L4_group;tma_issueBW;tma_issueSL;tma_issueSmSt;tma_l1_bound_group", "MetricName": "tma_fb_full", "MetricThreshold": "tma_fb_full > 0.3", - "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", + "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", "ScaleUnit": "100%" }, { @@ -262,16 +266,18 @@ "MetricExpr": "tma_frontend_bound - tma_fetch_latency", "MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", "MetricName": "tma_fetch_bandwidth", - "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_info_iptb, tma_lcp", + "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues", - "MetricExpr": "4 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE / tma_info_slots", + "MetricExpr": "4 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE / tma_info_thread_slots", "MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: RS_EVENTS.EMPTY_END", "ScaleUnit": "100%" }, @@ -322,10 +328,11 @@ }, { "BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend", - "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_slots", + "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_thread_slots", "MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound.", "ScaleUnit": "100%" }, @@ -335,440 +342,441 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_heavy_operations", "MetricThreshold": "tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to instruction cache misses.", - "MetricExpr": "ICACHE.IFDATA_STALL / tma_info_clks", + "MetricExpr": "ICACHE.IFDATA_STALL / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;IcMiss;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_icache_misses", "MetricThreshold": "tma_icache_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", "ScaleUnit": "100%" }, { - "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", - "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time", - "MetricGroup": "Power;Summary", - "MetricName": "tma_info_average_frequency" - }, - { - "BriefDescription": "Branch instructions per taken branch.", - "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", - "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_bptkbranch" - }, - { "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", - "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_slots / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_thread_slots / BR_MISP_RETIRED.ALL_BRANCHES", "MetricGroup": "Bad;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_branch_misprediction_cost", + "MetricName": "tma_info_bad_spec_branch_misprediction_cost", "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_mispredicts_resteers" }, { - "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "Pipeline", - "MetricName": "tma_info_clks" + "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", + "MetricExpr": "tma_info_inst_mix_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_indirect", + "MetricThreshold": "tma_info_bad_spec_ipmisp_indirect < 1e3" + }, + { + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BadSpec;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmispredict", + "MetricThreshold": "tma_info_bad_spec_ipmispredict < 200" }, { "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", - "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_clks))", + "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_thread_clks))", "MetricGroup": "SMT", - "MetricName": "tma_info_core_clks" + "MetricName": "tma_info_core_core_clks" }, { "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks", "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_coreipc" - }, - { - "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricExpr": "1 / tma_info_ipc", - "MetricGroup": "Mem;Pipeline", - "MetricName": "tma_info_cpi" + "MetricName": "tma_info_core_coreipc" }, { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": "HPC;Summary", - "MetricName": "tma_info_cpu_utilization" + "BriefDescription": "Floating Point Operations Per Cycle", + "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / tma_info_core_core_clks", + "MetricGroup": "Flops;Ret", + "MetricName": "tma_info_core_flopc" }, { - "BriefDescription": "Average Parallel L2 cache miss data reads", - "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", - "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_data_l2_mlp" + "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", + "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@) / (2 * tma_info_core_core_clks)", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_core_fp_arith_utilization", + "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." }, { - "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricExpr": "64 * (UNC_ARB_TRK_REQUESTS.ALL + UNC_ARB_COH_TRK_REQUESTS.ALL) / 1e6 / duration_time / 1e3", - "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", - "MetricName": "tma_info_dram_bw_use", - "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", + "MetricExpr": "UOPS_EXECUTED.THREAD / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC)", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", + "MetricName": "tma_info_core_ilp" }, { "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", "MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + LSD.UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)", "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", - "MetricName": "tma_info_dsb_coverage", - "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_iptb, tma_lcp" - }, - { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", - "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", - "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", - "MetricName": "tma_info_execute" + "MetricName": "tma_info_frontend_dsb_coverage", + "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 4 > 0.35", + "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_inst_mix_iptb, tma_lcp" }, { - "BriefDescription": "The ratio of Executed- by Issued-Uops", - "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", - "MetricGroup": "Cor;Pipeline", - "MetricName": "tma_info_execute_per_issue", - "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." - }, - { - "BriefDescription": "Floating Point Operations Per Cycle", - "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / tma_info_core_clks", - "MetricGroup": "Flops;Ret", - "MetricName": "tma_info_flopc" - }, - { - "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", - "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@) / (2 * tma_info_core_clks)", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_fp_arith_utilization", - "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." - }, - { - "BriefDescription": "Giga Floating Point Operations Per Second", - "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / 1e9 / duration_time", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_gflops", - "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_inst_mix_instructions / BACLEARS.ANY", + "MetricGroup": "Fed", + "MetricName": "tma_info_frontend_ipunknown_branch" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", - "MetricExpr": "UOPS_EXECUTED.THREAD / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC)", - "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", - "MetricName": "tma_info_ilp" + "BriefDescription": "Branch instructions per taken branch.", + "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;PGO", + "MetricName": "tma_info_inst_mix_bptkbranch" }, { "BriefDescription": "Total number of retired Instructions", "MetricExpr": "INST_RETIRED.ANY", "MetricGroup": "Summary;TmaL1;tma_L1_group", - "MetricName": "tma_info_instructions", + "MetricName": "tma_info_inst_mix_instructions", "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" }, { "BriefDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_iparith", - "MetricThreshold": "tma_info_iparith < 10", + "MetricName": "tma_info_inst_mix_iparith", + "MetricThreshold": "tma_info_inst_mix_iparith < 10", "PublicDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate). May undercount due to FMA double counting. Approximated prior to BDW." }, { "BriefDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx128", - "MetricThreshold": "tma_info_iparith_avx128 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx128", + "MetricThreshold": "tma_info_inst_mix_iparith_avx128 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx256", - "MetricThreshold": "tma_info_iparith_avx256 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx256", + "MetricThreshold": "tma_info_inst_mix_iparith_avx256 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_dp", - "MetricThreshold": "tma_info_iparith_scalar_dp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_dp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_dp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_SINGLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_sp", - "MetricThreshold": "tma_info_iparith_scalar_sp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_sp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_sp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per Branch (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Branches;Fed;InsType", - "MetricName": "tma_info_ipbranch", - "MetricThreshold": "tma_info_ipbranch < 8" - }, - { - "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": "Ret;Summary", - "MetricName": "tma_info_ipc" + "MetricName": "tma_info_inst_mix_ipbranch", + "MetricThreshold": "tma_info_inst_mix_ipbranch < 8" }, { "BriefDescription": "Instructions per (near) call (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_CALL", "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_ipcall", - "MetricThreshold": "tma_info_ipcall < 200" - }, - { - "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", - "MetricGroup": "Branches;OS", - "MetricName": "tma_info_ipfarbranch", - "MetricThreshold": "tma_info_ipfarbranch < 1e6" + "MetricName": "tma_info_inst_mix_ipcall", + "MetricThreshold": "tma_info_inst_mix_ipcall < 200" }, { "BriefDescription": "Instructions per Floating Point (FP) Operation (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_ipflop", - "MetricThreshold": "tma_info_ipflop < 10" + "MetricName": "tma_info_inst_mix_ipflop", + "MetricThreshold": "tma_info_inst_mix_ipflop < 10" }, { "BriefDescription": "Instructions per Load (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_LOADS", "MetricGroup": "InsType", - "MetricName": "tma_info_ipload", - "MetricThreshold": "tma_info_ipload < 3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", - "MetricExpr": "tma_info_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_indirect", - "MetricThreshold": "tma_info_ipmisp_indirect < 1e3" - }, - { - "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BadSpec;BrMispredicts", - "MetricName": "tma_info_ipmispredict", - "MetricThreshold": "tma_info_ipmispredict < 200" + "MetricName": "tma_info_inst_mix_ipload", + "MetricThreshold": "tma_info_inst_mix_ipload < 3" }, { "BriefDescription": "Instructions per Store (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_STORES", "MetricGroup": "InsType", - "MetricName": "tma_info_ipstore", - "MetricThreshold": "tma_info_ipstore < 8" + "MetricName": "tma_info_inst_mix_ipstore", + "MetricThreshold": "tma_info_inst_mix_ipstore < 8" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW;Frontend;PGO;tma_issueFB", - "MetricName": "tma_info_iptb", - "MetricThreshold": "tma_info_iptb < 9", - "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_lcp" - }, - { - "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", - "MetricExpr": "tma_info_instructions / BACLEARS.ANY", - "MetricGroup": "Fed", - "MetricName": "tma_info_ipunknown_branch" - }, - { - "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_cpi" - }, - { - "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_utilization", - "MetricThreshold": "tma_info_kernel_utilization > 0.05" + "MetricName": "tma_info_inst_mix_iptb", + "MetricThreshold": "tma_info_inst_mix_iptb < 9", + "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_lcp" }, { "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw" - }, - { - "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "tma_info_l1d_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw_1t" - }, - { - "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L1_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki" + "MetricName": "tma_info_memory_core_l1d_cache_fill_bw" }, { "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw" + "MetricName": "tma_info_memory_core_l2_cache_fill_bw" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "tma_info_l2_cache_fill_bw", + "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw_1t" + "MetricName": "tma_info_memory_core_l3_cache_fill_bw" + }, + { + "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L1_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l1mpki" }, { "BriefDescription": "L2 cache hits per kilo instruction for all request types (including speculative)", "MetricExpr": "1e3 * (L2_RQSTS.REFERENCES - L2_RQSTS.MISS) / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_all" + "MetricName": "tma_info_memory_l2hpki_all" }, { "BriefDescription": "L2 cache hits per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_HIT / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_load" + "MetricName": "tma_info_memory_l2hpki_load" }, { "BriefDescription": "L2 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L2_MISS / INST_RETIRED.ANY", "MetricGroup": "Backend;CacheMisses;Mem", - "MetricName": "tma_info_l2mpki" + "MetricName": "tma_info_memory_l2mpki" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all request types (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem;Offcore", - "MetricName": "tma_info_l2mpki_all" + "MetricName": "tma_info_memory_l2mpki_all" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2mpki_load" + "MetricName": "tma_info_memory_l2mpki_load" }, { - "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "0", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw_1t" + "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L3_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l3mpki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw" + "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_UOPS_RETIRED.L1_MISS + MEM_LOAD_UOPS_RETIRED.HIT_LFB)", + "MetricGroup": "Mem;MemoryBound;MemoryLat", + "MetricName": "tma_info_memory_load_miss_real_latency" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw_1t" + "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", + "MetricGroup": "Mem;MemoryBW;MemoryBound", + "MetricName": "tma_info_memory_mlp", + "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" }, { - "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L3_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l3mpki" + "BriefDescription": "Average Parallel L2 cache miss data reads", + "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", + "MetricGroup": "Memory_BW;Offcore", + "MetricName": "tma_info_memory_oro_data_l2_mlp" }, { "BriefDescription": "Average Latency for L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS.DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l2_miss_latency" + "MetricName": "tma_info_memory_oro_load_l2_miss_latency" }, { "BriefDescription": "Average Parallel L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_DATA_RD", "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_load_l2_mlp" + "MetricName": "tma_info_memory_oro_load_l2_mlp" }, { - "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_UOPS_RETIRED.L1_MISS + MEM_LOAD_UOPS_RETIRED.HIT_LFB)", - "MetricGroup": "Mem;MemoryBound;MemoryLat", - "MetricName": "tma_info_load_miss_real_latency" + "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l1d_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l1d_cache_fill_bw_1t" }, { - "BriefDescription": "Average number of parallel requests to external memory", - "MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_OCCUPANCY.CYCLES_WITH_ANY_REQUEST", - "MetricGroup": "Mem;SoC", - "MetricName": "tma_info_mem_parallel_requests", - "PublicDescription": "Average number of parallel requests to external memory. Accounts for all requests" + "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l2_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l2_cache_fill_bw_1t" }, { - "BriefDescription": "Average latency of all requests to external memory (in Uncore cycles)", - "MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_REQUESTS.ALL", - "MetricGroup": "Mem;SoC", - "MetricName": "tma_info_mem_request_latency" + "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "0", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_thread_l3_cache_access_bw_1t" }, { - "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", - "MetricGroup": "Mem;MemoryBW;MemoryBound", - "MetricName": "tma_info_mlp", - "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" + "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l3_cache_fill_bw_1t" }, { "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", - "MetricExpr": "(cpu@ITLB_MISSES.WALK_DURATION\\,cmask\\=1@ + cpu@DTLB_LOAD_MISSES.WALK_DURATION\\,cmask\\=1@ + cpu@DTLB_STORE_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * (DTLB_STORE_MISSES.WALK_COMPLETED + DTLB_LOAD_MISSES.WALK_COMPLETED + ITLB_MISSES.WALK_COMPLETED)) / tma_info_core_clks", + "MetricExpr": "(cpu@ITLB_MISSES.WALK_DURATION\\,cmask\\=1@ + cpu@DTLB_LOAD_MISSES.WALK_DURATION\\,cmask\\=1@ + cpu@DTLB_STORE_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * (DTLB_STORE_MISSES.WALK_COMPLETED + DTLB_LOAD_MISSES.WALK_COMPLETED + ITLB_MISSES.WALK_COMPLETED)) / tma_info_core_core_clks", "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_page_walks_utilization", - "MetricThreshold": "tma_info_page_walks_utilization > 0.5" + "MetricName": "tma_info_memory_tlb_page_walks_utilization", + "MetricThreshold": "tma_info_memory_tlb_page_walks_utilization > 0.5" + }, + { + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", + "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", + "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", + "MetricName": "tma_info_pipeline_execute" }, { "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@", "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_retire" + "MetricName": "tma_info_pipeline_retire" }, { - "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", - "MetricExpr": "4 * tma_info_core_clks", - "MetricGroup": "TmaL1;tma_L1_group", - "MetricName": "tma_info_slots" + "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time", + "MetricGroup": "Power;Summary", + "MetricName": "tma_info_system_average_frequency" + }, + { + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricGroup": "HPC;Summary", + "MetricName": "tma_info_system_cpu_utilization" + }, + { + "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", + "MetricExpr": "64 * (UNC_ARB_TRK_REQUESTS.ALL + UNC_ARB_COH_TRK_REQUESTS.ALL) / 1e6 / duration_time / 1e3", + "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", + "MetricName": "tma_info_system_dram_bw_use", + "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_mem_bandwidth, tma_sq_full" + }, + { + "BriefDescription": "Giga Floating Point Operations Per Second", + "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / 1e9 / duration_time", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_system_gflops", + "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + }, + { + "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", + "MetricGroup": "Branches;OS", + "MetricName": "tma_info_system_ipfarbranch", + "MetricThreshold": "tma_info_system_ipfarbranch < 1e6" + }, + { + "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_cpi" + }, + { + "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_utilization", + "MetricThreshold": "tma_info_system_kernel_utilization > 0.05" + }, + { + "BriefDescription": "Average number of parallel requests to external memory", + "MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_OCCUPANCY.CYCLES_WITH_ANY_REQUEST", + "MetricGroup": "Mem;SoC", + "MetricName": "tma_info_system_mem_parallel_requests", + "PublicDescription": "Average number of parallel requests to external memory. Accounts for all requests" + }, + { + "BriefDescription": "Average latency of all requests to external memory (in Uncore cycles)", + "MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_REQUESTS.ALL", + "MetricGroup": "Mem;SoC", + "MetricName": "tma_info_system_mem_request_latency" }, { "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / (CPU_CLK_UNHALTED.REF_XCLK_ANY / 2) if #SMT_on else 0)", "MetricGroup": "SMT", - "MetricName": "tma_info_smt_2t_utilization" + "MetricName": "tma_info_system_smt_2t_utilization" }, { "BriefDescription": "Socket actual clocks when any core is active on that socket", "MetricExpr": "UNC_CLOCK.SOCKET", "MetricGroup": "SoC", - "MetricName": "tma_info_socket_clks" + "MetricName": "tma_info_system_socket_clks" }, { "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC", "MetricGroup": "Power", - "MetricName": "tma_info_turbo_utilization" + "MetricName": "tma_info_system_turbo_utilization" + }, + { + "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "Pipeline", + "MetricName": "tma_info_thread_clks" + }, + { + "BriefDescription": "Cycles Per Instruction (per Logical Processor)", + "MetricExpr": "1 / tma_info_thread_ipc", + "MetricGroup": "Mem;Pipeline", + "MetricName": "tma_info_thread_cpi" + }, + { + "BriefDescription": "The ratio of Executed- by Issued-Uops", + "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", + "MetricGroup": "Cor;Pipeline", + "MetricName": "tma_info_thread_execute_per_issue", + "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." + }, + { + "BriefDescription": "Instructions Per Cycle (per Logical Processor)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks", + "MetricGroup": "Ret;Summary", + "MetricName": "tma_info_thread_ipc" + }, + { + "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", + "MetricExpr": "4 * tma_info_core_core_clks", + "MetricGroup": "TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots" }, { "BriefDescription": "Uops Per Instruction", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / INST_RETIRED.ANY", "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_uoppi", - "MetricThreshold": "tma_info_uoppi > 1.05" + "MetricName": "tma_info_thread_uoppi", + "MetricThreshold": "tma_info_thread_uoppi > 1.05" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW", - "MetricName": "tma_info_uptb", - "MetricThreshold": "tma_info_uptb < 6" + "MetricName": "tma_info_thread_uptb", + "MetricThreshold": "tma_info_thread_uptb < 6" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses", - "MetricExpr": "(14 * ITLB_MISSES.STLB_HIT + cpu@ITLB_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * ITLB_MISSES.WALK_COMPLETED) / tma_info_clks", + "MetricExpr": "(14 * ITLB_MISSES.STLB_HIT + cpu@ITLB_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * ITLB_MISSES.WALK_COMPLETED) / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -777,7 +785,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled without loads missing the L1 data cache", - "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_clks, 0)", + "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_thread_clks, 0)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_issueL1;tma_issueMC;tma_memory_bound_group", "MetricName": "tma_l1_bound", "MetricThreshold": "tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -786,7 +794,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled due to L2 cache accesses by loads", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l2_bound", "MetricThreshold": "tma_l2_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -796,7 +804,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS) * CYCLE_ACTIVITY.STALLS_L2_MISS / tma_info_clks", + "MetricExpr": "MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS) * CYCLE_ACTIVITY.STALLS_L2_MISS / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -806,7 +814,7 @@ { "BriefDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "29 * (MEM_LOAD_UOPS_RETIRED.L3_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) / tma_info_clks", + "MetricExpr": "29 * (MEM_LOAD_UOPS_RETIRED.L3_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) / tma_info_thread_clks", "MetricGroup": "MemoryLat;TopdownL4;tma_L4_group;tma_issueLat;tma_l3_bound_group", "MetricName": "tma_l3_hit_latency", "MetricThreshold": "tma_l3_hit_latency > 0.1 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -815,11 +823,11 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)", - "MetricExpr": "ILD_STALL.LCP / tma_info_clks", + "MetricExpr": "ILD_STALL.LCP / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_lcp", "MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb", + "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb", "ScaleUnit": "100%" }, { @@ -828,13 +836,14 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_light_operations", "MetricThreshold": "tma_light_operations > 0.6", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation). This correlates with total number of instructions used by the program. A uops-per-instruction (see UopPI metric) ratio of 1 or less should be expected for decently optimized software running on Intel Core/Xeon products. While this often indicates efficient X86 instructions were executed; high value does not necessarily mean better performance cannot be achieved. Sample with: INST_RETIRED.PREC_DIST", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Load operations", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 + UOPS_DISPATCHED_PORT.PORT_7 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 + UOPS_DISPATCHED_PORT.PORT_7 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_load_op_utilization", "MetricThreshold": "tma_load_op_utilization > 0.6", @@ -844,7 +853,7 @@ { "BriefDescription": "This metric represents fraction of cycles the CPU spent handling cache misses due to lock operations", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO) / tma_info_clks", + "MetricExpr": "MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO) / tma_info_thread_clks", "MetricGroup": "Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_l1_bound_group", "MetricName": "tma_lock_latency", "MetricThreshold": "tma_lock_latency > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -858,21 +867,22 @@ "MetricGroup": "BadSpec;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.1 & tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears. These slots are either wasted by uops fetched prior to the clear; or stalls the out-of-order portion of the machine needs to recover its state after the clear. For example; this can happen due to memory ordering Nukes (e.g. Memory Disambiguation) or Self-Modifying-Code (SMC) nukes. Sample with: MACHINE_CLEARS.COUNT. Related metrics: tma_clears_resteers, tma_contested_accesses, tma_data_sharing, tma_false_sharing, tma_l1_bound, tma_microcode_sequencer, tma_ms_switches, tma_remote_cache", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_clks", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW", "MetricName": "tma_mem_bandwidth", "MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_sq_full", + "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_sq_full", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat", "MetricName": "tma_mem_latency", "MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -882,16 +892,17 @@ { "BriefDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_MEM_ANY + RESOURCE_STALLS.SB) / (CYCLE_ACTIVITY.STALLS_TOTAL + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_MEM_ANY + RESOURCE_STALLS.SB) / (CYCLE_ACTIVITY.STALLS_TOTAL + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_thread_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", "MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck. Memory Bound estimates fraction of slots where pipeline is likely stalled due to demand load or store instructions. This accounts mainly for (1) non-completed in-flight memory demand loads which coincides with execution units starvation; in addition to (2) cases where stores could impose backpressure on the pipeline when many of them get buffered at the same time (less common out of the two).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_thread_slots", "MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS", "MetricName": "tma_microcode_sequencer", "MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1", @@ -904,21 +915,21 @@ "MetricGroup": "BadSpec;BrMispredicts;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueBM", "MetricName": "tma_mispredicts_resteers", "MetricThreshold": "tma_mispredicts_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline)", - "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_mite", - "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline). This pipeline is used for code that was not pre-cached in the DSB or LSD. For example; inefficiencies due to asymmetric decoders; use of long immediate or LCP can manifest as MITE fetch bandwidth bottleneck.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)", - "MetricExpr": "2 * IDQ.MS_SWITCHES / tma_info_clks", + "MetricExpr": "2 * IDQ.MS_SWITCHES / tma_info_thread_clks", "MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO", "MetricName": "tma_ms_switches", "MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -927,7 +938,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 0 ([SNB+] ALU; [HSW+] ALU and 2nd branch)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_core_clks", "MetricGroup": "Compute;TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_0", "MetricThreshold": "tma_port_0 > 0.6", @@ -936,7 +947,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 1 (ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_1", "MetricThreshold": "tma_port_1 > 0.6", @@ -945,7 +956,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 2 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_2", "MetricThreshold": "tma_port_2 > 0.6", @@ -954,7 +965,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 3 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_3", "MetricThreshold": "tma_port_3 > 0.6", @@ -972,7 +983,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 5 ([SNB+] Branches and ALU; [HSW+] ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_5", "MetricThreshold": "tma_port_5 > 0.6", @@ -981,7 +992,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 6 ([HSW+]Primary Branch and simple ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_6 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_6 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_6", "MetricThreshold": "tma_port_6 > 0.6", @@ -990,7 +1001,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 7 ([HSW+]simple Store-address)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_7 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_7 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_store_op_utilization_group", "MetricName": "tma_port_7", "MetricThreshold": "tma_port_7 > 0.6", @@ -1000,7 +1011,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_TOTAL + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_clks", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_TOTAL + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_thread_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_ports_utilization", "MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -1009,7 +1020,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed no uops on any execution port (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,inv\\,cmask\\=1@ / 2 if #SMT_on else (CYCLE_ACTIVITY.STALLS_TOTAL - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0)) / tma_info_core_clks)", + "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,inv\\,cmask\\=1@ / 2 if #SMT_on else (CYCLE_ACTIVITY.STALLS_TOTAL - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0)) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_0", "MetricThreshold": "tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1018,7 +1029,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the CPU executed total of 1 uop per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) / tma_info_core_clks)", + "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issueL1;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_1", "MetricThreshold": "tma_ports_utilized_1 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1027,7 +1038,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 2 uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC - UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_clks)", + "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC - UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issue2P;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_2", "MetricThreshold": "tma_ports_utilized_2 > 0.15 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1036,7 +1047,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 3 or more uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise).", - "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_clks", + "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_3m", "MetricThreshold": "tma_ports_utilized_3m > 0.7 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1044,17 +1055,18 @@ }, { "BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired. Ideally; all pipeline slots would be attributed to the Retiring category. Retiring of 100% would indicate the maximum Pipeline_Width throughput was achieved. Maximizing Retiring typically increases the Instructions-per-cycle (see IPC metric). Note that a high Retiring value does not necessary mean there is no room for more performance. For example; Heavy-operations or Microcode Assists are categorized under Retiring. They often indicate suboptimal performance and can often be optimized or avoided. Sample with: UOPS_RETIRED.RETIRE_SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles handling memory load split accesses - load that cross 64-byte cache line boundary", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "tma_info_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_split_loads", "MetricThreshold": "tma_split_loads > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1063,7 +1075,7 @@ }, { "BriefDescription": "This metric represents rate of split store accesses", - "MetricExpr": "2 * MEM_UOPS_RETIRED.SPLIT_STORES / tma_info_core_clks", + "MetricExpr": "2 * MEM_UOPS_RETIRED.SPLIT_STORES / tma_info_core_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_issueSpSt;tma_store_bound_group", "MetricName": "tma_split_stores", "MetricThreshold": "tma_split_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1072,16 +1084,16 @@ }, { "BriefDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors)", - "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_clks", + "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_core_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueBW;tma_l3_bound_group", "MetricName": "tma_sq_full", "MetricThreshold": "tma_sq_full > 0.3 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_mem_bandwidth", + "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_mem_bandwidth", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write", - "MetricExpr": "RESOURCE_STALLS.SB / tma_info_clks", + "MetricExpr": "RESOURCE_STALLS.SB / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1090,7 +1102,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of cycles when the memory subsystem had loads blocked since they could not forward data from earlier (in program order) overlapping stores", - "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_clks", + "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_store_fwd_blk", "MetricThreshold": "tma_store_fwd_blk > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1100,7 +1112,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU spent handling L1D store misses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(L2_RQSTS.RFO_HIT * 9 * (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) + (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_clks", + "MetricExpr": "(L2_RQSTS.RFO_HIT * 9 * (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) + (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_thread_clks", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_issueSL;tma_store_bound_group", "MetricName": "tma_store_latency", "MetricThreshold": "tma_store_latency > 0.1 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1109,7 +1121,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Store operations", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_core_clks", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_store_op_utilization", "MetricThreshold": "tma_store_op_utilization > 0.6", @@ -1126,7 +1138,7 @@ }, { "BriefDescription": "This metric serves as an approximation of legacy x87 usage", - "MetricExpr": "INST_RETIRED.X87 * tma_info_uoppi / UOPS_RETIRED.RETIRE_SLOTS", + "MetricExpr": "INST_RETIRED.X87 * tma_info_thread_uoppi / UOPS_RETIRED.RETIRE_SLOTS", "MetricGroup": "Compute;TopdownL4;tma_L4_group;tma_fp_arith_group", "MetricName": "tma_x87_use", "MetricThreshold": "tma_x87_use > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6)", diff --git a/tools/perf/pmu-events/arch/x86/broadwell/floating-point.json b/tools/perf/pmu-events/arch/x86/broadwell/floating-point.json index e4826dc7f797..986869252e71 100644 --- a/tools/perf/pmu-events/arch/x86/broadwell/floating-point.json +++ b/tools/perf/pmu-events/arch/x86/broadwell/floating-point.json @@ -32,6 +32,14 @@ "UMask": "0x20" }, { + "BriefDescription": "Number of SSE/AVX computational 128-bit packed single and 256-bit packed double precision FP instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, 1 for each element. Applies to SSE* and AVX* packed single precision and packed double precision FP instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB count twice as they perform 2 calculations per element.", + "EventCode": "0xc7", + "EventName": "FP_ARITH_INST_RETIRED.4_FLOPS", + "PublicDescription": "Number of SSE/AVX computational 128-bit packed single precision and 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point and packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", + "SampleAfterValue": "2000003", + "UMask": "0x18" + }, + { "BriefDescription": "Number of SSE/AVX computational double precision floating-point instructions retired; some instructions will count twice as noted below. Applies to SSE* and AVX* scalar and packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform multiple calculations per element.", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.DOUBLE", @@ -77,6 +85,13 @@ "UMask": "0x2a" }, { + "BriefDescription": "Number of any Vector retired FP arithmetic instructions", + "EventCode": "0xc7", + "EventName": "FP_ARITH_INST_RETIRED.VECTOR", + "SampleAfterValue": "2000003", + "UMask": "0xfc" + }, + { "BriefDescription": "Cycles with any input/output SSE or FP assist", "CounterMask": "1", "EventCode": "0xCA", diff --git a/tools/perf/pmu-events/arch/x86/broadwell/metricgroups.json b/tools/perf/pmu-events/arch/x86/broadwell/metricgroups.json new file mode 100644 index 000000000000..f6a0258e3241 --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/broadwell/metricgroups.json @@ -0,0 +1,107 @@ +{ + "Backend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Bad": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BadSpec": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BigFoot": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BrMispredicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Branches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CacheMisses": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Compute": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Cor": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSBmiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DataSharing": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Fed": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Flops": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpScalar": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Frontend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "HPC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IcMiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "InsType": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "L2Evicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "LSD": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MachineClears": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBound": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryTLB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_BW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_Lat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MicroSeq": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "OS": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Offcore": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PGO": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Pipeline": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PortsUtil": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Ret": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Retire": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SMT": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Server": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Snoop": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SoC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL1": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL2": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL3mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "TopdownL5": "Metrics for top-down breakdown at level 5", + "TopdownL6": "Metrics for top-down breakdown at level 6", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_L5_group": "Metrics for top-down breakdown at level 5", + "tma_L6_group": "Metrics for top-down breakdown at level 6", + "tma_alu_op_utilization_group": "Metrics contributing to tma_alu_op_utilization category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_branch_resteers_group": "Metrics contributing to tma_branch_resteers category", + "tma_core_bound_group": "Metrics contributing to tma_core_bound category", + "tma_dram_bound_group": "Metrics contributing to tma_dram_bound category", + "tma_dtlb_load_group": "Metrics contributing to tma_dtlb_load category", + "tma_dtlb_store_group": "Metrics contributing to tma_dtlb_store category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_fp_arith_group": "Metrics contributing to tma_fp_arith category", + "tma_fp_vector_group": "Metrics contributing to tma_fp_vector category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_heavy_operations_group": "Metrics contributing to tma_heavy_operations category", + "tma_issue2P": "Metrics related by the issue $issue2P", + "tma_issueBM": "Metrics related by the issue $issueBM", + "tma_issueBW": "Metrics related by the issue $issueBW", + "tma_issueFB": "Metrics related by the issue $issueFB", + "tma_issueL1": "Metrics related by the issue $issueL1", + "tma_issueLat": "Metrics related by the issue $issueLat", + "tma_issueMC": "Metrics related by the issue $issueMC", + "tma_issueMS": "Metrics related by the issue $issueMS", + "tma_issueMV": "Metrics related by the issue $issueMV", + "tma_issueRFO": "Metrics related by the issue $issueRFO", + "tma_issueSL": "Metrics related by the issue $issueSL", + "tma_issueSO": "Metrics related by the issue $issueSO", + "tma_issueSmSt": "Metrics related by the issue $issueSmSt", + "tma_issueSpSt": "Metrics related by the issue $issueSpSt", + "tma_issueSyncxn": "Metrics related by the issue $issueSyncxn", + "tma_issueTLB": "Metrics related by the issue $issueTLB", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_l3_bound_group": "Metrics contributing to tma_l3_bound category", + "tma_light_operations_group": "Metrics contributing to tma_light_operations category", + "tma_load_op_utilization_group": "Metrics contributing to tma_load_op_utilization category", + "tma_mem_latency_group": "Metrics contributing to tma_mem_latency category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_microcode_sequencer_group": "Metrics contributing to tma_microcode_sequencer category", + "tma_ports_utilization_group": "Metrics contributing to tma_ports_utilization category", + "tma_ports_utilized_0_group": "Metrics contributing to tma_ports_utilized_0 category", + "tma_ports_utilized_3m_group": "Metrics contributing to tma_ports_utilized_3m category", + "tma_retiring_group": "Metrics contributing to tma_retiring category", + "tma_serializing_operation_group": "Metrics contributing to tma_serializing_operation category", + "tma_store_bound_group": "Metrics contributing to tma_store_bound category", + "tma_store_op_utilization_group": "Metrics contributing to tma_store_op_utilization category" +} diff --git a/tools/perf/pmu-events/arch/x86/broadwellde/bdwde-metrics.json b/tools/perf/pmu-events/arch/x86/broadwellde/bdwde-metrics.json index fb57c7382408..8fc62b8f667d 100644 --- a/tools/perf/pmu-events/arch/x86/broadwellde/bdwde-metrics.json +++ b/tools/perf/pmu-events/arch/x86/broadwellde/bdwde-metrics.json @@ -65,7 +65,7 @@ }, { "BriefDescription": "This metric estimates how often memory load accesses were aliased by preceding stores (in program order) with a 4K address offset", - "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_clks", + "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_4k_aliasing", "MetricThreshold": "tma_4k_aliasing > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -75,7 +75,7 @@ { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution ports for ALU operations.", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5 + UOPS_DISPATCHED_PORT.PORT_6) / tma_info_slots", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5 + UOPS_DISPATCHED_PORT.PORT_6) / tma_info_thread_slots", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_alu_op_utilization", "MetricThreshold": "tma_alu_op_utilization > 0.6", @@ -83,7 +83,7 @@ }, { "BriefDescription": "This metric estimates fraction of slots the CPU retired uops delivered by the Microcode_Sequencer as a result of Assists", - "MetricExpr": "100 * OTHER_ASSISTS.ANY_WB_ASSIST / tma_info_slots", + "MetricExpr": "100 * OTHER_ASSISTS.ANY_WB_ASSIST / tma_info_thread_slots", "MetricGroup": "TopdownL4;tma_L4_group;tma_microcode_sequencer_group", "MetricName": "tma_assists", "MetricThreshold": "tma_assists > 0.1 & (tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1)", @@ -97,15 +97,17 @@ "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend. Backend is the portion of the processor core where the out-of-order scheduler dispatches ready uops into their respective execution units; and once completed these uops get retired according to program order. For example; stalls due to data-cache misses or stalls due to the divider unit being overloaded are both categorized under Backend Bound. Backend Bound is further divided into two main categories: Memory Bound and Core Bound. Sample with: TOPDOWN.BACKEND_BOUND_SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations", - "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_slots", + "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots wasted due to incorrect speculations. This include slots used to issue uops that do not eventually get retired and slots for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work due to miss-predicted branches are categorized under Bad Speculation category. Incorrect data speculation followed by Memory Ordering Nukes is another example.", "ScaleUnit": "100%" }, @@ -116,12 +118,13 @@ "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15", - "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: TOPDOWN.BR_MISPREDICT_SLOTS. Related metrics: tma_info_branch_misprediction_cost, tma_mispredicts_resteers", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: TOPDOWN.BR_MISPREDICT_SLOTS. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers", - "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_clks", + "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteers", "MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -150,7 +153,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to contested accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(60 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) + 43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS)))) / tma_info_clks", + "MetricExpr": "(60 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) + 43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS)))) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_contested_accesses", "MetricThreshold": "tma_contested_accesses > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -164,13 +167,14 @@ "MetricGroup": "Backend;Compute;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck. Shortage in hardware compute resources; or dependencies in software's instructions are both categorized under Core Bound. Hence it may indicate the machine ran out of an out-of-order resource; certain execution units are overloaded or dependencies in program's data- or instruction-flow are limiting the performance (e.g. FP-chained long-latency arithmetic operations).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to data-sharing accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) / tma_info_clks", + "MetricExpr": "43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) / tma_info_thread_clks", "MetricGroup": "Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_data_sharing", "MetricThreshold": "tma_data_sharing > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -179,7 +183,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the Divider unit was active", - "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_clks", + "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_core_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_divider", "MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -189,7 +193,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS)) * CYCLE_ACTIVITY.STALLS_L2_MISS / tma_info_clks", + "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS)) * CYCLE_ACTIVITY.STALLS_L2_MISS / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -198,25 +202,25 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline", - "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSB;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_dsb", - "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline. For example; inefficient utilization of the DSB cache structure or bank conflict when reading from it; are categorized here.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_dsb_switches", "MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb, tma_lcp", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses", - "MetricExpr": "(8 * DTLB_LOAD_MISSES.STLB_HIT + cpu@DTLB_LOAD_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * DTLB_LOAD_MISSES.WALK_COMPLETED) / tma_info_clks", + "MetricExpr": "(8 * DTLB_LOAD_MISSES.STLB_HIT + cpu@DTLB_LOAD_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * DTLB_LOAD_MISSES.WALK_COMPLETED) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group", "MetricName": "tma_dtlb_load", "MetricThreshold": "tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -225,7 +229,7 @@ }, { "BriefDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses", - "MetricExpr": "(8 * DTLB_STORE_MISSES.STLB_HIT + cpu@DTLB_STORE_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * DTLB_STORE_MISSES.WALK_COMPLETED) / tma_info_clks", + "MetricExpr": "(8 * DTLB_STORE_MISSES.STLB_HIT + cpu@DTLB_STORE_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * DTLB_STORE_MISSES.WALK_COMPLETED) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_store_bound_group", "MetricName": "tma_dtlb_store", "MetricThreshold": "tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -235,11 +239,11 @@ { "BriefDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "tma_info_load_miss_real_latency * cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@ / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@ / tma_info_thread_clks", "MetricGroup": "MemoryBW;TopdownL4;tma_L4_group;tma_issueBW;tma_issueSL;tma_issueSmSt;tma_l1_bound_group", "MetricName": "tma_fb_full", "MetricThreshold": "tma_fb_full > 0.3", - "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", + "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", "ScaleUnit": "100%" }, { @@ -247,16 +251,18 @@ "MetricExpr": "tma_frontend_bound - tma_fetch_latency", "MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", "MetricName": "tma_fetch_bandwidth", - "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_info_iptb, tma_lcp", + "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues", - "MetricExpr": "4 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE / tma_info_slots", + "MetricExpr": "4 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE / tma_info_thread_slots", "MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: FRONTEND_RETIRED.LATENCY_GE_16_PS;FRONTEND_RETIRED.LATENCY_GE_8_PS", "ScaleUnit": "100%" }, @@ -307,10 +313,11 @@ }, { "BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend", - "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_slots", + "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_thread_slots", "MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound. Sample with: FRONTEND_RETIRED.LATENCY_GE_4_PS", "ScaleUnit": "100%" }, @@ -320,12 +327,13 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_heavy_operations", "MetricThreshold": "tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to instruction cache misses", - "MetricExpr": "ICACHE.IFDATA_STALL / tma_info_clks", + "MetricExpr": "ICACHE.IFDATA_STALL / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;IcMiss;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_icache_misses", "MetricThreshold": "tma_icache_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -333,409 +341,409 @@ "ScaleUnit": "100%" }, { - "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", - "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time", - "MetricGroup": "Power;Summary", - "MetricName": "tma_info_average_frequency" - }, - { - "BriefDescription": "Branch instructions per taken branch.", - "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", - "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_bptkbranch" - }, - { "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", - "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_slots / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_thread_slots / BR_MISP_RETIRED.ALL_BRANCHES", "MetricGroup": "Bad;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_branch_misprediction_cost", + "MetricName": "tma_info_bad_spec_branch_misprediction_cost", "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_mispredicts_resteers" }, { - "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "Pipeline", - "MetricName": "tma_info_clks" + "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", + "MetricExpr": "tma_info_inst_mix_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_indirect", + "MetricThreshold": "tma_info_bad_spec_ipmisp_indirect < 1e3" + }, + { + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BadSpec;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmispredict", + "MetricThreshold": "tma_info_bad_spec_ipmispredict < 200" }, { "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", - "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_clks))", + "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_thread_clks))", "MetricGroup": "SMT", - "MetricName": "tma_info_core_clks" + "MetricName": "tma_info_core_core_clks" }, { "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks", "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_coreipc" - }, - { - "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricExpr": "1 / tma_info_ipc", - "MetricGroup": "Mem;Pipeline", - "MetricName": "tma_info_cpi" + "MetricName": "tma_info_core_coreipc" }, { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": "HPC;Summary", - "MetricName": "tma_info_cpu_utilization" + "BriefDescription": "Floating Point Operations Per Cycle", + "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / tma_info_core_core_clks", + "MetricGroup": "Flops;Ret", + "MetricName": "tma_info_core_flopc" }, { - "BriefDescription": "Average Parallel L2 cache miss data reads", - "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", - "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_data_l2_mlp" + "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", + "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@) / (2 * tma_info_core_core_clks)", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_core_fp_arith_utilization", + "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." }, { - "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricExpr": "64 * (arb@event\\=0x81\\,umask\\=0x1@ + arb@event\\=0x84\\,umask\\=0x1@) / 1e6 / duration_time / 1e3", - "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", - "MetricName": "tma_info_dram_bw_use", - "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", + "MetricExpr": "UOPS_EXECUTED.THREAD / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC)", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", + "MetricName": "tma_info_core_ilp" }, { "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", "MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + LSD.UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)", "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", - "MetricName": "tma_info_dsb_coverage", - "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_iptb, tma_lcp" + "MetricName": "tma_info_frontend_dsb_coverage", + "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 4 > 0.35", + "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_inst_mix_iptb, tma_lcp" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", - "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", - "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", - "MetricName": "tma_info_execute" - }, - { - "BriefDescription": "The ratio of Executed- by Issued-Uops", - "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", - "MetricGroup": "Cor;Pipeline", - "MetricName": "tma_info_execute_per_issue", - "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." - }, - { - "BriefDescription": "Floating Point Operations Per Cycle", - "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / tma_info_core_clks", - "MetricGroup": "Flops;Ret", - "MetricName": "tma_info_flopc" - }, - { - "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", - "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@) / (2 * tma_info_core_clks)", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_fp_arith_utilization", - "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." - }, - { - "BriefDescription": "Giga Floating Point Operations Per Second", - "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / 1e9 / duration_time", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_gflops", - "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_inst_mix_instructions / BACLEARS.ANY", + "MetricGroup": "Fed", + "MetricName": "tma_info_frontend_ipunknown_branch" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", - "MetricExpr": "UOPS_EXECUTED.THREAD / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC)", - "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", - "MetricName": "tma_info_ilp" + "BriefDescription": "Branch instructions per taken branch.", + "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;PGO", + "MetricName": "tma_info_inst_mix_bptkbranch" }, { "BriefDescription": "Total number of retired Instructions", "MetricExpr": "INST_RETIRED.ANY", "MetricGroup": "Summary;TmaL1;tma_L1_group", - "MetricName": "tma_info_instructions", + "MetricName": "tma_info_inst_mix_instructions", "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" }, { "BriefDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_iparith", - "MetricThreshold": "tma_info_iparith < 10", + "MetricName": "tma_info_inst_mix_iparith", + "MetricThreshold": "tma_info_inst_mix_iparith < 10", "PublicDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate). May undercount due to FMA double counting. Approximated prior to BDW." }, { "BriefDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx128", - "MetricThreshold": "tma_info_iparith_avx128 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx128", + "MetricThreshold": "tma_info_inst_mix_iparith_avx128 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx256", - "MetricThreshold": "tma_info_iparith_avx256 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx256", + "MetricThreshold": "tma_info_inst_mix_iparith_avx256 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_dp", - "MetricThreshold": "tma_info_iparith_scalar_dp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_dp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_dp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_SINGLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_sp", - "MetricThreshold": "tma_info_iparith_scalar_sp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_sp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_sp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per Branch (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Branches;Fed;InsType", - "MetricName": "tma_info_ipbranch", - "MetricThreshold": "tma_info_ipbranch < 8" - }, - { - "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": "Ret;Summary", - "MetricName": "tma_info_ipc" + "MetricName": "tma_info_inst_mix_ipbranch", + "MetricThreshold": "tma_info_inst_mix_ipbranch < 8" }, { "BriefDescription": "Instructions per (near) call (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_CALL", "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_ipcall", - "MetricThreshold": "tma_info_ipcall < 200" - }, - { - "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", - "MetricGroup": "Branches;OS", - "MetricName": "tma_info_ipfarbranch", - "MetricThreshold": "tma_info_ipfarbranch < 1e6" + "MetricName": "tma_info_inst_mix_ipcall", + "MetricThreshold": "tma_info_inst_mix_ipcall < 200" }, { "BriefDescription": "Instructions per Floating Point (FP) Operation (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_ipflop", - "MetricThreshold": "tma_info_ipflop < 10" + "MetricName": "tma_info_inst_mix_ipflop", + "MetricThreshold": "tma_info_inst_mix_ipflop < 10" }, { "BriefDescription": "Instructions per Load (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_LOADS", "MetricGroup": "InsType", - "MetricName": "tma_info_ipload", - "MetricThreshold": "tma_info_ipload < 3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", - "MetricExpr": "tma_info_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_indirect", - "MetricThreshold": "tma_info_ipmisp_indirect < 1e3" - }, - { - "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BadSpec;BrMispredicts", - "MetricName": "tma_info_ipmispredict", - "MetricThreshold": "tma_info_ipmispredict < 200" + "MetricName": "tma_info_inst_mix_ipload", + "MetricThreshold": "tma_info_inst_mix_ipload < 3" }, { "BriefDescription": "Instructions per Store (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_STORES", "MetricGroup": "InsType", - "MetricName": "tma_info_ipstore", - "MetricThreshold": "tma_info_ipstore < 8" + "MetricName": "tma_info_inst_mix_ipstore", + "MetricThreshold": "tma_info_inst_mix_ipstore < 8" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW;Frontend;PGO;tma_issueFB", - "MetricName": "tma_info_iptb", - "MetricThreshold": "tma_info_iptb < 9", - "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_lcp" - }, - { - "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", - "MetricExpr": "tma_info_instructions / BACLEARS.ANY", - "MetricGroup": "Fed", - "MetricName": "tma_info_ipunknown_branch" - }, - { - "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_cpi" - }, - { - "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_utilization", - "MetricThreshold": "tma_info_kernel_utilization > 0.05" + "MetricName": "tma_info_inst_mix_iptb", + "MetricThreshold": "tma_info_inst_mix_iptb < 9", + "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_lcp" }, { "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw" - }, - { - "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "tma_info_l1d_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw_1t" - }, - { - "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L1_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki" + "MetricName": "tma_info_memory_core_l1d_cache_fill_bw" }, { "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw" + "MetricName": "tma_info_memory_core_l2_cache_fill_bw" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "tma_info_l2_cache_fill_bw", + "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw_1t" + "MetricName": "tma_info_memory_core_l3_cache_fill_bw" + }, + { + "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L1_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l1mpki" }, { "BriefDescription": "L2 cache hits per kilo instruction for all request types (including speculative)", "MetricExpr": "1e3 * (L2_RQSTS.REFERENCES - L2_RQSTS.MISS) / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_all" + "MetricName": "tma_info_memory_l2hpki_all" }, { "BriefDescription": "L2 cache hits per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_HIT / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_load" + "MetricName": "tma_info_memory_l2hpki_load" }, { "BriefDescription": "L2 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L2_MISS / INST_RETIRED.ANY", "MetricGroup": "Backend;CacheMisses;Mem", - "MetricName": "tma_info_l2mpki" + "MetricName": "tma_info_memory_l2mpki" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all request types (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem;Offcore", - "MetricName": "tma_info_l2mpki_all" + "MetricName": "tma_info_memory_l2mpki_all" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2mpki_load" + "MetricName": "tma_info_memory_l2mpki_load" }, { - "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "0", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw_1t" + "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L3_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l3mpki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw" + "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_UOPS_RETIRED.L1_MISS + MEM_LOAD_UOPS_RETIRED.HIT_LFB)", + "MetricGroup": "Mem;MemoryBound;MemoryLat", + "MetricName": "tma_info_memory_load_miss_real_latency" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw_1t" + "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", + "MetricGroup": "Mem;MemoryBW;MemoryBound", + "MetricName": "tma_info_memory_mlp", + "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" }, { - "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L3_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l3mpki" + "BriefDescription": "Average Parallel L2 cache miss data reads", + "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", + "MetricGroup": "Memory_BW;Offcore", + "MetricName": "tma_info_memory_oro_data_l2_mlp" }, { "BriefDescription": "Average Latency for L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS.DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l2_miss_latency" + "MetricName": "tma_info_memory_oro_load_l2_miss_latency" }, { "BriefDescription": "Average Parallel L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_DATA_RD", "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_load_l2_mlp" + "MetricName": "tma_info_memory_oro_load_l2_mlp" }, { - "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_UOPS_RETIRED.L1_MISS + MEM_LOAD_UOPS_RETIRED.HIT_LFB)", - "MetricGroup": "Mem;MemoryBound;MemoryLat", - "MetricName": "tma_info_load_miss_real_latency" + "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l1d_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l1d_cache_fill_bw_1t" }, { - "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", - "MetricGroup": "Mem;MemoryBW;MemoryBound", - "MetricName": "tma_info_mlp", - "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" + "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l2_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l2_cache_fill_bw_1t" + }, + { + "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "0", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_thread_l3_cache_access_bw_1t" + }, + { + "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l3_cache_fill_bw_1t" }, { "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", - "MetricExpr": "(cpu@ITLB_MISSES.WALK_DURATION\\,cmask\\=1@ + cpu@DTLB_LOAD_MISSES.WALK_DURATION\\,cmask\\=1@ + cpu@DTLB_STORE_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * (DTLB_STORE_MISSES.WALK_COMPLETED + DTLB_LOAD_MISSES.WALK_COMPLETED + ITLB_MISSES.WALK_COMPLETED)) / tma_info_core_clks", + "MetricExpr": "(cpu@ITLB_MISSES.WALK_DURATION\\,cmask\\=1@ + cpu@DTLB_LOAD_MISSES.WALK_DURATION\\,cmask\\=1@ + cpu@DTLB_STORE_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * (DTLB_STORE_MISSES.WALK_COMPLETED + DTLB_LOAD_MISSES.WALK_COMPLETED + ITLB_MISSES.WALK_COMPLETED)) / tma_info_core_core_clks", "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_page_walks_utilization", - "MetricThreshold": "tma_info_page_walks_utilization > 0.5" + "MetricName": "tma_info_memory_tlb_page_walks_utilization", + "MetricThreshold": "tma_info_memory_tlb_page_walks_utilization > 0.5" + }, + { + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", + "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", + "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", + "MetricName": "tma_info_pipeline_execute" }, { "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@", "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_retire" + "MetricName": "tma_info_pipeline_retire" }, { - "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", - "MetricExpr": "4 * tma_info_core_clks", - "MetricGroup": "TmaL1;tma_L1_group", - "MetricName": "tma_info_slots" + "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time", + "MetricGroup": "Power;Summary", + "MetricName": "tma_info_system_average_frequency" + }, + { + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricGroup": "HPC;Summary", + "MetricName": "tma_info_system_cpu_utilization" + }, + { + "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", + "MetricExpr": "64 * (arb@event\\=0x81\\,umask\\=0x1@ + arb@event\\=0x84\\,umask\\=0x1@) / 1e6 / duration_time / 1e3", + "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", + "MetricName": "tma_info_system_dram_bw_use", + "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_mem_bandwidth, tma_sq_full" + }, + { + "BriefDescription": "Giga Floating Point Operations Per Second", + "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / 1e9 / duration_time", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_system_gflops", + "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + }, + { + "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", + "MetricGroup": "Branches;OS", + "MetricName": "tma_info_system_ipfarbranch", + "MetricThreshold": "tma_info_system_ipfarbranch < 1e6" + }, + { + "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_cpi" + }, + { + "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_utilization", + "MetricThreshold": "tma_info_system_kernel_utilization > 0.05" }, { "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / (CPU_CLK_UNHALTED.REF_XCLK_ANY / 2) if #SMT_on else 0)", "MetricGroup": "SMT", - "MetricName": "tma_info_smt_2t_utilization" + "MetricName": "tma_info_system_smt_2t_utilization" }, { "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC", "MetricGroup": "Power", - "MetricName": "tma_info_turbo_utilization" + "MetricName": "tma_info_system_turbo_utilization" + }, + { + "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "Pipeline", + "MetricName": "tma_info_thread_clks" + }, + { + "BriefDescription": "Cycles Per Instruction (per Logical Processor)", + "MetricExpr": "1 / tma_info_thread_ipc", + "MetricGroup": "Mem;Pipeline", + "MetricName": "tma_info_thread_cpi" + }, + { + "BriefDescription": "The ratio of Executed- by Issued-Uops", + "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", + "MetricGroup": "Cor;Pipeline", + "MetricName": "tma_info_thread_execute_per_issue", + "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." + }, + { + "BriefDescription": "Instructions Per Cycle (per Logical Processor)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks", + "MetricGroup": "Ret;Summary", + "MetricName": "tma_info_thread_ipc" + }, + { + "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", + "MetricExpr": "4 * tma_info_core_core_clks", + "MetricGroup": "TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots" }, { "BriefDescription": "Uops Per Instruction", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / INST_RETIRED.ANY", "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_uoppi", - "MetricThreshold": "tma_info_uoppi > 1.05" + "MetricName": "tma_info_thread_uoppi", + "MetricThreshold": "tma_info_thread_uoppi > 1.05" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW", - "MetricName": "tma_info_uptb", - "MetricThreshold": "tma_info_uptb < 6" + "MetricName": "tma_info_thread_uptb", + "MetricThreshold": "tma_info_thread_uptb < 6" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses", - "MetricExpr": "(14 * ITLB_MISSES.STLB_HIT + cpu@ITLB_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * ITLB_MISSES.WALK_COMPLETED) / tma_info_clks", + "MetricExpr": "(14 * ITLB_MISSES.STLB_HIT + cpu@ITLB_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * ITLB_MISSES.WALK_COMPLETED) / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -744,7 +752,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled without loads missing the L1 data cache", - "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_clks, 0)", + "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_thread_clks, 0)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_issueL1;tma_issueMC;tma_memory_bound_group", "MetricName": "tma_l1_bound", "MetricThreshold": "tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -753,7 +761,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled due to L2 cache accesses by loads", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l2_bound", "MetricThreshold": "tma_l2_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -763,7 +771,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS) * CYCLE_ACTIVITY.STALLS_L2_MISS / tma_info_clks", + "MetricExpr": "MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS) * CYCLE_ACTIVITY.STALLS_L2_MISS / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -773,7 +781,7 @@ { "BriefDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "29 * (MEM_LOAD_UOPS_RETIRED.L3_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) / tma_info_clks", + "MetricExpr": "29 * (MEM_LOAD_UOPS_RETIRED.L3_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) / tma_info_thread_clks", "MetricGroup": "MemoryLat;TopdownL4;tma_L4_group;tma_issueLat;tma_l3_bound_group", "MetricName": "tma_l3_hit_latency", "MetricThreshold": "tma_l3_hit_latency > 0.1 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -782,11 +790,11 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)", - "MetricExpr": "ILD_STALL.LCP / tma_info_clks", + "MetricExpr": "ILD_STALL.LCP / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_lcp", "MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb", + "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb", "ScaleUnit": "100%" }, { @@ -795,13 +803,14 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_light_operations", "MetricThreshold": "tma_light_operations > 0.6", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation). This correlates with total number of instructions used by the program. A uops-per-instruction (see UopPI metric) ratio of 1 or less should be expected for decently optimized software running on Intel Core/Xeon products. While this often indicates efficient X86 instructions were executed; high value does not necessarily mean better performance cannot be achieved. Sample with: INST_RETIRED.PREC_DIST", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Load operations", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 + UOPS_DISPATCHED_PORT.PORT_7 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 + UOPS_DISPATCHED_PORT.PORT_7 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_load_op_utilization", "MetricThreshold": "tma_load_op_utilization > 0.6", @@ -811,7 +820,7 @@ { "BriefDescription": "This metric represents fraction of cycles the CPU spent handling cache misses due to lock operations", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO) / tma_info_clks", + "MetricExpr": "MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO) / tma_info_thread_clks", "MetricGroup": "Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_l1_bound_group", "MetricName": "tma_lock_latency", "MetricThreshold": "tma_lock_latency > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -825,21 +834,22 @@ "MetricGroup": "BadSpec;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.1 & tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears. These slots are either wasted by uops fetched prior to the clear; or stalls the out-of-order portion of the machine needs to recover its state after the clear. For example; this can happen due to memory ordering Nukes (e.g. Memory Disambiguation) or Self-Modifying-Code (SMC) nukes. Sample with: MACHINE_CLEARS.COUNT. Related metrics: tma_clears_resteers, tma_contested_accesses, tma_data_sharing, tma_false_sharing, tma_l1_bound, tma_microcode_sequencer, tma_ms_switches, tma_remote_cache", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_clks", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW", "MetricName": "tma_mem_bandwidth", "MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_sq_full", + "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_sq_full", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat", "MetricName": "tma_mem_latency", "MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -849,16 +859,17 @@ { "BriefDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_MEM_ANY + RESOURCE_STALLS.SB) / (CYCLE_ACTIVITY.STALLS_TOTAL + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_MEM_ANY + RESOURCE_STALLS.SB) / (CYCLE_ACTIVITY.STALLS_TOTAL + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_thread_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", "MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck. Memory Bound estimates fraction of slots where pipeline is likely stalled due to demand load or store instructions. This accounts mainly for (1) non-completed in-flight memory demand loads which coincides with execution units starvation; in addition to (2) cases where stores could impose backpressure on the pipeline when many of them get buffered at the same time (less common out of the two).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_thread_slots", "MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS", "MetricName": "tma_microcode_sequencer", "MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1", @@ -871,21 +882,21 @@ "MetricGroup": "BadSpec;BrMispredicts;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueBM", "MetricName": "tma_mispredicts_resteers", "MetricThreshold": "tma_mispredicts_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline)", - "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_mite", - "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline). This pipeline is used for code that was not pre-cached in the DSB or LSD. For example; inefficiencies due to asymmetric decoders; use of long immediate or LCP can manifest as MITE fetch bandwidth bottleneck. Sample with: FRONTEND_RETIRED.ANY_DSB_MISS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)", - "MetricExpr": "2 * IDQ.MS_SWITCHES / tma_info_clks", + "MetricExpr": "2 * IDQ.MS_SWITCHES / tma_info_thread_clks", "MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO", "MetricName": "tma_ms_switches", "MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -894,7 +905,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 0 ([SNB+] ALU; [HSW+] ALU and 2nd branch)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_core_clks", "MetricGroup": "Compute;TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_0", "MetricThreshold": "tma_port_0 > 0.6", @@ -903,7 +914,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 1 (ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_1", "MetricThreshold": "tma_port_1 > 0.6", @@ -912,7 +923,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 2 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_2", "MetricThreshold": "tma_port_2 > 0.6", @@ -920,7 +931,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 3 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_3", "MetricThreshold": "tma_port_3 > 0.6", @@ -937,7 +948,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 5 ([SNB+] Branches and ALU; [HSW+] ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_5", "MetricThreshold": "tma_port_5 > 0.6", @@ -946,7 +957,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 6 ([HSW+]Primary Branch and simple ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_6 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_6 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_6", "MetricThreshold": "tma_port_6 > 0.6", @@ -955,7 +966,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 7 ([HSW+]simple Store-address)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_7 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_7 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_store_op_utilization_group", "MetricName": "tma_port_7", "MetricThreshold": "tma_port_7 > 0.6", @@ -964,7 +975,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_TOTAL + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_clks", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_TOTAL + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_thread_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_ports_utilization", "MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -973,7 +984,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed no uops on any execution port (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,inv\\,cmask\\=1@ / 2 if #SMT_on else (CYCLE_ACTIVITY.STALLS_TOTAL - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0)) / tma_info_core_clks)", + "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,inv\\,cmask\\=1@ / 2 if #SMT_on else (CYCLE_ACTIVITY.STALLS_TOTAL - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0)) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_0", "MetricThreshold": "tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -982,7 +993,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the CPU executed total of 1 uop per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) / tma_info_core_clks)", + "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issueL1;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_1", "MetricThreshold": "tma_ports_utilized_1 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -991,7 +1002,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 2 uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC - UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_clks)", + "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC - UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issue2P;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_2", "MetricThreshold": "tma_ports_utilized_2 > 0.15 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1000,7 +1011,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 3 or more uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_clks", + "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_3m", "MetricThreshold": "tma_ports_utilized_3m > 0.7 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1009,17 +1020,18 @@ }, { "BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired. Ideally; all pipeline slots would be attributed to the Retiring category. Retiring of 100% would indicate the maximum Pipeline_Width throughput was achieved. Maximizing Retiring typically increases the Instructions-per-cycle (see IPC metric). Note that a high Retiring value does not necessary mean there is no room for more performance. For example; Heavy-operations or Microcode Assists are categorized under Retiring. They often indicate suboptimal performance and can often be optimized or avoided. Sample with: UOPS_RETIRED.SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles handling memory load split accesses - load that cross 64-byte cache line boundary", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "tma_info_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_split_loads", "MetricThreshold": "tma_split_loads > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1028,7 +1040,7 @@ }, { "BriefDescription": "This metric represents rate of split store accesses", - "MetricExpr": "2 * MEM_UOPS_RETIRED.SPLIT_STORES / tma_info_core_clks", + "MetricExpr": "2 * MEM_UOPS_RETIRED.SPLIT_STORES / tma_info_core_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_issueSpSt;tma_store_bound_group", "MetricName": "tma_split_stores", "MetricThreshold": "tma_split_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1037,16 +1049,16 @@ }, { "BriefDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors)", - "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_clks", + "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_core_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueBW;tma_l3_bound_group", "MetricName": "tma_sq_full", "MetricThreshold": "tma_sq_full > 0.3 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_mem_bandwidth", + "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_mem_bandwidth", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write", - "MetricExpr": "RESOURCE_STALLS.SB / tma_info_clks", + "MetricExpr": "RESOURCE_STALLS.SB / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1055,7 +1067,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of cycles when the memory subsystem had loads blocked since they could not forward data from earlier (in program order) overlapping stores", - "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_clks", + "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_store_fwd_blk", "MetricThreshold": "tma_store_fwd_blk > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1065,7 +1077,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU spent handling L1D store misses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(L2_RQSTS.RFO_HIT * 9 * (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) + (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_clks", + "MetricExpr": "(L2_RQSTS.RFO_HIT * 9 * (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) + (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_thread_clks", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_issueSL;tma_store_bound_group", "MetricName": "tma_store_latency", "MetricThreshold": "tma_store_latency > 0.1 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1074,7 +1086,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Store operations", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_core_clks", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_store_op_utilization", "MetricThreshold": "tma_store_op_utilization > 0.6", @@ -1092,7 +1104,7 @@ }, { "BriefDescription": "This metric serves as an approximation of legacy x87 usage", - "MetricExpr": "INST_RETIRED.X87 * tma_info_uoppi / UOPS_RETIRED.RETIRE_SLOTS", + "MetricExpr": "INST_RETIRED.X87 * tma_info_thread_uoppi / UOPS_RETIRED.RETIRE_SLOTS", "MetricGroup": "Compute;TopdownL4;tma_L4_group;tma_fp_arith_group", "MetricName": "tma_x87_use", "MetricThreshold": "tma_x87_use > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6)", diff --git a/tools/perf/pmu-events/arch/x86/broadwellde/floating-point.json b/tools/perf/pmu-events/arch/x86/broadwellde/floating-point.json index e4826dc7f797..986869252e71 100644 --- a/tools/perf/pmu-events/arch/x86/broadwellde/floating-point.json +++ b/tools/perf/pmu-events/arch/x86/broadwellde/floating-point.json @@ -32,6 +32,14 @@ "UMask": "0x20" }, { + "BriefDescription": "Number of SSE/AVX computational 128-bit packed single and 256-bit packed double precision FP instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, 1 for each element. Applies to SSE* and AVX* packed single precision and packed double precision FP instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB count twice as they perform 2 calculations per element.", + "EventCode": "0xc7", + "EventName": "FP_ARITH_INST_RETIRED.4_FLOPS", + "PublicDescription": "Number of SSE/AVX computational 128-bit packed single precision and 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point and packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", + "SampleAfterValue": "2000003", + "UMask": "0x18" + }, + { "BriefDescription": "Number of SSE/AVX computational double precision floating-point instructions retired; some instructions will count twice as noted below. Applies to SSE* and AVX* scalar and packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform multiple calculations per element.", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.DOUBLE", @@ -77,6 +85,13 @@ "UMask": "0x2a" }, { + "BriefDescription": "Number of any Vector retired FP arithmetic instructions", + "EventCode": "0xc7", + "EventName": "FP_ARITH_INST_RETIRED.VECTOR", + "SampleAfterValue": "2000003", + "UMask": "0xfc" + }, + { "BriefDescription": "Cycles with any input/output SSE or FP assist", "CounterMask": "1", "EventCode": "0xCA", diff --git a/tools/perf/pmu-events/arch/x86/broadwellde/metricgroups.json b/tools/perf/pmu-events/arch/x86/broadwellde/metricgroups.json new file mode 100644 index 000000000000..f6a0258e3241 --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/broadwellde/metricgroups.json @@ -0,0 +1,107 @@ +{ + "Backend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Bad": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BadSpec": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BigFoot": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BrMispredicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Branches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CacheMisses": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Compute": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Cor": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSBmiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DataSharing": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Fed": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Flops": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpScalar": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Frontend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "HPC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IcMiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "InsType": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "L2Evicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "LSD": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MachineClears": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBound": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryTLB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_BW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_Lat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MicroSeq": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "OS": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Offcore": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PGO": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Pipeline": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PortsUtil": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Ret": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Retire": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SMT": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Server": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Snoop": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SoC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL1": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL2": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL3mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "TopdownL5": "Metrics for top-down breakdown at level 5", + "TopdownL6": "Metrics for top-down breakdown at level 6", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_L5_group": "Metrics for top-down breakdown at level 5", + "tma_L6_group": "Metrics for top-down breakdown at level 6", + "tma_alu_op_utilization_group": "Metrics contributing to tma_alu_op_utilization category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_branch_resteers_group": "Metrics contributing to tma_branch_resteers category", + "tma_core_bound_group": "Metrics contributing to tma_core_bound category", + "tma_dram_bound_group": "Metrics contributing to tma_dram_bound category", + "tma_dtlb_load_group": "Metrics contributing to tma_dtlb_load category", + "tma_dtlb_store_group": "Metrics contributing to tma_dtlb_store category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_fp_arith_group": "Metrics contributing to tma_fp_arith category", + "tma_fp_vector_group": "Metrics contributing to tma_fp_vector category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_heavy_operations_group": "Metrics contributing to tma_heavy_operations category", + "tma_issue2P": "Metrics related by the issue $issue2P", + "tma_issueBM": "Metrics related by the issue $issueBM", + "tma_issueBW": "Metrics related by the issue $issueBW", + "tma_issueFB": "Metrics related by the issue $issueFB", + "tma_issueL1": "Metrics related by the issue $issueL1", + "tma_issueLat": "Metrics related by the issue $issueLat", + "tma_issueMC": "Metrics related by the issue $issueMC", + "tma_issueMS": "Metrics related by the issue $issueMS", + "tma_issueMV": "Metrics related by the issue $issueMV", + "tma_issueRFO": "Metrics related by the issue $issueRFO", + "tma_issueSL": "Metrics related by the issue $issueSL", + "tma_issueSO": "Metrics related by the issue $issueSO", + "tma_issueSmSt": "Metrics related by the issue $issueSmSt", + "tma_issueSpSt": "Metrics related by the issue $issueSpSt", + "tma_issueSyncxn": "Metrics related by the issue $issueSyncxn", + "tma_issueTLB": "Metrics related by the issue $issueTLB", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_l3_bound_group": "Metrics contributing to tma_l3_bound category", + "tma_light_operations_group": "Metrics contributing to tma_light_operations category", + "tma_load_op_utilization_group": "Metrics contributing to tma_load_op_utilization category", + "tma_mem_latency_group": "Metrics contributing to tma_mem_latency category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_microcode_sequencer_group": "Metrics contributing to tma_microcode_sequencer category", + "tma_ports_utilization_group": "Metrics contributing to tma_ports_utilization category", + "tma_ports_utilized_0_group": "Metrics contributing to tma_ports_utilized_0 category", + "tma_ports_utilized_3m_group": "Metrics contributing to tma_ports_utilized_3m category", + "tma_retiring_group": "Metrics contributing to tma_retiring category", + "tma_serializing_operation_group": "Metrics contributing to tma_serializing_operation category", + "tma_store_bound_group": "Metrics contributing to tma_store_bound category", + "tma_store_op_utilization_group": "Metrics contributing to tma_store_op_utilization category" +} diff --git a/tools/perf/pmu-events/arch/x86/broadwellx/bdx-metrics.json b/tools/perf/pmu-events/arch/x86/broadwellx/bdx-metrics.json index 65ec0c9e55d1..b319e4edc238 100644 --- a/tools/perf/pmu-events/arch/x86/broadwellx/bdx-metrics.json +++ b/tools/perf/pmu-events/arch/x86/broadwellx/bdx-metrics.json @@ -50,11 +50,207 @@ }, { "BriefDescription": "Uncore frequency per die [GHZ]", - "MetricExpr": "tma_info_socket_clks / #num_dies / duration_time / 1e9", + "MetricExpr": "tma_info_system_socket_clks / #num_dies / duration_time / 1e9", "MetricGroup": "SoC", "MetricName": "UNCORE_FREQ" }, { + "BriefDescription": "Cycles per instruction retired; indicating how much time each executed instruction took; in units of cycles.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD / INST_RETIRED.ANY", + "MetricName": "cpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "CPU operating frequency (in GHz)", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD / CPU_CLK_UNHALTED.REF_TSC * #SYSTEM_TSC_FREQ / 1e9", + "MetricName": "cpu_operating_frequency", + "ScaleUnit": "1GHz" + }, + { + "BriefDescription": "Percentage of time spent in the active CPU power state C0", + "MetricExpr": "tma_info_system_cpu_utilization", + "MetricName": "cpu_utilization", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data loads to the total number of completed instructions", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "dtlb_load_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data loads to the total number of completed instructions. This implies it missed in the DTLB and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data stores to the total number of completed instructions", + "MetricExpr": "DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "dtlb_store_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data stores to the total number of completed instructions. This implies it missed in the DTLB and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Bandwidth of IO reads that are initiated by end device controllers that are requesting memory from the CPU.", + "MetricExpr": "cbox@UNC_C_TOR_INSERTS.OPCODE\\,filter_opc\\=0x19e@ * 64 / 1e6 / duration_time", + "MetricName": "io_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Bandwidth of IO writes that are initiated by end device controllers that are writing memory to the CPU.", + "MetricExpr": "(cbox@UNC_C_TOR_INSERTS.OPCODE\\,filter_opc\\=0x1c8\\,filter_tid\\=0x3e@ + cbox@UNC_C_TOR_INSERTS.OPCODE\\,filter_opc\\=0x180\\,filter_tid\\=0x3e@) * 64 / 1e6 / duration_time", + "MetricName": "io_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for 2 megabyte and 4 megabyte page sizes) caused by a code fetch to the total number of completed instructions", + "MetricExpr": "ITLB_MISSES.WALK_COMPLETED_2M_4M / INST_RETIRED.ANY", + "MetricName": "itlb_large_page_mpi", + "PublicDescription": "Ratio of number of completed page walks (for 2 megabyte and 4 megabyte page sizes) caused by a code fetch to the total number of completed instructions. This implies it missed in the Instruction Translation Lookaside Buffer (ITLB) and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by a code fetch to the total number of completed instructions", + "MetricExpr": "ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "itlb_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by a code fetch to the total number of completed instructions. This implies it missed in the ITLB (Instruction TLB) and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read requests missing in L1 instruction cache (includes prefetches) to the total number of completed instructions", + "MetricExpr": "L2_RQSTS.ALL_CODE_RD / INST_RETIRED.ANY", + "MetricName": "l1_i_code_read_misses_with_prefetches_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of demand load requests hitting in L1 data cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_UOPS_RETIRED.L1_HIT / INST_RETIRED.ANY", + "MetricName": "l1d_demand_data_read_hits_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of requests missing L1 data cache (includes data+rfo w/ prefetches) to the total number of completed instructions", + "MetricExpr": "L1D.REPLACEMENT / INST_RETIRED.ANY", + "MetricName": "l1d_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read request missing L2 cache to the total number of completed instructions", + "MetricExpr": "L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", + "MetricName": "l2_demand_code_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed demand load requests hitting in L2 cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_UOPS_RETIRED.L2_HIT / INST_RETIRED.ANY", + "MetricName": "l2_demand_data_read_hits_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed data read request missing L2 cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_UOPS_RETIRED.L2_MISS / INST_RETIRED.ANY", + "MetricName": "l2_demand_data_read_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of requests missing L2 cache (includes code+data+rfo w/ prefetches) to the total number of completed instructions", + "MetricExpr": "L2_LINES_IN.ALL / INST_RETIRED.ANY", + "MetricName": "l2_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read requests missing last level core cache (includes demand w/ prefetches) to the total number of completed instructions", + "MetricExpr": "(cbox@UNC_C_TOR_INSERTS.MISS_OPCODE\\,filter_opc\\=0x181@ + cbox@UNC_C_TOR_INSERTS.MISS_OPCODE\\,filter_opc\\=0x191@) / INST_RETIRED.ANY", + "MetricName": "llc_code_read_mpi_demand_plus_prefetch", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand and prefetch data read miss (read memory access) in nano seconds", + "MetricExpr": "1e9 * (cbox@UNC_C_TOR_OCCUPANCY.MISS_OPCODE\\,filter_opc\\=0x182@ / cbox@UNC_C_TOR_INSERTS.MISS_OPCODE\\,filter_opc\\=0x182@) / (UNC_C_CLOCKTICKS / (#num_cores / #num_packages * #num_packages)) * duration_time", + "MetricName": "llc_data_read_demand_plus_prefetch_miss_latency", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand and prefetch data read miss (read memory access) addressed to local memory in nano seconds", + "MetricExpr": "1e9 * (cbox@UNC_C_TOR_OCCUPANCY.MISS_LOCAL_OPCODE\\,filter_opc\\=0x182@ / cbox@UNC_C_TOR_INSERTS.MISS_LOCAL_OPCODE\\,filter_opc\\=0x182@) / (UNC_C_CLOCKTICKS / (#num_cores / #num_packages * #num_packages)) * duration_time", + "MetricName": "llc_data_read_demand_plus_prefetch_miss_latency_for_local_requests", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand and prefetch data read miss (read memory access) addressed to remote memory in nano seconds", + "MetricExpr": "1e9 * (cbox@UNC_C_TOR_OCCUPANCY.MISS_REMOTE_OPCODE\\,filter_opc\\=0x182@ / cbox@UNC_C_TOR_INSERTS.MISS_REMOTE_OPCODE\\,filter_opc\\=0x182@) / (UNC_C_CLOCKTICKS / (#num_cores / #num_packages * #num_packages)) * duration_time", + "MetricName": "llc_data_read_demand_plus_prefetch_miss_latency_for_remote_requests", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Ratio of number of data read requests missing last level core cache (includes demand w/ prefetches) to the total number of completed instructions", + "MetricExpr": "(cbox@UNC_C_TOR_INSERTS.MISS_OPCODE\\,filter_opc\\=0x182@ + cbox@UNC_C_TOR_INSERTS.MISS_OPCODE\\,filter_opc\\=0x192@) / INST_RETIRED.ANY", + "MetricName": "llc_data_read_mpi_demand_plus_prefetch", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "The ratio of number of completed memory load instructions to the total number completed instructions", + "MetricExpr": "MEM_UOPS_RETIRED.ALL_LOADS / INST_RETIRED.ANY", + "MetricName": "loads_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "DDR memory read bandwidth (MB/sec)", + "MetricExpr": "UNC_M_CAS_COUNT.RD * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "DDR memory bandwidth (MB/sec)", + "MetricExpr": "(UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_total", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "DDR memory write bandwidth (MB/sec)", + "MetricExpr": "UNC_M_CAS_COUNT.WR * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Memory read that miss the last level cache (LLC) addressed to local DRAM as a percentage of total memory read accesses, does not include LLC prefetches.", + "MetricExpr": "cbox@UNC_C_TOR_INSERTS.MISS_LOCAL_OPCODE\\,filter_opc\\=0x182@ / (cbox@UNC_C_TOR_INSERTS.MISS_LOCAL_OPCODE\\,filter_opc\\=0x182@ + cbox@UNC_C_TOR_INSERTS.MISS_REMOTE_OPCODE\\,filter_opc\\=0x182@)", + "MetricName": "numa_reads_addressed_to_local_dram", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Memory reads that miss the last level cache (LLC) addressed to remote DRAM as a percentage of total memory read accesses, does not include LLC prefetches.", + "MetricExpr": "cbox@UNC_C_TOR_INSERTS.MISS_REMOTE_OPCODE\\,filter_opc\\=0x182@ / (cbox@UNC_C_TOR_INSERTS.MISS_LOCAL_OPCODE\\,filter_opc\\=0x182@ + cbox@UNC_C_TOR_INSERTS.MISS_REMOTE_OPCODE\\,filter_opc\\=0x182@)", + "MetricName": "numa_reads_addressed_to_remote_dram", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from decoded instruction cache (decoded stream buffer or DSB) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.DSB_UOPS / UOPS_ISSUED.ANY", + "MetricName": "percent_uops_delivered_from_decoded_icache", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from legacy decode pipeline (Micro-instruction Translation Engine or MITE) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.MITE_UOPS / UOPS_ISSUED.ANY", + "MetricName": "percent_uops_delivered_from_legacy_decode_pipeline", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from loop stream detector(LSD) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "LSD.UOPS / UOPS_ISSUED.ANY", + "MetricName": "percent_uops_delivered_from_loop_stream_detector", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from microcode sequencer (MS) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.MS_UOPS / UOPS_ISSUED.ANY", + "MetricName": "percent_uops_delivered_from_microcode_sequencer", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Intel(R) Quick Path Interconnect (QPI) data transmit bandwidth (MB/sec)", + "MetricExpr": "UNC_Q_TxL_FLITS_G0.DATA * 8 / 1e6 / duration_time", + "MetricName": "qpi_data_transmit_bw", + "ScaleUnit": "1MB/s" + }, + { "BriefDescription": "Percentage of cycles spent in System Management Interrupts.", "MetricExpr": "((msr@aperf@ - cycles) / msr@aperf@ if msr@smi@ > 0 else 0)", "MetricGroup": "smi", @@ -70,8 +266,14 @@ "ScaleUnit": "1SMI#" }, { + "BriefDescription": "The ratio of number of completed memory store instructions to the total number completed instructions", + "MetricExpr": "MEM_UOPS_RETIRED.ALL_STORES / INST_RETIRED.ANY", + "MetricName": "stores_per_instr", + "ScaleUnit": "1per_instr" + }, + { "BriefDescription": "This metric estimates how often memory load accesses were aliased by preceding stores (in program order) with a 4K address offset", - "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_clks", + "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_4k_aliasing", "MetricThreshold": "tma_4k_aliasing > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -81,7 +283,7 @@ { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution ports for ALU operations.", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5 + UOPS_DISPATCHED_PORT.PORT_6) / tma_info_slots", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5 + UOPS_DISPATCHED_PORT.PORT_6) / tma_info_thread_slots", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_alu_op_utilization", "MetricThreshold": "tma_alu_op_utilization > 0.6", @@ -89,7 +291,7 @@ }, { "BriefDescription": "This metric estimates fraction of slots the CPU retired uops delivered by the Microcode_Sequencer as a result of Assists", - "MetricExpr": "100 * OTHER_ASSISTS.ANY_WB_ASSIST / tma_info_slots", + "MetricExpr": "100 * OTHER_ASSISTS.ANY_WB_ASSIST / tma_info_thread_slots", "MetricGroup": "TopdownL4;tma_L4_group;tma_microcode_sequencer_group", "MetricName": "tma_assists", "MetricThreshold": "tma_assists > 0.1 & (tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1)", @@ -103,15 +305,17 @@ "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend. Backend is the portion of the processor core where the out-of-order scheduler dispatches ready uops into their respective execution units; and once completed these uops get retired according to program order. For example; stalls due to data-cache misses or stalls due to the divider unit being overloaded are both categorized under Backend Bound. Backend Bound is further divided into two main categories: Memory Bound and Core Bound.", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations", - "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_slots", + "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots wasted due to incorrect speculations. This include slots used to issue uops that do not eventually get retired and slots for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work due to miss-predicted branches are categorized under Bad Speculation category. Incorrect data speculation followed by Memory Ordering Nukes is another example.", "ScaleUnit": "100%" }, @@ -122,12 +326,13 @@ "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15", - "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_branch_misprediction_cost, tma_mispredicts_resteers", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers", - "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_clks", + "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteers", "MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -156,7 +361,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to contested accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(60 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) + 43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD)))) / tma_info_clks", + "MetricExpr": "(60 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) + 43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD)))) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_contested_accesses", "MetricThreshold": "tma_contested_accesses > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -170,13 +375,14 @@ "MetricGroup": "Backend;Compute;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck. Shortage in hardware compute resources; or dependencies in software's instructions are both categorized under Core Bound. Hence it may indicate the machine ran out of an out-of-order resource; certain execution units are overloaded or dependencies in program's data- or instruction-flow are limiting the performance (e.g. FP-chained long-latency arithmetic operations).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to data-sharing accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) / tma_info_clks", + "MetricExpr": "43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) / tma_info_thread_clks", "MetricGroup": "Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_data_sharing", "MetricThreshold": "tma_data_sharing > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -185,7 +391,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the Divider unit was active", - "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_clks", + "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_core_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_divider", "MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -195,7 +401,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS)) * CYCLE_ACTIVITY.STALLS_L2_MISS / tma_info_clks", + "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS)) * CYCLE_ACTIVITY.STALLS_L2_MISS / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -204,25 +410,25 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline", - "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSB;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_dsb", - "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline. For example; inefficient utilization of the DSB cache structure or bank conflict when reading from it; are categorized here.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_dsb_switches", "MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb, tma_lcp", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses", - "MetricExpr": "(8 * DTLB_LOAD_MISSES.STLB_HIT + cpu@DTLB_LOAD_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * DTLB_LOAD_MISSES.WALK_COMPLETED) / tma_info_clks", + "MetricExpr": "(8 * DTLB_LOAD_MISSES.STLB_HIT + cpu@DTLB_LOAD_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * DTLB_LOAD_MISSES.WALK_COMPLETED) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group", "MetricName": "tma_dtlb_load", "MetricThreshold": "tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -231,7 +437,7 @@ }, { "BriefDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses", - "MetricExpr": "(8 * DTLB_STORE_MISSES.STLB_HIT + cpu@DTLB_STORE_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * DTLB_STORE_MISSES.WALK_COMPLETED) / tma_info_clks", + "MetricExpr": "(8 * DTLB_STORE_MISSES.STLB_HIT + cpu@DTLB_STORE_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * DTLB_STORE_MISSES.WALK_COMPLETED) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_store_bound_group", "MetricName": "tma_dtlb_store", "MetricThreshold": "tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -240,7 +446,7 @@ }, { "BriefDescription": "This metric roughly estimates how often CPU was handling synchronizations due to False Sharing", - "MetricExpr": "(200 * OFFCORE_RESPONSE.DEMAND_RFO.LLC_MISS.REMOTE_HITM + 60 * OFFCORE_RESPONSE.DEMAND_RFO.LLC_HIT.HITM_OTHER_CORE) / tma_info_clks", + "MetricExpr": "(200 * OFFCORE_RESPONSE.DEMAND_RFO.LLC_MISS.REMOTE_HITM + 60 * OFFCORE_RESPONSE.DEMAND_RFO.LLC_HIT.HITM_OTHER_CORE) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_store_bound_group", "MetricName": "tma_false_sharing", "MetricThreshold": "tma_false_sharing > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -250,11 +456,11 @@ { "BriefDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "tma_info_load_miss_real_latency * cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@ / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@ / tma_info_thread_clks", "MetricGroup": "MemoryBW;TopdownL4;tma_L4_group;tma_issueBW;tma_issueSL;tma_issueSmSt;tma_l1_bound_group", "MetricName": "tma_fb_full", "MetricThreshold": "tma_fb_full > 0.3", - "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", + "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", "ScaleUnit": "100%" }, { @@ -262,16 +468,18 @@ "MetricExpr": "tma_frontend_bound - tma_fetch_latency", "MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", "MetricName": "tma_fetch_bandwidth", - "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_info_iptb, tma_lcp", + "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues", - "MetricExpr": "4 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE / tma_info_slots", + "MetricExpr": "4 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE / tma_info_thread_slots", "MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: RS_EVENTS.EMPTY_END", "ScaleUnit": "100%" }, @@ -322,10 +530,11 @@ }, { "BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend", - "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_slots", + "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_thread_slots", "MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound.", "ScaleUnit": "100%" }, @@ -335,441 +544,442 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_heavy_operations", "MetricThreshold": "tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to instruction cache misses.", - "MetricExpr": "ICACHE.IFDATA_STALL / tma_info_clks", + "MetricExpr": "ICACHE.IFDATA_STALL / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;IcMiss;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_icache_misses", "MetricThreshold": "tma_icache_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", "ScaleUnit": "100%" }, { - "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", - "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time", - "MetricGroup": "Power;Summary", - "MetricName": "tma_info_average_frequency" - }, - { - "BriefDescription": "Branch instructions per taken branch.", - "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", - "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_bptkbranch" - }, - { "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", - "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_slots / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_thread_slots / BR_MISP_RETIRED.ALL_BRANCHES", "MetricGroup": "Bad;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_branch_misprediction_cost", + "MetricName": "tma_info_bad_spec_branch_misprediction_cost", "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_mispredicts_resteers" }, { - "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "Pipeline", - "MetricName": "tma_info_clks" + "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", + "MetricExpr": "tma_info_inst_mix_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_indirect", + "MetricThreshold": "tma_info_bad_spec_ipmisp_indirect < 1e3" + }, + { + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BadSpec;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmispredict", + "MetricThreshold": "tma_info_bad_spec_ipmispredict < 200" }, { "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", - "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_clks))", + "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_thread_clks))", "MetricGroup": "SMT", - "MetricName": "tma_info_core_clks" + "MetricName": "tma_info_core_core_clks" }, { "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks", "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_coreipc" + "MetricName": "tma_info_core_coreipc" }, { - "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricExpr": "1 / tma_info_ipc", - "MetricGroup": "Mem;Pipeline", - "MetricName": "tma_info_cpi" - }, - { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": "HPC;Summary", - "MetricName": "tma_info_cpu_utilization" + "BriefDescription": "Floating Point Operations Per Cycle", + "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / tma_info_core_core_clks", + "MetricGroup": "Flops;Ret", + "MetricName": "tma_info_core_flopc" }, { - "BriefDescription": "Average Parallel L2 cache miss data reads", - "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", - "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_data_l2_mlp" + "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", + "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@) / (2 * tma_info_core_core_clks)", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_core_fp_arith_utilization", + "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." }, { - "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time", - "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", - "MetricName": "tma_info_dram_bw_use", - "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", + "MetricExpr": "UOPS_EXECUTED.THREAD / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC)", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", + "MetricName": "tma_info_core_ilp" }, { "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", "MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + LSD.UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)", "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", - "MetricName": "tma_info_dsb_coverage", - "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_iptb, tma_lcp" - }, - { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", - "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", - "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", - "MetricName": "tma_info_execute" + "MetricName": "tma_info_frontend_dsb_coverage", + "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 4 > 0.35", + "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_inst_mix_iptb, tma_lcp" }, { - "BriefDescription": "The ratio of Executed- by Issued-Uops", - "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", - "MetricGroup": "Cor;Pipeline", - "MetricName": "tma_info_execute_per_issue", - "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." - }, - { - "BriefDescription": "Floating Point Operations Per Cycle", - "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / tma_info_core_clks", - "MetricGroup": "Flops;Ret", - "MetricName": "tma_info_flopc" - }, - { - "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", - "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@) / (2 * tma_info_core_clks)", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_fp_arith_utilization", - "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." - }, - { - "BriefDescription": "Giga Floating Point Operations Per Second", - "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / 1e9 / duration_time", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_gflops", - "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_inst_mix_instructions / BACLEARS.ANY", + "MetricGroup": "Fed", + "MetricName": "tma_info_frontend_ipunknown_branch" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", - "MetricExpr": "UOPS_EXECUTED.THREAD / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC)", - "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", - "MetricName": "tma_info_ilp" + "BriefDescription": "Branch instructions per taken branch.", + "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;PGO", + "MetricName": "tma_info_inst_mix_bptkbranch" }, { "BriefDescription": "Total number of retired Instructions", "MetricExpr": "INST_RETIRED.ANY", "MetricGroup": "Summary;TmaL1;tma_L1_group", - "MetricName": "tma_info_instructions", + "MetricName": "tma_info_inst_mix_instructions", "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" }, { "BriefDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_iparith", - "MetricThreshold": "tma_info_iparith < 10", + "MetricName": "tma_info_inst_mix_iparith", + "MetricThreshold": "tma_info_inst_mix_iparith < 10", "PublicDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate). May undercount due to FMA double counting. Approximated prior to BDW." }, { "BriefDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx128", - "MetricThreshold": "tma_info_iparith_avx128 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx128", + "MetricThreshold": "tma_info_inst_mix_iparith_avx128 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx256", - "MetricThreshold": "tma_info_iparith_avx256 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx256", + "MetricThreshold": "tma_info_inst_mix_iparith_avx256 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_dp", - "MetricThreshold": "tma_info_iparith_scalar_dp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_dp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_dp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_SINGLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_sp", - "MetricThreshold": "tma_info_iparith_scalar_sp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_sp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_sp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per Branch (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Branches;Fed;InsType", - "MetricName": "tma_info_ipbranch", - "MetricThreshold": "tma_info_ipbranch < 8" - }, - { - "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": "Ret;Summary", - "MetricName": "tma_info_ipc" + "MetricName": "tma_info_inst_mix_ipbranch", + "MetricThreshold": "tma_info_inst_mix_ipbranch < 8" }, { "BriefDescription": "Instructions per (near) call (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_CALL", "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_ipcall", - "MetricThreshold": "tma_info_ipcall < 200" - }, - { - "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", - "MetricGroup": "Branches;OS", - "MetricName": "tma_info_ipfarbranch", - "MetricThreshold": "tma_info_ipfarbranch < 1e6" + "MetricName": "tma_info_inst_mix_ipcall", + "MetricThreshold": "tma_info_inst_mix_ipcall < 200" }, { "BriefDescription": "Instructions per Floating Point (FP) Operation (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_ipflop", - "MetricThreshold": "tma_info_ipflop < 10" + "MetricName": "tma_info_inst_mix_ipflop", + "MetricThreshold": "tma_info_inst_mix_ipflop < 10" }, { "BriefDescription": "Instructions per Load (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_LOADS", "MetricGroup": "InsType", - "MetricName": "tma_info_ipload", - "MetricThreshold": "tma_info_ipload < 3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", - "MetricExpr": "tma_info_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_indirect", - "MetricThreshold": "tma_info_ipmisp_indirect < 1e3" - }, - { - "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BadSpec;BrMispredicts", - "MetricName": "tma_info_ipmispredict", - "MetricThreshold": "tma_info_ipmispredict < 200" + "MetricName": "tma_info_inst_mix_ipload", + "MetricThreshold": "tma_info_inst_mix_ipload < 3" }, { "BriefDescription": "Instructions per Store (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_STORES", "MetricGroup": "InsType", - "MetricName": "tma_info_ipstore", - "MetricThreshold": "tma_info_ipstore < 8" + "MetricName": "tma_info_inst_mix_ipstore", + "MetricThreshold": "tma_info_inst_mix_ipstore < 8" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW;Frontend;PGO;tma_issueFB", - "MetricName": "tma_info_iptb", - "MetricThreshold": "tma_info_iptb < 9", - "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_lcp" - }, - { - "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", - "MetricExpr": "tma_info_instructions / BACLEARS.ANY", - "MetricGroup": "Fed", - "MetricName": "tma_info_ipunknown_branch" - }, - { - "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_cpi" - }, - { - "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_utilization", - "MetricThreshold": "tma_info_kernel_utilization > 0.05" + "MetricName": "tma_info_inst_mix_iptb", + "MetricThreshold": "tma_info_inst_mix_iptb < 9", + "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_lcp" }, { "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw" - }, - { - "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "tma_info_l1d_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw_1t" - }, - { - "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L1_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki" + "MetricName": "tma_info_memory_core_l1d_cache_fill_bw" }, { "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw" + "MetricName": "tma_info_memory_core_l2_cache_fill_bw" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "tma_info_l2_cache_fill_bw", + "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw_1t" + "MetricName": "tma_info_memory_core_l3_cache_fill_bw" + }, + { + "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L1_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l1mpki" }, { "BriefDescription": "L2 cache hits per kilo instruction for all request types (including speculative)", "MetricExpr": "1e3 * (L2_RQSTS.REFERENCES - L2_RQSTS.MISS) / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_all" + "MetricName": "tma_info_memory_l2hpki_all" }, { "BriefDescription": "L2 cache hits per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_HIT / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_load" + "MetricName": "tma_info_memory_l2hpki_load" }, { "BriefDescription": "L2 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L2_MISS / INST_RETIRED.ANY", "MetricGroup": "Backend;CacheMisses;Mem", - "MetricName": "tma_info_l2mpki" + "MetricName": "tma_info_memory_l2mpki" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all request types (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem;Offcore", - "MetricName": "tma_info_l2mpki_all" + "MetricName": "tma_info_memory_l2mpki_all" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2mpki_load" + "MetricName": "tma_info_memory_l2mpki_load" }, { - "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "0", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw_1t" + "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L3_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l3mpki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw" + "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_UOPS_RETIRED.L1_MISS + MEM_LOAD_UOPS_RETIRED.HIT_LFB)", + "MetricGroup": "Mem;MemoryBound;MemoryLat", + "MetricName": "tma_info_memory_load_miss_real_latency" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw_1t" + "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", + "MetricGroup": "Mem;MemoryBW;MemoryBound", + "MetricName": "tma_info_memory_mlp", + "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" }, { - "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L3_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l3mpki" + "BriefDescription": "Average Parallel L2 cache miss data reads", + "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", + "MetricGroup": "Memory_BW;Offcore", + "MetricName": "tma_info_memory_oro_data_l2_mlp" }, { "BriefDescription": "Average Latency for L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS.DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l2_miss_latency" + "MetricName": "tma_info_memory_oro_load_l2_miss_latency" }, { "BriefDescription": "Average Parallel L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_DATA_RD", "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_load_l2_mlp" + "MetricName": "tma_info_memory_oro_load_l2_mlp" }, { - "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_UOPS_RETIRED.L1_MISS + MEM_LOAD_UOPS_RETIRED.HIT_LFB)", - "MetricGroup": "Mem;MemoryBound;MemoryLat", - "MetricName": "tma_info_load_miss_real_latency" + "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l1d_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l1d_cache_fill_bw_1t" }, { - "BriefDescription": "Average number of parallel data read requests to external memory", - "MetricExpr": "UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182\\,thresh\\=1@", - "MetricGroup": "Mem;MemoryBW;SoC", - "MetricName": "tma_info_mem_parallel_reads", - "PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches" + "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l2_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l2_cache_fill_bw_1t" }, { - "BriefDescription": "Average latency of data read request to external memory (in nanoseconds)", - "MetricExpr": "1e9 * (UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_INSERTS.MISS_OPCODE@filter_opc\\=0x182@) / (tma_info_socket_clks / duration_time)", - "MetricGroup": "Mem;MemoryLat;SoC", - "MetricName": "tma_info_mem_read_latency", - "PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)" + "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "0", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_thread_l3_cache_access_bw_1t" }, { - "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", - "MetricGroup": "Mem;MemoryBW;MemoryBound", - "MetricName": "tma_info_mlp", - "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" + "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l3_cache_fill_bw_1t" }, { "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", - "MetricExpr": "(ITLB_MISSES.WALK_DURATION + DTLB_LOAD_MISSES.WALK_DURATION + DTLB_STORE_MISSES.WALK_DURATION + 7 * (DTLB_STORE_MISSES.WALK_COMPLETED + DTLB_LOAD_MISSES.WALK_COMPLETED + ITLB_MISSES.WALK_COMPLETED)) / (2 * tma_info_core_clks)", + "MetricExpr": "(ITLB_MISSES.WALK_DURATION + DTLB_LOAD_MISSES.WALK_DURATION + DTLB_STORE_MISSES.WALK_DURATION + 7 * (DTLB_STORE_MISSES.WALK_COMPLETED + DTLB_LOAD_MISSES.WALK_COMPLETED + ITLB_MISSES.WALK_COMPLETED)) / (2 * tma_info_core_core_clks)", "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_page_walks_utilization", - "MetricThreshold": "tma_info_page_walks_utilization > 0.5" + "MetricName": "tma_info_memory_tlb_page_walks_utilization", + "MetricThreshold": "tma_info_memory_tlb_page_walks_utilization > 0.5" + }, + { + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", + "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", + "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", + "MetricName": "tma_info_pipeline_execute" }, { "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@", "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_retire" + "MetricName": "tma_info_pipeline_retire" }, { - "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", - "MetricExpr": "4 * tma_info_core_clks", - "MetricGroup": "TmaL1;tma_L1_group", - "MetricName": "tma_info_slots" + "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time", + "MetricGroup": "Power;Summary", + "MetricName": "tma_info_system_average_frequency" + }, + { + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricGroup": "HPC;Summary", + "MetricName": "tma_info_system_cpu_utilization" + }, + { + "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", + "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time", + "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", + "MetricName": "tma_info_system_dram_bw_use", + "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_mem_bandwidth, tma_sq_full" + }, + { + "BriefDescription": "Giga Floating Point Operations Per Second", + "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / 1e9 / duration_time", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_system_gflops", + "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + }, + { + "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", + "MetricGroup": "Branches;OS", + "MetricName": "tma_info_system_ipfarbranch", + "MetricThreshold": "tma_info_system_ipfarbranch < 1e6" + }, + { + "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_cpi" + }, + { + "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_utilization", + "MetricThreshold": "tma_info_system_kernel_utilization > 0.05" + }, + { + "BriefDescription": "Average number of parallel data read requests to external memory", + "MetricExpr": "UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182\\,thresh\\=1@", + "MetricGroup": "Mem;MemoryBW;SoC", + "MetricName": "tma_info_system_mem_parallel_reads", + "PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches" + }, + { + "BriefDescription": "Average latency of data read request to external memory (in nanoseconds)", + "MetricExpr": "1e9 * (UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_INSERTS.MISS_OPCODE@filter_opc\\=0x182@) / (tma_info_system_socket_clks / duration_time)", + "MetricGroup": "Mem;MemoryLat;SoC", + "MetricName": "tma_info_system_mem_read_latency", + "PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)" }, { "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / (CPU_CLK_UNHALTED.REF_XCLK_ANY / 2) if #SMT_on else 0)", "MetricGroup": "SMT", - "MetricName": "tma_info_smt_2t_utilization" + "MetricName": "tma_info_system_smt_2t_utilization" }, { "BriefDescription": "Socket actual clocks when any core is active on that socket", "MetricExpr": "cbox_0@event\\=0x0@", "MetricGroup": "SoC", - "MetricName": "tma_info_socket_clks" + "MetricName": "tma_info_system_socket_clks" }, { "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC", "MetricGroup": "Power", - "MetricName": "tma_info_turbo_utilization" + "MetricName": "tma_info_system_turbo_utilization" + }, + { + "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "Pipeline", + "MetricName": "tma_info_thread_clks" + }, + { + "BriefDescription": "Cycles Per Instruction (per Logical Processor)", + "MetricExpr": "1 / tma_info_thread_ipc", + "MetricGroup": "Mem;Pipeline", + "MetricName": "tma_info_thread_cpi" + }, + { + "BriefDescription": "The ratio of Executed- by Issued-Uops", + "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", + "MetricGroup": "Cor;Pipeline", + "MetricName": "tma_info_thread_execute_per_issue", + "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." + }, + { + "BriefDescription": "Instructions Per Cycle (per Logical Processor)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks", + "MetricGroup": "Ret;Summary", + "MetricName": "tma_info_thread_ipc" + }, + { + "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", + "MetricExpr": "4 * tma_info_core_core_clks", + "MetricGroup": "TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots" }, { "BriefDescription": "Uops Per Instruction", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / INST_RETIRED.ANY", "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_uoppi", - "MetricThreshold": "tma_info_uoppi > 1.05" + "MetricName": "tma_info_thread_uoppi", + "MetricThreshold": "tma_info_thread_uoppi > 1.05" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW", - "MetricName": "tma_info_uptb", - "MetricThreshold": "tma_info_uptb < 6" + "MetricName": "tma_info_thread_uptb", + "MetricThreshold": "tma_info_thread_uptb < 6" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses", - "MetricExpr": "(14 * ITLB_MISSES.STLB_HIT + cpu@ITLB_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * ITLB_MISSES.WALK_COMPLETED) / tma_info_clks", + "MetricExpr": "(14 * ITLB_MISSES.STLB_HIT + cpu@ITLB_MISSES.WALK_DURATION\\,cmask\\=1@ + 7 * ITLB_MISSES.WALK_COMPLETED) / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -778,7 +988,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled without loads missing the L1 data cache", - "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_clks, 0)", + "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_thread_clks, 0)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_issueL1;tma_issueMC;tma_memory_bound_group", "MetricName": "tma_l1_bound", "MetricThreshold": "tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -787,7 +997,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled due to L2 cache accesses by loads", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l2_bound", "MetricThreshold": "tma_l2_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -797,7 +1007,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS) * CYCLE_ACTIVITY.STALLS_L2_MISS / tma_info_clks", + "MetricExpr": "MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS) * CYCLE_ACTIVITY.STALLS_L2_MISS / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -807,7 +1017,7 @@ { "BriefDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "41 * (MEM_LOAD_UOPS_RETIRED.L3_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) / tma_info_clks", + "MetricExpr": "41 * (MEM_LOAD_UOPS_RETIRED.L3_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) / tma_info_thread_clks", "MetricGroup": "MemoryLat;TopdownL4;tma_L4_group;tma_issueLat;tma_l3_bound_group", "MetricName": "tma_l3_hit_latency", "MetricThreshold": "tma_l3_hit_latency > 0.1 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -816,11 +1026,11 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)", - "MetricExpr": "ILD_STALL.LCP / tma_info_clks", + "MetricExpr": "ILD_STALL.LCP / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_lcp", "MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb", + "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb", "ScaleUnit": "100%" }, { @@ -829,13 +1039,14 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_light_operations", "MetricThreshold": "tma_light_operations > 0.6", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation). This correlates with total number of instructions used by the program. A uops-per-instruction (see UopPI metric) ratio of 1 or less should be expected for decently optimized software running on Intel Core/Xeon products. While this often indicates efficient X86 instructions were executed; high value does not necessarily mean better performance cannot be achieved. Sample with: INST_RETIRED.PREC_DIST", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Load operations", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 + UOPS_DISPATCHED_PORT.PORT_7 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 + UOPS_DISPATCHED_PORT.PORT_7 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_load_op_utilization", "MetricThreshold": "tma_load_op_utilization > 0.6", @@ -845,7 +1056,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from local memory", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "200 * (MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) / tma_info_clks", + "MetricExpr": "200 * (MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) / tma_info_thread_clks", "MetricGroup": "Server;TopdownL5;tma_L5_group;tma_mem_latency_group", "MetricName": "tma_local_dram", "MetricThreshold": "tma_local_dram > 0.1 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -855,7 +1066,7 @@ { "BriefDescription": "This metric represents fraction of cycles the CPU spent handling cache misses due to lock operations", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO) / tma_info_clks", + "MetricExpr": "MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO) / tma_info_thread_clks", "MetricGroup": "Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_l1_bound_group", "MetricName": "tma_lock_latency", "MetricThreshold": "tma_lock_latency > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -869,21 +1080,22 @@ "MetricGroup": "BadSpec;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.1 & tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears. These slots are either wasted by uops fetched prior to the clear; or stalls the out-of-order portion of the machine needs to recover its state after the clear. For example; this can happen due to memory ordering Nukes (e.g. Memory Disambiguation) or Self-Modifying-Code (SMC) nukes. Sample with: MACHINE_CLEARS.COUNT. Related metrics: tma_clears_resteers, tma_contested_accesses, tma_data_sharing, tma_false_sharing, tma_l1_bound, tma_microcode_sequencer, tma_ms_switches, tma_remote_cache", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_clks", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW", "MetricName": "tma_mem_bandwidth", "MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_sq_full", + "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_sq_full", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat", "MetricName": "tma_mem_latency", "MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -893,16 +1105,17 @@ { "BriefDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_MEM_ANY + RESOURCE_STALLS.SB) / (CYCLE_ACTIVITY.STALLS_TOTAL + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_MEM_ANY + RESOURCE_STALLS.SB) / (CYCLE_ACTIVITY.STALLS_TOTAL + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_thread_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", "MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck. Memory Bound estimates fraction of slots where pipeline is likely stalled due to demand load or store instructions. This accounts mainly for (1) non-completed in-flight memory demand loads which coincides with execution units starvation; in addition to (2) cases where stores could impose backpressure on the pipeline when many of them get buffered at the same time (less common out of the two).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_thread_slots", "MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS", "MetricName": "tma_microcode_sequencer", "MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1", @@ -915,21 +1128,21 @@ "MetricGroup": "BadSpec;BrMispredicts;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueBM", "MetricName": "tma_mispredicts_resteers", "MetricThreshold": "tma_mispredicts_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline)", - "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_mite", - "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline). This pipeline is used for code that was not pre-cached in the DSB or LSD. For example; inefficiencies due to asymmetric decoders; use of long immediate or LCP can manifest as MITE fetch bandwidth bottleneck.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)", - "MetricExpr": "2 * IDQ.MS_SWITCHES / tma_info_clks", + "MetricExpr": "2 * IDQ.MS_SWITCHES / tma_info_thread_clks", "MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO", "MetricName": "tma_ms_switches", "MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -938,7 +1151,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 0 ([SNB+] ALU; [HSW+] ALU and 2nd branch)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_core_clks", "MetricGroup": "Compute;TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_0", "MetricThreshold": "tma_port_0 > 0.6", @@ -947,7 +1160,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 1 (ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_1", "MetricThreshold": "tma_port_1 > 0.6", @@ -956,7 +1169,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 2 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_2", "MetricThreshold": "tma_port_2 > 0.6", @@ -965,7 +1178,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 3 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_3", "MetricThreshold": "tma_port_3 > 0.6", @@ -983,7 +1196,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 5 ([SNB+] Branches and ALU; [HSW+] ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_5", "MetricThreshold": "tma_port_5 > 0.6", @@ -992,7 +1205,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 6 ([HSW+]Primary Branch and simple ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_6 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_6 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_6", "MetricThreshold": "tma_port_6 > 0.6", @@ -1001,7 +1214,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 7 ([HSW+]simple Store-address)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_7 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_7 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_store_op_utilization_group", "MetricName": "tma_port_7", "MetricThreshold": "tma_port_7 > 0.6", @@ -1011,7 +1224,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_TOTAL + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_clks", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_TOTAL + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_thread_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_ports_utilization", "MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -1020,7 +1233,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed no uops on any execution port (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,inv\\,cmask\\=1@ / 2 if #SMT_on else (CYCLE_ACTIVITY.STALLS_TOTAL - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0)) / tma_info_core_clks)", + "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,inv\\,cmask\\=1@ / 2 if #SMT_on else (CYCLE_ACTIVITY.STALLS_TOTAL - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0)) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_0", "MetricThreshold": "tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1029,7 +1242,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the CPU executed total of 1 uop per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) / tma_info_core_clks)", + "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issueL1;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_1", "MetricThreshold": "tma_ports_utilized_1 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1038,7 +1251,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 2 uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC - UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_clks)", + "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC - UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issue2P;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_2", "MetricThreshold": "tma_ports_utilized_2 > 0.15 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1047,7 +1260,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 3 or more uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise).", - "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_clks", + "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_3m", "MetricThreshold": "tma_ports_utilized_3m > 0.7 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1056,7 +1269,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from remote cache in other sockets including synchronizations issues", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(200 * (MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) + 180 * (MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD)))) / tma_info_clks", + "MetricExpr": "(200 * (MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) + 180 * (MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD)))) / tma_info_thread_clks", "MetricGroup": "Offcore;Server;Snoop;TopdownL5;tma_L5_group;tma_issueSyncxn;tma_mem_latency_group", "MetricName": "tma_remote_cache", "MetricThreshold": "tma_remote_cache > 0.05 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1066,7 +1279,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from remote memory", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "310 * (MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) / tma_info_clks", + "MetricExpr": "310 * (MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) / tma_info_thread_clks", "MetricGroup": "Server;Snoop;TopdownL5;tma_L5_group;tma_mem_latency_group", "MetricName": "tma_remote_dram", "MetricThreshold": "tma_remote_dram > 0.1 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1075,17 +1288,18 @@ }, { "BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired. Ideally; all pipeline slots would be attributed to the Retiring category. Retiring of 100% would indicate the maximum Pipeline_Width throughput was achieved. Maximizing Retiring typically increases the Instructions-per-cycle (see IPC metric). Note that a high Retiring value does not necessary mean there is no room for more performance. For example; Heavy-operations or Microcode Assists are categorized under Retiring. They often indicate suboptimal performance and can often be optimized or avoided. Sample with: UOPS_RETIRED.RETIRE_SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles handling memory load split accesses - load that cross 64-byte cache line boundary", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "tma_info_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_split_loads", "MetricThreshold": "tma_split_loads > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1094,7 +1308,7 @@ }, { "BriefDescription": "This metric represents rate of split store accesses", - "MetricExpr": "2 * MEM_UOPS_RETIRED.SPLIT_STORES / tma_info_core_clks", + "MetricExpr": "2 * MEM_UOPS_RETIRED.SPLIT_STORES / tma_info_core_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_issueSpSt;tma_store_bound_group", "MetricName": "tma_split_stores", "MetricThreshold": "tma_split_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1103,16 +1317,16 @@ }, { "BriefDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors)", - "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_clks", + "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_core_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueBW;tma_l3_bound_group", "MetricName": "tma_sq_full", "MetricThreshold": "tma_sq_full > 0.3 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_mem_bandwidth", + "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_mem_bandwidth", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write", - "MetricExpr": "RESOURCE_STALLS.SB / tma_info_clks", + "MetricExpr": "RESOURCE_STALLS.SB / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1121,7 +1335,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of cycles when the memory subsystem had loads blocked since they could not forward data from earlier (in program order) overlapping stores", - "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_clks", + "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_store_fwd_blk", "MetricThreshold": "tma_store_fwd_blk > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1131,7 +1345,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU spent handling L1D store misses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(L2_RQSTS.RFO_HIT * 9 * (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) + (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_clks", + "MetricExpr": "(L2_RQSTS.RFO_HIT * 9 * (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) + (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_thread_clks", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_issueSL;tma_store_bound_group", "MetricName": "tma_store_latency", "MetricThreshold": "tma_store_latency > 0.1 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1140,7 +1354,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Store operations", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_core_clks", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_store_op_utilization", "MetricThreshold": "tma_store_op_utilization > 0.6", @@ -1157,11 +1371,17 @@ }, { "BriefDescription": "This metric serves as an approximation of legacy x87 usage", - "MetricExpr": "INST_RETIRED.X87 * tma_info_uoppi / UOPS_RETIRED.RETIRE_SLOTS", + "MetricExpr": "INST_RETIRED.X87 * tma_info_thread_uoppi / UOPS_RETIRED.RETIRE_SLOTS", "MetricGroup": "Compute;TopdownL4;tma_L4_group;tma_fp_arith_group", "MetricName": "tma_x87_use", "MetricThreshold": "tma_x87_use > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6)", "PublicDescription": "This metric serves as an approximation of legacy x87 usage. It accounts for instructions beyond X87 FP arithmetic operations; hence may be used as a thermometer to avoid X87 high usage and preferably upgrade to modern ISA. See Tip under Tuning Hint.", "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uncore operating frequency in GHz", + "MetricExpr": "UNC_C_CLOCKTICKS / (#num_cores / #num_packages * #num_packages) / 1e9 / duration_time", + "MetricName": "uncore_frequency", + "ScaleUnit": "1GHz" } ] diff --git a/tools/perf/pmu-events/arch/x86/broadwellx/floating-point.json b/tools/perf/pmu-events/arch/x86/broadwellx/floating-point.json index e4826dc7f797..986869252e71 100644 --- a/tools/perf/pmu-events/arch/x86/broadwellx/floating-point.json +++ b/tools/perf/pmu-events/arch/x86/broadwellx/floating-point.json @@ -32,6 +32,14 @@ "UMask": "0x20" }, { + "BriefDescription": "Number of SSE/AVX computational 128-bit packed single and 256-bit packed double precision FP instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, 1 for each element. Applies to SSE* and AVX* packed single precision and packed double precision FP instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB count twice as they perform 2 calculations per element.", + "EventCode": "0xc7", + "EventName": "FP_ARITH_INST_RETIRED.4_FLOPS", + "PublicDescription": "Number of SSE/AVX computational 128-bit packed single precision and 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point and packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", + "SampleAfterValue": "2000003", + "UMask": "0x18" + }, + { "BriefDescription": "Number of SSE/AVX computational double precision floating-point instructions retired; some instructions will count twice as noted below. Applies to SSE* and AVX* scalar and packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform multiple calculations per element.", "EventCode": "0xc7", "EventName": "FP_ARITH_INST_RETIRED.DOUBLE", @@ -77,6 +85,13 @@ "UMask": "0x2a" }, { + "BriefDescription": "Number of any Vector retired FP arithmetic instructions", + "EventCode": "0xc7", + "EventName": "FP_ARITH_INST_RETIRED.VECTOR", + "SampleAfterValue": "2000003", + "UMask": "0xfc" + }, + { "BriefDescription": "Cycles with any input/output SSE or FP assist", "CounterMask": "1", "EventCode": "0xCA", diff --git a/tools/perf/pmu-events/arch/x86/broadwellx/metricgroups.json b/tools/perf/pmu-events/arch/x86/broadwellx/metricgroups.json new file mode 100644 index 000000000000..f6a0258e3241 --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/broadwellx/metricgroups.json @@ -0,0 +1,107 @@ +{ + "Backend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Bad": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BadSpec": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BigFoot": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BrMispredicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Branches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CacheMisses": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Compute": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Cor": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSBmiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DataSharing": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Fed": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Flops": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpScalar": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Frontend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "HPC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IcMiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "InsType": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "L2Evicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "LSD": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MachineClears": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBound": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryTLB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_BW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_Lat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MicroSeq": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "OS": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Offcore": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PGO": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Pipeline": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PortsUtil": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Ret": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Retire": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SMT": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Server": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Snoop": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SoC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL1": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL2": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL3mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "TopdownL5": "Metrics for top-down breakdown at level 5", + "TopdownL6": "Metrics for top-down breakdown at level 6", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_L5_group": "Metrics for top-down breakdown at level 5", + "tma_L6_group": "Metrics for top-down breakdown at level 6", + "tma_alu_op_utilization_group": "Metrics contributing to tma_alu_op_utilization category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_branch_resteers_group": "Metrics contributing to tma_branch_resteers category", + "tma_core_bound_group": "Metrics contributing to tma_core_bound category", + "tma_dram_bound_group": "Metrics contributing to tma_dram_bound category", + "tma_dtlb_load_group": "Metrics contributing to tma_dtlb_load category", + "tma_dtlb_store_group": "Metrics contributing to tma_dtlb_store category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_fp_arith_group": "Metrics contributing to tma_fp_arith category", + "tma_fp_vector_group": "Metrics contributing to tma_fp_vector category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_heavy_operations_group": "Metrics contributing to tma_heavy_operations category", + "tma_issue2P": "Metrics related by the issue $issue2P", + "tma_issueBM": "Metrics related by the issue $issueBM", + "tma_issueBW": "Metrics related by the issue $issueBW", + "tma_issueFB": "Metrics related by the issue $issueFB", + "tma_issueL1": "Metrics related by the issue $issueL1", + "tma_issueLat": "Metrics related by the issue $issueLat", + "tma_issueMC": "Metrics related by the issue $issueMC", + "tma_issueMS": "Metrics related by the issue $issueMS", + "tma_issueMV": "Metrics related by the issue $issueMV", + "tma_issueRFO": "Metrics related by the issue $issueRFO", + "tma_issueSL": "Metrics related by the issue $issueSL", + "tma_issueSO": "Metrics related by the issue $issueSO", + "tma_issueSmSt": "Metrics related by the issue $issueSmSt", + "tma_issueSpSt": "Metrics related by the issue $issueSpSt", + "tma_issueSyncxn": "Metrics related by the issue $issueSyncxn", + "tma_issueTLB": "Metrics related by the issue $issueTLB", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_l3_bound_group": "Metrics contributing to tma_l3_bound category", + "tma_light_operations_group": "Metrics contributing to tma_light_operations category", + "tma_load_op_utilization_group": "Metrics contributing to tma_load_op_utilization category", + "tma_mem_latency_group": "Metrics contributing to tma_mem_latency category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_microcode_sequencer_group": "Metrics contributing to tma_microcode_sequencer category", + "tma_ports_utilization_group": "Metrics contributing to tma_ports_utilization category", + "tma_ports_utilized_0_group": "Metrics contributing to tma_ports_utilized_0 category", + "tma_ports_utilized_3m_group": "Metrics contributing to tma_ports_utilized_3m category", + "tma_retiring_group": "Metrics contributing to tma_retiring category", + "tma_serializing_operation_group": "Metrics contributing to tma_serializing_operation category", + "tma_store_bound_group": "Metrics contributing to tma_store_bound category", + "tma_store_op_utilization_group": "Metrics contributing to tma_store_op_utilization category" +} diff --git a/tools/perf/pmu-events/arch/x86/cascadelakex/clx-metrics.json b/tools/perf/pmu-events/arch/x86/cascadelakex/clx-metrics.json index 8f7dc72accd0..0e2e446ced7a 100644 --- a/tools/perf/pmu-events/arch/x86/cascadelakex/clx-metrics.json +++ b/tools/perf/pmu-events/arch/x86/cascadelakex/clx-metrics.json @@ -50,11 +50,238 @@ }, { "BriefDescription": "Uncore frequency per die [GHZ]", - "MetricExpr": "tma_info_socket_clks / #num_dies / duration_time / 1e9", + "MetricExpr": "tma_info_system_socket_clks / #num_dies / duration_time / 1e9", "MetricGroup": "SoC", "MetricName": "UNCORE_FREQ" }, { + "BriefDescription": "Cycles per instruction retired; indicating how much time each executed instruction took; in units of cycles.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD / INST_RETIRED.ANY", + "MetricName": "cpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "CPU operating frequency (in GHz)", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD / CPU_CLK_UNHALTED.REF_TSC * #SYSTEM_TSC_FREQ / 1e9", + "MetricName": "cpu_operating_frequency", + "ScaleUnit": "1GHz" + }, + { + "BriefDescription": "Percentage of time spent in the active CPU power state C0", + "MetricExpr": "tma_info_system_cpu_utilization", + "MetricName": "cpu_utilization", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for 2 megabyte page sizes) caused by demand data loads to the total number of completed instructions", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_COMPLETED_2M_4M / INST_RETIRED.ANY", + "MetricName": "dtlb_2mb_large_page_load_mpi", + "PublicDescription": "Ratio of number of completed page walks (for 2 megabyte page sizes) caused by demand data loads to the total number of completed instructions. This implies it missed in the Data Translation Lookaside Buffer (DTLB) and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data loads to the total number of completed instructions", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "dtlb_load_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data loads to the total number of completed instructions. This implies it missed in the DTLB and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data stores to the total number of completed instructions", + "MetricExpr": "DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "dtlb_store_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data stores to the total number of completed instructions. This implies it missed in the DTLB and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Bandwidth of IO reads that are initiated by end device controllers that are requesting memory from the CPU.", + "MetricExpr": "(UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART0 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART1 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART2 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART3) * 4 / 1e6 / duration_time", + "MetricName": "io_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Bandwidth of IO writes that are initiated by end device controllers that are writing memory to the CPU.", + "MetricExpr": "(UNC_IIO_PAYLOAD_BYTES_IN.MEM_WRITE.PART0 + UNC_IIO_PAYLOAD_BYTES_IN.MEM_WRITE.PART1 + UNC_IIO_PAYLOAD_BYTES_IN.MEM_WRITE.PART2 + UNC_IIO_PAYLOAD_BYTES_IN.MEM_WRITE.PART3) * 4 / 1e6 / duration_time", + "MetricName": "io_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for 2 megabyte and 4 megabyte page sizes) caused by a code fetch to the total number of completed instructions", + "MetricExpr": "ITLB_MISSES.WALK_COMPLETED_2M_4M / INST_RETIRED.ANY", + "MetricName": "itlb_large_page_mpi", + "PublicDescription": "Ratio of number of completed page walks (for 2 megabyte and 4 megabyte page sizes) caused by a code fetch to the total number of completed instructions. This implies it missed in the Instruction Translation Lookaside Buffer (ITLB) and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by a code fetch to the total number of completed instructions", + "MetricExpr": "ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "itlb_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by a code fetch to the total number of completed instructions. This implies it missed in the ITLB (Instruction TLB) and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read requests missing in L1 instruction cache (includes prefetches) to the total number of completed instructions", + "MetricExpr": "L2_RQSTS.ALL_CODE_RD / INST_RETIRED.ANY", + "MetricName": "l1_i_code_read_misses_with_prefetches_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of demand load requests hitting in L1 data cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_RETIRED.L1_HIT / INST_RETIRED.ANY", + "MetricName": "l1d_demand_data_read_hits_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of requests missing L1 data cache (includes data+rfo w/ prefetches) to the total number of completed instructions", + "MetricExpr": "L1D.REPLACEMENT / INST_RETIRED.ANY", + "MetricName": "l1d_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read request missing L2 cache to the total number of completed instructions", + "MetricExpr": "L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", + "MetricName": "l2_demand_code_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed demand load requests hitting in L2 cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_RETIRED.L2_HIT / INST_RETIRED.ANY", + "MetricName": "l2_demand_data_read_hits_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed data read request missing L2 cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_RETIRED.L2_MISS / INST_RETIRED.ANY", + "MetricName": "l2_demand_data_read_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of requests missing L2 cache (includes code+data+rfo w/ prefetches) to the total number of completed instructions", + "MetricExpr": "L2_LINES_IN.ALL / INST_RETIRED.ANY", + "MetricName": "l2_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read requests missing last level core cache (includes demand w/ prefetches) to the total number of completed instructions", + "MetricExpr": "cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x12CC0233@ / INST_RETIRED.ANY", + "MetricName": "llc_code_read_mpi_demand_plus_prefetch", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand and prefetch data read miss (read memory access) in nano seconds", + "MetricExpr": "1e9 * (cha@UNC_CHA_TOR_OCCUPANCY.IA_MISS\\,config1\\=0x40433@ / cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40433@) / (UNC_CHA_CLOCKTICKS / (source_count(UNC_CHA_CLOCKTICKS) * #num_packages)) * duration_time", + "MetricName": "llc_data_read_demand_plus_prefetch_miss_latency", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand and prefetch data read miss (read memory access) addressed to local memory in nano seconds", + "MetricExpr": "1e9 * (cha@UNC_CHA_TOR_OCCUPANCY.IA_MISS\\,config1\\=0x40432@ / cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40432@) / (UNC_CHA_CLOCKTICKS / (source_count(UNC_CHA_CLOCKTICKS) * #num_packages)) * duration_time", + "MetricName": "llc_data_read_demand_plus_prefetch_miss_latency_for_local_requests", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand and prefetch data read miss (read memory access) addressed to remote memory in nano seconds", + "MetricExpr": "1e9 * (cha@UNC_CHA_TOR_OCCUPANCY.IA_MISS\\,config1\\=0x40431@ / cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40431@) / (UNC_CHA_CLOCKTICKS / (source_count(UNC_CHA_CLOCKTICKS) * #num_packages)) * duration_time", + "MetricName": "llc_data_read_demand_plus_prefetch_miss_latency_for_remote_requests", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Ratio of number of data read requests missing last level core cache (includes demand w/ prefetches) to the total number of completed instructions", + "MetricExpr": "cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x12D40433@ / INST_RETIRED.ANY", + "MetricName": "llc_data_read_mpi_demand_plus_prefetch", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Bandwidth (MB/sec) of read requests that miss the last level cache (LLC) and go to local memory.", + "MetricExpr": "UNC_CHA_REQUESTS.READS_LOCAL * 64 / 1e6 / duration_time", + "MetricName": "llc_miss_local_memory_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Bandwidth (MB/sec) of write requests that miss the last level cache (LLC) and go to local memory.", + "MetricExpr": "UNC_CHA_REQUESTS.WRITES_LOCAL * 64 / 1e6 / duration_time", + "MetricName": "llc_miss_local_memory_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Bandwidth (MB/sec) of read requests that miss the last level cache (LLC) and go to remote memory.", + "MetricExpr": "UNC_CHA_REQUESTS.READS_REMOTE * 64 / 1e6 / duration_time", + "MetricName": "llc_miss_remote_memory_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "The ratio of number of completed memory load instructions to the total number completed instructions", + "MetricExpr": "MEM_INST_RETIRED.ALL_LOADS / INST_RETIRED.ANY", + "MetricName": "loads_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "DDR memory read bandwidth (MB/sec)", + "MetricExpr": "UNC_M_CAS_COUNT.RD * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "DDR memory bandwidth (MB/sec)", + "MetricExpr": "(UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_total", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "DDR memory write bandwidth (MB/sec)", + "MetricExpr": "UNC_M_CAS_COUNT.WR * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Memory read that miss the last level cache (LLC) addressed to local DRAM as a percentage of total memory read accesses, does not include LLC prefetches.", + "MetricExpr": "cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40432@ / (cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40432@ + cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40431@)", + "MetricName": "numa_reads_addressed_to_local_dram", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Memory reads that miss the last level cache (LLC) addressed to remote DRAM as a percentage of total memory read accesses, does not include LLC prefetches.", + "MetricExpr": "cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40431@ / (cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40432@ + cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40431@)", + "MetricName": "numa_reads_addressed_to_remote_dram", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from decoded instruction cache (decoded stream buffer or DSB) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS + LSD.UOPS)", + "MetricName": "percent_uops_delivered_from_decoded_icache", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from legacy decode pipeline (Micro-instruction Translation Engine or MITE) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.MITE_UOPS / (IDQ.DSB_UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS + LSD.UOPS)", + "MetricName": "percent_uops_delivered_from_legacy_decode_pipeline", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from microcode sequencer (MS) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.MS_UOPS / (IDQ.DSB_UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS + LSD.UOPS)", + "MetricName": "percent_uops_delivered_from_microcode_sequencer", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Intel(R) Optane(TM) Persistent Memory(PMEM) memory read bandwidth (MB/sec)", + "MetricExpr": "UNC_M_PMM_RPQ_INSERTS * 64 / 1e6 / duration_time", + "MetricName": "pmem_memory_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Intel(R) Optane(TM) Persistent Memory(PMEM) memory bandwidth (MB/sec)", + "MetricExpr": "(UNC_M_PMM_RPQ_INSERTS + UNC_M_PMM_WPQ_INSERTS) * 64 / 1e6 / duration_time", + "MetricName": "pmem_memory_bandwidth_total", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Intel(R) Optane(TM) Persistent Memory(PMEM) memory write bandwidth (MB/sec)", + "MetricExpr": "UNC_M_PMM_WPQ_INSERTS * 64 / 1e6 / duration_time", + "MetricName": "pmem_memory_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { "BriefDescription": "Percentage of cycles spent in System Management Interrupts.", "MetricExpr": "((msr@aperf@ - cycles) / msr@aperf@ if msr@smi@ > 0 else 0)", "MetricGroup": "smi", @@ -70,8 +297,14 @@ "ScaleUnit": "1SMI#" }, { + "BriefDescription": "The ratio of number of completed memory store instructions to the total number completed instructions", + "MetricExpr": "MEM_INST_RETIRED.ALL_STORES / INST_RETIRED.ANY", + "MetricName": "stores_per_instr", + "ScaleUnit": "1per_instr" + }, + { "BriefDescription": "This metric estimates how often memory load accesses were aliased by preceding stores (in program order) with a 4K address offset", - "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_clks", + "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_4k_aliasing", "MetricThreshold": "tma_4k_aliasing > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -80,7 +313,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution ports for ALU operations.", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5 + UOPS_DISPATCHED_PORT.PORT_6) / tma_info_slots", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5 + UOPS_DISPATCHED_PORT.PORT_6) / tma_info_thread_slots", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_alu_op_utilization", "MetricThreshold": "tma_alu_op_utilization > 0.6", @@ -88,7 +321,7 @@ }, { "BriefDescription": "This metric estimates fraction of slots the CPU retired uops delivered by the Microcode_Sequencer as a result of Assists", - "MetricExpr": "100 * (FP_ASSIST.ANY + OTHER_ASSISTS.ANY) / tma_info_slots", + "MetricExpr": "100 * (FP_ASSIST.ANY + OTHER_ASSISTS.ANY) / tma_info_thread_slots", "MetricGroup": "TopdownL4;tma_L4_group;tma_microcode_sequencer_group", "MetricName": "tma_assists", "MetricThreshold": "tma_assists > 0.1 & (tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1)", @@ -97,19 +330,21 @@ }, { "BriefDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend", - "MetricExpr": "1 - tma_frontend_bound - (UOPS_ISSUED.ANY + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_slots", + "MetricExpr": "1 - tma_frontend_bound - (UOPS_ISSUED.ANY + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend. Backend is the portion of the processor core where the out-of-order scheduler dispatches ready uops into their respective execution units; and once completed these uops get retired according to program order. For example; stalls due to data-cache misses or stalls due to the divider unit being overloaded are both categorized under Backend Bound. Backend Bound is further divided into two main categories: Memory Bound and Core Bound.", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations", - "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_slots", + "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots wasted due to incorrect speculations. This include slots used to issue uops that do not eventually get retired and slots for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work due to miss-predicted branches are categorized under Bad Speculation category. Incorrect data speculation followed by Memory Ordering Nukes is another example.", "ScaleUnit": "100%" }, @@ -120,12 +355,13 @@ "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15", - "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_branch_misprediction_cost, tma_info_mispredictions, tma_mispredicts_resteers", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_info_bottleneck_mispredictions, tma_mispredicts_resteers", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers", - "MetricExpr": "INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks + tma_unknown_branches", + "MetricExpr": "INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks + tma_unknown_branches", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteers", "MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -143,7 +379,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Machine Clears", - "MetricExpr": "(1 - BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT)) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks", + "MetricExpr": "(1 - BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT)) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks", "MetricGroup": "BadSpec;MachineClears;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueMC", "MetricName": "tma_clears_resteers", "MetricThreshold": "tma_clears_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", @@ -153,7 +389,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to contested accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(44 * tma_info_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_HITM * (OCR.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE / (OCR.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE + OCR.DEMAND_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD))) + 44 * tma_info_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "(44 * tma_info_system_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_HITM * (OCR.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE / (OCR.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE + OCR.DEMAND_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD))) + 44 * tma_info_system_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_contested_accesses", "MetricThreshold": "tma_contested_accesses > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -167,13 +403,14 @@ "MetricGroup": "Backend;Compute;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck. Shortage in hardware compute resources; or dependencies in software's instructions are both categorized under Core Bound. Hence it may indicate the machine ran out of an out-of-order resource; certain execution units are overloaded or dependencies in program's data- or instruction-flow are limiting the performance (e.g. FP-chained long-latency arithmetic operations).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to data-sharing accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "44 * tma_info_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_L3_HIT_RETIRED.XSNP_HITM * (1 - OCR.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE / (OCR.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE + OCR.DEMAND_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD))) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "44 * tma_info_system_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_L3_HIT_RETIRED.XSNP_HITM * (1 - OCR.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE / (OCR.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE + OCR.DEMAND_DATA_RD.L3_HIT.HIT_OTHER_CORE_FWD))) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_data_sharing", "MetricThreshold": "tma_data_sharing > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -182,16 +419,16 @@ }, { "BriefDescription": "This metric represents fraction of cycles where decoder-0 was the only active decoder", - "MetricExpr": "(cpu@INST_DECODED.DECODERS\\,cmask\\=1@ - cpu@INST_DECODED.DECODERS\\,cmask\\=2@) / tma_info_core_clks / 2", + "MetricExpr": "(cpu@INST_DECODED.DECODERS\\,cmask\\=1@ - cpu@INST_DECODED.DECODERS\\,cmask\\=2@) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL4;tma_L4_group;tma_issueD0;tma_mite_group", "MetricName": "tma_decoder0_alone", - "MetricThreshold": "tma_decoder0_alone > 0.1 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35))", + "MetricThreshold": "tma_decoder0_alone > 0.1 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35))", "PublicDescription": "This metric represents fraction of cycles where decoder-0 was the only active decoder. Related metrics: tma_few_uops_instructions", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles where the Divider unit was active", - "MetricExpr": "ARITH.DIVIDER_ACTIVE / tma_info_clks", + "MetricExpr": "ARITH.DIVIDER_ACTIVE / tma_info_thread_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_divider", "MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -201,7 +438,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks - tma_l2_bound - tma_pmm_bound if #has_pmem > 0 else CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks - tma_l2_bound)", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_thread_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks - tma_l2_bound - tma_pmm_bound if #has_pmem > 0 else CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_thread_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks - tma_l2_bound)", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -210,45 +447,45 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline", - "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSB;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_dsb", - "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline. For example; inefficient utilization of the DSB cache structure or bank conflict when reading from it; are categorized here.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_dsb_switches", "MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb, tma_lcp", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "min(9 * cpu@DTLB_LOAD_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_LOAD_MISSES.WALK_ACTIVE, max(CYCLE_ACTIVITY.CYCLES_MEM_ANY - CYCLE_ACTIVITY.CYCLES_L1D_MISS, 0)) / tma_info_clks", + "MetricExpr": "min(9 * cpu@DTLB_LOAD_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_LOAD_MISSES.WALK_ACTIVE, max(CYCLE_ACTIVITY.CYCLES_MEM_ANY - CYCLE_ACTIVITY.CYCLES_L1D_MISS, 0)) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group", "MetricName": "tma_dtlb_load", "MetricThreshold": "tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses. TLBs (Translation Look-aside Buffers) are processor caches for recently used entries out of the Page Tables that are used to map virtual- to physical-addresses by the operating system. This metric approximates the potential delay of demand loads missing the first-level data TLB (assuming worst case scenario with back to back misses to different pages). This includes hitting in the second-level TLB (STLB) as well as performing a hardware page walk on an STLB miss. Sample with: MEM_INST_RETIRED.STLB_MISS_LOADS_PS. Related metrics: tma_dtlb_store, tma_info_memory_data_tlbs", + "PublicDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses. TLBs (Translation Look-aside Buffers) are processor caches for recently used entries out of the Page Tables that are used to map virtual- to physical-addresses by the operating system. This metric approximates the potential delay of demand loads missing the first-level data TLB (assuming worst case scenario with back to back misses to different pages). This includes hitting in the second-level TLB (STLB) as well as performing a hardware page walk on an STLB miss. Sample with: MEM_INST_RETIRED.STLB_MISS_LOADS_PS. Related metrics: tma_dtlb_store, tma_info_bottleneck_memory_data_tlbs", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses", - "MetricExpr": "(9 * cpu@DTLB_STORE_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_STORE_MISSES.WALK_ACTIVE) / tma_info_core_clks", + "MetricExpr": "(9 * cpu@DTLB_STORE_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_STORE_MISSES.WALK_ACTIVE) / tma_info_core_core_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_store_bound_group", "MetricName": "tma_dtlb_store", "MetricThreshold": "tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses. As with ordinary data caching; focus on improving data locality and reducing working-set size to reduce DTLB overhead. Additionally; consider using profile-guided optimization (PGO) to collocate frequently-used data on the same page. Try using larger page sizes for large amounts of frequently-used data. Sample with: MEM_INST_RETIRED.STLB_MISS_STORES_PS. Related metrics: tma_dtlb_load, tma_info_memory_data_tlbs", + "PublicDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses. As with ordinary data caching; focus on improving data locality and reducing working-set size to reduce DTLB overhead. Additionally; consider using profile-guided optimization (PGO) to collocate frequently-used data on the same page. Try using larger page sizes for large amounts of frequently-used data. Sample with: MEM_INST_RETIRED.STLB_MISS_STORES_PS. Related metrics: tma_dtlb_load, tma_info_bottleneck_memory_data_tlbs", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates how often CPU was handling synchronizations due to False Sharing", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(110 * tma_info_average_frequency * (OCR.DEMAND_RFO.L3_MISS.REMOTE_HITM + OCR.PF_L2_RFO.L3_MISS.REMOTE_HITM) + 47.5 * tma_info_average_frequency * (OCR.DEMAND_RFO.L3_HIT.HITM_OTHER_CORE + OCR.PF_L2_RFO.L3_HIT.HITM_OTHER_CORE)) / tma_info_clks", + "MetricExpr": "(110 * tma_info_system_average_frequency * (OCR.DEMAND_RFO.L3_MISS.REMOTE_HITM + OCR.PF_L2_RFO.L3_MISS.REMOTE_HITM) + 47.5 * tma_info_system_average_frequency * (OCR.DEMAND_RFO.L3_HIT.HITM_OTHER_CORE + OCR.PF_L2_RFO.L3_HIT.HITM_OTHER_CORE)) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_store_bound_group", "MetricName": "tma_false_sharing", "MetricThreshold": "tma_false_sharing > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -258,11 +495,11 @@ { "BriefDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "tma_info_load_miss_real_latency * cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@ / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@ / tma_info_thread_clks", "MetricGroup": "MemoryBW;TopdownL4;tma_L4_group;tma_issueBW;tma_issueSL;tma_issueSmSt;tma_l1_bound_group", "MetricName": "tma_fb_full", "MetricThreshold": "tma_fb_full > 0.3", - "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", + "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", "ScaleUnit": "100%" }, { @@ -270,16 +507,18 @@ "MetricExpr": "tma_frontend_bound - tma_fetch_latency", "MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", "MetricName": "tma_fetch_bandwidth", - "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb, tma_lcp", + "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues", - "MetricExpr": "4 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE / tma_info_slots", + "MetricExpr": "4 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE / tma_info_thread_slots", "MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: FRONTEND_RETIRED.LATENCY_GE_16_PS;FRONTEND_RETIRED.LATENCY_GE_8_PS", "ScaleUnit": "100%" }, @@ -350,10 +589,11 @@ }, { "BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend", - "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_slots", + "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_thread_slots", "MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound. Sample with: FRONTEND_RETIRED.LATENCY_GE_4_PS", "ScaleUnit": "100%" }, @@ -368,16 +608,17 @@ }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences", - "MetricExpr": "(UOPS_RETIRED.RETIRE_SLOTS + UOPS_RETIRED.MACRO_FUSED - INST_RETIRED.ANY) / tma_info_slots", + "MetricExpr": "(UOPS_RETIRED.RETIRE_SLOTS + UOPS_RETIRED.MACRO_FUSED - INST_RETIRED.ANY) / tma_info_thread_slots", "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_heavy_operations", "MetricThreshold": "tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to instruction cache misses", - "MetricExpr": "(ICACHE_16B.IFDATA_STALL + 2 * cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@) / tma_info_clks", + "MetricExpr": "(ICACHE_16B.IFDATA_STALL + 2 * cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@) / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;IcMiss;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_icache_misses", "MetricThreshold": "tma_icache_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -385,705 +626,711 @@ "ScaleUnit": "100%" }, { - "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", - "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time", - "MetricGroup": "Power;Summary", - "MetricName": "tma_info_average_frequency" + "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", + "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_thread_slots / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BrMispredicts;tma_issueBM", + "MetricName": "tma_info_bad_spec_branch_misprediction_cost", + "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_info_bottleneck_mispredictions, tma_mispredicts_resteers" + }, + { + "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", + "MetricExpr": "tma_info_inst_mix_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_indirect", + "MetricThreshold": "tma_info_bad_spec_ipmisp_indirect < 1e3" + }, + { + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_core_ipmispredict", + "MetricGroup": "Bad;BadSpec;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmispredict", + "MetricThreshold": "tma_info_bad_spec_ipmispredict < 200" + }, + { + "BriefDescription": "Probability of Core Bound bottleneck hidden by SMT-profiling artifacts", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(100 * (1 - tma_core_bound / tma_ports_utilization if tma_core_bound < tma_ports_utilization else 1) if tma_info_system_smt_2t_utilization > 0.5 else 0)", + "MetricGroup": "Cor;SMT", + "MetricName": "tma_info_botlnk_l0_core_bound_likely", + "MetricThreshold": "tma_info_botlnk_l0_core_bound_likely > 0.5" + }, + { + "BriefDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_fetch_latency * tma_dsb_switches / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches) + tma_fetch_bandwidth * tma_mite / (tma_dsb + tma_mite))", + "MetricGroup": "DSBmiss;Fed;tma_issueFB", + "MetricName": "tma_info_botlnk_l2_dsb_misses", + "MetricThreshold": "tma_info_botlnk_l2_dsb_misses > 10", + "PublicDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp" + }, + { + "BriefDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck", + "MetricExpr": "100 * (tma_fetch_latency * tma_icache_misses / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", + "MetricGroup": "Fed;FetchLat;IcMiss;tma_issueFL", + "MetricName": "tma_info_botlnk_l2_ic_misses", + "MetricThreshold": "tma_info_botlnk_l2_ic_misses > 5", + "PublicDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck. Related metrics: " }, { "BriefDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses)", "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "100 * tma_fetch_latency * (tma_itlb_misses + tma_icache_misses + tma_unknown_branches) / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)", "MetricGroup": "BigFoot;Fed;Frontend;IcMiss;MemoryTLB;tma_issueBC", - "MetricName": "tma_info_big_code", - "MetricThreshold": "tma_info_big_code > 20", - "PublicDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses). Related metrics: tma_info_branching_overhead" + "MetricName": "tma_info_bottleneck_big_code", + "MetricThreshold": "tma_info_bottleneck_big_code > 20", + "PublicDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses). Related metrics: tma_info_bottleneck_branching_overhead" }, { - "BriefDescription": "Branch instructions per taken branch.", - "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", - "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_bptkbranch" + "BriefDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls)", + "MetricExpr": "100 * ((BR_INST_RETIRED.CONDITIONAL + 3 * BR_INST_RETIRED.NEAR_CALL + (BR_INST_RETIRED.NEAR_TAKEN - (BR_INST_RETIRED.CONDITIONAL - BR_INST_RETIRED.NOT_TAKEN) - 2 * BR_INST_RETIRED.NEAR_CALL)) / tma_info_thread_slots)", + "MetricGroup": "Ret;tma_issueBC", + "MetricName": "tma_info_bottleneck_branching_overhead", + "MetricThreshold": "tma_info_bottleneck_branching_overhead > 10", + "PublicDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls). Related metrics: tma_info_bottleneck_big_code" }, { - "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", - "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_slots / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_branch_misprediction_cost", - "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_info_mispredictions, tma_mispredicts_resteers" + "BriefDescription": "Total pipeline cost of instruction fetch bandwidth related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_frontend_bound - tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) - tma_info_bottleneck_big_code", + "MetricGroup": "Fed;FetchBW;Frontend", + "MetricName": "tma_info_bottleneck_instruction_fetch_bw", + "MetricThreshold": "tma_info_bottleneck_instruction_fetch_bw > 20" }, { - "BriefDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls)", - "MetricExpr": "100 * ((BR_INST_RETIRED.CONDITIONAL + 3 * BR_INST_RETIRED.NEAR_CALL + (BR_INST_RETIRED.NEAR_TAKEN - (BR_INST_RETIRED.CONDITIONAL - BR_INST_RETIRED.NOT_TAKEN) - 2 * BR_INST_RETIRED.NEAR_CALL)) / tma_info_slots)", - "MetricGroup": "Ret;tma_issueBC", - "MetricName": "tma_info_branching_overhead", - "MetricThreshold": "tma_info_branching_overhead > 10", - "PublicDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls). Related metrics: tma_info_big_code" + "BriefDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_mem_bandwidth / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_sq_full / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full))) + tma_l1_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_fb_full / (tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk))", + "MetricGroup": "Mem;MemoryBW;Offcore;tma_issueBW", + "MetricName": "tma_info_bottleneck_memory_bandwidth", + "MetricThreshold": "tma_info_bottleneck_memory_bandwidth > 20", + "PublicDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks. Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full" }, { - "BriefDescription": "Fraction of branches that are CALL or RET", - "MetricExpr": "(BR_INST_RETIRED.NEAR_CALL + BR_INST_RETIRED.NEAR_RETURN) / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_callret" + "BriefDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_l1_bound / max(tma_memory_bound, tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_dtlb_load / max(tma_l1_bound, tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk)) + tma_store_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_dtlb_store / (tma_dtlb_store + tma_false_sharing + tma_split_stores + tma_store_latency)))", + "MetricGroup": "Mem;MemoryTLB;Offcore;tma_issueTLB", + "MetricName": "tma_info_bottleneck_memory_data_tlbs", + "MetricThreshold": "tma_info_bottleneck_memory_data_tlbs > 20", + "PublicDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs). Related metrics: tma_dtlb_load, tma_dtlb_store" }, { - "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "Pipeline", - "MetricName": "tma_info_clks" + "BriefDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_mem_latency / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_l3_hit_latency / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full)) + tma_l2_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound))", + "MetricGroup": "Mem;MemoryLat;Offcore;tma_issueLat", + "MetricName": "tma_info_bottleneck_memory_latency", + "MetricThreshold": "tma_info_bottleneck_memory_latency > 20", + "PublicDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches). Related metrics: tma_l3_hit_latency, tma_mem_latency" }, { - "BriefDescription": "STLB (2nd level TLB) code speculative misses per kilo instruction (misses of any page-size that complete the page walk)", - "MetricExpr": "1e3 * ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", - "MetricGroup": "Fed;MemoryTLB", - "MetricName": "tma_info_code_stlb_mpki" + "BriefDescription": "Total pipeline cost of Branch Misprediction related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", + "MetricGroup": "Bad;BadSpec;BrMispredicts;tma_issueBM", + "MetricName": "tma_info_bottleneck_mispredictions", + "MetricThreshold": "tma_info_bottleneck_mispredictions > 20", + "PublicDescription": "Total pipeline cost of Branch Misprediction related bottlenecks. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers" + }, + { + "BriefDescription": "Fraction of branches that are CALL or RET", + "MetricExpr": "(BR_INST_RETIRED.NEAR_CALL + BR_INST_RETIRED.NEAR_RETURN) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_callret" }, { "BriefDescription": "Fraction of branches that are non-taken conditionals", "MetricExpr": "BR_INST_RETIRED.NOT_TAKEN / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Bad;Branches;CodeGen;PGO", - "MetricName": "tma_info_cond_nt" + "MetricName": "tma_info_branches_cond_nt" }, { "BriefDescription": "Fraction of branches that are taken conditionals", "MetricExpr": "(BR_INST_RETIRED.CONDITIONAL - BR_INST_RETIRED.NOT_TAKEN) / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Bad;Branches;CodeGen;PGO", - "MetricName": "tma_info_cond_tk" + "MetricName": "tma_info_branches_cond_tk" }, { - "BriefDescription": "Probability of Core Bound bottleneck hidden by SMT-profiling artifacts", + "BriefDescription": "Fraction of branches that are unconditional (direct or indirect) jumps", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(100 * (1 - tma_core_bound / tma_ports_utilization if tma_core_bound < tma_ports_utilization else 1) if tma_info_smt_2t_utilization > 0.5 else 0)", - "MetricGroup": "Cor;SMT", - "MetricName": "tma_info_core_bound_likely", - "MetricThreshold": "tma_info_core_bound_likely > 0.5" + "MetricExpr": "(BR_INST_RETIRED.NEAR_TAKEN - (BR_INST_RETIRED.CONDITIONAL - BR_INST_RETIRED.NOT_TAKEN) - 2 * BR_INST_RETIRED.NEAR_CALL) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_jump" }, { "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", - "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_clks))", + "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_thread_clks))", "MetricGroup": "SMT", - "MetricName": "tma_info_core_clks" + "MetricName": "tma_info_core_core_clks" }, { "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks", "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_coreipc" + "MetricName": "tma_info_core_coreipc" }, { - "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricExpr": "1 / tma_info_ipc", - "MetricGroup": "Mem;Pipeline", - "MetricName": "tma_info_cpi" + "BriefDescription": "Floating Point Operations Per Cycle", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * (FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE) + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / tma_info_core_core_clks", + "MetricGroup": "Flops;Ret", + "MetricName": "tma_info_core_flopc" }, { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": "HPC;Summary", - "MetricName": "tma_info_cpu_utilization" + "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", + "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@) / (2 * tma_info_core_core_clks)", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_core_fp_arith_utilization", + "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." }, { - "BriefDescription": "Average Parallel L2 cache miss data reads", - "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", - "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_data_l2_mlp" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", + "MetricExpr": "UOPS_EXECUTED.THREAD / (UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1)", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", + "MetricName": "tma_info_core_ilp" }, { - "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time", - "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", - "MetricName": "tma_info_dram_bw_use", - "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_info_memory_bandwidth, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear)", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BadSpec;BrMispredicts;TopdownL1;tma_L1_group", + "MetricName": "tma_info_core_ipmispredict", + "MetricgroupNoGroup": "TopdownL1" }, { "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", "MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)", "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", - "MetricName": "tma_info_dsb_coverage", - "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_misses, tma_info_iptb, tma_lcp" - }, - { - "BriefDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * (tma_fetch_latency * tma_dsb_switches / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches) + tma_fetch_bandwidth * tma_mite / (tma_dsb + tma_mite))", - "MetricGroup": "DSBmiss;Fed;tma_issueFB", - "MetricName": "tma_info_dsb_misses", - "MetricThreshold": "tma_info_dsb_misses > 10", - "PublicDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb, tma_lcp" + "MetricName": "tma_info_frontend_dsb_coverage", + "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 4 > 0.35", + "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_inst_mix_iptb, tma_lcp" }, { "BriefDescription": "Average number of cycles of a switch from the DSB fetch-unit to MITE fetch unit - see DSB_Switches tree node for details.", "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / DSB2MITE_SWITCHES.COUNT", "MetricGroup": "DSBmiss", - "MetricName": "tma_info_dsb_switch_cost" - }, - { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", - "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", - "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", - "MetricName": "tma_info_execute" - }, - { - "BriefDescription": "The ratio of Executed- by Issued-Uops", - "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", - "MetricGroup": "Cor;Pipeline", - "MetricName": "tma_info_execute_per_issue", - "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." - }, - { - "BriefDescription": "Fill Buffer (FB) hits per kilo instructions for retired demand loads (L1D misses that merge into ongoing miss-handling entries)", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.FB_HIT / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_fb_hpki" + "MetricName": "tma_info_frontend_dsb_switch_cost" }, { "BriefDescription": "Average number of Uops issued by front-end when it issued something", "MetricExpr": "UOPS_ISSUED.ANY / cpu@UOPS_ISSUED.ANY\\,cmask\\=1@", "MetricGroup": "Fed;FetchBW", - "MetricName": "tma_info_fetch_upc" + "MetricName": "tma_info_frontend_fetch_upc" }, { - "BriefDescription": "Floating Point Operations Per Cycle", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * (FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE) + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / tma_info_core_clks", - "MetricGroup": "Flops;Ret", - "MetricName": "tma_info_flopc" - }, - { - "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@) / (2 * tma_info_core_clks)", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_fp_arith_utilization", - "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." + "BriefDescription": "Average Latency for L1 instruction cache misses", + "MetricExpr": "ICACHE_16B.IFDATA_STALL / cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@ + 2", + "MetricGroup": "Fed;FetchLat;IcMiss", + "MetricName": "tma_info_frontend_icache_miss_latency" }, { - "BriefDescription": "Giga Floating Point Operations Per Second", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * (FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE) + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / 1e9 / duration_time", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_gflops", - "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + "BriefDescription": "Instructions per non-speculative DSB miss (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / FRONTEND_RETIRED.ANY_DSB_MISS", + "MetricGroup": "DSBmiss;Fed", + "MetricName": "tma_info_frontend_ipdsb_miss_ret", + "MetricThreshold": "tma_info_frontend_ipdsb_miss_ret < 50" }, { - "BriefDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck", - "MetricExpr": "100 * (tma_fetch_latency * tma_icache_misses / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", - "MetricGroup": "Fed;FetchLat;IcMiss;tma_issueFL", - "MetricName": "tma_info_ic_misses", - "MetricThreshold": "tma_info_ic_misses > 5", - "PublicDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck. Related metrics: " + "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_inst_mix_instructions / BACLEARS.ANY", + "MetricGroup": "Fed", + "MetricName": "tma_info_frontend_ipunknown_branch" }, { - "BriefDescription": "Average Latency for L1 instruction cache misses", - "MetricExpr": "ICACHE_16B.IFDATA_STALL / cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@ + 2", - "MetricGroup": "Fed;FetchLat;IcMiss", - "MetricName": "tma_info_icache_miss_latency" + "BriefDescription": "L2 cache true code cacheline misses per kilo instruction", + "MetricExpr": "1e3 * FRONTEND_RETIRED.L2_MISS / INST_RETIRED.ANY", + "MetricGroup": "IcMiss", + "MetricName": "tma_info_frontend_l2mpki_code" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", - "MetricExpr": "UOPS_EXECUTED.THREAD / (UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1)", - "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", - "MetricName": "tma_info_ilp" + "BriefDescription": "L2 cache speculative code cacheline misses per kilo instruction", + "MetricExpr": "1e3 * L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", + "MetricGroup": "IcMiss", + "MetricName": "tma_info_frontend_l2mpki_code_all" }, { - "BriefDescription": "Total pipeline cost of instruction fetch bandwidth related bottlenecks", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * (tma_frontend_bound - tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) - tma_info_big_code", - "MetricGroup": "Fed;FetchBW;Frontend", - "MetricName": "tma_info_instruction_fetch_bw", - "MetricThreshold": "tma_info_instruction_fetch_bw > 20" + "BriefDescription": "Branch instructions per taken branch.", + "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;PGO", + "MetricName": "tma_info_inst_mix_bptkbranch" }, { "BriefDescription": "Total number of retired Instructions", "MetricExpr": "INST_RETIRED.ANY", "MetricGroup": "Summary;TmaL1;tma_L1_group", - "MetricName": "tma_info_instructions", + "MetricName": "tma_info_inst_mix_instructions", "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" }, { - "BriefDescription": "Average IO (network or disk) Bandwidth Use for Reads [GB / sec]", - "MetricExpr": "(UNC_IIO_DATA_REQ_OF_CPU.MEM_WRITE.PART0 + UNC_IIO_DATA_REQ_OF_CPU.MEM_WRITE.PART1 + UNC_IIO_DATA_REQ_OF_CPU.MEM_WRITE.PART2 + UNC_IIO_DATA_REQ_OF_CPU.MEM_WRITE.PART3) * 4 / 1e9 / duration_time", - "MetricGroup": "IoBW;Mem;Server;SoC", - "MetricName": "tma_info_io_read_bw" - }, - { - "BriefDescription": "Average IO (network or disk) Bandwidth Use for Writes [GB / sec]", - "MetricExpr": "(UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART0 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART1 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART2 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART3) * 4 / 1e9 / duration_time", - "MetricGroup": "IoBW;Mem;Server;SoC", - "MetricName": "tma_info_io_write_bw" - }, - { "BriefDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate)", "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "INST_RETIRED.ANY / (cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_iparith", - "MetricThreshold": "tma_info_iparith < 10", + "MetricName": "tma_info_inst_mix_iparith", + "MetricThreshold": "tma_info_inst_mix_iparith < 10", "PublicDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate). May undercount due to FMA double counting. Approximated prior to BDW." }, { "BriefDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx128", - "MetricThreshold": "tma_info_iparith_avx128 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx128", + "MetricThreshold": "tma_info_inst_mix_iparith_avx128 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx256", - "MetricThreshold": "tma_info_iparith_avx256 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx256", + "MetricThreshold": "tma_info_inst_mix_iparith_avx256 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic AVX 512-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx512", - "MetricThreshold": "tma_info_iparith_avx512 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx512", + "MetricThreshold": "tma_info_inst_mix_iparith_avx512 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX 512-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_dp", - "MetricThreshold": "tma_info_iparith_scalar_dp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_dp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_dp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_SINGLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_sp", - "MetricThreshold": "tma_info_iparith_scalar_sp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_sp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_sp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per Branch (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Branches;Fed;InsType", - "MetricName": "tma_info_ipbranch", - "MetricThreshold": "tma_info_ipbranch < 8" - }, - { - "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": "Ret;Summary", - "MetricName": "tma_info_ipc" + "MetricName": "tma_info_inst_mix_ipbranch", + "MetricThreshold": "tma_info_inst_mix_ipbranch < 8" }, { "BriefDescription": "Instructions per (near) call (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_CALL", "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_ipcall", - "MetricThreshold": "tma_info_ipcall < 200" - }, - { - "BriefDescription": "Instructions per non-speculative DSB miss (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / FRONTEND_RETIRED.ANY_DSB_MISS", - "MetricGroup": "DSBmiss;Fed", - "MetricName": "tma_info_ipdsb_miss_ret", - "MetricThreshold": "tma_info_ipdsb_miss_ret < 50" - }, - { - "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", - "MetricGroup": "Branches;OS", - "MetricName": "tma_info_ipfarbranch", - "MetricThreshold": "tma_info_ipfarbranch < 1e6" + "MetricName": "tma_info_inst_mix_ipcall", + "MetricThreshold": "tma_info_inst_mix_ipcall < 200" }, { "BriefDescription": "Instructions per Floating Point (FP) Operation (lower number means higher occurrence rate)", "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * (FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE) + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_ipflop", - "MetricThreshold": "tma_info_ipflop < 10" + "MetricName": "tma_info_inst_mix_ipflop", + "MetricThreshold": "tma_info_inst_mix_ipflop < 10" }, { "BriefDescription": "Instructions per Load (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_LOADS", "MetricGroup": "InsType", - "MetricName": "tma_info_ipload", - "MetricThreshold": "tma_info_ipload < 3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", - "MetricExpr": "tma_info_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_indirect", - "MetricThreshold": "tma_info_ipmisp_indirect < 1e3" - }, - { - "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BadSpec;BrMispredicts", - "MetricName": "tma_info_ipmispredict", - "MetricThreshold": "tma_info_ipmispredict < 200" + "MetricName": "tma_info_inst_mix_ipload", + "MetricThreshold": "tma_info_inst_mix_ipload < 3" }, { "BriefDescription": "Instructions per Store (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_STORES", "MetricGroup": "InsType", - "MetricName": "tma_info_ipstore", - "MetricThreshold": "tma_info_ipstore < 8" + "MetricName": "tma_info_inst_mix_ipstore", + "MetricThreshold": "tma_info_inst_mix_ipstore < 8" }, { "BriefDescription": "Instructions per Software prefetch instruction (of any type: NTA/T0/T1/T2/Prefetch) (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / cpu@SW_PREFETCH_ACCESS.T0\\,umask\\=0xF@", "MetricGroup": "Prefetches", - "MetricName": "tma_info_ipswpf", - "MetricThreshold": "tma_info_ipswpf < 100" + "MetricName": "tma_info_inst_mix_ipswpf", + "MetricThreshold": "tma_info_inst_mix_ipswpf < 100" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW;Frontend;PGO;tma_issueFB", - "MetricName": "tma_info_iptb", - "MetricThreshold": "tma_info_iptb < 9", - "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_lcp" + "MetricName": "tma_info_inst_mix_iptb", + "MetricThreshold": "tma_info_inst_mix_iptb < 9", + "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_lcp" }, { - "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", - "MetricExpr": "tma_info_instructions / BACLEARS.ANY", - "MetricGroup": "Fed", - "MetricName": "tma_info_ipunknown_branch" + "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_core_l1d_cache_fill_bw" }, { - "BriefDescription": "Fraction of branches that are unconditional (direct or indirect) jumps", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(BR_INST_RETIRED.NEAR_TAKEN - (BR_INST_RETIRED.CONDITIONAL - BR_INST_RETIRED.NOT_TAKEN) - 2 * BR_INST_RETIRED.NEAR_CALL) / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_jump" + "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_core_l2_cache_fill_bw" }, { - "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_cpi" + "BriefDescription": "Rate of non silent evictions from the L2 cache per Kilo instruction", + "MetricExpr": "1e3 * L2_LINES_OUT.NON_SILENT / tma_info_inst_mix_instructions", + "MetricGroup": "L2Evicts;Mem;Server", + "MetricName": "tma_info_memory_core_l2_evictions_nonsilent_pki" }, { - "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_utilization", - "MetricThreshold": "tma_info_kernel_utilization > 0.05" + "BriefDescription": "Rate of silent evictions from the L2 cache per Kilo instruction where the evicted lines are dropped (no writeback to L3 or memory)", + "MetricExpr": "1e3 * L2_LINES_OUT.SILENT / tma_info_inst_mix_instructions", + "MetricGroup": "L2Evicts;Mem;Server", + "MetricName": "tma_info_memory_core_l2_evictions_silent_pki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw" + "BriefDescription": "Average per-core data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * OFFCORE_REQUESTS.ALL_REQUESTS / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_core_l3_cache_access_bw" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "tma_info_l1d_cache_fill_bw", + "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw_1t" + "MetricName": "tma_info_memory_core_l3_cache_fill_bw" + }, + { + "BriefDescription": "Fill Buffer (FB) hits per kilo instructions for retired demand loads (L1D misses that merge into ongoing miss-handling entries)", + "MetricExpr": "1e3 * MEM_LOAD_RETIRED.FB_HIT / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_fb_hpki" }, { "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L1_MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki" + "MetricName": "tma_info_memory_l1mpki" }, { "BriefDescription": "L1 cache true misses per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.ALL_DEMAND_DATA_RD / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki_load" - }, - { - "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw" - }, - { - "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "tma_info_l2_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw_1t" - }, - { - "BriefDescription": "Rate of non silent evictions from the L2 cache per Kilo instruction", - "MetricExpr": "1e3 * L2_LINES_OUT.NON_SILENT / tma_info_instructions", - "MetricGroup": "L2Evicts;Mem;Server", - "MetricName": "tma_info_l2_evictions_nonsilent_pki" - }, - { - "BriefDescription": "Rate of silent evictions from the L2 cache per Kilo instruction where the evicted lines are dropped (no writeback to L3 or memory)", - "MetricExpr": "1e3 * L2_LINES_OUT.SILENT / tma_info_instructions", - "MetricGroup": "L2Evicts;Mem;Server", - "MetricName": "tma_info_l2_evictions_silent_pki" + "MetricName": "tma_info_memory_l1mpki_load" }, { "BriefDescription": "L2 cache hits per kilo instruction for all request types (including speculative)", "MetricExpr": "1e3 * (L2_RQSTS.REFERENCES - L2_RQSTS.MISS) / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_all" + "MetricName": "tma_info_memory_l2hpki_all" }, { "BriefDescription": "L2 cache hits per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_HIT / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_load" + "MetricName": "tma_info_memory_l2hpki_load" }, { "BriefDescription": "L2 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L2_MISS / INST_RETIRED.ANY", "MetricGroup": "Backend;CacheMisses;Mem", - "MetricName": "tma_info_l2mpki" + "MetricName": "tma_info_memory_l2mpki" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all request types (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem;Offcore", - "MetricName": "tma_info_l2mpki_all" - }, - { - "BriefDescription": "L2 cache true code cacheline misses per kilo instruction", - "MetricExpr": "1e3 * FRONTEND_RETIRED.L2_MISS / INST_RETIRED.ANY", - "MetricGroup": "IcMiss", - "MetricName": "tma_info_l2mpki_code" - }, - { - "BriefDescription": "L2 cache speculative code cacheline misses per kilo instruction", - "MetricExpr": "1e3 * L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", - "MetricGroup": "IcMiss", - "MetricName": "tma_info_l2mpki_code_all" + "MetricName": "tma_info_memory_l2mpki_all" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2mpki_load" + "MetricName": "tma_info_memory_l2mpki_load" }, { - "BriefDescription": "Average per-core data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * OFFCORE_REQUESTS.ALL_REQUESTS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw" - }, - { - "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_access_bw", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw_1t" + "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L3_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l3mpki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw" + "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", + "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_RETIRED.L1_MISS + MEM_LOAD_RETIRED.FB_HIT)", + "MetricGroup": "Mem;MemoryBound;MemoryLat", + "MetricName": "tma_info_memory_load_miss_real_latency" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw_1t" + "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", + "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", + "MetricGroup": "Mem;MemoryBW;MemoryBound", + "MetricName": "tma_info_memory_mlp", + "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" }, { - "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L3_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l3mpki" + "BriefDescription": "Average Parallel L2 cache miss data reads", + "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", + "MetricGroup": "Memory_BW;Offcore", + "MetricName": "tma_info_memory_oro_data_l2_mlp" }, { "BriefDescription": "Average Latency for L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS.DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l2_miss_latency" + "MetricName": "tma_info_memory_oro_load_l2_miss_latency" }, { "BriefDescription": "Average Parallel L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_DATA_RD", "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_load_l2_mlp" + "MetricName": "tma_info_memory_oro_load_l2_mlp" }, { - "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", - "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_RETIRED.L1_MISS + MEM_LOAD_RETIRED.FB_HIT)", - "MetricGroup": "Mem;MemoryBound;MemoryLat", - "MetricName": "tma_info_load_miss_real_latency" + "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l1d_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l1d_cache_fill_bw_1t" + }, + { + "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l2_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l2_cache_fill_bw_1t" + }, + { + "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_access_bw", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_thread_l3_cache_access_bw_1t" + }, + { + "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l3_cache_fill_bw_1t" + }, + { + "BriefDescription": "STLB (2nd level TLB) code speculative misses per kilo instruction (misses of any page-size that complete the page walk)", + "MetricExpr": "1e3 * ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricGroup": "Fed;MemoryTLB", + "MetricName": "tma_info_memory_tlb_code_stlb_mpki" }, { "BriefDescription": "STLB (2nd level TLB) data load speculative misses per kilo instruction (misses of any page-size that complete the page walk)", "MetricExpr": "1e3 * DTLB_LOAD_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_load_stlb_mpki" + "MetricName": "tma_info_memory_tlb_load_stlb_mpki" + }, + { + "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", + "MetricConstraint": "NO_GROUP_EVENTS_NMI", + "MetricExpr": "(ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING + EPT.WALK_PENDING) / (2 * tma_info_core_core_clks)", + "MetricGroup": "Mem;MemoryTLB", + "MetricName": "tma_info_memory_tlb_page_walks_utilization", + "MetricThreshold": "tma_info_memory_tlb_page_walks_utilization > 0.5" + }, + { + "BriefDescription": "STLB (2nd level TLB) data store speculative misses per kilo instruction (misses of any page-size that complete the page walk)", + "MetricExpr": "1e3 * DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricGroup": "Mem;MemoryTLB", + "MetricName": "tma_info_memory_tlb_store_stlb_mpki" + }, + { + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", + "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", + "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", + "MetricName": "tma_info_pipeline_execute" + }, + { + "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@", + "MetricGroup": "Pipeline;Ret", + "MetricName": "tma_info_pipeline_retire" + }, + { + "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time", + "MetricGroup": "Power;Summary", + "MetricName": "tma_info_system_average_frequency" + }, + { + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricGroup": "HPC;Summary", + "MetricName": "tma_info_system_cpu_utilization" + }, + { + "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", + "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time", + "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", + "MetricName": "tma_info_system_dram_bw_use", + "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_mem_bandwidth, tma_sq_full" + }, + { + "BriefDescription": "Giga Floating Point Operations Per Second", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * (FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE) + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / 1e9 / duration_time", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_system_gflops", + "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + }, + { + "BriefDescription": "Average IO (network or disk) Bandwidth Use for Reads [GB / sec]", + "MetricExpr": "(UNC_IIO_DATA_REQ_OF_CPU.MEM_WRITE.PART0 + UNC_IIO_DATA_REQ_OF_CPU.MEM_WRITE.PART1 + UNC_IIO_DATA_REQ_OF_CPU.MEM_WRITE.PART2 + UNC_IIO_DATA_REQ_OF_CPU.MEM_WRITE.PART3) * 4 / 1e9 / duration_time", + "MetricGroup": "IoBW;Mem;Server;SoC", + "MetricName": "tma_info_system_io_read_bw" + }, + { + "BriefDescription": "Average IO (network or disk) Bandwidth Use for Writes [GB / sec]", + "MetricExpr": "(UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART0 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART1 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART2 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART3) * 4 / 1e9 / duration_time", + "MetricGroup": "IoBW;Mem;Server;SoC", + "MetricName": "tma_info_system_io_write_bw" + }, + { + "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", + "MetricGroup": "Branches;OS", + "MetricName": "tma_info_system_ipfarbranch", + "MetricThreshold": "tma_info_system_ipfarbranch < 1e6" + }, + { + "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_cpi" + }, + { + "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_utilization", + "MetricThreshold": "tma_info_system_kernel_utilization > 0.05" }, { "BriefDescription": "Average latency of data read request to external DRAM memory [in nanoseconds]", "MetricExpr": "1e9 * (UNC_M_RPQ_OCCUPANCY / UNC_M_RPQ_INSERTS) / imc_0@event\\=0x0@", "MetricGroup": "Mem;MemoryLat;Server;SoC", - "MetricName": "tma_info_mem_dram_read_latency", + "MetricName": "tma_info_system_mem_dram_read_latency", "PublicDescription": "Average latency of data read request to external DRAM memory [in nanoseconds]. Accounts for demand loads and L1/L2 data-read prefetches" }, { "BriefDescription": "Average number of parallel data read requests to external memory", "MetricExpr": "UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD / UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD@thresh\\=1@", "MetricGroup": "Mem;MemoryBW;SoC", - "MetricName": "tma_info_mem_parallel_reads", + "MetricName": "tma_info_system_mem_parallel_reads", "PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches" }, { "BriefDescription": "Average latency of data read request to external 3D X-Point memory [in nanoseconds]", "MetricExpr": "(1e9 * (UNC_M_PMM_RPQ_OCCUPANCY.ALL / UNC_M_PMM_RPQ_INSERTS) / imc_0@event\\=0x0@ if #has_pmem > 0 else 0)", "MetricGroup": "Mem;MemoryLat;Server;SoC", - "MetricName": "tma_info_mem_pmm_read_latency", + "MetricName": "tma_info_system_mem_pmm_read_latency", "PublicDescription": "Average latency of data read request to external 3D X-Point memory [in nanoseconds]. Accounts for demand loads and L1/L2 data-read prefetches" }, { "BriefDescription": "Average latency of data read request to external memory (in nanoseconds)", - "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD / UNC_CHA_TOR_INSERTS.IA_MISS_DRD) / (tma_info_socket_clks / duration_time)", + "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD / UNC_CHA_TOR_INSERTS.IA_MISS_DRD) / (tma_info_system_socket_clks / duration_time)", "MetricGroup": "Mem;MemoryLat;SoC", - "MetricName": "tma_info_mem_read_latency", + "MetricName": "tma_info_system_mem_read_latency", "PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)" }, { - "BriefDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_mem_bandwidth / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_sq_full / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full))) + tma_l1_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_fb_full / (tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk))", - "MetricGroup": "Mem;MemoryBW;Offcore;tma_issueBW", - "MetricName": "tma_info_memory_bandwidth", - "MetricThreshold": "tma_info_memory_bandwidth > 20", - "PublicDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks. Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_mem_bandwidth, tma_sq_full" - }, - { - "BriefDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * tma_memory_bound * (tma_l1_bound / max(tma_memory_bound, tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_dtlb_load / max(tma_l1_bound, tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk)) + tma_store_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_dtlb_store / (tma_dtlb_store + tma_false_sharing + tma_split_stores + tma_store_latency)))", - "MetricGroup": "Mem;MemoryTLB;Offcore;tma_issueTLB", - "MetricName": "tma_info_memory_data_tlbs", - "MetricThreshold": "tma_info_memory_data_tlbs > 20", - "PublicDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs). Related metrics: tma_dtlb_load, tma_dtlb_store" - }, - { - "BriefDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_mem_latency / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_l3_hit_latency / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full)) + tma_l2_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound))", - "MetricGroup": "Mem;MemoryLat;Offcore;tma_issueLat", - "MetricName": "tma_info_memory_latency", - "MetricThreshold": "tma_info_memory_latency > 20", - "PublicDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches). Related metrics: tma_l3_hit_latency, tma_mem_latency" - }, - { - "BriefDescription": "Total pipeline cost of Branch Misprediction related bottlenecks", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * (tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", - "MetricGroup": "Bad;BadSpec;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_mispredictions", - "MetricThreshold": "tma_info_mispredictions > 20", - "PublicDescription": "Total pipeline cost of Branch Misprediction related bottlenecks. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost, tma_mispredicts_resteers" - }, - { - "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", - "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", - "MetricGroup": "Mem;MemoryBW;MemoryBound", - "MetricName": "tma_info_mlp", - "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" - }, - { - "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", - "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING + EPT.WALK_PENDING) / (2 * tma_info_core_clks)", - "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_page_walks_utilization", - "MetricThreshold": "tma_info_page_walks_utilization > 0.5" - }, - { "BriefDescription": "Average 3DXP Memory Bandwidth Use for reads [GB / sec]", "MetricExpr": "(64 * UNC_M_PMM_RPQ_INSERTS / 1e9 / duration_time if #has_pmem > 0 else 0)", "MetricGroup": "Mem;MemoryBW;Server;SoC", - "MetricName": "tma_info_pmm_read_bw" + "MetricName": "tma_info_system_pmm_read_bw" }, { "BriefDescription": "Average 3DXP Memory Bandwidth Use for Writes [GB / sec]", "MetricExpr": "(64 * UNC_M_PMM_WPQ_INSERTS / 1e9 / duration_time if #has_pmem > 0 else 0)", "MetricGroup": "Mem;MemoryBW;Server;SoC", - "MetricName": "tma_info_pmm_write_bw" + "MetricName": "tma_info_system_pmm_write_bw" }, { "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for baseline license level 0", - "MetricExpr": "(CORE_POWER.LVL0_TURBO_LICENSE / 2 / tma_info_core_clks if #SMT_on else CORE_POWER.LVL0_TURBO_LICENSE / tma_info_core_clks)", + "MetricExpr": "(CORE_POWER.LVL0_TURBO_LICENSE / 2 / tma_info_core_core_clks if #SMT_on else CORE_POWER.LVL0_TURBO_LICENSE / tma_info_core_core_clks)", "MetricGroup": "Power", - "MetricName": "tma_info_power_license0_utilization", + "MetricName": "tma_info_system_power_license0_utilization", "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for baseline license level 0. This includes non-AVX codes, SSE, AVX 128-bit, and low-current AVX 256-bit codes." }, { "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 1", - "MetricExpr": "(CORE_POWER.LVL1_TURBO_LICENSE / 2 / tma_info_core_clks if #SMT_on else CORE_POWER.LVL1_TURBO_LICENSE / tma_info_core_clks)", + "MetricExpr": "(CORE_POWER.LVL1_TURBO_LICENSE / 2 / tma_info_core_core_clks if #SMT_on else CORE_POWER.LVL1_TURBO_LICENSE / tma_info_core_core_clks)", "MetricGroup": "Power", - "MetricName": "tma_info_power_license1_utilization", - "MetricThreshold": "tma_info_power_license1_utilization > 0.5", + "MetricName": "tma_info_system_power_license1_utilization", + "MetricThreshold": "tma_info_system_power_license1_utilization > 0.5", "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 1. This includes high current AVX 256-bit instructions as well as low current AVX 512-bit instructions." }, { "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 2 (introduced in SKX)", - "MetricExpr": "(CORE_POWER.LVL2_TURBO_LICENSE / 2 / tma_info_core_clks if #SMT_on else CORE_POWER.LVL2_TURBO_LICENSE / tma_info_core_clks)", + "MetricExpr": "(CORE_POWER.LVL2_TURBO_LICENSE / 2 / tma_info_core_core_clks if #SMT_on else CORE_POWER.LVL2_TURBO_LICENSE / tma_info_core_core_clks)", "MetricGroup": "Power", - "MetricName": "tma_info_power_license2_utilization", - "MetricThreshold": "tma_info_power_license2_utilization > 0.5", + "MetricName": "tma_info_system_power_license2_utilization", + "MetricThreshold": "tma_info_system_power_license2_utilization > 0.5", "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 2 (introduced in SKX). This includes high current AVX 512-bit instructions." }, { - "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@", - "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_retire" - }, - { - "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", - "MetricExpr": "4 * tma_info_core_clks", - "MetricGroup": "TmaL1;tma_L1_group", - "MetricName": "tma_info_slots" - }, - { "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / (CPU_CLK_UNHALTED.REF_XCLK_ANY / 2) if #SMT_on else 0)", "MetricGroup": "SMT", - "MetricName": "tma_info_smt_2t_utilization" + "MetricName": "tma_info_system_smt_2t_utilization" }, { "BriefDescription": "Socket actual clocks when any core is active on that socket", "MetricExpr": "cha_0@event\\=0x0@", "MetricGroup": "SoC", - "MetricName": "tma_info_socket_clks" - }, - { - "BriefDescription": "STLB (2nd level TLB) data store speculative misses per kilo instruction (misses of any page-size that complete the page walk)", - "MetricExpr": "1e3 * DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", - "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_store_stlb_mpki" + "MetricName": "tma_info_system_socket_clks" }, { "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC", "MetricGroup": "Power", - "MetricName": "tma_info_turbo_utilization" + "MetricName": "tma_info_system_turbo_utilization" + }, + { + "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "Pipeline", + "MetricName": "tma_info_thread_clks" + }, + { + "BriefDescription": "Cycles Per Instruction (per Logical Processor)", + "MetricExpr": "1 / tma_info_thread_ipc", + "MetricGroup": "Mem;Pipeline", + "MetricName": "tma_info_thread_cpi" + }, + { + "BriefDescription": "The ratio of Executed- by Issued-Uops", + "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", + "MetricGroup": "Cor;Pipeline", + "MetricName": "tma_info_thread_execute_per_issue", + "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." + }, + { + "BriefDescription": "Instructions Per Cycle (per Logical Processor)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks", + "MetricGroup": "Ret;Summary", + "MetricName": "tma_info_thread_ipc" + }, + { + "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", + "MetricExpr": "4 * tma_info_core_core_clks", + "MetricGroup": "TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots" }, { "BriefDescription": "Uops Per Instruction", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / INST_RETIRED.ANY", "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_uoppi", - "MetricThreshold": "tma_info_uoppi > 1.05" + "MetricName": "tma_info_thread_uoppi", + "MetricThreshold": "tma_info_thread_uoppi > 1.05" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW", - "MetricName": "tma_info_uptb", - "MetricThreshold": "tma_info_uptb < 6" + "MetricName": "tma_info_thread_uptb", + "MetricThreshold": "tma_info_thread_uptb < 6" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses", - "MetricExpr": "ICACHE_64B.IFTAG_STALL / tma_info_clks", + "MetricExpr": "ICACHE_64B.IFTAG_STALL / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -1092,7 +1339,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled without loads missing the L1 data cache", - "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_clks, 0)", + "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_thread_clks, 0)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_issueL1;tma_issueMC;tma_memory_bound_group", "MetricName": "tma_l1_bound", "MetricThreshold": "tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1102,7 +1349,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled due to L2 cache accesses by loads", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) / (MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@) * ((CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks)", + "MetricExpr": "MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) / (MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@) * ((CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l2_bound", "MetricThreshold": "tma_l2_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1111,7 +1358,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L2_MISS - CYCLE_ACTIVITY.STALLS_L3_MISS) / tma_info_clks", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L2_MISS - CYCLE_ACTIVITY.STALLS_L3_MISS) / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1120,20 +1367,20 @@ }, { "BriefDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited)", - "MetricExpr": "17 * tma_info_average_frequency * MEM_LOAD_RETIRED.L3_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "17 * tma_info_system_average_frequency * MEM_LOAD_RETIRED.L3_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "MemoryLat;TopdownL4;tma_L4_group;tma_issueLat;tma_l3_bound_group", "MetricName": "tma_l3_hit_latency", "MetricThreshold": "tma_l3_hit_latency > 0.1 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited). Avoiding private cache misses (i.e. L2 misses/L3 hits) will improve the latency; reduce contention with sibling physical cores and increase performance. Note the value of this node may overlap with its siblings. Sample with: MEM_LOAD_RETIRED.L3_HIT_PS. Related metrics: tma_info_memory_latency, tma_mem_latency", + "PublicDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited). Avoiding private cache misses (i.e. L2 misses/L3 hits) will improve the latency; reduce contention with sibling physical cores and increase performance. Note the value of this node may overlap with its siblings. Sample with: MEM_LOAD_RETIRED.L3_HIT_PS. Related metrics: tma_info_bottleneck_memory_latency, tma_mem_latency", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)", - "MetricExpr": "ILD_STALL.LCP / tma_info_clks", + "MetricExpr": "ILD_STALL.LCP / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_lcp", "MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb", + "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb", "ScaleUnit": "100%" }, { @@ -1142,12 +1389,13 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_light_operations", "MetricThreshold": "tma_light_operations > 0.6", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation). This correlates with total number of instructions used by the program. A uops-per-instruction (see UopPI metric) ratio of 1 or less should be expected for decently optimized software running on Intel Core/Xeon products. While this often indicates efficient X86 instructions were executed; high value does not necessarily mean better performance cannot be achieved. Sample with: INST_RETIRED.PREC_DIST", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Load operations", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 + UOPS_DISPATCHED_PORT.PORT_7 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 + UOPS_DISPATCHED_PORT.PORT_7 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_load_op_utilization", "MetricThreshold": "tma_load_op_utilization > 0.6", @@ -1165,7 +1413,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles where the Second-level TLB (STLB) was missed by load accesses, performing a hardware page walk", - "MetricExpr": "DTLB_LOAD_MISSES.WALK_ACTIVE / tma_info_clks", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_ACTIVE / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL5;tma_L5_group;tma_dtlb_load_group", "MetricName": "tma_load_stlb_miss", "MetricThreshold": "tma_load_stlb_miss > 0.05 & (tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1173,7 +1421,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from local memory", - "MetricExpr": "59.5 * tma_info_average_frequency * MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "59.5 * tma_info_system_average_frequency * MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Server;TopdownL5;tma_L5_group;tma_mem_latency_group", "MetricName": "tma_local_dram", "MetricThreshold": "tma_local_dram > 0.1 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1182,7 +1430,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU spent handling cache misses due to lock operations", - "MetricExpr": "(12 * max(0, MEM_INST_RETIRED.LOCK_LOADS - L2_RQSTS.ALL_RFO) + MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES * (11 * L2_RQSTS.RFO_HIT + min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO))) / tma_info_clks", + "MetricExpr": "(12 * max(0, MEM_INST_RETIRED.LOCK_LOADS - L2_RQSTS.ALL_RFO) + MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES * (11 * L2_RQSTS.RFO_HIT + min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO))) / tma_info_thread_clks", "MetricGroup": "Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_l1_bound_group", "MetricName": "tma_lock_latency", "MetricThreshold": "tma_lock_latency > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1196,25 +1444,26 @@ "MetricGroup": "BadSpec;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.1 & tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears. These slots are either wasted by uops fetched prior to the clear; or stalls the out-of-order portion of the machine needs to recover its state after the clear. For example; this can happen due to memory ordering Nukes (e.g. Memory Disambiguation) or Self-Modifying-Code (SMC) nukes. Sample with: MACHINE_CLEARS.COUNT. Related metrics: tma_clears_resteers, tma_contested_accesses, tma_data_sharing, tma_false_sharing, tma_l1_bound, tma_microcode_sequencer, tma_ms_switches, tma_remote_cache", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_clks", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW", "MetricName": "tma_mem_bandwidth", "MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_sq_full", + "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_sq_full", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat", "MetricName": "tma_mem_latency", "MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM). This metric does not aggregate requests from other Logical Processors/Physical Cores/sockets (see Uncore counters for that). Related metrics: tma_info_memory_latency, tma_l3_hit_latency", + "PublicDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM). This metric does not aggregate requests from other Logical Processors/Physical Cores/sockets (see Uncore counters for that). Related metrics: tma_info_bottleneck_memory_latency, tma_l3_hit_latency", "ScaleUnit": "100%" }, { @@ -1224,6 +1473,7 @@ "MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck. Memory Bound estimates fraction of slots where pipeline is likely stalled due to demand load or store instructions. This accounts mainly for (1) non-completed in-flight memory demand loads which coincides with execution units starvation; in addition to (2) cases where stores could impose backpressure on the pipeline when many of them get buffered at the same time (less common out of the two).", "ScaleUnit": "100%" }, @@ -1237,7 +1487,7 @@ }, { "BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_thread_slots", "MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS", "MetricName": "tma_microcode_sequencer", "MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1", @@ -1246,19 +1496,19 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage", - "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks", + "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks", "MetricGroup": "BadSpec;BrMispredicts;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueBM", "MetricName": "tma_mispredicts_resteers", "MetricThreshold": "tma_mispredicts_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost, tma_info_mispredictions", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost, tma_info_bottleneck_mispredictions", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline)", - "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_mite", - "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline). This pipeline is used for code that was not pre-cached in the DSB or LSD. For example; inefficiencies due to asymmetric decoders; use of long immediate or LCP can manifest as MITE fetch bandwidth bottleneck. Sample with: FRONTEND_RETIRED.ANY_DSB_MISS", "ScaleUnit": "100%" }, @@ -1273,7 +1523,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)", - "MetricExpr": "2 * IDQ.MS_SWITCHES / tma_info_clks", + "MetricExpr": "2 * IDQ.MS_SWITCHES / tma_info_thread_clks", "MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO", "MetricName": "tma_ms_switches", "MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -1310,7 +1560,7 @@ { "BriefDescription": "This metric roughly estimates (based on idle latencies) how often the CPU was stalled on accesses to external 3D-Xpoint (Crystal Ridge, a.k.a", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(((1 - ((19 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + 10 * (MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS))) / (19 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + 10 * (MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + (25 * (MEM_LOAD_RETIRED.LOCAL_PMM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) if #has_pmem > 0 else 0) + 33 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_PMM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) if #has_pmem > 0 else 0))) if #has_pmem > 0 else 0)) * (CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks - tma_l2_bound) if 1e6 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_PMM + MEM_LOAD_RETIRED.LOCAL_PMM) > MEM_LOAD_RETIRED.L1_MISS else 0) if #has_pmem > 0 else 0)", + "MetricExpr": "(((1 - ((19 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + 10 * (MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS))) / (19 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + 10 * (MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + (25 * (MEM_LOAD_RETIRED.LOCAL_PMM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) if #has_pmem > 0 else 0) + 33 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_PMM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) if #has_pmem > 0 else 0))) if #has_pmem > 0 else 0)) * (CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_thread_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks - tma_l2_bound) if 1e6 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_PMM + MEM_LOAD_RETIRED.LOCAL_PMM) > MEM_LOAD_RETIRED.L1_MISS else 0) if #has_pmem > 0 else 0)", "MetricGroup": "MemoryBound;Server;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_pmm_bound", "MetricThreshold": "tma_pmm_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1319,7 +1569,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 0 ([SNB+] ALU; [HSW+] ALU and 2nd branch)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_core_clks", "MetricGroup": "Compute;TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_0", "MetricThreshold": "tma_port_0 > 0.6", @@ -1328,7 +1578,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 1 (ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_1", "MetricThreshold": "tma_port_1 > 0.6", @@ -1337,7 +1587,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 2 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_2", "MetricThreshold": "tma_port_2 > 0.6", @@ -1346,7 +1596,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 3 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_3", "MetricThreshold": "tma_port_3 > 0.6", @@ -1364,7 +1614,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 5 ([SNB+] Branches and ALU; [HSW+] ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_5", "MetricThreshold": "tma_port_5 > 0.6", @@ -1373,7 +1623,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 6 ([HSW+]Primary Branch and simple ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_6 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_6 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_6", "MetricThreshold": "tma_port_6 > 0.6", @@ -1382,7 +1632,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 7 ([HSW+]simple Store-address)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_7 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_7 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_store_op_utilization_group", "MetricName": "tma_port_7", "MetricThreshold": "tma_port_7 > 0.6", @@ -1391,7 +1641,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)", - "MetricExpr": "((EXE_ACTIVITY.EXE_BOUND_0_PORTS + (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL)) / tma_info_clks if ARITH.DIVIDER_ACTIVE < CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY else (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_clks)", + "MetricExpr": "((EXE_ACTIVITY.EXE_BOUND_0_PORTS + (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL)) / tma_info_thread_clks if ARITH.DIVIDER_ACTIVE < CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY else (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_thread_clks)", "MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_ports_utilization", "MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -1400,7 +1650,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed no uops on any execution port (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "(UOPS_EXECUTED.CORE_CYCLES_NONE / 2 if #SMT_on else CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_core_clks", + "MetricExpr": "(UOPS_EXECUTED.CORE_CYCLES_NONE / 2 if #SMT_on else CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_0", "MetricThreshold": "tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1409,7 +1659,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the CPU executed total of 1 uop per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((UOPS_EXECUTED.CORE_CYCLES_GE_1 - UOPS_EXECUTED.CORE_CYCLES_GE_2) / 2 if #SMT_on else EXE_ACTIVITY.1_PORTS_UTIL) / tma_info_core_clks", + "MetricExpr": "((UOPS_EXECUTED.CORE_CYCLES_GE_1 - UOPS_EXECUTED.CORE_CYCLES_GE_2) / 2 if #SMT_on else EXE_ACTIVITY.1_PORTS_UTIL) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issueL1;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_1", "MetricThreshold": "tma_ports_utilized_1 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1418,7 +1668,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 2 uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((UOPS_EXECUTED.CORE_CYCLES_GE_2 - UOPS_EXECUTED.CORE_CYCLES_GE_3) / 2 if #SMT_on else EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_core_clks", + "MetricExpr": "((UOPS_EXECUTED.CORE_CYCLES_GE_2 - UOPS_EXECUTED.CORE_CYCLES_GE_3) / 2 if #SMT_on else EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issue2P;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_2", "MetricThreshold": "tma_ports_utilized_2 > 0.15 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1427,7 +1677,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 3 or more uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise).", - "MetricExpr": "(UOPS_EXECUTED.CORE_CYCLES_GE_3 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_3) / tma_info_core_clks", + "MetricExpr": "(UOPS_EXECUTED.CORE_CYCLES_GE_3 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_3) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_3m", "MetricThreshold": "tma_ports_utilized_3m > 0.7 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1436,7 +1686,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from remote cache in other sockets including synchronizations issues", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(89.5 * tma_info_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM + 89.5 * tma_info_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "(89.5 * tma_info_system_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM + 89.5 * tma_info_system_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Offcore;Server;Snoop;TopdownL5;tma_L5_group;tma_issueSyncxn;tma_mem_latency_group", "MetricName": "tma_remote_cache", "MetricThreshold": "tma_remote_cache > 0.05 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1445,7 +1695,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from remote memory", - "MetricExpr": "127 * tma_info_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "127 * tma_info_system_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Server;Snoop;TopdownL5;tma_L5_group;tma_mem_latency_group", "MetricName": "tma_remote_dram", "MetricThreshold": "tma_remote_dram > 0.1 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1454,16 +1704,17 @@ }, { "BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired. Ideally; all pipeline slots would be attributed to the Retiring category. Retiring of 100% would indicate the maximum Pipeline_Width throughput was achieved. Maximizing Retiring typically increases the Instructions-per-cycle (see IPC metric). Note that a high Retiring value does not necessary mean there is no room for more performance. For example; Heavy-operations or Microcode Assists are categorized under Retiring. They often indicate suboptimal performance and can often be optimized or avoided. Sample with: UOPS_RETIRED.RETIRE_SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU issue-pipeline was stalled due to serializing operations", - "MetricExpr": "PARTIAL_RAT_STALLS.SCOREBOARD / tma_info_clks", + "MetricExpr": "PARTIAL_RAT_STALLS.SCOREBOARD / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL5;tma_L5_group;tma_issueSO;tma_ports_utilized_0_group", "MetricName": "tma_serializing_operation", "MetricThreshold": "tma_serializing_operation > 0.1 & (tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)))", @@ -1472,7 +1723,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to PAUSE Instructions", - "MetricExpr": "40 * ROB_MISC_EVENTS.PAUSE_INST / tma_info_clks", + "MetricExpr": "40 * ROB_MISC_EVENTS.PAUSE_INST / tma_info_thread_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_serializing_operation_group", "MetricName": "tma_slow_pause", "MetricThreshold": "tma_slow_pause > 0.05 & (tma_serializing_operation > 0.1 & (tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))))", @@ -1482,7 +1733,7 @@ { "BriefDescription": "This metric estimates fraction of cycles handling memory load split accesses - load that cross 64-byte cache line boundary", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "tma_info_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_split_loads", "MetricThreshold": "tma_split_loads > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1491,7 +1742,7 @@ }, { "BriefDescription": "This metric represents rate of split store accesses", - "MetricExpr": "MEM_INST_RETIRED.SPLIT_STORES / tma_info_core_clks", + "MetricExpr": "MEM_INST_RETIRED.SPLIT_STORES / tma_info_core_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_issueSpSt;tma_store_bound_group", "MetricName": "tma_split_stores", "MetricThreshold": "tma_split_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1500,16 +1751,16 @@ }, { "BriefDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors)", - "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_clks", + "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_core_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueBW;tma_l3_bound_group", "MetricName": "tma_sq_full", "MetricThreshold": "tma_sq_full > 0.3 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_mem_bandwidth", + "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_mem_bandwidth", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write", - "MetricExpr": "EXE_ACTIVITY.BOUND_ON_STORES / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.BOUND_ON_STORES / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1518,7 +1769,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of cycles when the memory subsystem had loads blocked since they could not forward data from earlier (in program order) overlapping stores", - "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_clks", + "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_store_fwd_blk", "MetricThreshold": "tma_store_fwd_blk > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1528,7 +1779,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU spent handling L1D store misses", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(L2_RQSTS.RFO_HIT * 11 * (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) + (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_clks", + "MetricExpr": "(L2_RQSTS.RFO_HIT * 11 * (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) + (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_thread_clks", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_issueSL;tma_store_bound_group", "MetricName": "tma_store_latency", "MetricThreshold": "tma_store_latency > 0.1 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1537,7 +1788,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Store operations", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_core_clks", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_store_op_utilization", "MetricThreshold": "tma_store_op_utilization > 0.6", @@ -1553,7 +1804,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles where the STLB was missed by store accesses, performing a hardware page walk", - "MetricExpr": "DTLB_STORE_MISSES.WALK_ACTIVE / tma_info_core_clks", + "MetricExpr": "DTLB_STORE_MISSES.WALK_ACTIVE / tma_info_core_core_clks", "MetricGroup": "MemoryTLB;TopdownL5;tma_L5_group;tma_dtlb_store_group", "MetricName": "tma_store_stlb_miss", "MetricThreshold": "tma_store_stlb_miss > 0.05 & (tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1561,7 +1812,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to new branch address clears", - "MetricExpr": "9 * BACLEARS.ANY / tma_info_clks", + "MetricExpr": "9 * BACLEARS.ANY / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;TopdownL4;tma_L4_group;tma_branch_resteers_group", "MetricName": "tma_unknown_branches", "MetricThreshold": "tma_unknown_branches > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", @@ -1604,5 +1855,17 @@ "MetricGroup": "transaction", "MetricName": "tsx_transactional_cycles", "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uncore operating frequency in GHz", + "MetricExpr": "UNC_CHA_CLOCKTICKS / (source_count(UNC_CHA_CLOCKTICKS) * #num_packages) / 1e9 / duration_time", + "MetricName": "uncore_frequency", + "ScaleUnit": "1GHz" + }, + { + "BriefDescription": "Intel(R) Ultra Path Interconnect (UPI) data transmit bandwidth (MB/sec)", + "MetricExpr": "UNC_UPI_TxL_FLITS.ALL_DATA * 7.111111111111111 / 1e6 / duration_time", + "MetricName": "upi_data_transmit_bw", + "ScaleUnit": "1MB/s" } ] diff --git a/tools/perf/pmu-events/arch/x86/cascadelakex/floating-point.json b/tools/perf/pmu-events/arch/x86/cascadelakex/floating-point.json index 1f46e6b33856..bb4d5101f962 100644 --- a/tools/perf/pmu-events/arch/x86/cascadelakex/floating-point.json +++ b/tools/perf/pmu-events/arch/x86/cascadelakex/floating-point.json @@ -32,6 +32,14 @@ "UMask": "0x20" }, { + "BriefDescription": "Number of SSE/AVX computational 128-bit packed single and 256-bit packed double precision FP instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, 1 for each element. Applies to SSE* and AVX* packed single precision and packed double precision FP instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB count twice as they perform 2 calculations per element.", + "EventCode": "0xC7", + "EventName": "FP_ARITH_INST_RETIRED.4_FLOPS", + "PublicDescription": "Number of SSE/AVX computational 128-bit packed single precision and 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point and packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", + "SampleAfterValue": "1000003", + "UMask": "0x18" + }, + { "BriefDescription": "Number of SSE/AVX computational 512-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", "EventCode": "0xC7", "EventName": "FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE", @@ -48,6 +56,22 @@ "UMask": "0x80" }, { + "BriefDescription": "Number of SSE/AVX computational 256-bit packed single precision and 512-bit packed double precision FP instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, 1 for each element. Applies to SSE* and AVX* packed single precision and double precision FP instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RSQRT14 RCP RCP14 DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB count twice as they perform 2 calculations per element.", + "EventCode": "0xC7", + "EventName": "FP_ARITH_INST_RETIRED.8_FLOPS", + "PublicDescription": "Number of SSE/AVX computational 256-bit packed single precision and 512-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed single precision and double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RSQRT14 RCP RCP14 DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", + "SampleAfterValue": "1000003", + "UMask": "0x18" + }, + { + "BriefDescription": "Counts once for most SIMD scalar computational floating-point instructions retired. Counts twice for DPP and FM(N)ADD/SUB instructions retired.", + "EventCode": "0xC7", + "EventName": "FP_ARITH_INST_RETIRED.SCALAR", + "PublicDescription": "Counts once for most SIMD scalar computational single precision and double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SIMD scalar single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT RCP FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", + "SampleAfterValue": "2000003", + "UMask": "0x3" + }, + { "BriefDescription": "Counts once for most SIMD scalar computational double precision floating-point instructions retired. Counts twice for DPP and FM(N)ADD/SUB instructions retired.", "EventCode": "0xC7", "EventName": "FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", @@ -64,6 +88,13 @@ "UMask": "0x2" }, { + "BriefDescription": "Number of any Vector retired FP arithmetic instructions", + "EventCode": "0xC7", + "EventName": "FP_ARITH_INST_RETIRED.VECTOR", + "SampleAfterValue": "2000003", + "UMask": "0xfc" + }, + { "BriefDescription": "Intel AVX-512 computational 512-bit packed BFloat16 instructions retired.", "EventCode": "0xCF", "EventName": "FP_ARITH_INST_RETIRED2.128BIT_PACKED_BF16", diff --git a/tools/perf/pmu-events/arch/x86/cascadelakex/metricgroups.json b/tools/perf/pmu-events/arch/x86/cascadelakex/metricgroups.json new file mode 100644 index 000000000000..bc6a9a4d27a9 --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/cascadelakex/metricgroups.json @@ -0,0 +1,114 @@ +{ + "Backend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Bad": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BadSpec": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BigFoot": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BrMispredicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Branches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CacheMisses": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CodeGen": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Compute": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Cor": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSBmiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DataSharing": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Fed": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Flops": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpScalar": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Frontend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "HPC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IcMiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "InsType": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IoBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "L2Evicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "LSD": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MachineClears": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBound": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryTLB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_BW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_Lat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MicroSeq": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "OS": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Offcore": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PGO": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Pipeline": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PortsUtil": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Prefetches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Ret": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Retire": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SMT": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Server": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Snoop": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SoC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL1": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL2": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL3mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "TopdownL5": "Metrics for top-down breakdown at level 5", + "TopdownL6": "Metrics for top-down breakdown at level 6", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_L5_group": "Metrics for top-down breakdown at level 5", + "tma_L6_group": "Metrics for top-down breakdown at level 6", + "tma_alu_op_utilization_group": "Metrics contributing to tma_alu_op_utilization category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_branch_resteers_group": "Metrics contributing to tma_branch_resteers category", + "tma_core_bound_group": "Metrics contributing to tma_core_bound category", + "tma_dram_bound_group": "Metrics contributing to tma_dram_bound category", + "tma_dtlb_load_group": "Metrics contributing to tma_dtlb_load category", + "tma_dtlb_store_group": "Metrics contributing to tma_dtlb_store category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_fp_arith_group": "Metrics contributing to tma_fp_arith category", + "tma_fp_vector_group": "Metrics contributing to tma_fp_vector category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_heavy_operations_group": "Metrics contributing to tma_heavy_operations category", + "tma_issue2P": "Metrics related by the issue $issue2P", + "tma_issueBC": "Metrics related by the issue $issueBC", + "tma_issueBM": "Metrics related by the issue $issueBM", + "tma_issueBW": "Metrics related by the issue $issueBW", + "tma_issueD0": "Metrics related by the issue $issueD0", + "tma_issueFB": "Metrics related by the issue $issueFB", + "tma_issueFL": "Metrics related by the issue $issueFL", + "tma_issueL1": "Metrics related by the issue $issueL1", + "tma_issueLat": "Metrics related by the issue $issueLat", + "tma_issueMC": "Metrics related by the issue $issueMC", + "tma_issueMS": "Metrics related by the issue $issueMS", + "tma_issueMV": "Metrics related by the issue $issueMV", + "tma_issueRFO": "Metrics related by the issue $issueRFO", + "tma_issueSL": "Metrics related by the issue $issueSL", + "tma_issueSO": "Metrics related by the issue $issueSO", + "tma_issueSmSt": "Metrics related by the issue $issueSmSt", + "tma_issueSpSt": "Metrics related by the issue $issueSpSt", + "tma_issueSyncxn": "Metrics related by the issue $issueSyncxn", + "tma_issueTLB": "Metrics related by the issue $issueTLB", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_l3_bound_group": "Metrics contributing to tma_l3_bound category", + "tma_light_operations_group": "Metrics contributing to tma_light_operations category", + "tma_load_op_utilization_group": "Metrics contributing to tma_load_op_utilization category", + "tma_mem_latency_group": "Metrics contributing to tma_mem_latency category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_microcode_sequencer_group": "Metrics contributing to tma_microcode_sequencer category", + "tma_mite_group": "Metrics contributing to tma_mite category", + "tma_ports_utilization_group": "Metrics contributing to tma_ports_utilization category", + "tma_ports_utilized_0_group": "Metrics contributing to tma_ports_utilized_0 category", + "tma_ports_utilized_3m_group": "Metrics contributing to tma_ports_utilized_3m category", + "tma_retiring_group": "Metrics contributing to tma_retiring category", + "tma_serializing_operation_group": "Metrics contributing to tma_serializing_operation category", + "tma_store_bound_group": "Metrics contributing to tma_store_bound category", + "tma_store_op_utilization_group": "Metrics contributing to tma_store_op_utilization category" +} diff --git a/tools/perf/pmu-events/arch/x86/cascadelakex/pipeline.json b/tools/perf/pmu-events/arch/x86/cascadelakex/pipeline.json index 0f06e314fe36..31a1663d57f8 100644 --- a/tools/perf/pmu-events/arch/x86/cascadelakex/pipeline.json +++ b/tools/perf/pmu-events/arch/x86/cascadelakex/pipeline.json @@ -26,12 +26,21 @@ "UMask": "0x4" }, { - "BriefDescription": "Conditional branch instructions retired.", + "BriefDescription": "Conditional branch instructions retired. [This event is alias to BR_INST_RETIRED.CONDITIONAL]", + "Errata": "SKL091", + "EventCode": "0xC4", + "EventName": "BR_INST_RETIRED.COND", + "PublicDescription": "This event counts conditional branch instructions retired. [This event is alias to BR_INST_RETIRED.CONDITIONAL]", + "SampleAfterValue": "400009", + "UMask": "0x1" + }, + { + "BriefDescription": "Conditional branch instructions retired. [This event is alias to BR_INST_RETIRED.COND]", "Errata": "SKL091", "EventCode": "0xC4", "EventName": "BR_INST_RETIRED.CONDITIONAL", "PEBS": "1", - "PublicDescription": "This event counts conditional branch instructions retired.", + "PublicDescription": "This event counts conditional branch instructions retired. [This event is alias to BR_INST_RETIRED.COND]", "SampleAfterValue": "400009", "UMask": "0x1" }, @@ -414,6 +423,16 @@ "UMask": "0x1" }, { + "BriefDescription": "Clears speculative count", + "CounterMask": "1", + "EdgeDetect": "1", + "EventCode": "0x0D", + "EventName": "INT_MISC.CLEARS_COUNT", + "PublicDescription": "Counts the number of speculative clears due to any type of branch misprediction or machine clears", + "SampleAfterValue": "2000003", + "UMask": "0x1" + }, + { "BriefDescription": "Cycles the issue-stage is waiting for front-end to fetch from resteered path following branch misprediction or machine clear events.", "EventCode": "0x0D", "EventName": "INT_MISC.CLEAR_RESTEER_CYCLES", diff --git a/tools/perf/pmu-events/arch/x86/elkhartlake/cache.json b/tools/perf/pmu-events/arch/x86/elkhartlake/cache.json index 0ab90e3bf76b..c6be60584522 100644 --- a/tools/perf/pmu-events/arch/x86/elkhartlake/cache.json +++ b/tools/perf/pmu-events/arch/x86/elkhartlake/cache.json @@ -72,6 +72,7 @@ "BriefDescription": "Counts the number of cycles the core is stalled due to an instruction cache or TLB miss which hit in the L2, LLC, DRAM or MMIO (Non-DRAM).", "EventCode": "0x34", "EventName": "MEM_BOUND_STALLS.IFETCH", + "PublicDescription": "Counts the number of cycles the core is stalled due to an instruction cache or translation lookaside buffer (TLB) miss which hit in the L2, LLC, DRAM or MMIO (Non-DRAM).", "SampleAfterValue": "200003", "UMask": "0x38" }, @@ -437,6 +438,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT", "MSRIndex": "0x1a6,0x1a7", @@ -446,6 +448,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_HITM", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7", @@ -455,6 +458,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_HIT_NO_FWD", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_NO_FWD", "MSRIndex": "0x1a6,0x1a7", @@ -464,6 +468,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_HIT_WITH_FWD", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7", @@ -473,6 +478,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_MISS", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7", @@ -482,6 +488,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_NOT_NEEDED", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_NOT_NEEDED", "MSRIndex": "0x1a6,0x1a7", diff --git a/tools/perf/pmu-events/arch/x86/elkhartlake/memory.json b/tools/perf/pmu-events/arch/x86/elkhartlake/memory.json index 18621909d1a9..c02eb0e836ad 100644 --- a/tools/perf/pmu-events/arch/x86/elkhartlake/memory.json +++ b/tools/perf/pmu-events/arch/x86/elkhartlake/memory.json @@ -96,6 +96,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_MISS", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.L3_MISS", "MSRIndex": "0x1a6,0x1a7", @@ -105,6 +106,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_MISS_LOCAL", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.L3_MISS_LOCAL", "MSRIndex": "0x1a6,0x1a7", diff --git a/tools/perf/pmu-events/arch/x86/elkhartlake/other.json b/tools/perf/pmu-events/arch/x86/elkhartlake/other.json index 00ae180ded25..fefbc383b840 100644 --- a/tools/perf/pmu-events/arch/x86/elkhartlake/other.json +++ b/tools/perf/pmu-events/arch/x86/elkhartlake/other.json @@ -1,6 +1,7 @@ [ { "BriefDescription": "This event is deprecated. Refer to new event BUS_LOCK.SELF_LOCKS", + "Deprecated": "1", "EdgeDetect": "1", "EventCode": "0x63", "EventName": "BUS_LOCK.ALL", @@ -16,6 +17,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event BUS_LOCK.BLOCK_CYCLES", + "Deprecated": "1", "EventCode": "0x63", "EventName": "BUS_LOCK.CYCLES_OTHER_BLOCK", "SampleAfterValue": "200003", @@ -23,6 +25,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event BUS_LOCK.LOCK_CYCLES", + "Deprecated": "1", "EventCode": "0x63", "EventName": "BUS_LOCK.CYCLES_SELF_BLOCK", "SampleAfterValue": "200003", @@ -46,6 +49,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event MEM_BOUND_STALLS.LOAD_DRAM_HIT", + "Deprecated": "1", "EventCode": "0x34", "EventName": "C0_STALLS.LOAD_DRAM_HIT", "SampleAfterValue": "200003", @@ -53,6 +57,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event MEM_BOUND_STALLS.LOAD_L2_HIT", + "Deprecated": "1", "EventCode": "0x34", "EventName": "C0_STALLS.LOAD_L2_HIT", "SampleAfterValue": "200003", @@ -60,6 +65,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event MEM_BOUND_STALLS.LOAD_LLC_HIT", + "Deprecated": "1", "EventCode": "0x34", "EventName": "C0_STALLS.LOAD_LLC_HIT", "SampleAfterValue": "200003", @@ -207,6 +213,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.ANY_RESPONSE", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7", @@ -216,6 +223,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.DRAM", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.DRAM", "MSRIndex": "0x1a6,0x1a7", @@ -225,6 +233,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.LOCAL_DRAM", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.LOCAL_DRAM", "MSRIndex": "0x1a6,0x1a7", @@ -234,6 +243,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.OUTSTANDING", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.OUTSTANDING", "MSRIndex": "0x1a6", diff --git a/tools/perf/pmu-events/arch/x86/elkhartlake/pipeline.json b/tools/perf/pmu-events/arch/x86/elkhartlake/pipeline.json index 9dd8c909facc..c483c0838e08 100644 --- a/tools/perf/pmu-events/arch/x86/elkhartlake/pipeline.json +++ b/tools/perf/pmu-events/arch/x86/elkhartlake/pipeline.json @@ -165,6 +165,7 @@ }, { "BriefDescription": "This event is deprecated.", + "Deprecated": "1", "EventCode": "0xcd", "EventName": "CYCLES_DIV_BUSY.ANY", "SampleAfterValue": "2000003" @@ -283,6 +284,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event TOPDOWN_BAD_SPECULATION.FASTNUKE", + "Deprecated": "1", "EventCode": "0x73", "EventName": "TOPDOWN_BAD_SPECULATION.MONUKE", "SampleAfterValue": "1000003", @@ -338,6 +340,7 @@ }, { "BriefDescription": "This event is deprecated.", + "Deprecated": "1", "EventCode": "0x74", "EventName": "TOPDOWN_BE_BOUND.STORE_BUFFER", "SampleAfterValue": "1000003", diff --git a/tools/perf/pmu-events/arch/x86/haswell/hsw-metrics.json b/tools/perf/pmu-events/arch/x86/haswell/hsw-metrics.json index 2528418200bb..79d89c263677 100644 --- a/tools/perf/pmu-events/arch/x86/haswell/hsw-metrics.json +++ b/tools/perf/pmu-events/arch/x86/haswell/hsw-metrics.json @@ -50,7 +50,7 @@ }, { "BriefDescription": "Uncore frequency per die [GHZ]", - "MetricExpr": "tma_info_socket_clks / #num_dies / duration_time / 1e9", + "MetricExpr": "tma_info_system_socket_clks / #num_dies / duration_time / 1e9", "MetricGroup": "SoC", "MetricName": "UNCORE_FREQ" }, @@ -71,7 +71,7 @@ }, { "BriefDescription": "This metric estimates how often memory load accesses were aliased by preceding stores (in program order) with a 4K address offset", - "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_clks", + "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_4k_aliasing", "MetricThreshold": "tma_4k_aliasing > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -81,7 +81,7 @@ { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution ports for ALU operations.", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5 + UOPS_DISPATCHED_PORT.PORT_6) / tma_info_slots", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5 + UOPS_DISPATCHED_PORT.PORT_6) / tma_info_thread_slots", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_alu_op_utilization", "MetricThreshold": "tma_alu_op_utilization > 0.6", @@ -89,7 +89,7 @@ }, { "BriefDescription": "This metric estimates fraction of slots the CPU retired uops delivered by the Microcode_Sequencer as a result of Assists", - "MetricExpr": "100 * OTHER_ASSISTS.ANY_WB_ASSIST / tma_info_slots", + "MetricExpr": "100 * OTHER_ASSISTS.ANY_WB_ASSIST / tma_info_thread_slots", "MetricGroup": "TopdownL4;tma_L4_group;tma_microcode_sequencer_group", "MetricName": "tma_assists", "MetricThreshold": "tma_assists > 0.1 & (tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1)", @@ -103,15 +103,17 @@ "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend. Backend is the portion of the processor core where the out-of-order scheduler dispatches ready uops into their respective execution units; and once completed these uops get retired according to program order. For example; stalls due to data-cache misses or stalls due to the divider unit being overloaded are both categorized under Backend Bound. Backend Bound is further divided into two main categories: Memory Bound and Core Bound.", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations", - "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_slots", + "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots wasted due to incorrect speculations. This include slots used to issue uops that do not eventually get retired and slots for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work due to miss-predicted branches are categorized under Bad Speculation category. Incorrect data speculation followed by Memory Ordering Nukes is another example.", "ScaleUnit": "100%" }, @@ -122,12 +124,13 @@ "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15", - "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_branch_misprediction_cost, tma_mispredicts_resteers", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers", - "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_clks", + "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteers", "MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -147,7 +150,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to contested accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(60 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) + 43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS)))) / tma_info_clks", + "MetricExpr": "(60 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) + 43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS)))) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_contested_accesses", "MetricThreshold": "tma_contested_accesses > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -161,13 +164,14 @@ "MetricGroup": "Backend;Compute;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck. Shortage in hardware compute resources; or dependencies in software's instructions are both categorized under Core Bound. Hence it may indicate the machine ran out of an out-of-order resource; certain execution units are overloaded or dependencies in program's data- or instruction-flow are limiting the performance (e.g. FP-chained long-latency arithmetic operations).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to data-sharing accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) / tma_info_clks", + "MetricExpr": "43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) / tma_info_thread_clks", "MetricGroup": "Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_data_sharing", "MetricThreshold": "tma_data_sharing > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -176,7 +180,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the Divider unit was active", - "MetricExpr": "10 * ARITH.DIVIDER_UOPS / tma_info_core_clks", + "MetricExpr": "10 * ARITH.DIVIDER_UOPS / tma_info_core_core_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_divider", "MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -186,7 +190,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS)) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_clks", + "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS)) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -195,25 +199,25 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline", - "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSB;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_dsb", - "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline. For example; inefficient utilization of the DSB cache structure or bank conflict when reading from it; are categorized here.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_dsb_switches", "MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb, tma_lcp", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses", - "MetricExpr": "(8 * DTLB_LOAD_MISSES.STLB_HIT + DTLB_LOAD_MISSES.WALK_DURATION) / tma_info_clks", + "MetricExpr": "(8 * DTLB_LOAD_MISSES.STLB_HIT + DTLB_LOAD_MISSES.WALK_DURATION) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group", "MetricName": "tma_dtlb_load", "MetricThreshold": "tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -222,7 +226,7 @@ }, { "BriefDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses", - "MetricExpr": "(8 * DTLB_STORE_MISSES.STLB_HIT + DTLB_STORE_MISSES.WALK_DURATION) / tma_info_clks", + "MetricExpr": "(8 * DTLB_STORE_MISSES.STLB_HIT + DTLB_STORE_MISSES.WALK_DURATION) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_store_bound_group", "MetricName": "tma_dtlb_store", "MetricThreshold": "tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -231,7 +235,7 @@ }, { "BriefDescription": "This metric roughly estimates how often CPU was handling synchronizations due to False Sharing", - "MetricExpr": "60 * OFFCORE_RESPONSE.DEMAND_RFO.L3_HIT.HITM_OTHER_CORE / tma_info_clks", + "MetricExpr": "60 * OFFCORE_RESPONSE.DEMAND_RFO.L3_HIT.HITM_OTHER_CORE / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_store_bound_group", "MetricName": "tma_false_sharing", "MetricThreshold": "tma_false_sharing > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -241,11 +245,11 @@ { "BriefDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "tma_info_load_miss_real_latency * cpu@L1D_PEND_MISS.REQUEST_FB_FULL\\,cmask\\=1@ / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * cpu@L1D_PEND_MISS.REQUEST_FB_FULL\\,cmask\\=1@ / tma_info_thread_clks", "MetricGroup": "MemoryBW;TopdownL4;tma_L4_group;tma_issueBW;tma_issueSL;tma_issueSmSt;tma_l1_bound_group", "MetricName": "tma_fb_full", "MetricThreshold": "tma_fb_full > 0.3", - "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", + "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", "ScaleUnit": "100%" }, { @@ -253,25 +257,28 @@ "MetricExpr": "tma_frontend_bound - tma_fetch_latency", "MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", "MetricName": "tma_fetch_bandwidth", - "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_info_iptb, tma_lcp", + "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues", - "MetricExpr": "4 * min(CPU_CLK_UNHALTED.THREAD, IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE) / tma_info_slots", + "MetricExpr": "4 * min(CPU_CLK_UNHALTED.THREAD, IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE) / tma_info_thread_slots", "MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: RS_EVENTS.EMPTY_END", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend", - "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_slots", + "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_thread_slots", "MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound.", "ScaleUnit": "100%" }, @@ -281,329 +288,330 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_heavy_operations", "MetricThreshold": "tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to instruction cache misses.", - "MetricExpr": "ICACHE.IFDATA_STALL / tma_info_clks", + "MetricExpr": "ICACHE.IFDATA_STALL / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;IcMiss;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_icache_misses", "MetricThreshold": "tma_icache_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", "ScaleUnit": "100%" }, { - "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", - "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time", - "MetricGroup": "Power;Summary", - "MetricName": "tma_info_average_frequency" - }, - { - "BriefDescription": "Branch instructions per taken branch.", - "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", - "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_bptkbranch" + "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", + "MetricExpr": "tma_info_inst_mix_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_indirect", + "MetricThreshold": "tma_info_bad_spec_ipmisp_indirect < 1e3" }, { - "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "Pipeline", - "MetricName": "tma_info_clks" + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BadSpec;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmispredict", + "MetricThreshold": "tma_info_bad_spec_ipmispredict < 200" }, { "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", - "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_clks))", + "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_thread_clks))", "MetricGroup": "SMT", - "MetricName": "tma_info_core_clks" + "MetricName": "tma_info_core_core_clks" }, { "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks", "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_coreipc" + "MetricName": "tma_info_core_coreipc" }, { - "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricExpr": "1 / tma_info_ipc", - "MetricGroup": "Mem;Pipeline", - "MetricName": "tma_info_cpi" - }, - { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": "HPC;Summary", - "MetricName": "tma_info_cpu_utilization" - }, - { - "BriefDescription": "Average Parallel L2 cache miss data reads", - "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", - "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_data_l2_mlp" - }, - { - "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricExpr": "64 * (UNC_ARB_TRK_REQUESTS.ALL + UNC_ARB_COH_TRK_REQUESTS.ALL) / 1e6 / duration_time / 1e3", - "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", - "MetricName": "tma_info_dram_bw_use", - "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", + "MetricExpr": "(UOPS_EXECUTED.CORE / 2 / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@) if #SMT_on else UOPS_EXECUTED.CORE / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@))", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", + "MetricName": "tma_info_core_ilp" }, { "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", "MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + LSD.UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)", "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", - "MetricName": "tma_info_dsb_coverage", - "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_iptb, tma_lcp" + "MetricName": "tma_info_frontend_dsb_coverage", + "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 4 > 0.35", + "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_inst_mix_iptb, tma_lcp" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", - "MetricExpr": "(UOPS_EXECUTED.CORE / 2 / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@) if #SMT_on else UOPS_EXECUTED.CORE / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@))", - "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", - "MetricName": "tma_info_ilp" + "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_inst_mix_instructions / BACLEARS.ANY", + "MetricGroup": "Fed", + "MetricName": "tma_info_frontend_ipunknown_branch" + }, + { + "BriefDescription": "Branch instructions per taken branch.", + "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;PGO", + "MetricName": "tma_info_inst_mix_bptkbranch" }, { "BriefDescription": "Total number of retired Instructions", "MetricExpr": "INST_RETIRED.ANY", "MetricGroup": "Summary;TmaL1;tma_L1_group", - "MetricName": "tma_info_instructions", + "MetricName": "tma_info_inst_mix_instructions", "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" }, { "BriefDescription": "Instructions per Branch (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Branches;Fed;InsType", - "MetricName": "tma_info_ipbranch", - "MetricThreshold": "tma_info_ipbranch < 8" - }, - { - "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": "Ret;Summary", - "MetricName": "tma_info_ipc" + "MetricName": "tma_info_inst_mix_ipbranch", + "MetricThreshold": "tma_info_inst_mix_ipbranch < 8" }, { "BriefDescription": "Instructions per (near) call (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_CALL", "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_ipcall", - "MetricThreshold": "tma_info_ipcall < 200" - }, - { - "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", - "MetricGroup": "Branches;OS", - "MetricName": "tma_info_ipfarbranch", - "MetricThreshold": "tma_info_ipfarbranch < 1e6" + "MetricName": "tma_info_inst_mix_ipcall", + "MetricThreshold": "tma_info_inst_mix_ipcall < 200" }, { "BriefDescription": "Instructions per Load (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_LOADS", "MetricGroup": "InsType", - "MetricName": "tma_info_ipload", - "MetricThreshold": "tma_info_ipload < 3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", - "MetricExpr": "tma_info_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_indirect", - "MetricThreshold": "tma_info_ipmisp_indirect < 1e3" - }, - { - "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BadSpec;BrMispredicts", - "MetricName": "tma_info_ipmispredict", - "MetricThreshold": "tma_info_ipmispredict < 200" + "MetricName": "tma_info_inst_mix_ipload", + "MetricThreshold": "tma_info_inst_mix_ipload < 3" }, { "BriefDescription": "Instructions per Store (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_STORES", "MetricGroup": "InsType", - "MetricName": "tma_info_ipstore", - "MetricThreshold": "tma_info_ipstore < 8" + "MetricName": "tma_info_inst_mix_ipstore", + "MetricThreshold": "tma_info_inst_mix_ipstore < 8" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW;Frontend;PGO;tma_issueFB", - "MetricName": "tma_info_iptb", - "MetricThreshold": "tma_info_iptb < 9", - "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_lcp" - }, - { - "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", - "MetricExpr": "tma_info_instructions / BACLEARS.ANY", - "MetricGroup": "Fed", - "MetricName": "tma_info_ipunknown_branch" - }, - { - "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_cpi" - }, - { - "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_utilization", - "MetricThreshold": "tma_info_kernel_utilization > 0.05" + "MetricName": "tma_info_inst_mix_iptb", + "MetricThreshold": "tma_info_inst_mix_iptb < 9", + "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_lcp" }, { "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw" - }, - { - "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "tma_info_l1d_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw_1t" - }, - { - "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L1_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki" + "MetricName": "tma_info_memory_core_l1d_cache_fill_bw" }, { "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw" + "MetricName": "tma_info_memory_core_l2_cache_fill_bw" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "tma_info_l2_cache_fill_bw", + "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw_1t" + "MetricName": "tma_info_memory_core_l3_cache_fill_bw" + }, + { + "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L1_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l1mpki" }, { "BriefDescription": "L2 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L2_MISS / INST_RETIRED.ANY", "MetricGroup": "Backend;CacheMisses;Mem", - "MetricName": "tma_info_l2mpki" + "MetricName": "tma_info_memory_l2mpki" }, { - "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "0", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw_1t" + "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L3_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l3mpki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw" + "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_UOPS_RETIRED.L1_MISS + MEM_LOAD_UOPS_RETIRED.HIT_LFB)", + "MetricGroup": "Mem;MemoryBound;MemoryLat", + "MetricName": "tma_info_memory_load_miss_real_latency" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw_1t" + "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", + "MetricGroup": "Mem;MemoryBW;MemoryBound", + "MetricName": "tma_info_memory_mlp", + "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" }, { - "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L3_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l3mpki" + "BriefDescription": "Average Parallel L2 cache miss data reads", + "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", + "MetricGroup": "Memory_BW;Offcore", + "MetricName": "tma_info_memory_oro_data_l2_mlp" }, { "BriefDescription": "Average Latency for L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS.DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l2_miss_latency" + "MetricName": "tma_info_memory_oro_load_l2_miss_latency" }, { "BriefDescription": "Average Parallel L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_DATA_RD", "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_load_l2_mlp" + "MetricName": "tma_info_memory_oro_load_l2_mlp" }, { - "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_UOPS_RETIRED.L1_MISS + MEM_LOAD_UOPS_RETIRED.HIT_LFB)", - "MetricGroup": "Mem;MemoryBound;MemoryLat", - "MetricName": "tma_info_load_miss_real_latency" + "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l1d_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l1d_cache_fill_bw_1t" }, { - "BriefDescription": "Average number of parallel requests to external memory", - "MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_OCCUPANCY.CYCLES_WITH_ANY_REQUEST", - "MetricGroup": "Mem;SoC", - "MetricName": "tma_info_mem_parallel_requests", - "PublicDescription": "Average number of parallel requests to external memory. Accounts for all requests" + "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l2_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l2_cache_fill_bw_1t" }, { - "BriefDescription": "Average latency of all requests to external memory (in Uncore cycles)", - "MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_REQUESTS.ALL", - "MetricGroup": "Mem;SoC", - "MetricName": "tma_info_mem_request_latency" + "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "0", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_thread_l3_cache_access_bw_1t" }, { - "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", - "MetricGroup": "Mem;MemoryBW;MemoryBound", - "MetricName": "tma_info_mlp", - "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" + "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l3_cache_fill_bw_1t" }, { "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", - "MetricExpr": "(ITLB_MISSES.WALK_DURATION + DTLB_LOAD_MISSES.WALK_DURATION + DTLB_STORE_MISSES.WALK_DURATION) / tma_info_core_clks", + "MetricExpr": "(ITLB_MISSES.WALK_DURATION + DTLB_LOAD_MISSES.WALK_DURATION + DTLB_STORE_MISSES.WALK_DURATION) / tma_info_core_core_clks", "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_page_walks_utilization", - "MetricThreshold": "tma_info_page_walks_utilization > 0.5" + "MetricName": "tma_info_memory_tlb_page_walks_utilization", + "MetricThreshold": "tma_info_memory_tlb_page_walks_utilization > 0.5" }, { "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@", "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_retire" + "MetricName": "tma_info_pipeline_retire" }, { - "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", - "MetricExpr": "4 * tma_info_core_clks", - "MetricGroup": "TmaL1;tma_L1_group", - "MetricName": "tma_info_slots" + "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time", + "MetricGroup": "Power;Summary", + "MetricName": "tma_info_system_average_frequency" + }, + { + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricGroup": "HPC;Summary", + "MetricName": "tma_info_system_cpu_utilization" + }, + { + "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", + "MetricExpr": "64 * (UNC_ARB_TRK_REQUESTS.ALL + UNC_ARB_COH_TRK_REQUESTS.ALL) / 1e6 / duration_time / 1e3", + "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", + "MetricName": "tma_info_system_dram_bw_use", + "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_mem_bandwidth, tma_sq_full" + }, + { + "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", + "MetricGroup": "Branches;OS", + "MetricName": "tma_info_system_ipfarbranch", + "MetricThreshold": "tma_info_system_ipfarbranch < 1e6" + }, + { + "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_cpi" + }, + { + "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_utilization", + "MetricThreshold": "tma_info_system_kernel_utilization > 0.05" + }, + { + "BriefDescription": "Average number of parallel requests to external memory", + "MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_OCCUPANCY.CYCLES_WITH_ANY_REQUEST", + "MetricGroup": "Mem;SoC", + "MetricName": "tma_info_system_mem_parallel_requests", + "PublicDescription": "Average number of parallel requests to external memory. Accounts for all requests" + }, + { + "BriefDescription": "Average latency of all requests to external memory (in Uncore cycles)", + "MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_REQUESTS.ALL", + "MetricGroup": "Mem;SoC", + "MetricName": "tma_info_system_mem_request_latency" }, { "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / (CPU_CLK_UNHALTED.REF_XCLK_ANY / 2) if #SMT_on else 0)", "MetricGroup": "SMT", - "MetricName": "tma_info_smt_2t_utilization" + "MetricName": "tma_info_system_smt_2t_utilization" }, { "BriefDescription": "Socket actual clocks when any core is active on that socket", "MetricExpr": "UNC_CLOCK.SOCKET", "MetricGroup": "SoC", - "MetricName": "tma_info_socket_clks" + "MetricName": "tma_info_system_socket_clks" }, { "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC", "MetricGroup": "Power", - "MetricName": "tma_info_turbo_utilization" + "MetricName": "tma_info_system_turbo_utilization" + }, + { + "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "Pipeline", + "MetricName": "tma_info_thread_clks" + }, + { + "BriefDescription": "Cycles Per Instruction (per Logical Processor)", + "MetricExpr": "1 / tma_info_thread_ipc", + "MetricGroup": "Mem;Pipeline", + "MetricName": "tma_info_thread_cpi" + }, + { + "BriefDescription": "Instructions Per Cycle (per Logical Processor)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks", + "MetricGroup": "Ret;Summary", + "MetricName": "tma_info_thread_ipc" + }, + { + "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", + "MetricExpr": "4 * tma_info_core_core_clks", + "MetricGroup": "TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots" }, { "BriefDescription": "Uops Per Instruction", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / INST_RETIRED.ANY", "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_uoppi", - "MetricThreshold": "tma_info_uoppi > 1.05" + "MetricName": "tma_info_thread_uoppi", + "MetricThreshold": "tma_info_thread_uoppi > 1.05" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW", - "MetricName": "tma_info_uptb", - "MetricThreshold": "tma_info_uptb < 6" + "MetricName": "tma_info_thread_uptb", + "MetricThreshold": "tma_info_thread_uptb < 6" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses", - "MetricExpr": "(14 * ITLB_MISSES.STLB_HIT + ITLB_MISSES.WALK_DURATION) / tma_info_clks", + "MetricExpr": "(14 * ITLB_MISSES.STLB_HIT + ITLB_MISSES.WALK_DURATION) / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -612,7 +620,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled without loads missing the L1 data cache", - "MetricExpr": "max((min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING) - CYCLE_ACTIVITY.STALLS_L1D_PENDING) / tma_info_clks, 0)", + "MetricExpr": "max((min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING) - CYCLE_ACTIVITY.STALLS_L1D_PENDING) / tma_info_thread_clks, 0)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_issueL1;tma_issueMC;tma_memory_bound_group", "MetricName": "tma_l1_bound", "MetricThreshold": "tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -621,7 +629,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled due to L2 cache accesses by loads", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L1D_PENDING - CYCLE_ACTIVITY.STALLS_L2_PENDING) / tma_info_clks", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L1D_PENDING - CYCLE_ACTIVITY.STALLS_L2_PENDING) / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l2_bound", "MetricThreshold": "tma_l2_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -631,7 +639,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_clks", + "MetricExpr": "MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -641,7 +649,7 @@ { "BriefDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "29 * (MEM_LOAD_UOPS_RETIRED.L3_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) / tma_info_clks", + "MetricExpr": "29 * (MEM_LOAD_UOPS_RETIRED.L3_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.L3_MISS))) / tma_info_thread_clks", "MetricGroup": "MemoryLat;TopdownL4;tma_L4_group;tma_issueLat;tma_l3_bound_group", "MetricName": "tma_l3_hit_latency", "MetricThreshold": "tma_l3_hit_latency > 0.1 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -650,11 +658,11 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)", - "MetricExpr": "ILD_STALL.LCP / tma_info_clks", + "MetricExpr": "ILD_STALL.LCP / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_lcp", "MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb", + "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb", "ScaleUnit": "100%" }, { @@ -663,13 +671,14 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_light_operations", "MetricThreshold": "tma_light_operations > 0.6", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation). This correlates with total number of instructions used by the program. A uops-per-instruction (see UopPI metric) ratio of 1 or less should be expected for decently optimized software running on Intel Core/Xeon products. While this often indicates efficient X86 instructions were executed; high value does not necessarily mean better performance cannot be achieved. Sample with: INST_RETIRED.PREC_DIST", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Load operations", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 + UOPS_DISPATCHED_PORT.PORT_7 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 + UOPS_DISPATCHED_PORT.PORT_7 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_load_op_utilization", "MetricThreshold": "tma_load_op_utilization > 0.6", @@ -679,7 +688,7 @@ { "BriefDescription": "This metric represents fraction of cycles the CPU spent handling cache misses due to lock operations", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO) / tma_info_clks", + "MetricExpr": "MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO) / tma_info_thread_clks", "MetricGroup": "Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_l1_bound_group", "MetricName": "tma_lock_latency", "MetricThreshold": "tma_lock_latency > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -693,21 +702,22 @@ "MetricGroup": "BadSpec;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.1 & tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears. These slots are either wasted by uops fetched prior to the clear; or stalls the out-of-order portion of the machine needs to recover its state after the clear. For example; this can happen due to memory ordering Nukes (e.g. Memory Disambiguation) or Self-Modifying-Code (SMC) nukes. Sample with: MACHINE_CLEARS.COUNT. Related metrics: tma_clears_resteers, tma_contested_accesses, tma_data_sharing, tma_false_sharing, tma_l1_bound, tma_microcode_sequencer, tma_ms_switches, tma_remote_cache", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=6@) / tma_info_clks", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=6@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW", "MetricName": "tma_mem_bandwidth", "MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_sq_full", + "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_sq_full", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat", "MetricName": "tma_mem_latency", "MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -717,16 +727,17 @@ { "BriefDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "((min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING) + RESOURCE_STALLS.SB) / (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - (cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ if tma_info_ipc > 1.8 else cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@)) / 2 - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) if #SMT_on else min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - (cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ if tma_info_ipc > 1.8 else cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", + "MetricExpr": "((min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING) + RESOURCE_STALLS.SB) / (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - (cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ if tma_info_thread_ipc > 1.8 else cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@)) / 2 - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) if #SMT_on else min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - (cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ if tma_info_thread_ipc > 1.8 else cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", "MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck. Memory Bound estimates fraction of slots where pipeline is likely stalled due to demand load or store instructions. This accounts mainly for (1) non-completed in-flight memory demand loads which coincides with execution units starvation; in addition to (2) cases where stores could impose backpressure on the pipeline when many of them get buffered at the same time (less common out of the two).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_thread_slots", "MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS", "MetricName": "tma_microcode_sequencer", "MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1", @@ -735,16 +746,16 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline)", - "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_mite", - "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline). This pipeline is used for code that was not pre-cached in the DSB or LSD. For example; inefficiencies due to asymmetric decoders; use of long immediate or LCP can manifest as MITE fetch bandwidth bottleneck.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)", - "MetricExpr": "2 * IDQ.MS_SWITCHES / tma_info_clks", + "MetricExpr": "2 * IDQ.MS_SWITCHES / tma_info_thread_clks", "MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO", "MetricName": "tma_ms_switches", "MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -753,7 +764,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 0 ([SNB+] ALU; [HSW+] ALU and 2nd branch)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_core_clks", "MetricGroup": "Compute;TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_0", "MetricThreshold": "tma_port_0 > 0.6", @@ -762,7 +773,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 1 (ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_1", "MetricThreshold": "tma_port_1 > 0.6", @@ -771,7 +782,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 2 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_2", "MetricThreshold": "tma_port_2 > 0.6", @@ -780,7 +791,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 3 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_3", "MetricThreshold": "tma_port_3 > 0.6", @@ -798,7 +809,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 5 ([SNB+] Branches and ALU; [HSW+] ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_5", "MetricThreshold": "tma_port_5 > 0.6", @@ -807,7 +818,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 6 ([HSW+]Primary Branch and simple ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_6 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_6 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_6", "MetricThreshold": "tma_port_6 > 0.6", @@ -816,7 +827,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 7 ([HSW+]simple Store-address)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_7 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_7 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_store_op_utilization_group", "MetricName": "tma_port_7", "MetricThreshold": "tma_port_7 > 0.6", @@ -826,7 +837,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - (cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ if tma_info_ipc > 1.8 else cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@)) / 2 - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB if #SMT_on else min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - (cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ if tma_info_ipc > 1.8 else cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING)) / tma_info_clks", + "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - (cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ if tma_info_thread_ipc > 1.8 else cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@)) / 2 - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB if #SMT_on else min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - (cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ if tma_info_thread_ipc > 1.8 else cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING)) / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_ports_utilization", "MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -835,7 +846,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed no uops on any execution port (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,inv\\,cmask\\=1@ / 2 if #SMT_on else (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0)) / tma_info_core_clks)", + "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,inv\\,cmask\\=1@ / 2 if #SMT_on else (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0)) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_0", "MetricThreshold": "tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -844,7 +855,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the CPU executed total of 1 uop per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / 2 if #SMT_on else (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / tma_info_core_clks)", + "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / 2 if #SMT_on else (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issueL1;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_1", "MetricThreshold": "tma_ports_utilized_1 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -853,7 +864,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 2 uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / 2 if #SMT_on else (cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / tma_info_core_clks)", + "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / 2 if #SMT_on else (cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issue2P;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_2", "MetricThreshold": "tma_ports_utilized_2 > 0.15 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -862,7 +873,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 3 or more uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise).", - "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ / 2 if #SMT_on else cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / tma_info_core_clks", + "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ / 2 if #SMT_on else cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_3m", "MetricThreshold": "tma_ports_utilized_3m > 0.7 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -870,17 +881,18 @@ }, { "BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired. Ideally; all pipeline slots would be attributed to the Retiring category. Retiring of 100% would indicate the maximum Pipeline_Width throughput was achieved. Maximizing Retiring typically increases the Instructions-per-cycle (see IPC metric). Note that a high Retiring value does not necessary mean there is no room for more performance. For example; Heavy-operations or Microcode Assists are categorized under Retiring. They often indicate suboptimal performance and can often be optimized or avoided. Sample with: UOPS_RETIRED.RETIRE_SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles handling memory load split accesses - load that cross 64-byte cache line boundary", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "tma_info_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_split_loads", "MetricThreshold": "tma_split_loads > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -889,7 +901,7 @@ }, { "BriefDescription": "This metric represents rate of split store accesses", - "MetricExpr": "2 * MEM_UOPS_RETIRED.SPLIT_STORES / tma_info_core_clks", + "MetricExpr": "2 * MEM_UOPS_RETIRED.SPLIT_STORES / tma_info_core_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_issueSpSt;tma_store_bound_group", "MetricName": "tma_split_stores", "MetricThreshold": "tma_split_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -898,16 +910,16 @@ }, { "BriefDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors)", - "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_clks", + "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_core_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueBW;tma_l3_bound_group", "MetricName": "tma_sq_full", "MetricThreshold": "tma_sq_full > 0.3 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_mem_bandwidth", + "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_mem_bandwidth", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write", - "MetricExpr": "RESOURCE_STALLS.SB / tma_info_clks", + "MetricExpr": "RESOURCE_STALLS.SB / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -916,7 +928,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of cycles when the memory subsystem had loads blocked since they could not forward data from earlier (in program order) overlapping stores", - "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_clks", + "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_store_fwd_blk", "MetricThreshold": "tma_store_fwd_blk > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -926,7 +938,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU spent handling L1D store misses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(L2_RQSTS.RFO_HIT * 9 * (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) + (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_clks", + "MetricExpr": "(L2_RQSTS.RFO_HIT * 9 * (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) + (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_thread_clks", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_issueSL;tma_store_bound_group", "MetricName": "tma_store_latency", "MetricThreshold": "tma_store_latency > 0.1 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -935,7 +947,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Store operations", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_core_clks", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_store_op_utilization", "MetricThreshold": "tma_store_op_utilization > 0.6", @@ -943,7 +955,7 @@ }, { "BriefDescription": "This metric serves as an approximation of legacy x87 usage", - "MetricExpr": "INST_RETIRED.X87 * tma_info_uoppi / UOPS_RETIRED.RETIRE_SLOTS", + "MetricExpr": "INST_RETIRED.X87 * tma_info_thread_uoppi / UOPS_RETIRED.RETIRE_SLOTS", "MetricGroup": "Compute;TopdownL4;tma_L4_group;tma_fp_arith_group", "MetricName": "tma_x87_use", "MetricThreshold": "tma_x87_use > 0.1", diff --git a/tools/perf/pmu-events/arch/x86/haswell/metricgroups.json b/tools/perf/pmu-events/arch/x86/haswell/metricgroups.json new file mode 100644 index 000000000000..f6a0258e3241 --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/haswell/metricgroups.json @@ -0,0 +1,107 @@ +{ + "Backend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Bad": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BadSpec": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BigFoot": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BrMispredicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Branches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CacheMisses": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Compute": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Cor": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSBmiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DataSharing": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Fed": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Flops": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpScalar": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Frontend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "HPC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IcMiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "InsType": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "L2Evicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "LSD": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MachineClears": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBound": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryTLB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_BW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_Lat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MicroSeq": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "OS": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Offcore": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PGO": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Pipeline": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PortsUtil": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Ret": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Retire": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SMT": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Server": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Snoop": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SoC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL1": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL2": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL3mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "TopdownL5": "Metrics for top-down breakdown at level 5", + "TopdownL6": "Metrics for top-down breakdown at level 6", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_L5_group": "Metrics for top-down breakdown at level 5", + "tma_L6_group": "Metrics for top-down breakdown at level 6", + "tma_alu_op_utilization_group": "Metrics contributing to tma_alu_op_utilization category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_branch_resteers_group": "Metrics contributing to tma_branch_resteers category", + "tma_core_bound_group": "Metrics contributing to tma_core_bound category", + "tma_dram_bound_group": "Metrics contributing to tma_dram_bound category", + "tma_dtlb_load_group": "Metrics contributing to tma_dtlb_load category", + "tma_dtlb_store_group": "Metrics contributing to tma_dtlb_store category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_fp_arith_group": "Metrics contributing to tma_fp_arith category", + "tma_fp_vector_group": "Metrics contributing to tma_fp_vector category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_heavy_operations_group": "Metrics contributing to tma_heavy_operations category", + "tma_issue2P": "Metrics related by the issue $issue2P", + "tma_issueBM": "Metrics related by the issue $issueBM", + "tma_issueBW": "Metrics related by the issue $issueBW", + "tma_issueFB": "Metrics related by the issue $issueFB", + "tma_issueL1": "Metrics related by the issue $issueL1", + "tma_issueLat": "Metrics related by the issue $issueLat", + "tma_issueMC": "Metrics related by the issue $issueMC", + "tma_issueMS": "Metrics related by the issue $issueMS", + "tma_issueMV": "Metrics related by the issue $issueMV", + "tma_issueRFO": "Metrics related by the issue $issueRFO", + "tma_issueSL": "Metrics related by the issue $issueSL", + "tma_issueSO": "Metrics related by the issue $issueSO", + "tma_issueSmSt": "Metrics related by the issue $issueSmSt", + "tma_issueSpSt": "Metrics related by the issue $issueSpSt", + "tma_issueSyncxn": "Metrics related by the issue $issueSyncxn", + "tma_issueTLB": "Metrics related by the issue $issueTLB", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_l3_bound_group": "Metrics contributing to tma_l3_bound category", + "tma_light_operations_group": "Metrics contributing to tma_light_operations category", + "tma_load_op_utilization_group": "Metrics contributing to tma_load_op_utilization category", + "tma_mem_latency_group": "Metrics contributing to tma_mem_latency category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_microcode_sequencer_group": "Metrics contributing to tma_microcode_sequencer category", + "tma_ports_utilization_group": "Metrics contributing to tma_ports_utilization category", + "tma_ports_utilized_0_group": "Metrics contributing to tma_ports_utilized_0 category", + "tma_ports_utilized_3m_group": "Metrics contributing to tma_ports_utilized_3m category", + "tma_retiring_group": "Metrics contributing to tma_retiring category", + "tma_serializing_operation_group": "Metrics contributing to tma_serializing_operation category", + "tma_store_bound_group": "Metrics contributing to tma_store_bound category", + "tma_store_op_utilization_group": "Metrics contributing to tma_store_op_utilization category" +} diff --git a/tools/perf/pmu-events/arch/x86/haswellx/hsx-metrics.json b/tools/perf/pmu-events/arch/x86/haswellx/hsx-metrics.json index 11f152c346eb..5f451948c893 100644 --- a/tools/perf/pmu-events/arch/x86/haswellx/hsx-metrics.json +++ b/tools/perf/pmu-events/arch/x86/haswellx/hsx-metrics.json @@ -50,11 +50,207 @@ }, { "BriefDescription": "Uncore frequency per die [GHZ]", - "MetricExpr": "tma_info_socket_clks / #num_dies / duration_time / 1e9", + "MetricExpr": "tma_info_system_socket_clks / #num_dies / duration_time / 1e9", "MetricGroup": "SoC", "MetricName": "UNCORE_FREQ" }, { + "BriefDescription": "Cycles per instruction retired; indicating how much time each executed instruction took; in units of cycles.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD / INST_RETIRED.ANY", + "MetricName": "cpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "CPU operating frequency (in GHz)", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD / CPU_CLK_UNHALTED.REF_TSC * #SYSTEM_TSC_FREQ / 1e9", + "MetricName": "cpu_operating_frequency", + "ScaleUnit": "1GHz" + }, + { + "BriefDescription": "Percentage of time spent in the active CPU power state C0", + "MetricExpr": "tma_info_system_cpu_utilization", + "MetricName": "cpu_utilization", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data loads to the total number of completed instructions", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "dtlb_load_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data loads to the total number of completed instructions. This implies it missed in the DTLB and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data stores to the total number of completed instructions", + "MetricExpr": "DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "dtlb_store_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data stores to the total number of completed instructions. This implies it missed in the DTLB and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Bandwidth of IO reads that are initiated by end device controllers that are requesting memory from the CPU.", + "MetricExpr": "cbox@UNC_C_TOR_INSERTS.OPCODE\\,filter_opc\\=0x19e@ * 64 / 1e6 / duration_time", + "MetricName": "io_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Bandwidth of IO writes that are initiated by end device controllers that are writing memory to the CPU.", + "MetricExpr": "cbox@UNC_C_TOR_INSERTS.OPCODE\\,filter_opc\\=0x1c8\\,filter_tid\\=0x3e@ * 64 / 1e6 / duration_time", + "MetricName": "io_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for 2 megabyte and 4 megabyte page sizes) caused by a code fetch to the total number of completed instructions", + "MetricExpr": "ITLB_MISSES.WALK_COMPLETED_2M_4M / INST_RETIRED.ANY", + "MetricName": "itlb_large_page_mpi", + "PublicDescription": "Ratio of number of completed page walks (for 2 megabyte and 4 megabyte page sizes) caused by a code fetch to the total number of completed instructions. This implies it missed in the Instruction Translation Lookaside Buffer (ITLB) and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by a code fetch to the total number of completed instructions", + "MetricExpr": "ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "itlb_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by a code fetch to the total number of completed instructions. This implies it missed in the ITLB (Instruction TLB) and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read requests missing in L1 instruction cache (includes prefetches) to the total number of completed instructions", + "MetricExpr": "L2_RQSTS.ALL_CODE_RD / INST_RETIRED.ANY", + "MetricName": "l1_i_code_read_misses_with_prefetches_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of demand load requests hitting in L1 data cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_UOPS_RETIRED.L1_HIT / INST_RETIRED.ANY", + "MetricName": "l1d_demand_data_read_hits_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of requests missing L1 data cache (includes data+rfo w/ prefetches) to the total number of completed instructions", + "MetricExpr": "L1D.REPLACEMENT / INST_RETIRED.ANY", + "MetricName": "l1d_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read request missing L2 cache to the total number of completed instructions", + "MetricExpr": "L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", + "MetricName": "l2_demand_code_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed demand load requests hitting in L2 cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_UOPS_RETIRED.L2_HIT / INST_RETIRED.ANY", + "MetricName": "l2_demand_data_read_hits_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed data read request missing L2 cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_UOPS_RETIRED.L2_MISS / INST_RETIRED.ANY", + "MetricName": "l2_demand_data_read_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of requests missing L2 cache (includes code+data+rfo w/ prefetches) to the total number of completed instructions", + "MetricExpr": "L2_LINES_IN.ALL / INST_RETIRED.ANY", + "MetricName": "l2_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read requests missing last level core cache (includes demand w/ prefetches) to the total number of completed instructions", + "MetricExpr": "(cbox@UNC_C_TOR_INSERTS.MISS_OPCODE\\,filter_opc\\=0x181@ + cbox@UNC_C_TOR_INSERTS.MISS_OPCODE\\,filter_opc\\=0x191@) / INST_RETIRED.ANY", + "MetricName": "llc_code_read_mpi_demand_plus_prefetch", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand and prefetch data read miss (read memory access) in nano seconds", + "MetricExpr": "1e9 * (cbox@UNC_C_TOR_OCCUPANCY.MISS_OPCODE\\,filter_opc\\=0x182@ / cbox@UNC_C_TOR_INSERTS.MISS_OPCODE\\,filter_opc\\=0x182@) / (UNC_C_CLOCKTICKS / (#num_cores / #num_packages * #num_packages)) * duration_time", + "MetricName": "llc_data_read_demand_plus_prefetch_miss_latency", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand and prefetch data read miss (read memory access) addressed to local memory in nano seconds", + "MetricExpr": "1e9 * (cbox@UNC_C_TOR_OCCUPANCY.MISS_LOCAL_OPCODE\\,filter_opc\\=0x182@ / cbox@UNC_C_TOR_INSERTS.MISS_LOCAL_OPCODE\\,filter_opc\\=0x182@) / (UNC_C_CLOCKTICKS / (#num_cores / #num_packages * #num_packages)) * duration_time", + "MetricName": "llc_data_read_demand_plus_prefetch_miss_latency_for_local_requests", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand and prefetch data read miss (read memory access) addressed to remote memory in nano seconds", + "MetricExpr": "1e9 * (cbox@UNC_C_TOR_OCCUPANCY.MISS_REMOTE_OPCODE\\,filter_opc\\=0x182@ / cbox@UNC_C_TOR_INSERTS.MISS_REMOTE_OPCODE\\,filter_opc\\=0x182@) / (UNC_C_CLOCKTICKS / (#num_cores / #num_packages * #num_packages)) * duration_time", + "MetricName": "llc_data_read_demand_plus_prefetch_miss_latency_for_remote_requests", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Ratio of number of data read requests missing last level core cache (includes demand w/ prefetches) to the total number of completed instructions", + "MetricExpr": "(cbox@UNC_C_TOR_INSERTS.MISS_OPCODE\\,filter_opc\\=0x182@ + cbox@UNC_C_TOR_INSERTS.MISS_OPCODE\\,filter_opc\\=0x192@) / INST_RETIRED.ANY", + "MetricName": "llc_data_read_mpi_demand_plus_prefetch", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "The ratio of number of completed memory load instructions to the total number completed instructions", + "MetricExpr": "MEM_UOPS_RETIRED.ALL_LOADS / INST_RETIRED.ANY", + "MetricName": "loads_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "DDR memory read bandwidth (MB/sec)", + "MetricExpr": "UNC_M_CAS_COUNT.RD * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "DDR memory bandwidth (MB/sec)", + "MetricExpr": "(UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_total", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "DDR memory write bandwidth (MB/sec)", + "MetricExpr": "UNC_M_CAS_COUNT.WR * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Memory read that miss the last level cache (LLC) addressed to local DRAM as a percentage of total memory read accesses, does not include LLC prefetches.", + "MetricExpr": "cbox@UNC_C_TOR_INSERTS.MISS_LOCAL_OPCODE\\,filter_opc\\=0x182@ / (cbox@UNC_C_TOR_INSERTS.MISS_LOCAL_OPCODE\\,filter_opc\\=0x182@ + cbox@UNC_C_TOR_INSERTS.MISS_REMOTE_OPCODE\\,filter_opc\\=0x182@)", + "MetricName": "numa_reads_addressed_to_local_dram", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Memory reads that miss the last level cache (LLC) addressed to remote DRAM as a percentage of total memory read accesses, does not include LLC prefetches.", + "MetricExpr": "cbox@UNC_C_TOR_INSERTS.MISS_REMOTE_OPCODE\\,filter_opc\\=0x182@ / (cbox@UNC_C_TOR_INSERTS.MISS_LOCAL_OPCODE\\,filter_opc\\=0x182@ + cbox@UNC_C_TOR_INSERTS.MISS_REMOTE_OPCODE\\,filter_opc\\=0x182@)", + "MetricName": "numa_reads_addressed_to_remote_dram", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from decoded instruction cache (decoded stream buffer or DSB) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.DSB_UOPS / UOPS_ISSUED.ANY", + "MetricName": "percent_uops_delivered_from_decoded_icache", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from legacy decode pipeline (Micro-instruction Translation Engine or MITE) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.MITE_UOPS / UOPS_ISSUED.ANY", + "MetricName": "percent_uops_delivered_from_legacy_decode_pipeline", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from loop stream detector(LSD) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "(UOPS_ISSUED.ANY - IDQ.MITE_UOPS - IDQ.MS_UOPS - IDQ.DSB_UOPS) / UOPS_ISSUED.ANY", + "MetricName": "percent_uops_delivered_from_loop_stream_detector", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from microcode sequencer (MS) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.MS_UOPS / UOPS_ISSUED.ANY", + "MetricName": "percent_uops_delivered_from_microcode_sequencer", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Intel(R) Quick Path Interconnect (QPI) data transmit bandwidth (MB/sec)", + "MetricExpr": "UNC_Q_TxL_FLITS_G0.DATA * 8 / 1e6 / duration_time", + "MetricName": "qpi_data_transmit_bw", + "ScaleUnit": "1MB/s" + }, + { "BriefDescription": "Percentage of cycles spent in System Management Interrupts.", "MetricExpr": "((msr@aperf@ - cycles) / msr@aperf@ if msr@smi@ > 0 else 0)", "MetricGroup": "smi", @@ -70,8 +266,14 @@ "ScaleUnit": "1SMI#" }, { + "BriefDescription": "The ratio of number of completed memory store instructions to the total number completed instructions", + "MetricExpr": "MEM_UOPS_RETIRED.ALL_STORES / INST_RETIRED.ANY", + "MetricName": "stores_per_instr", + "ScaleUnit": "1per_instr" + }, + { "BriefDescription": "This metric estimates how often memory load accesses were aliased by preceding stores (in program order) with a 4K address offset", - "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_clks", + "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_4k_aliasing", "MetricThreshold": "tma_4k_aliasing > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -81,7 +283,7 @@ { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution ports for ALU operations.", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5 + UOPS_DISPATCHED_PORT.PORT_6) / tma_info_slots", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5 + UOPS_DISPATCHED_PORT.PORT_6) / tma_info_thread_slots", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_alu_op_utilization", "MetricThreshold": "tma_alu_op_utilization > 0.6", @@ -89,7 +291,7 @@ }, { "BriefDescription": "This metric estimates fraction of slots the CPU retired uops delivered by the Microcode_Sequencer as a result of Assists", - "MetricExpr": "100 * OTHER_ASSISTS.ANY_WB_ASSIST / tma_info_slots", + "MetricExpr": "100 * OTHER_ASSISTS.ANY_WB_ASSIST / tma_info_thread_slots", "MetricGroup": "TopdownL4;tma_L4_group;tma_microcode_sequencer_group", "MetricName": "tma_assists", "MetricThreshold": "tma_assists > 0.1 & (tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1)", @@ -103,15 +305,17 @@ "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend. Backend is the portion of the processor core where the out-of-order scheduler dispatches ready uops into their respective execution units; and once completed these uops get retired according to program order. For example; stalls due to data-cache misses or stalls due to the divider unit being overloaded are both categorized under Backend Bound. Backend Bound is further divided into two main categories: Memory Bound and Core Bound.", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations", - "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_slots", + "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots wasted due to incorrect speculations. This include slots used to issue uops that do not eventually get retired and slots for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work due to miss-predicted branches are categorized under Bad Speculation category. Incorrect data speculation followed by Memory Ordering Nukes is another example.", "ScaleUnit": "100%" }, @@ -122,12 +326,13 @@ "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15", - "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_branch_misprediction_cost, tma_mispredicts_resteers", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers", - "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_clks", + "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteers", "MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -147,7 +352,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to contested accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(60 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) + 43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD)))) / tma_info_clks", + "MetricExpr": "(60 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) + 43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD)))) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_contested_accesses", "MetricThreshold": "tma_contested_accesses > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -161,13 +366,14 @@ "MetricGroup": "Backend;Compute;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck. Shortage in hardware compute resources; or dependencies in software's instructions are both categorized under Core Bound. Hence it may indicate the machine ran out of an out-of-order resource; certain execution units are overloaded or dependencies in program's data- or instruction-flow are limiting the performance (e.g. FP-chained long-latency arithmetic operations).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to data-sharing accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) / tma_info_clks", + "MetricExpr": "43 * (MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) / tma_info_thread_clks", "MetricGroup": "Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_data_sharing", "MetricThreshold": "tma_data_sharing > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -176,7 +382,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the Divider unit was active", - "MetricExpr": "10 * ARITH.DIVIDER_UOPS / tma_info_core_clks", + "MetricExpr": "10 * ARITH.DIVIDER_UOPS / tma_info_core_core_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_divider", "MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -186,7 +392,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS)) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_clks", + "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS)) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -195,25 +401,25 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline", - "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSB;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_dsb", - "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline. For example; inefficient utilization of the DSB cache structure or bank conflict when reading from it; are categorized here.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_dsb_switches", "MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb, tma_lcp", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses", - "MetricExpr": "(8 * DTLB_LOAD_MISSES.STLB_HIT + DTLB_LOAD_MISSES.WALK_DURATION) / tma_info_clks", + "MetricExpr": "(8 * DTLB_LOAD_MISSES.STLB_HIT + DTLB_LOAD_MISSES.WALK_DURATION) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group", "MetricName": "tma_dtlb_load", "MetricThreshold": "tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -222,7 +428,7 @@ }, { "BriefDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses", - "MetricExpr": "(8 * DTLB_STORE_MISSES.STLB_HIT + DTLB_STORE_MISSES.WALK_DURATION) / tma_info_clks", + "MetricExpr": "(8 * DTLB_STORE_MISSES.STLB_HIT + DTLB_STORE_MISSES.WALK_DURATION) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_store_bound_group", "MetricName": "tma_dtlb_store", "MetricThreshold": "tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -231,7 +437,7 @@ }, { "BriefDescription": "This metric roughly estimates how often CPU was handling synchronizations due to False Sharing", - "MetricExpr": "(200 * OFFCORE_RESPONSE.DEMAND_RFO.LLC_MISS.REMOTE_HITM + 60 * OFFCORE_RESPONSE.DEMAND_RFO.LLC_HIT.HITM_OTHER_CORE) / tma_info_clks", + "MetricExpr": "(200 * OFFCORE_RESPONSE.DEMAND_RFO.LLC_MISS.REMOTE_HITM + 60 * OFFCORE_RESPONSE.DEMAND_RFO.LLC_HIT.HITM_OTHER_CORE) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_store_bound_group", "MetricName": "tma_false_sharing", "MetricThreshold": "tma_false_sharing > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -241,11 +447,11 @@ { "BriefDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "tma_info_load_miss_real_latency * cpu@L1D_PEND_MISS.REQUEST_FB_FULL\\,cmask\\=1@ / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * cpu@L1D_PEND_MISS.REQUEST_FB_FULL\\,cmask\\=1@ / tma_info_thread_clks", "MetricGroup": "MemoryBW;TopdownL4;tma_L4_group;tma_issueBW;tma_issueSL;tma_issueSmSt;tma_l1_bound_group", "MetricName": "tma_fb_full", "MetricThreshold": "tma_fb_full > 0.3", - "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", + "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", "ScaleUnit": "100%" }, { @@ -253,25 +459,28 @@ "MetricExpr": "tma_frontend_bound - tma_fetch_latency", "MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", "MetricName": "tma_fetch_bandwidth", - "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_info_iptb, tma_lcp", + "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues", - "MetricExpr": "4 * min(CPU_CLK_UNHALTED.THREAD, IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE) / tma_info_slots", + "MetricExpr": "4 * min(CPU_CLK_UNHALTED.THREAD, IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE) / tma_info_thread_slots", "MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: RS_EVENTS.EMPTY_END", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend", - "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_slots", + "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_thread_slots", "MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound.", "ScaleUnit": "100%" }, @@ -281,330 +490,331 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_heavy_operations", "MetricThreshold": "tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to instruction cache misses.", - "MetricExpr": "ICACHE.IFDATA_STALL / tma_info_clks", + "MetricExpr": "ICACHE.IFDATA_STALL / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;IcMiss;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_icache_misses", "MetricThreshold": "tma_icache_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", "ScaleUnit": "100%" }, { - "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", - "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time", - "MetricGroup": "Power;Summary", - "MetricName": "tma_info_average_frequency" - }, - { - "BriefDescription": "Branch instructions per taken branch.", - "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", - "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_bptkbranch" + "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", + "MetricExpr": "tma_info_inst_mix_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_indirect", + "MetricThreshold": "tma_info_bad_spec_ipmisp_indirect < 1e3" }, { - "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "Pipeline", - "MetricName": "tma_info_clks" + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BadSpec;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmispredict", + "MetricThreshold": "tma_info_bad_spec_ipmispredict < 200" }, { "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", - "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_clks))", + "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_thread_clks))", "MetricGroup": "SMT", - "MetricName": "tma_info_core_clks" + "MetricName": "tma_info_core_core_clks" }, { "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks", "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_coreipc" + "MetricName": "tma_info_core_coreipc" }, { - "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricExpr": "1 / tma_info_ipc", - "MetricGroup": "Mem;Pipeline", - "MetricName": "tma_info_cpi" - }, - { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": "HPC;Summary", - "MetricName": "tma_info_cpu_utilization" - }, - { - "BriefDescription": "Average Parallel L2 cache miss data reads", - "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", - "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_data_l2_mlp" - }, - { - "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time", - "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", - "MetricName": "tma_info_dram_bw_use", - "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", + "MetricExpr": "(UOPS_EXECUTED.CORE / 2 / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@) if #SMT_on else UOPS_EXECUTED.CORE / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@))", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", + "MetricName": "tma_info_core_ilp" }, { "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", "MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + LSD.UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)", "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", - "MetricName": "tma_info_dsb_coverage", - "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_iptb, tma_lcp" + "MetricName": "tma_info_frontend_dsb_coverage", + "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 4 > 0.35", + "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_inst_mix_iptb, tma_lcp" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", - "MetricExpr": "(UOPS_EXECUTED.CORE / 2 / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@) if #SMT_on else UOPS_EXECUTED.CORE / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@))", - "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", - "MetricName": "tma_info_ilp" + "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_inst_mix_instructions / BACLEARS.ANY", + "MetricGroup": "Fed", + "MetricName": "tma_info_frontend_ipunknown_branch" + }, + { + "BriefDescription": "Branch instructions per taken branch.", + "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;PGO", + "MetricName": "tma_info_inst_mix_bptkbranch" }, { "BriefDescription": "Total number of retired Instructions", "MetricExpr": "INST_RETIRED.ANY", "MetricGroup": "Summary;TmaL1;tma_L1_group", - "MetricName": "tma_info_instructions", + "MetricName": "tma_info_inst_mix_instructions", "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" }, { "BriefDescription": "Instructions per Branch (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Branches;Fed;InsType", - "MetricName": "tma_info_ipbranch", - "MetricThreshold": "tma_info_ipbranch < 8" - }, - { - "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": "Ret;Summary", - "MetricName": "tma_info_ipc" + "MetricName": "tma_info_inst_mix_ipbranch", + "MetricThreshold": "tma_info_inst_mix_ipbranch < 8" }, { "BriefDescription": "Instructions per (near) call (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_CALL", "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_ipcall", - "MetricThreshold": "tma_info_ipcall < 200" - }, - { - "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", - "MetricGroup": "Branches;OS", - "MetricName": "tma_info_ipfarbranch", - "MetricThreshold": "tma_info_ipfarbranch < 1e6" + "MetricName": "tma_info_inst_mix_ipcall", + "MetricThreshold": "tma_info_inst_mix_ipcall < 200" }, { "BriefDescription": "Instructions per Load (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_LOADS", "MetricGroup": "InsType", - "MetricName": "tma_info_ipload", - "MetricThreshold": "tma_info_ipload < 3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", - "MetricExpr": "tma_info_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_indirect", - "MetricThreshold": "tma_info_ipmisp_indirect < 1e3" - }, - { - "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BadSpec;BrMispredicts", - "MetricName": "tma_info_ipmispredict", - "MetricThreshold": "tma_info_ipmispredict < 200" + "MetricName": "tma_info_inst_mix_ipload", + "MetricThreshold": "tma_info_inst_mix_ipload < 3" }, { "BriefDescription": "Instructions per Store (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_STORES", "MetricGroup": "InsType", - "MetricName": "tma_info_ipstore", - "MetricThreshold": "tma_info_ipstore < 8" + "MetricName": "tma_info_inst_mix_ipstore", + "MetricThreshold": "tma_info_inst_mix_ipstore < 8" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW;Frontend;PGO;tma_issueFB", - "MetricName": "tma_info_iptb", - "MetricThreshold": "tma_info_iptb < 9", - "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_lcp" - }, - { - "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", - "MetricExpr": "tma_info_instructions / BACLEARS.ANY", - "MetricGroup": "Fed", - "MetricName": "tma_info_ipunknown_branch" - }, - { - "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_cpi" - }, - { - "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_utilization", - "MetricThreshold": "tma_info_kernel_utilization > 0.05" + "MetricName": "tma_info_inst_mix_iptb", + "MetricThreshold": "tma_info_inst_mix_iptb < 9", + "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_lcp" }, { "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw" - }, - { - "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "tma_info_l1d_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw_1t" - }, - { - "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L1_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki" + "MetricName": "tma_info_memory_core_l1d_cache_fill_bw" }, { "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw" + "MetricName": "tma_info_memory_core_l2_cache_fill_bw" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "tma_info_l2_cache_fill_bw", + "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw_1t" + "MetricName": "tma_info_memory_core_l3_cache_fill_bw" + }, + { + "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L1_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l1mpki" }, { "BriefDescription": "L2 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L2_MISS / INST_RETIRED.ANY", "MetricGroup": "Backend;CacheMisses;Mem", - "MetricName": "tma_info_l2mpki" + "MetricName": "tma_info_memory_l2mpki" }, { - "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "0", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw_1t" + "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L3_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l3mpki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw" + "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_UOPS_RETIRED.L1_MISS + MEM_LOAD_UOPS_RETIRED.HIT_LFB)", + "MetricGroup": "Mem;MemoryBound;MemoryLat", + "MetricName": "tma_info_memory_load_miss_real_latency" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw_1t" + "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", + "MetricGroup": "Mem;MemoryBW;MemoryBound", + "MetricName": "tma_info_memory_mlp", + "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" }, { - "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L3_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l3mpki" + "BriefDescription": "Average Parallel L2 cache miss data reads", + "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", + "MetricGroup": "Memory_BW;Offcore", + "MetricName": "tma_info_memory_oro_data_l2_mlp" }, { "BriefDescription": "Average Latency for L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS.DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l2_miss_latency" + "MetricName": "tma_info_memory_oro_load_l2_miss_latency" }, { "BriefDescription": "Average Parallel L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_DATA_RD", "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_load_l2_mlp" + "MetricName": "tma_info_memory_oro_load_l2_mlp" }, { - "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_UOPS_RETIRED.L1_MISS + MEM_LOAD_UOPS_RETIRED.HIT_LFB)", - "MetricGroup": "Mem;MemoryBound;MemoryLat", - "MetricName": "tma_info_load_miss_real_latency" + "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l1d_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l1d_cache_fill_bw_1t" }, { - "BriefDescription": "Average number of parallel data read requests to external memory", - "MetricExpr": "UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182\\,thresh\\=1@", - "MetricGroup": "Mem;MemoryBW;SoC", - "MetricName": "tma_info_mem_parallel_reads", - "PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches" + "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l2_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l2_cache_fill_bw_1t" }, { - "BriefDescription": "Average latency of data read request to external memory (in nanoseconds)", - "MetricExpr": "1e9 * (UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_INSERTS.MISS_OPCODE@filter_opc\\=0x182@) / (tma_info_socket_clks / duration_time)", - "MetricGroup": "Mem;MemoryLat;SoC", - "MetricName": "tma_info_mem_read_latency", - "PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)" + "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "0", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_thread_l3_cache_access_bw_1t" }, { - "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", - "MetricGroup": "Mem;MemoryBW;MemoryBound", - "MetricName": "tma_info_mlp", - "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" + "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l3_cache_fill_bw_1t" }, { "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", - "MetricExpr": "(ITLB_MISSES.WALK_DURATION + DTLB_LOAD_MISSES.WALK_DURATION + DTLB_STORE_MISSES.WALK_DURATION) / tma_info_core_clks", + "MetricExpr": "(ITLB_MISSES.WALK_DURATION + DTLB_LOAD_MISSES.WALK_DURATION + DTLB_STORE_MISSES.WALK_DURATION) / tma_info_core_core_clks", "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_page_walks_utilization", - "MetricThreshold": "tma_info_page_walks_utilization > 0.5" + "MetricName": "tma_info_memory_tlb_page_walks_utilization", + "MetricThreshold": "tma_info_memory_tlb_page_walks_utilization > 0.5" }, { "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@", "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_retire" + "MetricName": "tma_info_pipeline_retire" }, { - "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", - "MetricExpr": "4 * tma_info_core_clks", - "MetricGroup": "TmaL1;tma_L1_group", - "MetricName": "tma_info_slots" + "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time", + "MetricGroup": "Power;Summary", + "MetricName": "tma_info_system_average_frequency" + }, + { + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricGroup": "HPC;Summary", + "MetricName": "tma_info_system_cpu_utilization" + }, + { + "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", + "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time", + "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", + "MetricName": "tma_info_system_dram_bw_use", + "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_mem_bandwidth, tma_sq_full" + }, + { + "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", + "MetricGroup": "Branches;OS", + "MetricName": "tma_info_system_ipfarbranch", + "MetricThreshold": "tma_info_system_ipfarbranch < 1e6" + }, + { + "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_cpi" + }, + { + "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_utilization", + "MetricThreshold": "tma_info_system_kernel_utilization > 0.05" + }, + { + "BriefDescription": "Average number of parallel data read requests to external memory", + "MetricExpr": "UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182\\,thresh\\=1@", + "MetricGroup": "Mem;MemoryBW;SoC", + "MetricName": "tma_info_system_mem_parallel_reads", + "PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches" + }, + { + "BriefDescription": "Average latency of data read request to external memory (in nanoseconds)", + "MetricExpr": "1e9 * (UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_INSERTS.MISS_OPCODE@filter_opc\\=0x182@) / (tma_info_system_socket_clks / duration_time)", + "MetricGroup": "Mem;MemoryLat;SoC", + "MetricName": "tma_info_system_mem_read_latency", + "PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)" }, { "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / (CPU_CLK_UNHALTED.REF_XCLK_ANY / 2) if #SMT_on else 0)", "MetricGroup": "SMT", - "MetricName": "tma_info_smt_2t_utilization" + "MetricName": "tma_info_system_smt_2t_utilization" }, { "BriefDescription": "Socket actual clocks when any core is active on that socket", "MetricExpr": "cbox_0@event\\=0x0@", "MetricGroup": "SoC", - "MetricName": "tma_info_socket_clks" + "MetricName": "tma_info_system_socket_clks" }, { "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC", "MetricGroup": "Power", - "MetricName": "tma_info_turbo_utilization" + "MetricName": "tma_info_system_turbo_utilization" + }, + { + "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "Pipeline", + "MetricName": "tma_info_thread_clks" + }, + { + "BriefDescription": "Cycles Per Instruction (per Logical Processor)", + "MetricExpr": "1 / tma_info_thread_ipc", + "MetricGroup": "Mem;Pipeline", + "MetricName": "tma_info_thread_cpi" + }, + { + "BriefDescription": "Instructions Per Cycle (per Logical Processor)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks", + "MetricGroup": "Ret;Summary", + "MetricName": "tma_info_thread_ipc" + }, + { + "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", + "MetricExpr": "4 * tma_info_core_core_clks", + "MetricGroup": "TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots" }, { "BriefDescription": "Uops Per Instruction", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / INST_RETIRED.ANY", "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_uoppi", - "MetricThreshold": "tma_info_uoppi > 1.05" + "MetricName": "tma_info_thread_uoppi", + "MetricThreshold": "tma_info_thread_uoppi > 1.05" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW", - "MetricName": "tma_info_uptb", - "MetricThreshold": "tma_info_uptb < 6" + "MetricName": "tma_info_thread_uptb", + "MetricThreshold": "tma_info_thread_uptb < 6" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses", - "MetricExpr": "(14 * ITLB_MISSES.STLB_HIT + ITLB_MISSES.WALK_DURATION) / tma_info_clks", + "MetricExpr": "(14 * ITLB_MISSES.STLB_HIT + ITLB_MISSES.WALK_DURATION) / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -613,7 +823,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled without loads missing the L1 data cache", - "MetricExpr": "max((min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING) - CYCLE_ACTIVITY.STALLS_L1D_PENDING) / tma_info_clks, 0)", + "MetricExpr": "max((min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING) - CYCLE_ACTIVITY.STALLS_L1D_PENDING) / tma_info_thread_clks, 0)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_issueL1;tma_issueMC;tma_memory_bound_group", "MetricName": "tma_l1_bound", "MetricThreshold": "tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -622,7 +832,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled due to L2 cache accesses by loads", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L1D_PENDING - CYCLE_ACTIVITY.STALLS_L2_PENDING) / tma_info_clks", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L1D_PENDING - CYCLE_ACTIVITY.STALLS_L2_PENDING) / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l2_bound", "MetricThreshold": "tma_l2_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -632,7 +842,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_clks", + "MetricExpr": "MEM_LOAD_UOPS_RETIRED.L3_HIT / (MEM_LOAD_UOPS_RETIRED.L3_HIT + 7 * MEM_LOAD_UOPS_RETIRED.L3_MISS) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -642,7 +852,7 @@ { "BriefDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "41 * (MEM_LOAD_UOPS_RETIRED.L3_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) / tma_info_clks", + "MetricExpr": "41 * (MEM_LOAD_UOPS_RETIRED.L3_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) / tma_info_thread_clks", "MetricGroup": "MemoryLat;TopdownL4;tma_L4_group;tma_issueLat;tma_l3_bound_group", "MetricName": "tma_l3_hit_latency", "MetricThreshold": "tma_l3_hit_latency > 0.1 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -651,11 +861,11 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)", - "MetricExpr": "ILD_STALL.LCP / tma_info_clks", + "MetricExpr": "ILD_STALL.LCP / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_lcp", "MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb", + "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb", "ScaleUnit": "100%" }, { @@ -664,13 +874,14 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_light_operations", "MetricThreshold": "tma_light_operations > 0.6", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation). This correlates with total number of instructions used by the program. A uops-per-instruction (see UopPI metric) ratio of 1 or less should be expected for decently optimized software running on Intel Core/Xeon products. While this often indicates efficient X86 instructions were executed; high value does not necessarily mean better performance cannot be achieved. Sample with: INST_RETIRED.PREC_DIST", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Load operations", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 + UOPS_DISPATCHED_PORT.PORT_7 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 + UOPS_DISPATCHED_PORT.PORT_7 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_load_op_utilization", "MetricThreshold": "tma_load_op_utilization > 0.6", @@ -680,7 +891,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from local memory", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "200 * (MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) / tma_info_clks", + "MetricExpr": "200 * (MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) / tma_info_thread_clks", "MetricGroup": "Server;TopdownL5;tma_L5_group;tma_mem_latency_group", "MetricName": "tma_local_dram", "MetricThreshold": "tma_local_dram > 0.1 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -690,7 +901,7 @@ { "BriefDescription": "This metric represents fraction of cycles the CPU spent handling cache misses due to lock operations", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO) / tma_info_clks", + "MetricExpr": "MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO) / tma_info_thread_clks", "MetricGroup": "Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_l1_bound_group", "MetricName": "tma_lock_latency", "MetricThreshold": "tma_lock_latency > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -704,21 +915,22 @@ "MetricGroup": "BadSpec;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.1 & tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears. These slots are either wasted by uops fetched prior to the clear; or stalls the out-of-order portion of the machine needs to recover its state after the clear. For example; this can happen due to memory ordering Nukes (e.g. Memory Disambiguation) or Self-Modifying-Code (SMC) nukes. Sample with: MACHINE_CLEARS.COUNT. Related metrics: tma_clears_resteers, tma_contested_accesses, tma_data_sharing, tma_false_sharing, tma_l1_bound, tma_microcode_sequencer, tma_ms_switches, tma_remote_cache", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=6@) / tma_info_clks", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=6@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW", "MetricName": "tma_mem_bandwidth", "MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_sq_full", + "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_sq_full", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat", "MetricName": "tma_mem_latency", "MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -728,16 +940,17 @@ { "BriefDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "((min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING) + RESOURCE_STALLS.SB) / (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - (cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ if tma_info_ipc > 1.8 else cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@)) / 2 - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) if #SMT_on else min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - (cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ if tma_info_ipc > 1.8 else cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", + "MetricExpr": "((min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING) + RESOURCE_STALLS.SB) / (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - (cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ if tma_info_thread_ipc > 1.8 else cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@)) / 2 - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) if #SMT_on else min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - (cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ if tma_info_thread_ipc > 1.8 else cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", "MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck. Memory Bound estimates fraction of slots where pipeline is likely stalled due to demand load or store instructions. This accounts mainly for (1) non-completed in-flight memory demand loads which coincides with execution units starvation; in addition to (2) cases where stores could impose backpressure on the pipeline when many of them get buffered at the same time (less common out of the two).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_thread_slots", "MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS", "MetricName": "tma_microcode_sequencer", "MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1", @@ -746,16 +959,16 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline)", - "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_mite", - "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline). This pipeline is used for code that was not pre-cached in the DSB or LSD. For example; inefficiencies due to asymmetric decoders; use of long immediate or LCP can manifest as MITE fetch bandwidth bottleneck.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)", - "MetricExpr": "2 * IDQ.MS_SWITCHES / tma_info_clks", + "MetricExpr": "2 * IDQ.MS_SWITCHES / tma_info_thread_clks", "MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO", "MetricName": "tma_ms_switches", "MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -764,7 +977,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 0 ([SNB+] ALU; [HSW+] ALU and 2nd branch)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_core_clks", "MetricGroup": "Compute;TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_0", "MetricThreshold": "tma_port_0 > 0.6", @@ -773,7 +986,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 1 (ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_1", "MetricThreshold": "tma_port_1 > 0.6", @@ -782,7 +995,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 2 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_2", "MetricThreshold": "tma_port_2 > 0.6", @@ -791,7 +1004,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 3 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_3", "MetricThreshold": "tma_port_3 > 0.6", @@ -809,7 +1022,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 5 ([SNB+] Branches and ALU; [HSW+] ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_5", "MetricThreshold": "tma_port_5 > 0.6", @@ -818,7 +1031,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 6 ([HSW+]Primary Branch and simple ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_6 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_6 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_6", "MetricThreshold": "tma_port_6 > 0.6", @@ -827,7 +1040,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 7 ([HSW+]simple Store-address)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_7 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_7 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_store_op_utilization_group", "MetricName": "tma_port_7", "MetricThreshold": "tma_port_7 > 0.6", @@ -837,7 +1050,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - (cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ if tma_info_ipc > 1.8 else cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@)) / 2 - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB if #SMT_on else min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - (cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ if tma_info_ipc > 1.8 else cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING)) / tma_info_clks", + "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - (cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ if tma_info_thread_ipc > 1.8 else cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@)) / 2 - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB if #SMT_on else min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - (cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ if tma_info_thread_ipc > 1.8 else cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING)) / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_ports_utilization", "MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -846,7 +1059,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed no uops on any execution port (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,inv\\,cmask\\=1@ / 2 if #SMT_on else (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0)) / tma_info_core_clks)", + "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,inv\\,cmask\\=1@ / 2 if #SMT_on else (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0)) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_0", "MetricThreshold": "tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -855,7 +1068,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the CPU executed total of 1 uop per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / 2 if #SMT_on else (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / tma_info_core_clks)", + "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / 2 if #SMT_on else (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issueL1;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_1", "MetricThreshold": "tma_ports_utilized_1 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -864,7 +1077,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 2 uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / 2 if #SMT_on else (cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / tma_info_core_clks)", + "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / 2 if #SMT_on else (cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issue2P;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_2", "MetricThreshold": "tma_ports_utilized_2 > 0.15 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -873,7 +1086,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 3 or more uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise).", - "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ / 2 if #SMT_on else cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / tma_info_core_clks", + "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ / 2 if #SMT_on else cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_3m", "MetricThreshold": "tma_ports_utilized_3m > 0.7 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -882,7 +1095,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from remote cache in other sockets including synchronizations issues", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(200 * (MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) + 180 * (MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD)))) / tma_info_clks", + "MetricExpr": "(200 * (MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) + 180 * (MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD)))) / tma_info_thread_clks", "MetricGroup": "Offcore;Server;Snoop;TopdownL5;tma_L5_group;tma_issueSyncxn;tma_mem_latency_group", "MetricName": "tma_remote_cache", "MetricThreshold": "tma_remote_cache > 0.05 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -892,7 +1105,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from remote memory", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "310 * (MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) / tma_info_clks", + "MetricExpr": "310 * (MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.L3_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_L3_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_L3_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_L3_MISS_RETIRED.REMOTE_FWD))) / tma_info_thread_clks", "MetricGroup": "Server;Snoop;TopdownL5;tma_L5_group;tma_mem_latency_group", "MetricName": "tma_remote_dram", "MetricThreshold": "tma_remote_dram > 0.1 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -901,17 +1114,18 @@ }, { "BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired. Ideally; all pipeline slots would be attributed to the Retiring category. Retiring of 100% would indicate the maximum Pipeline_Width throughput was achieved. Maximizing Retiring typically increases the Instructions-per-cycle (see IPC metric). Note that a high Retiring value does not necessary mean there is no room for more performance. For example; Heavy-operations or Microcode Assists are categorized under Retiring. They often indicate suboptimal performance and can often be optimized or avoided. Sample with: UOPS_RETIRED.RETIRE_SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles handling memory load split accesses - load that cross 64-byte cache line boundary", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "tma_info_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_split_loads", "MetricThreshold": "tma_split_loads > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -920,7 +1134,7 @@ }, { "BriefDescription": "This metric represents rate of split store accesses", - "MetricExpr": "2 * MEM_UOPS_RETIRED.SPLIT_STORES / tma_info_core_clks", + "MetricExpr": "2 * MEM_UOPS_RETIRED.SPLIT_STORES / tma_info_core_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_issueSpSt;tma_store_bound_group", "MetricName": "tma_split_stores", "MetricThreshold": "tma_split_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -929,16 +1143,16 @@ }, { "BriefDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors)", - "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_clks", + "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_core_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueBW;tma_l3_bound_group", "MetricName": "tma_sq_full", "MetricThreshold": "tma_sq_full > 0.3 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_mem_bandwidth", + "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_mem_bandwidth", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write", - "MetricExpr": "RESOURCE_STALLS.SB / tma_info_clks", + "MetricExpr": "RESOURCE_STALLS.SB / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -947,7 +1161,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of cycles when the memory subsystem had loads blocked since they could not forward data from earlier (in program order) overlapping stores", - "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_clks", + "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_store_fwd_blk", "MetricThreshold": "tma_store_fwd_blk > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -957,7 +1171,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU spent handling L1D store misses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(L2_RQSTS.RFO_HIT * 9 * (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) + (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_clks", + "MetricExpr": "(L2_RQSTS.RFO_HIT * 9 * (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) + (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_thread_clks", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_issueSL;tma_store_bound_group", "MetricName": "tma_store_latency", "MetricThreshold": "tma_store_latency > 0.1 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -966,7 +1180,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Store operations", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_core_clks", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_store_op_utilization", "MetricThreshold": "tma_store_op_utilization > 0.6", @@ -974,11 +1188,17 @@ }, { "BriefDescription": "This metric serves as an approximation of legacy x87 usage", - "MetricExpr": "INST_RETIRED.X87 * tma_info_uoppi / UOPS_RETIRED.RETIRE_SLOTS", + "MetricExpr": "INST_RETIRED.X87 * tma_info_thread_uoppi / UOPS_RETIRED.RETIRE_SLOTS", "MetricGroup": "Compute;TopdownL4;tma_L4_group;tma_fp_arith_group", "MetricName": "tma_x87_use", "MetricThreshold": "tma_x87_use > 0.1", "PublicDescription": "This metric serves as an approximation of legacy x87 usage. It accounts for instructions beyond X87 FP arithmetic operations; hence may be used as a thermometer to avoid X87 high usage and preferably upgrade to modern ISA. See Tip under Tuning Hint.", "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uncore operating frequency in GHz", + "MetricExpr": "UNC_C_CLOCKTICKS / (#num_cores / #num_packages * #num_packages) / 1e9 / duration_time", + "MetricName": "uncore_frequency", + "ScaleUnit": "1GHz" } ] diff --git a/tools/perf/pmu-events/arch/x86/haswellx/metricgroups.json b/tools/perf/pmu-events/arch/x86/haswellx/metricgroups.json new file mode 100644 index 000000000000..f6a0258e3241 --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/haswellx/metricgroups.json @@ -0,0 +1,107 @@ +{ + "Backend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Bad": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BadSpec": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BigFoot": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BrMispredicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Branches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CacheMisses": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Compute": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Cor": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSBmiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DataSharing": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Fed": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Flops": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpScalar": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Frontend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "HPC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IcMiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "InsType": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "L2Evicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "LSD": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MachineClears": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBound": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryTLB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_BW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_Lat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MicroSeq": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "OS": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Offcore": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PGO": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Pipeline": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PortsUtil": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Ret": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Retire": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SMT": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Server": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Snoop": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SoC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL1": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL2": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL3mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "TopdownL5": "Metrics for top-down breakdown at level 5", + "TopdownL6": "Metrics for top-down breakdown at level 6", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_L5_group": "Metrics for top-down breakdown at level 5", + "tma_L6_group": "Metrics for top-down breakdown at level 6", + "tma_alu_op_utilization_group": "Metrics contributing to tma_alu_op_utilization category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_branch_resteers_group": "Metrics contributing to tma_branch_resteers category", + "tma_core_bound_group": "Metrics contributing to tma_core_bound category", + "tma_dram_bound_group": "Metrics contributing to tma_dram_bound category", + "tma_dtlb_load_group": "Metrics contributing to tma_dtlb_load category", + "tma_dtlb_store_group": "Metrics contributing to tma_dtlb_store category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_fp_arith_group": "Metrics contributing to tma_fp_arith category", + "tma_fp_vector_group": "Metrics contributing to tma_fp_vector category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_heavy_operations_group": "Metrics contributing to tma_heavy_operations category", + "tma_issue2P": "Metrics related by the issue $issue2P", + "tma_issueBM": "Metrics related by the issue $issueBM", + "tma_issueBW": "Metrics related by the issue $issueBW", + "tma_issueFB": "Metrics related by the issue $issueFB", + "tma_issueL1": "Metrics related by the issue $issueL1", + "tma_issueLat": "Metrics related by the issue $issueLat", + "tma_issueMC": "Metrics related by the issue $issueMC", + "tma_issueMS": "Metrics related by the issue $issueMS", + "tma_issueMV": "Metrics related by the issue $issueMV", + "tma_issueRFO": "Metrics related by the issue $issueRFO", + "tma_issueSL": "Metrics related by the issue $issueSL", + "tma_issueSO": "Metrics related by the issue $issueSO", + "tma_issueSmSt": "Metrics related by the issue $issueSmSt", + "tma_issueSpSt": "Metrics related by the issue $issueSpSt", + "tma_issueSyncxn": "Metrics related by the issue $issueSyncxn", + "tma_issueTLB": "Metrics related by the issue $issueTLB", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_l3_bound_group": "Metrics contributing to tma_l3_bound category", + "tma_light_operations_group": "Metrics contributing to tma_light_operations category", + "tma_load_op_utilization_group": "Metrics contributing to tma_load_op_utilization category", + "tma_mem_latency_group": "Metrics contributing to tma_mem_latency category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_microcode_sequencer_group": "Metrics contributing to tma_microcode_sequencer category", + "tma_ports_utilization_group": "Metrics contributing to tma_ports_utilization category", + "tma_ports_utilized_0_group": "Metrics contributing to tma_ports_utilized_0 category", + "tma_ports_utilized_3m_group": "Metrics contributing to tma_ports_utilized_3m category", + "tma_retiring_group": "Metrics contributing to tma_retiring category", + "tma_serializing_operation_group": "Metrics contributing to tma_serializing_operation category", + "tma_store_bound_group": "Metrics contributing to tma_store_bound category", + "tma_store_op_utilization_group": "Metrics contributing to tma_store_op_utilization category" +} diff --git a/tools/perf/pmu-events/arch/x86/icelake/cache.json b/tools/perf/pmu-events/arch/x86/icelake/cache.json index a9174a0837f0..79b9f02a4b63 100644 --- a/tools/perf/pmu-events/arch/x86/icelake/cache.json +++ b/tools/perf/pmu-events/arch/x86/icelake/cache.json @@ -339,6 +339,16 @@ "UMask": "0x8" }, { + "BriefDescription": "Retired instructions with at least 1 uncacheable load or Bus Lock.", + "Data_LA": "1", + "EventCode": "0xd4", + "EventName": "MEM_LOAD_MISC_RETIRED.UC", + "PEBS": "1", + "PublicDescription": "Retired instructions with at least one load to uncacheable memory-type, or at least one cache-line split locked access (Bus Lock).", + "SampleAfterValue": "100007", + "UMask": "0x4" + }, + { "BriefDescription": "Number of completed demand load requests that missed the L1, but hit the FB(fill buffer), because a preceding miss to the same cacheline initiated the line to be brought into L1, but data is not yet ready in L1.", "Data_LA": "1", "EventCode": "0xd1", @@ -834,6 +844,14 @@ "UMask": "0x4" }, { + "BriefDescription": "Counts bus locks, accounts for cache line split locks and UC locks.", + "EventCode": "0xF4", + "EventName": "SQ_MISC.BUS_LOCK", + "PublicDescription": "Counts the more expensive bus lock needed to enforce cache coherency for certain memory accesses that need to be done atomically. Can be created by issuing an atomic instruction (via the LOCK prefix) which causes a cache line split or accesses uncacheable memory.", + "SampleAfterValue": "100003", + "UMask": "0x10" + }, + { "BriefDescription": "Cycles the queue waiting for offcore responses is full.", "EventCode": "0xf4", "EventName": "SQ_MISC.SQ_FULL", diff --git a/tools/perf/pmu-events/arch/x86/icelake/icl-metrics.json b/tools/perf/pmu-events/arch/x86/icelake/icl-metrics.json index f45ae3483df4..cc4edf855064 100644 --- a/tools/perf/pmu-events/arch/x86/icelake/icl-metrics.json +++ b/tools/perf/pmu-events/arch/x86/icelake/icl-metrics.json @@ -64,7 +64,7 @@ }, { "BriefDescription": "Uncore frequency per die [GHZ]", - "MetricExpr": "tma_info_socket_clks / #num_dies / duration_time / 1e9", + "MetricExpr": "tma_info_system_socket_clks / #num_dies / duration_time / 1e9", "MetricGroup": "SoC", "MetricName": "UNCORE_FREQ" }, @@ -85,7 +85,7 @@ }, { "BriefDescription": "This metric estimates how often memory load accesses were aliased by preceding stores (in program order) with a 4K address offset", - "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_clks", + "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_4k_aliasing", "MetricThreshold": "tma_4k_aliasing > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -94,7 +94,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution ports for ALU operations.", - "MetricExpr": "(UOPS_DISPATCHED.PORT_0 + UOPS_DISPATCHED.PORT_1 + UOPS_DISPATCHED.PORT_5 + UOPS_DISPATCHED.PORT_6) / (4 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED.PORT_0 + UOPS_DISPATCHED.PORT_1 + UOPS_DISPATCHED.PORT_5 + UOPS_DISPATCHED.PORT_6) / (4 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_alu_op_utilization", "MetricThreshold": "tma_alu_op_utilization > 0.6", @@ -102,7 +102,7 @@ }, { "BriefDescription": "This metric estimates fraction of slots the CPU retired uops delivered by the Microcode_Sequencer as a result of Assists", - "MetricExpr": "100 * ASSISTS.ANY / tma_info_slots", + "MetricExpr": "100 * ASSISTS.ANY / tma_info_thread_slots", "MetricGroup": "TopdownL4;tma_L4_group;tma_microcode_sequencer_group", "MetricName": "tma_assists", "MetricThreshold": "tma_assists > 0.1 & (tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1)", @@ -111,25 +111,29 @@ }, { "BriefDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend", - "MetricExpr": "topdown\\-be\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 5 * cpu@INT_MISC.RECOVERY_CYCLES\\,cmask\\=1\\,edge@ / tma_info_slots", - "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "topdown\\-be\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 5 * cpu@INT_MISC.RECOVERY_CYCLES\\,cmask\\=1\\,edge@ / tma_info_thread_slots", + "MetricGroup": "Default;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend. Backend is the portion of the processor core where the out-of-order scheduler dispatches ready uops into their respective execution units; and once completed these uops get retired according to program order. For example; stalls due to data-cache misses or stalls due to the divider unit being overloaded are both categorized under Backend Bound. Backend Bound is further divided into two main categories: Memory Bound and Core Bound. Sample with: TOPDOWN.BACKEND_BOUND_SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations", + "DefaultMetricgroupName": "TopdownL1", "MetricExpr": "max(1 - (tma_frontend_bound + tma_backend_bound + tma_retiring), 0)", - "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", + "MetricGroup": "Default;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots wasted due to incorrect speculations. This include slots used to issue uops that do not eventually get retired and slots for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work due to miss-predicted branches are categorized under Bad Speculation category. Incorrect data speculation followed by Memory Ordering Nukes is another example.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring branch instructions.", - "MetricExpr": "tma_light_operations * BR_INST_RETIRED.ALL_BRANCHES / (tma_retiring * tma_info_slots)", + "MetricExpr": "tma_light_operations * BR_INST_RETIRED.ALL_BRANCHES / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_branch_instructions", "MetricThreshold": "tma_branch_instructions > 0.1 & tma_light_operations > 0.6", @@ -141,12 +145,13 @@ "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15", - "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_branch_misprediction_cost, tma_info_mispredictions, tma_mispredicts_resteers", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_info_bottleneck_mispredictions, tma_mispredicts_resteers", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers", - "MetricExpr": "INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks + tma_unknown_branches", + "MetricExpr": "INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks + tma_unknown_branches", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteers", "MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -164,7 +169,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Machine Clears", - "MetricExpr": "(1 - BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT)) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks", + "MetricExpr": "(1 - BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT)) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks", "MetricGroup": "BadSpec;MachineClears;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueMC", "MetricName": "tma_clears_resteers", "MetricThreshold": "tma_clears_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", @@ -174,7 +179,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to contested accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(29 * tma_info_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_HITM + 23.5 * tma_info_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "(29 * tma_info_system_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_HITM + 23.5 * tma_info_system_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_contested_accesses", "MetricThreshold": "tma_contested_accesses > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -187,13 +192,14 @@ "MetricGroup": "Backend;Compute;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck. Shortage in hardware compute resources; or dependencies in software's instructions are both categorized under Core Bound. Hence it may indicate the machine ran out of an out-of-order resource; certain execution units are overloaded or dependencies in program's data- or instruction-flow are limiting the performance (e.g. FP-chained long-latency arithmetic operations).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to data-sharing accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "23.5 * tma_info_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "23.5 * tma_info_system_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_data_sharing", "MetricThreshold": "tma_data_sharing > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -202,16 +208,16 @@ }, { "BriefDescription": "This metric represents fraction of cycles where decoder-0 was the only active decoder", - "MetricExpr": "(cpu@INST_DECODED.DECODERS\\,cmask\\=1@ - cpu@INST_DECODED.DECODERS\\,cmask\\=2@) / tma_info_core_clks / 2", + "MetricExpr": "(cpu@INST_DECODED.DECODERS\\,cmask\\=1@ - cpu@INST_DECODED.DECODERS\\,cmask\\=2@) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL4;tma_L4_group;tma_issueD0;tma_mite_group", "MetricName": "tma_decoder0_alone", - "MetricThreshold": "tma_decoder0_alone > 0.1 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 5 > 0.35))", + "MetricThreshold": "tma_decoder0_alone > 0.1 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 5 > 0.35))", "PublicDescription": "This metric represents fraction of cycles where decoder-0 was the only active decoder. Related metrics: tma_few_uops_instructions", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles where the Divider unit was active", - "MetricExpr": "ARITH.DIVIDER_ACTIVE / tma_info_clks", + "MetricExpr": "ARITH.DIVIDER_ACTIVE / tma_info_thread_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_divider", "MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -221,7 +227,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks - tma_l2_bound", + "MetricExpr": "CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_thread_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks - tma_l2_bound", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -230,43 +236,43 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline", - "MetricExpr": "(IDQ.DSB_CYCLES_ANY - IDQ.DSB_CYCLES_OK) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.DSB_CYCLES_ANY - IDQ.DSB_CYCLES_OK) / tma_info_core_core_clks / 2", "MetricGroup": "DSB;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_dsb", - "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 5 > 0.35)", + "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 5 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline. For example; inefficient utilization of the DSB cache structure or bank conflict when reading from it; are categorized here.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_dsb_switches", "MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb, tma_lcp", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses", - "MetricExpr": "min(7 * cpu@DTLB_LOAD_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_LOAD_MISSES.WALK_ACTIVE, max(CYCLE_ACTIVITY.CYCLES_MEM_ANY - CYCLE_ACTIVITY.CYCLES_L1D_MISS, 0)) / tma_info_clks", + "MetricExpr": "min(7 * cpu@DTLB_LOAD_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_LOAD_MISSES.WALK_ACTIVE, max(CYCLE_ACTIVITY.CYCLES_MEM_ANY - CYCLE_ACTIVITY.CYCLES_L1D_MISS, 0)) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group", "MetricName": "tma_dtlb_load", "MetricThreshold": "tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses. TLBs (Translation Look-aside Buffers) are processor caches for recently used entries out of the Page Tables that are used to map virtual- to physical-addresses by the operating system. This metric approximates the potential delay of demand loads missing the first-level data TLB (assuming worst case scenario with back to back misses to different pages). This includes hitting in the second-level TLB (STLB) as well as performing a hardware page walk on an STLB miss. Sample with: MEM_INST_RETIRED.STLB_MISS_LOADS_PS. Related metrics: tma_dtlb_store, tma_info_memory_data_tlbs", + "PublicDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses. TLBs (Translation Look-aside Buffers) are processor caches for recently used entries out of the Page Tables that are used to map virtual- to physical-addresses by the operating system. This metric approximates the potential delay of demand loads missing the first-level data TLB (assuming worst case scenario with back to back misses to different pages). This includes hitting in the second-level TLB (STLB) as well as performing a hardware page walk on an STLB miss. Sample with: MEM_INST_RETIRED.STLB_MISS_LOADS_PS. Related metrics: tma_dtlb_store, tma_info_bottleneck_memory_data_tlbs", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses", - "MetricExpr": "(7 * cpu@DTLB_STORE_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_STORE_MISSES.WALK_ACTIVE) / tma_info_core_clks", + "MetricExpr": "(7 * cpu@DTLB_STORE_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_STORE_MISSES.WALK_ACTIVE) / tma_info_core_core_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_store_bound_group", "MetricName": "tma_dtlb_store", "MetricThreshold": "tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses. As with ordinary data caching; focus on improving data locality and reducing working-set size to reduce DTLB overhead. Additionally; consider using profile-guided optimization (PGO) to collocate frequently-used data on the same page. Try using larger page sizes for large amounts of frequently-used data. Sample with: MEM_INST_RETIRED.STLB_MISS_STORES_PS. Related metrics: tma_dtlb_load, tma_info_memory_data_tlbs", + "PublicDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses. As with ordinary data caching; focus on improving data locality and reducing working-set size to reduce DTLB overhead. Additionally; consider using profile-guided optimization (PGO) to collocate frequently-used data on the same page. Try using larger page sizes for large amounts of frequently-used data. Sample with: MEM_INST_RETIRED.STLB_MISS_STORES_PS. Related metrics: tma_dtlb_load, tma_info_bottleneck_memory_data_tlbs", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates how often CPU was handling synchronizations due to False Sharing", - "MetricExpr": "32.5 * tma_info_average_frequency * OCR.DEMAND_RFO.L3_HIT.SNOOP_HITM / tma_info_clks", + "MetricExpr": "32.5 * tma_info_system_average_frequency * OCR.DEMAND_RFO.L3_HIT.SNOOP_HITM / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_store_bound_group", "MetricName": "tma_false_sharing", "MetricThreshold": "tma_false_sharing > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -275,11 +281,11 @@ }, { "BriefDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed", - "MetricExpr": "L1D_PEND_MISS.FB_FULL / tma_info_clks", + "MetricExpr": "L1D_PEND_MISS.FB_FULL / tma_info_thread_clks", "MetricGroup": "MemoryBW;TopdownL4;tma_L4_group;tma_issueBW;tma_issueSL;tma_issueSmSt;tma_l1_bound_group", "MetricName": "tma_fb_full", "MetricThreshold": "tma_fb_full > 0.3", - "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", + "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", "ScaleUnit": "100%" }, { @@ -287,16 +293,18 @@ "MetricExpr": "max(0, tma_frontend_bound - tma_fetch_latency)", "MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", "MetricName": "tma_fetch_bandwidth", - "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 5 > 0.35", - "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb, tma_lcp", + "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 5 > 0.35", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues", - "MetricExpr": "(5 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE - INT_MISC.UOP_DROPPING) / tma_info_slots", + "MetricExpr": "(5 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE - INT_MISC.UOP_DROPPING) / tma_info_thread_slots", "MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: FRONTEND_RETIRED.LATENCY_GE_16_PS;FRONTEND_RETIRED.LATENCY_GE_8_PS", "ScaleUnit": "100%" }, @@ -311,6 +319,7 @@ }, { "BriefDescription": "This metric represents overall arithmetic floating-point (FP) operations fraction the CPU has executed (retired)", + "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "tma_x87_use + tma_fp_scalar + tma_fp_vector", "MetricGroup": "HPC;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_fp_arith", @@ -320,7 +329,7 @@ }, { "BriefDescription": "This metric approximates arithmetic floating-point (FP) scalar uops fraction the CPU has retired", - "MetricExpr": "cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ / (tma_retiring * tma_info_slots)", + "MetricExpr": "cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL4;tma_L4_group;tma_fp_arith_group;tma_issue2P", "MetricName": "tma_fp_scalar", "MetricThreshold": "tma_fp_scalar > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6)", @@ -329,7 +338,7 @@ }, { "BriefDescription": "This metric approximates arithmetic floating-point (FP) vector uops fraction the CPU has retired aggregated across all vector widths", - "MetricExpr": "cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@ / (tma_retiring * tma_info_slots)", + "MetricExpr": "cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@ / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL4;tma_L4_group;tma_fp_arith_group;tma_issue2P", "MetricName": "tma_fp_vector", "MetricThreshold": "tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6)", @@ -338,7 +347,7 @@ }, { "BriefDescription": "This metric approximates arithmetic FP vector uops fraction the CPU has retired for 128-bit wide vectors", - "MetricExpr": "(FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL5;tma_L5_group;tma_fp_vector_group;tma_issue2P", "MetricName": "tma_fp_vector_128b", "MetricThreshold": "tma_fp_vector_128b > 0.1 & (tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6))", @@ -347,7 +356,7 @@ }, { "BriefDescription": "This metric approximates arithmetic FP vector uops fraction the CPU has retired for 256-bit wide vectors", - "MetricExpr": "(FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL5;tma_L5_group;tma_fp_vector_group;tma_issue2P", "MetricName": "tma_fp_vector_256b", "MetricThreshold": "tma_fp_vector_256b > 0.1 & (tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6))", @@ -356,7 +365,7 @@ }, { "BriefDescription": "This metric approximates arithmetic FP vector uops fraction the CPU has retired for 512-bit wide vectors", - "MetricExpr": "(FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL5;tma_L5_group;tma_fp_vector_group;tma_issue2P", "MetricName": "tma_fp_vector_512b", "MetricThreshold": "tma_fp_vector_512b > 0.1 & (tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6))", @@ -365,10 +374,12 @@ }, { "BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend", - "MetricExpr": "topdown\\-fe\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) - INT_MISC.UOP_DROPPING / tma_info_slots", - "MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "topdown\\-fe\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) - INT_MISC.UOP_DROPPING / tma_info_thread_slots", + "MetricGroup": "Default;PGO;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound. Sample with: FRONTEND_RETIRED.LATENCY_GE_4_PS", "ScaleUnit": "100%" }, @@ -378,12 +389,13 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_heavy_operations", "MetricThreshold": "tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to instruction cache misses", - "MetricExpr": "ICACHE_16B.IFDATA_STALL / tma_info_clks", + "MetricExpr": "ICACHE_16B.IFDATA_STALL / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;IcMiss;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_icache_misses", "MetricThreshold": "tma_icache_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -391,668 +403,676 @@ "ScaleUnit": "100%" }, { - "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", - "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time", - "MetricGroup": "Power;Summary", - "MetricName": "tma_info_average_frequency" + "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_thread_slots / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BrMispredicts;tma_issueBM", + "MetricName": "tma_info_bad_spec_branch_misprediction_cost", + "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_info_bottleneck_mispredictions, tma_mispredicts_resteers" + }, + { + "BriefDescription": "Instructions per retired mispredicts for conditional non-taken branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_NTAKEN", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_cond_ntaken", + "MetricThreshold": "tma_info_bad_spec_ipmisp_cond_ntaken < 200" + }, + { + "BriefDescription": "Instructions per retired mispredicts for conditional taken branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_TAKEN", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_cond_taken", + "MetricThreshold": "tma_info_bad_spec_ipmisp_cond_taken < 200" + }, + { + "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.INDIRECT", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_indirect", + "MetricThreshold": "tma_info_bad_spec_ipmisp_indirect < 1e3" + }, + { + "BriefDescription": "Instructions per retired mispredicts for return branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.RET", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_ret", + "MetricThreshold": "tma_info_bad_spec_ipmisp_ret < 500" + }, + { + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BadSpec;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmispredict", + "MetricThreshold": "tma_info_bad_spec_ipmispredict < 200" + }, + { + "BriefDescription": "Probability of Core Bound bottleneck hidden by SMT-profiling artifacts", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(100 * (1 - tma_core_bound / tma_ports_utilization if tma_core_bound < tma_ports_utilization else 1) if tma_info_system_smt_2t_utilization > 0.5 else 0)", + "MetricGroup": "Cor;SMT", + "MetricName": "tma_info_botlnk_l0_core_bound_likely", + "MetricThreshold": "tma_info_botlnk_l0_core_bound_likely > 0.5" + }, + { + "BriefDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_fetch_latency * tma_dsb_switches / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches) + tma_fetch_bandwidth * tma_mite / (tma_dsb + tma_lsd + tma_mite))", + "MetricGroup": "DSBmiss;Fed;tma_issueFB", + "MetricName": "tma_info_botlnk_l2_dsb_misses", + "MetricThreshold": "tma_info_botlnk_l2_dsb_misses > 10", + "PublicDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp" + }, + { + "BriefDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck", + "MetricExpr": "100 * (tma_fetch_latency * tma_icache_misses / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", + "MetricGroup": "Fed;FetchLat;IcMiss;tma_issueFL", + "MetricName": "tma_info_botlnk_l2_ic_misses", + "MetricThreshold": "tma_info_botlnk_l2_ic_misses > 5", + "PublicDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck. Related metrics: " }, { "BriefDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses)", "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "100 * tma_fetch_latency * (tma_itlb_misses + tma_icache_misses + tma_unknown_branches) / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)", "MetricGroup": "BigFoot;Fed;Frontend;IcMiss;MemoryTLB;tma_issueBC", - "MetricName": "tma_info_big_code", - "MetricThreshold": "tma_info_big_code > 20", - "PublicDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses). Related metrics: tma_info_branching_overhead" + "MetricName": "tma_info_bottleneck_big_code", + "MetricThreshold": "tma_info_bottleneck_big_code > 20", + "PublicDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses). Related metrics: tma_info_bottleneck_branching_overhead" }, { - "BriefDescription": "Branch instructions per taken branch.", - "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", - "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_bptkbranch" + "BriefDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls)", + "MetricExpr": "100 * ((BR_INST_RETIRED.COND + 3 * BR_INST_RETIRED.NEAR_CALL + (BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL)) / tma_info_thread_slots)", + "MetricGroup": "Ret;tma_issueBC", + "MetricName": "tma_info_bottleneck_branching_overhead", + "MetricThreshold": "tma_info_bottleneck_branching_overhead > 10", + "PublicDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls). Related metrics: tma_info_bottleneck_big_code" }, { - "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", - "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_slots / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_branch_misprediction_cost", - "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_info_mispredictions, tma_mispredicts_resteers" + "BriefDescription": "Total pipeline cost of instruction fetch bandwidth related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_frontend_bound - tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) - tma_info_bottleneck_big_code", + "MetricGroup": "Fed;FetchBW;Frontend", + "MetricName": "tma_info_bottleneck_instruction_fetch_bw", + "MetricThreshold": "tma_info_bottleneck_instruction_fetch_bw > 20" }, { - "BriefDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls)", - "MetricExpr": "100 * ((BR_INST_RETIRED.COND + 3 * BR_INST_RETIRED.NEAR_CALL + (BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL)) / tma_info_slots)", - "MetricGroup": "Ret;tma_issueBC", - "MetricName": "tma_info_branching_overhead", - "MetricThreshold": "tma_info_branching_overhead > 10", - "PublicDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls). Related metrics: tma_info_big_code" + "BriefDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks", + "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_bandwidth / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_sq_full / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full))) + tma_l1_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_fb_full / (tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk))", + "MetricGroup": "Mem;MemoryBW;Offcore;tma_issueBW", + "MetricName": "tma_info_bottleneck_memory_bandwidth", + "MetricThreshold": "tma_info_bottleneck_memory_bandwidth > 20", + "PublicDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks. Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full" }, { - "BriefDescription": "Fraction of branches that are CALL or RET", - "MetricExpr": "(BR_INST_RETIRED.NEAR_CALL + BR_INST_RETIRED.NEAR_RETURN) / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_callret" + "BriefDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_l1_bound / max(tma_memory_bound, tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_load / max(tma_l1_bound, tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk)) + tma_store_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_store / (tma_dtlb_store + tma_false_sharing + tma_split_stores + tma_store_latency + tma_streaming_stores)))", + "MetricGroup": "Mem;MemoryTLB;Offcore;tma_issueTLB", + "MetricName": "tma_info_bottleneck_memory_data_tlbs", + "MetricThreshold": "tma_info_bottleneck_memory_data_tlbs > 20", + "PublicDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs). Related metrics: tma_dtlb_load, tma_dtlb_store" }, { - "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "Pipeline", - "MetricName": "tma_info_clks" + "BriefDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_latency / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_l3_hit_latency / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full)) + tma_l2_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound))", + "MetricGroup": "Mem;MemoryLat;Offcore;tma_issueLat", + "MetricName": "tma_info_bottleneck_memory_latency", + "MetricThreshold": "tma_info_bottleneck_memory_latency > 20", + "PublicDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches). Related metrics: tma_l3_hit_latency, tma_mem_latency" }, { - "BriefDescription": "STLB (2nd level TLB) code speculative misses per kilo instruction (misses of any page-size that complete the page walk)", - "MetricExpr": "1e3 * ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", - "MetricGroup": "Fed;MemoryTLB", - "MetricName": "tma_info_code_stlb_mpki" + "BriefDescription": "Total pipeline cost of Branch Misprediction related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", + "MetricGroup": "Bad;BadSpec;BrMispredicts;tma_issueBM", + "MetricName": "tma_info_bottleneck_mispredictions", + "MetricThreshold": "tma_info_bottleneck_mispredictions > 20", + "PublicDescription": "Total pipeline cost of Branch Misprediction related bottlenecks. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers" + }, + { + "BriefDescription": "Fraction of branches that are CALL or RET", + "MetricExpr": "(BR_INST_RETIRED.NEAR_CALL + BR_INST_RETIRED.NEAR_RETURN) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_callret" }, { "BriefDescription": "Fraction of branches that are non-taken conditionals", "MetricExpr": "BR_INST_RETIRED.COND_NTAKEN / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Bad;Branches;CodeGen;PGO", - "MetricName": "tma_info_cond_nt" + "MetricName": "tma_info_branches_cond_nt" }, { "BriefDescription": "Fraction of branches that are taken conditionals", "MetricExpr": "BR_INST_RETIRED.COND_TAKEN / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Bad;Branches;CodeGen;PGO", - "MetricName": "tma_info_cond_tk" + "MetricName": "tma_info_branches_cond_tk" }, { - "BriefDescription": "Probability of Core Bound bottleneck hidden by SMT-profiling artifacts", - "MetricExpr": "(100 * (1 - tma_core_bound / tma_ports_utilization if tma_core_bound < tma_ports_utilization else 1) if tma_info_smt_2t_utilization > 0.5 else 0)", - "MetricGroup": "Cor;SMT", - "MetricName": "tma_info_core_bound_likely", - "MetricThreshold": "tma_info_core_bound_likely > 0.5" + "BriefDescription": "Fraction of branches that are unconditional (direct or indirect) jumps", + "MetricExpr": "(BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_jump" + }, + { + "BriefDescription": "Fraction of branches of other types (not individually covered by other metrics in Info.Branches group)", + "MetricExpr": "1 - (tma_info_branches_cond_nt + tma_info_branches_cond_tk + tma_info_branches_callret + tma_info_branches_jump)", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_other_branches" }, { "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", "MetricExpr": "CPU_CLK_UNHALTED.DISTRIBUTED", "MetricGroup": "SMT", - "MetricName": "tma_info_core_clks" + "MetricName": "tma_info_core_core_clks" }, { "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks", "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_coreipc" - }, - { - "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricExpr": "1 / tma_info_ipc", - "MetricGroup": "Mem;Pipeline", - "MetricName": "tma_info_cpi" + "MetricName": "tma_info_core_coreipc" }, { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": "HPC;Summary", - "MetricName": "tma_info_cpu_utilization" + "BriefDescription": "Floating Point Operations Per Cycle", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * cpu@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE\\,umask\\=0x18@ + 8 * cpu@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE\\,umask\\=0x60@ + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / tma_info_core_core_clks", + "MetricGroup": "Flops;Ret", + "MetricName": "tma_info_core_flopc" }, { - "BriefDescription": "Average Parallel L2 cache miss data reads", - "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", - "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_data_l2_mlp" + "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@) / (2 * tma_info_core_core_clks)", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_core_fp_arith_utilization", + "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." }, { - "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricExpr": "64 * (UNC_ARB_TRK_REQUESTS.ALL + UNC_ARB_COH_TRK_REQUESTS.ALL) / 1e6 / duration_time / 1e3", - "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", - "MetricName": "tma_info_dram_bw_use", - "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_info_memory_bandwidth, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", + "MetricExpr": "UOPS_EXECUTED.THREAD / (UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1)", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", + "MetricName": "tma_info_core_ilp" }, { "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", "MetricExpr": "IDQ.DSB_UOPS / UOPS_ISSUED.ANY", "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", - "MetricName": "tma_info_dsb_coverage", - "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 5 > 0.35", - "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_misses, tma_info_iptb, tma_lcp" - }, - { - "BriefDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck", - "MetricExpr": "100 * (tma_fetch_latency * tma_dsb_switches / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches) + tma_fetch_bandwidth * tma_mite / (tma_dsb + tma_lsd + tma_mite))", - "MetricGroup": "DSBmiss;Fed;tma_issueFB", - "MetricName": "tma_info_dsb_misses", - "MetricThreshold": "tma_info_dsb_misses > 10", - "PublicDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb, tma_lcp" + "MetricName": "tma_info_frontend_dsb_coverage", + "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 5 > 0.35", + "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_inst_mix_iptb, tma_lcp" }, { "BriefDescription": "Average number of cycles of a switch from the DSB fetch-unit to MITE fetch unit - see DSB_Switches tree node for details.", "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / cpu@DSB2MITE_SWITCHES.PENALTY_CYCLES\\,cmask\\=1\\,edge@", "MetricGroup": "DSBmiss", - "MetricName": "tma_info_dsb_switch_cost" - }, - { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", - "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", - "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", - "MetricName": "tma_info_execute" - }, - { - "BriefDescription": "The ratio of Executed- by Issued-Uops", - "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", - "MetricGroup": "Cor;Pipeline", - "MetricName": "tma_info_execute_per_issue", - "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." - }, - { - "BriefDescription": "Fill Buffer (FB) hits per kilo instructions for retired demand loads (L1D misses that merge into ongoing miss-handling entries)", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.FB_HIT / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_fb_hpki" + "MetricName": "tma_info_frontend_dsb_switch_cost" }, { "BriefDescription": "Average number of Uops issued by front-end when it issued something", "MetricExpr": "UOPS_ISSUED.ANY / cpu@UOPS_ISSUED.ANY\\,cmask\\=1@", "MetricGroup": "Fed;FetchBW", - "MetricName": "tma_info_fetch_upc" + "MetricName": "tma_info_frontend_fetch_upc" }, { - "BriefDescription": "Floating Point Operations Per Cycle", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * cpu@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE\\,umask\\=0x18@ + 8 * cpu@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE\\,umask\\=0x60@ + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / tma_info_core_clks", - "MetricGroup": "Flops;Ret", - "MetricName": "tma_info_flopc" + "BriefDescription": "Average Latency for L1 instruction cache misses", + "MetricExpr": "ICACHE_16B.IFDATA_STALL / cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@", + "MetricGroup": "Fed;FetchLat;IcMiss", + "MetricName": "tma_info_frontend_icache_miss_latency" }, { - "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@) / (2 * tma_info_core_clks)", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_fp_arith_utilization", - "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." + "BriefDescription": "Instructions per non-speculative DSB miss (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / FRONTEND_RETIRED.ANY_DSB_MISS", + "MetricGroup": "DSBmiss;Fed", + "MetricName": "tma_info_frontend_ipdsb_miss_ret", + "MetricThreshold": "tma_info_frontend_ipdsb_miss_ret < 50" }, { - "BriefDescription": "Giga Floating Point Operations Per Second", - "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * cpu@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE\\,umask\\=0x18@ + 8 * cpu@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE\\,umask\\=0x60@ + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / 1e9 / duration_time", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_gflops", - "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_inst_mix_instructions / BACLEARS.ANY", + "MetricGroup": "Fed", + "MetricName": "tma_info_frontend_ipunknown_branch" }, { - "BriefDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck", - "MetricExpr": "100 * (tma_fetch_latency * tma_icache_misses / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", - "MetricGroup": "Fed;FetchLat;IcMiss;tma_issueFL", - "MetricName": "tma_info_ic_misses", - "MetricThreshold": "tma_info_ic_misses > 5", - "PublicDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck. Related metrics: " + "BriefDescription": "L2 cache true code cacheline misses per kilo instruction", + "MetricExpr": "1e3 * FRONTEND_RETIRED.L2_MISS / INST_RETIRED.ANY", + "MetricGroup": "IcMiss", + "MetricName": "tma_info_frontend_l2mpki_code" }, { - "BriefDescription": "Average Latency for L1 instruction cache misses", - "MetricExpr": "ICACHE_16B.IFDATA_STALL / cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@", - "MetricGroup": "Fed;FetchLat;IcMiss", - "MetricName": "tma_info_icache_miss_latency" + "BriefDescription": "L2 cache speculative code cacheline misses per kilo instruction", + "MetricExpr": "1e3 * L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", + "MetricGroup": "IcMiss", + "MetricName": "tma_info_frontend_l2mpki_code_all" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", - "MetricExpr": "UOPS_EXECUTED.THREAD / (UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1)", - "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", - "MetricName": "tma_info_ilp" + "BriefDescription": "Fraction of Uops delivered by the LSD (Loop Stream Detector; aka Loop Cache)", + "MetricExpr": "LSD.UOPS / UOPS_ISSUED.ANY", + "MetricGroup": "Fed;LSD", + "MetricName": "tma_info_frontend_lsd_coverage" }, { - "BriefDescription": "Total pipeline cost of instruction fetch bandwidth related bottlenecks", - "MetricExpr": "100 * (tma_frontend_bound - tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) - tma_info_big_code", - "MetricGroup": "Fed;FetchBW;Frontend", - "MetricName": "tma_info_instruction_fetch_bw", - "MetricThreshold": "tma_info_instruction_fetch_bw > 20" + "BriefDescription": "Branch instructions per taken branch.", + "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;PGO", + "MetricName": "tma_info_inst_mix_bptkbranch" }, { "BriefDescription": "Total number of retired Instructions", "MetricExpr": "INST_RETIRED.ANY", "MetricGroup": "Summary;TmaL1;tma_L1_group", - "MetricName": "tma_info_instructions", + "MetricName": "tma_info_inst_mix_instructions", "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" }, { "BriefDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_iparith", - "MetricThreshold": "tma_info_iparith < 10", + "MetricName": "tma_info_inst_mix_iparith", + "MetricThreshold": "tma_info_inst_mix_iparith < 10", "PublicDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate). May undercount due to FMA double counting. Approximated prior to BDW." }, { "BriefDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx128", - "MetricThreshold": "tma_info_iparith_avx128 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx128", + "MetricThreshold": "tma_info_inst_mix_iparith_avx128 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx256", - "MetricThreshold": "tma_info_iparith_avx256 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx256", + "MetricThreshold": "tma_info_inst_mix_iparith_avx256 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic AVX 512-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx512", - "MetricThreshold": "tma_info_iparith_avx512 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx512", + "MetricThreshold": "tma_info_inst_mix_iparith_avx512 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX 512-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_dp", - "MetricThreshold": "tma_info_iparith_scalar_dp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_dp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_dp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_SINGLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_sp", - "MetricThreshold": "tma_info_iparith_scalar_sp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_sp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_sp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per Branch (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Branches;Fed;InsType", - "MetricName": "tma_info_ipbranch", - "MetricThreshold": "tma_info_ipbranch < 8" - }, - { - "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": "Ret;Summary", - "MetricName": "tma_info_ipc" + "MetricName": "tma_info_inst_mix_ipbranch", + "MetricThreshold": "tma_info_inst_mix_ipbranch < 8" }, { "BriefDescription": "Instructions per (near) call (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_CALL", "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_ipcall", - "MetricThreshold": "tma_info_ipcall < 200" - }, - { - "BriefDescription": "Instructions per non-speculative DSB miss (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / FRONTEND_RETIRED.ANY_DSB_MISS", - "MetricGroup": "DSBmiss;Fed", - "MetricName": "tma_info_ipdsb_miss_ret", - "MetricThreshold": "tma_info_ipdsb_miss_ret < 50" - }, - { - "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", - "MetricGroup": "Branches;OS", - "MetricName": "tma_info_ipfarbranch", - "MetricThreshold": "tma_info_ipfarbranch < 1e6" + "MetricName": "tma_info_inst_mix_ipcall", + "MetricThreshold": "tma_info_inst_mix_ipcall < 200" }, { "BriefDescription": "Instructions per Floating Point (FP) Operation (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * cpu@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE\\,umask\\=0x18@ + 8 * cpu@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE\\,umask\\=0x60@ + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_ipflop", - "MetricThreshold": "tma_info_ipflop < 10" + "MetricName": "tma_info_inst_mix_ipflop", + "MetricThreshold": "tma_info_inst_mix_ipflop < 10" }, { "BriefDescription": "Instructions per Load (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_LOADS", "MetricGroup": "InsType", - "MetricName": "tma_info_ipload", - "MetricThreshold": "tma_info_ipload < 3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for conditional non-taken branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_NTAKEN", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_cond_ntaken", - "MetricThreshold": "tma_info_ipmisp_cond_ntaken < 200" - }, - { - "BriefDescription": "Instructions per retired mispredicts for conditional taken branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_TAKEN", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_cond_taken", - "MetricThreshold": "tma_info_ipmisp_cond_taken < 200" - }, - { - "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.INDIRECT", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_indirect", - "MetricThreshold": "tma_info_ipmisp_indirect < 1e3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for return branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.RET", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_ret", - "MetricThreshold": "tma_info_ipmisp_ret < 500" - }, - { - "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BadSpec;BrMispredicts", - "MetricName": "tma_info_ipmispredict", - "MetricThreshold": "tma_info_ipmispredict < 200" + "MetricName": "tma_info_inst_mix_ipload", + "MetricThreshold": "tma_info_inst_mix_ipload < 3" }, { "BriefDescription": "Instructions per Store (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_STORES", "MetricGroup": "InsType", - "MetricName": "tma_info_ipstore", - "MetricThreshold": "tma_info_ipstore < 8" + "MetricName": "tma_info_inst_mix_ipstore", + "MetricThreshold": "tma_info_inst_mix_ipstore < 8" }, { "BriefDescription": "Instructions per Software prefetch instruction (of any type: NTA/T0/T1/T2/Prefetch) (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / cpu@SW_PREFETCH_ACCESS.T0\\,umask\\=0xF@", "MetricGroup": "Prefetches", - "MetricName": "tma_info_ipswpf", - "MetricThreshold": "tma_info_ipswpf < 100" + "MetricName": "tma_info_inst_mix_ipswpf", + "MetricThreshold": "tma_info_inst_mix_ipswpf < 100" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW;Frontend;PGO;tma_issueFB", - "MetricName": "tma_info_iptb", - "MetricThreshold": "tma_info_iptb < 11", - "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_lcp" + "MetricName": "tma_info_inst_mix_iptb", + "MetricThreshold": "tma_info_inst_mix_iptb < 11", + "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_lcp" }, { - "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", - "MetricExpr": "tma_info_instructions / BACLEARS.ANY", - "MetricGroup": "Fed", - "MetricName": "tma_info_ipunknown_branch" - }, - { - "BriefDescription": "Fraction of branches that are unconditional (direct or indirect) jumps", - "MetricExpr": "(BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL) / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_jump" + "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_core_l1d_cache_fill_bw" }, { - "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_cpi" + "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_core_l2_cache_fill_bw" }, { - "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_utilization", - "MetricThreshold": "tma_info_kernel_utilization > 0.05" + "BriefDescription": "Average per-core data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * OFFCORE_REQUESTS.ALL_REQUESTS / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_core_l3_cache_access_bw" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", + "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw" + "MetricName": "tma_info_memory_core_l3_cache_fill_bw" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "tma_info_l1d_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw_1t" + "BriefDescription": "Fill Buffer (FB) hits per kilo instructions for retired demand loads (L1D misses that merge into ongoing miss-handling entries)", + "MetricExpr": "1e3 * MEM_LOAD_RETIRED.FB_HIT / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_fb_hpki" }, { "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L1_MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki" + "MetricName": "tma_info_memory_l1mpki" }, { "BriefDescription": "L1 cache true misses per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.ALL_DEMAND_DATA_RD / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki_load" - }, - { - "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw" - }, - { - "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "tma_info_l2_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw_1t" + "MetricName": "tma_info_memory_l1mpki_load" }, { "BriefDescription": "L2 cache hits per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_HIT / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_load" + "MetricName": "tma_info_memory_l2hpki_load" }, { "BriefDescription": "L2 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L2_MISS / INST_RETIRED.ANY", "MetricGroup": "Backend;CacheMisses;Mem", - "MetricName": "tma_info_l2mpki" + "MetricName": "tma_info_memory_l2mpki" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all request types (including speculative)", - "MetricExpr": "1e3 * (OFFCORE_REQUESTS.ALL_DATA_RD - OFFCORE_REQUESTS.DEMAND_DATA_RD + L2_RQSTS.ALL_DEMAND_MISS + L2_RQSTS.SWPF_MISS) / tma_info_instructions", + "MetricExpr": "1e3 * (OFFCORE_REQUESTS.ALL_DATA_RD - OFFCORE_REQUESTS.DEMAND_DATA_RD + L2_RQSTS.ALL_DEMAND_MISS + L2_RQSTS.SWPF_MISS) / tma_info_inst_mix_instructions", "MetricGroup": "CacheMisses;Mem;Offcore", - "MetricName": "tma_info_l2mpki_all" - }, - { - "BriefDescription": "L2 cache true code cacheline misses per kilo instruction", - "MetricExpr": "1e3 * FRONTEND_RETIRED.L2_MISS / INST_RETIRED.ANY", - "MetricGroup": "IcMiss", - "MetricName": "tma_info_l2mpki_code" - }, - { - "BriefDescription": "L2 cache speculative code cacheline misses per kilo instruction", - "MetricExpr": "1e3 * L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", - "MetricGroup": "IcMiss", - "MetricName": "tma_info_l2mpki_code_all" + "MetricName": "tma_info_memory_l2mpki_all" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2mpki_load" - }, - { - "BriefDescription": "Average per-core data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * OFFCORE_REQUESTS.ALL_REQUESTS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw" + "MetricName": "tma_info_memory_l2mpki_load" }, { - "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_access_bw", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw_1t" + "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L3_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l3mpki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw" + "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", + "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_RETIRED.L1_MISS + MEM_LOAD_RETIRED.FB_HIT)", + "MetricGroup": "Mem;MemoryBound;MemoryLat", + "MetricName": "tma_info_memory_load_miss_real_latency" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw_1t" + "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", + "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", + "MetricGroup": "Mem;MemoryBW;MemoryBound", + "MetricName": "tma_info_memory_mlp", + "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" }, { - "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L3_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l3mpki" + "BriefDescription": "Average Parallel L2 cache miss data reads", + "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", + "MetricGroup": "Memory_BW;Offcore", + "MetricName": "tma_info_memory_oro_data_l2_mlp" }, { "BriefDescription": "Average Latency for L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS.DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l2_miss_latency" + "MetricName": "tma_info_memory_oro_load_l2_miss_latency" }, { "BriefDescription": "Average Parallel L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / cpu@OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD\\,cmask\\=1@", "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_load_l2_mlp" + "MetricName": "tma_info_memory_oro_load_l2_mlp" }, { "BriefDescription": "Average Latency for L3 cache miss demand Loads", "MetricExpr": "cpu@OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD\\,umask\\=0x10@ / OFFCORE_REQUESTS.L3_MISS_DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l3_miss_latency" + "MetricName": "tma_info_memory_oro_load_l3_miss_latency" }, { - "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", - "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_RETIRED.L1_MISS + MEM_LOAD_RETIRED.FB_HIT)", - "MetricGroup": "Mem;MemoryBound;MemoryLat", - "MetricName": "tma_info_load_miss_real_latency" + "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l1d_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l1d_cache_fill_bw_1t" + }, + { + "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l2_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l2_cache_fill_bw_1t" + }, + { + "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_access_bw", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_thread_l3_cache_access_bw_1t" + }, + { + "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l3_cache_fill_bw_1t" + }, + { + "BriefDescription": "STLB (2nd level TLB) code speculative misses per kilo instruction (misses of any page-size that complete the page walk)", + "MetricExpr": "1e3 * ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricGroup": "Fed;MemoryTLB", + "MetricName": "tma_info_memory_tlb_code_stlb_mpki" }, { "BriefDescription": "STLB (2nd level TLB) data load speculative misses per kilo instruction (misses of any page-size that complete the page walk)", "MetricExpr": "1e3 * DTLB_LOAD_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_load_stlb_mpki" + "MetricName": "tma_info_memory_tlb_load_stlb_mpki" }, { - "BriefDescription": "Fraction of Uops delivered by the LSD (Loop Stream Detector; aka Loop Cache)", - "MetricExpr": "LSD.UOPS / UOPS_ISSUED.ANY", - "MetricGroup": "Fed;LSD", - "MetricName": "tma_info_lsd_coverage" + "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", + "MetricExpr": "(ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING) / (2 * tma_info_core_core_clks)", + "MetricGroup": "Mem;MemoryTLB", + "MetricName": "tma_info_memory_tlb_page_walks_utilization", + "MetricThreshold": "tma_info_memory_tlb_page_walks_utilization > 0.5" }, { - "BriefDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks", - "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_bandwidth / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_sq_full / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full))) + tma_l1_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_fb_full / (tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk))", - "MetricGroup": "Mem;MemoryBW;Offcore;tma_issueBW", - "MetricName": "tma_info_memory_bandwidth", - "MetricThreshold": "tma_info_memory_bandwidth > 20", - "PublicDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks. Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "STLB (2nd level TLB) data store speculative misses per kilo instruction (misses of any page-size that complete the page walk)", + "MetricExpr": "1e3 * DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricGroup": "Mem;MemoryTLB", + "MetricName": "tma_info_memory_tlb_store_stlb_mpki" }, { - "BriefDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs)", - "MetricExpr": "100 * tma_memory_bound * (tma_l1_bound / max(tma_memory_bound, tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_load / max(tma_l1_bound, tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk)) + tma_store_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_store / (tma_dtlb_store + tma_false_sharing + tma_split_stores + tma_store_latency + tma_streaming_stores)))", - "MetricGroup": "Mem;MemoryTLB;Offcore;tma_issueTLB", - "MetricName": "tma_info_memory_data_tlbs", - "MetricThreshold": "tma_info_memory_data_tlbs > 20", - "PublicDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs). Related metrics: tma_dtlb_load, tma_dtlb_store" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", + "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", + "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", + "MetricName": "tma_info_pipeline_execute" }, { - "BriefDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches)", - "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_latency / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_l3_hit_latency / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full)) + tma_l2_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound))", - "MetricGroup": "Mem;MemoryLat;Offcore;tma_issueLat", - "MetricName": "tma_info_memory_latency", - "MetricThreshold": "tma_info_memory_latency > 20", - "PublicDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches). Related metrics: tma_l3_hit_latency, tma_mem_latency" + "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "tma_retiring * tma_info_thread_slots / cpu@UOPS_RETIRED.SLOTS\\,cmask\\=1@", + "MetricGroup": "Pipeline;Ret", + "MetricName": "tma_info_pipeline_retire" }, { - "BriefDescription": "Total pipeline cost of Branch Misprediction related bottlenecks", - "MetricExpr": "100 * (tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", - "MetricGroup": "Bad;BadSpec;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_mispredictions", - "MetricThreshold": "tma_info_mispredictions > 20", - "PublicDescription": "Total pipeline cost of Branch Misprediction related bottlenecks. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost, tma_mispredicts_resteers" + "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time", + "MetricGroup": "Power;Summary", + "MetricName": "tma_info_system_average_frequency" }, { - "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", - "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", - "MetricGroup": "Mem;MemoryBW;MemoryBound", - "MetricName": "tma_info_mlp", - "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricGroup": "HPC;Summary", + "MetricName": "tma_info_system_cpu_utilization" }, { - "BriefDescription": "Fraction of branches of other types (not individually covered by other metrics in Info.Branches group)", - "MetricExpr": "1 - (tma_info_cond_nt + tma_info_cond_tk + tma_info_callret + tma_info_jump)", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_other_branches" + "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", + "MetricExpr": "64 * (UNC_ARB_TRK_REQUESTS.ALL + UNC_ARB_COH_TRK_REQUESTS.ALL) / 1e6 / duration_time / 1e3", + "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", + "MetricName": "tma_info_system_dram_bw_use", + "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_mem_bandwidth, tma_sq_full" }, { - "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", - "MetricExpr": "(ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING) / (2 * tma_info_core_clks)", - "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_page_walks_utilization", - "MetricThreshold": "tma_info_page_walks_utilization > 0.5" + "BriefDescription": "Giga Floating Point Operations Per Second", + "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * cpu@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE\\,umask\\=0x18@ + 8 * cpu@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE\\,umask\\=0x60@ + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / 1e9 / duration_time", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_system_gflops", + "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + }, + { + "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", + "MetricGroup": "Branches;OS", + "MetricName": "tma_info_system_ipfarbranch", + "MetricThreshold": "tma_info_system_ipfarbranch < 1e6" + }, + { + "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_cpi" + }, + { + "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_utilization", + "MetricThreshold": "tma_info_system_kernel_utilization > 0.05" }, { "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for baseline license level 0", - "MetricExpr": "CORE_POWER.LVL0_TURBO_LICENSE / tma_info_core_clks", + "MetricExpr": "CORE_POWER.LVL0_TURBO_LICENSE / tma_info_core_core_clks", "MetricGroup": "Power", - "MetricName": "tma_info_power_license0_utilization", + "MetricName": "tma_info_system_power_license0_utilization", "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for baseline license level 0. This includes non-AVX codes, SSE, AVX 128-bit, and low-current AVX 256-bit codes." }, { "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 1", - "MetricExpr": "CORE_POWER.LVL1_TURBO_LICENSE / tma_info_core_clks", + "MetricExpr": "CORE_POWER.LVL1_TURBO_LICENSE / tma_info_core_core_clks", "MetricGroup": "Power", - "MetricName": "tma_info_power_license1_utilization", - "MetricThreshold": "tma_info_power_license1_utilization > 0.5", + "MetricName": "tma_info_system_power_license1_utilization", + "MetricThreshold": "tma_info_system_power_license1_utilization > 0.5", "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 1. This includes high current AVX 256-bit instructions as well as low current AVX 512-bit instructions." }, { "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 2 (introduced in SKX)", - "MetricExpr": "CORE_POWER.LVL2_TURBO_LICENSE / tma_info_core_clks", + "MetricExpr": "CORE_POWER.LVL2_TURBO_LICENSE / tma_info_core_core_clks", "MetricGroup": "Power", - "MetricName": "tma_info_power_license2_utilization", - "MetricThreshold": "tma_info_power_license2_utilization > 0.5", + "MetricName": "tma_info_system_power_license2_utilization", + "MetricThreshold": "tma_info_system_power_license2_utilization > 0.5", "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 2 (introduced in SKX). This includes high current AVX 512-bit instructions." }, { - "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", - "MetricExpr": "tma_retiring * tma_info_slots / cpu@UOPS_RETIRED.SLOTS\\,cmask\\=1@", - "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_retire" - }, - { - "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", - "MetricExpr": "TOPDOWN.SLOTS", - "MetricGroup": "TmaL1;tma_L1_group", - "MetricName": "tma_info_slots" - }, - { - "BriefDescription": "Fraction of Physical Core issue-slots utilized by this Logical Processor", - "MetricExpr": "(tma_info_slots / (TOPDOWN.SLOTS / 2) if #SMT_on else 1)", - "MetricGroup": "SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_slots_utilization" - }, - { "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_DISTRIBUTED if #SMT_on else 0)", "MetricGroup": "SMT", - "MetricName": "tma_info_smt_2t_utilization" + "MetricName": "tma_info_system_smt_2t_utilization" }, { "BriefDescription": "Socket actual clocks when any core is active on that socket", "MetricExpr": "UNC_CLOCK.SOCKET", "MetricGroup": "SoC", - "MetricName": "tma_info_socket_clks" - }, - { - "BriefDescription": "STLB (2nd level TLB) data store speculative misses per kilo instruction (misses of any page-size that complete the page walk)", - "MetricExpr": "1e3 * DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", - "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_store_stlb_mpki" + "MetricName": "tma_info_system_socket_clks" }, { "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC", "MetricGroup": "Power", - "MetricName": "tma_info_turbo_utilization" + "MetricName": "tma_info_system_turbo_utilization" + }, + { + "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "Pipeline", + "MetricName": "tma_info_thread_clks" + }, + { + "BriefDescription": "Cycles Per Instruction (per Logical Processor)", + "MetricExpr": "1 / tma_info_thread_ipc", + "MetricGroup": "Mem;Pipeline", + "MetricName": "tma_info_thread_cpi" + }, + { + "BriefDescription": "The ratio of Executed- by Issued-Uops", + "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", + "MetricGroup": "Cor;Pipeline", + "MetricName": "tma_info_thread_execute_per_issue", + "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." + }, + { + "BriefDescription": "Instructions Per Cycle (per Logical Processor)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks", + "MetricGroup": "Ret;Summary", + "MetricName": "tma_info_thread_ipc" + }, + { + "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", + "MetricExpr": "TOPDOWN.SLOTS", + "MetricGroup": "TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots" + }, + { + "BriefDescription": "Fraction of Physical Core issue-slots utilized by this Logical Processor", + "MetricExpr": "(tma_info_thread_slots / (TOPDOWN.SLOTS / 2) if #SMT_on else 1)", + "MetricGroup": "SMT;TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots_utilization" }, { "BriefDescription": "Uops Per Instruction", - "MetricExpr": "tma_retiring * tma_info_slots / INST_RETIRED.ANY", + "MetricExpr": "tma_retiring * tma_info_thread_slots / INST_RETIRED.ANY", "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_uoppi", - "MetricThreshold": "tma_info_uoppi > 1.05" + "MetricName": "tma_info_thread_uoppi", + "MetricThreshold": "tma_info_thread_uoppi > 1.05" }, { "BriefDescription": "Instruction per taken branch", - "MetricExpr": "tma_retiring * tma_info_slots / BR_INST_RETIRED.NEAR_TAKEN", + "MetricExpr": "tma_retiring * tma_info_thread_slots / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW", - "MetricName": "tma_info_uptb", - "MetricThreshold": "tma_info_uptb < 7.5" + "MetricName": "tma_info_thread_uptb", + "MetricThreshold": "tma_info_thread_uptb < 7.5" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses", - "MetricExpr": "ICACHE_64B.IFTAG_STALL / tma_info_clks", + "MetricExpr": "ICACHE_64B.IFTAG_STALL / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -1061,7 +1081,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled without loads missing the L1 data cache", - "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_clks, 0)", + "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_thread_clks, 0)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_issueL1;tma_issueMC;tma_memory_bound_group", "MetricName": "tma_l1_bound", "MetricThreshold": "tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1071,7 +1091,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled due to L2 cache accesses by loads", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) / (MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + L1D_PEND_MISS.FB_FULL_PERIODS) * ((CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks)", + "MetricExpr": "MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) / (MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + L1D_PEND_MISS.FB_FULL_PERIODS) * ((CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l2_bound", "MetricThreshold": "tma_l2_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1080,7 +1100,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L2_MISS - CYCLE_ACTIVITY.STALLS_L3_MISS) / tma_info_clks", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L2_MISS - CYCLE_ACTIVITY.STALLS_L3_MISS) / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1089,20 +1109,20 @@ }, { "BriefDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited)", - "MetricExpr": "9 * tma_info_average_frequency * MEM_LOAD_RETIRED.L3_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "9 * tma_info_system_average_frequency * MEM_LOAD_RETIRED.L3_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "MemoryLat;TopdownL4;tma_L4_group;tma_issueLat;tma_l3_bound_group", "MetricName": "tma_l3_hit_latency", "MetricThreshold": "tma_l3_hit_latency > 0.1 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited). Avoiding private cache misses (i.e. L2 misses/L3 hits) will improve the latency; reduce contention with sibling physical cores and increase performance. Note the value of this node may overlap with its siblings. Sample with: MEM_LOAD_RETIRED.L3_HIT_PS. Related metrics: tma_info_memory_latency, tma_mem_latency", + "PublicDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited). Avoiding private cache misses (i.e. L2 misses/L3 hits) will improve the latency; reduce contention with sibling physical cores and increase performance. Note the value of this node may overlap with its siblings. Sample with: MEM_LOAD_RETIRED.L3_HIT_PS. Related metrics: tma_info_bottleneck_memory_latency, tma_mem_latency", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)", - "MetricExpr": "ILD_STALL.LCP / tma_info_clks", + "MetricExpr": "ILD_STALL.LCP / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_lcp", "MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb", + "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb", "ScaleUnit": "100%" }, { @@ -1111,12 +1131,13 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_light_operations", "MetricThreshold": "tma_light_operations > 0.6", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation). This correlates with total number of instructions used by the program. A uops-per-instruction (see UopPI metric) ratio of 1 or less should be expected for decently optimized software running on Intel Core/Xeon products. While this often indicates efficient X86 instructions were executed; high value does not necessarily mean better performance cannot be achieved. Sample with: INST_RETIRED.PREC_DIST", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Load operations", - "MetricExpr": "UOPS_DISPATCHED.PORT_2_3 / (2 * tma_info_core_clks)", + "MetricExpr": "UOPS_DISPATCHED.PORT_2_3 / (2 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_load_op_utilization", "MetricThreshold": "tma_load_op_utilization > 0.6", @@ -1133,7 +1154,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles where the Second-level TLB (STLB) was missed by load accesses, performing a hardware page walk", - "MetricExpr": "DTLB_LOAD_MISSES.WALK_ACTIVE / tma_info_clks", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_ACTIVE / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL5;tma_L5_group;tma_dtlb_load_group", "MetricName": "tma_load_stlb_miss", "MetricThreshold": "tma_load_stlb_miss > 0.05 & (tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1142,7 +1163,7 @@ { "BriefDescription": "This metric represents fraction of cycles the CPU spent handling cache misses due to lock operations", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(16 * max(0, MEM_INST_RETIRED.LOCK_LOADS - L2_RQSTS.ALL_RFO) + MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES * (10 * L2_RQSTS.RFO_HIT + min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO))) / tma_info_clks", + "MetricExpr": "(16 * max(0, MEM_INST_RETIRED.LOCK_LOADS - L2_RQSTS.ALL_RFO) + MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES * (10 * L2_RQSTS.RFO_HIT + min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO))) / tma_info_thread_clks", "MetricGroup": "Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_l1_bound_group", "MetricName": "tma_lock_latency", "MetricThreshold": "tma_lock_latency > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1151,10 +1172,10 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to LSD (Loop Stream Detector) unit", - "MetricExpr": "(LSD.CYCLES_ACTIVE - LSD.CYCLES_OK) / tma_info_core_clks / 2", + "MetricExpr": "(LSD.CYCLES_ACTIVE - LSD.CYCLES_OK) / tma_info_core_core_clks / 2", "MetricGroup": "FetchBW;LSD;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_lsd", - "MetricThreshold": "tma_lsd > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 5 > 0.35)", + "MetricThreshold": "tma_lsd > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 5 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to LSD (Loop Stream Detector) unit. LSD typically does well sustaining Uop supply. However; in some rare cases; optimal uop-delivery could not be reached for small loops whose size (in terms of number of uops) does not suit well the LSD structure.", "ScaleUnit": "100%" }, @@ -1164,25 +1185,26 @@ "MetricGroup": "BadSpec;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.1 & tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears. These slots are either wasted by uops fetched prior to the clear; or stalls the out-of-order portion of the machine needs to recover its state after the clear. For example; this can happen due to memory ordering Nukes (e.g. Memory Disambiguation) or Self-Modifying-Code (SMC) nukes. Sample with: MACHINE_CLEARS.COUNT. Related metrics: tma_clears_resteers, tma_contested_accesses, tma_data_sharing, tma_false_sharing, tma_l1_bound, tma_microcode_sequencer, tma_ms_switches, tma_remote_cache", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_clks", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW", "MetricName": "tma_mem_bandwidth", "MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_sq_full", + "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_sq_full", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat", "MetricName": "tma_mem_latency", "MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM). This metric does not aggregate requests from other Logical Processors/Physical Cores/sockets (see Uncore counters for that). Related metrics: tma_info_memory_latency, tma_l3_hit_latency", + "PublicDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM). This metric does not aggregate requests from other Logical Processors/Physical Cores/sockets (see Uncore counters for that). Related metrics: tma_info_bottleneck_memory_latency, tma_l3_hit_latency", "ScaleUnit": "100%" }, { @@ -1191,11 +1213,13 @@ "MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck. Memory Bound estimates fraction of slots where pipeline is likely stalled due to demand load or store instructions. This accounts mainly for (1) non-completed in-flight memory demand loads which coincides with execution units starvation; in addition to (2) cases where stores could impose backpressure on the pipeline when many of them get buffered at the same time (less common out of the two).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring memory operations -- uops for memory load or store accesses.", + "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "tma_light_operations * MEM_INST_RETIRED.ANY / INST_RETIRED.ANY", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_memory_operations", @@ -1204,7 +1228,7 @@ }, { "BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit", - "MetricExpr": "tma_retiring * tma_info_slots / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_slots", + "MetricExpr": "tma_retiring * tma_info_thread_slots / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_thread_slots", "MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS", "MetricName": "tma_microcode_sequencer", "MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1", @@ -1213,28 +1237,28 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage", - "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks", + "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks", "MetricGroup": "BadSpec;BrMispredicts;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueBM", "MetricName": "tma_mispredicts_resteers", "MetricThreshold": "tma_mispredicts_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost, tma_info_mispredictions", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost, tma_info_bottleneck_mispredictions", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline)", - "MetricExpr": "(IDQ.MITE_CYCLES_ANY - IDQ.MITE_CYCLES_OK) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.MITE_CYCLES_ANY - IDQ.MITE_CYCLES_OK) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_mite", - "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 5 > 0.35)", + "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 5 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline). This pipeline is used for code that was not pre-cached in the DSB or LSD. For example; inefficiencies due to asymmetric decoders; use of long immediate or LCP can manifest as MITE fetch bandwidth bottleneck. Sample with: FRONTEND_RETIRED.ANY_DSB_MISS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles where (only) 4 uops were delivered by the MITE pipeline", - "MetricExpr": "(cpu@IDQ.MITE_UOPS\\,cmask\\=4@ - cpu@IDQ.MITE_UOPS\\,cmask\\=5@) / tma_info_clks", + "MetricExpr": "(cpu@IDQ.MITE_UOPS\\,cmask\\=4@ - cpu@IDQ.MITE_UOPS\\,cmask\\=5@) / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchBW;TopdownL4;tma_L4_group;tma_mite_group", "MetricName": "tma_mite_4wide", - "MetricThreshold": "tma_mite_4wide > 0.05 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 5 > 0.35))", + "MetricThreshold": "tma_mite_4wide > 0.05 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 5 > 0.35))", "ScaleUnit": "100%" }, { @@ -1248,7 +1272,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)", - "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_clks", + "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_thread_clks", "MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO", "MetricName": "tma_ms_switches", "MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -1257,7 +1281,7 @@ }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring NOP (no op) instructions", - "MetricExpr": "tma_light_operations * INST_RETIRED.NOP / (tma_retiring * tma_info_slots)", + "MetricExpr": "tma_light_operations * INST_RETIRED.NOP / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_nop_instructions", "MetricThreshold": "tma_nop_instructions > 0.1 & tma_light_operations > 0.6", @@ -1266,6 +1290,7 @@ }, { "BriefDescription": "This metric represents the remaining light uops fraction the CPU has executed - remaining means not covered by other sibling nodes", + "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "max(0, tma_light_operations - (tma_fp_arith + tma_memory_operations + tma_branch_instructions + tma_nop_instructions))", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_other_light_ops", @@ -1275,7 +1300,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 0 ([SNB+] ALU; [HSW+] ALU and 2nd branch)", - "MetricExpr": "UOPS_DISPATCHED.PORT_0 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_0 / tma_info_core_core_clks", "MetricGroup": "Compute;TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_0", "MetricThreshold": "tma_port_0 > 0.6", @@ -1284,7 +1309,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 1 (ALU)", - "MetricExpr": "UOPS_DISPATCHED.PORT_1 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_1 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_1", "MetricThreshold": "tma_port_1 > 0.6", @@ -1293,7 +1318,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 5 ([SNB+] Branches and ALU; [HSW+] ALU)", - "MetricExpr": "UOPS_DISPATCHED.PORT_5 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_5 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_5", "MetricThreshold": "tma_port_5 > 0.6", @@ -1302,7 +1327,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 6 ([HSW+]Primary Branch and simple ALU)", - "MetricExpr": "UOPS_DISPATCHED.PORT_6 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_6 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_6", "MetricThreshold": "tma_port_6 > 0.6", @@ -1311,7 +1336,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)", - "MetricExpr": "((cpu@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) + (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL)) / tma_info_clks if ARITH.DIVIDER_ACTIVE < CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY else (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_clks)", + "MetricExpr": "((cpu@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) + (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL)) / tma_info_thread_clks if ARITH.DIVIDER_ACTIVE < CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY else (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_thread_clks)", "MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_ports_utilization", "MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -1320,7 +1345,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed no uops on any execution port (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "cpu@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ / tma_info_clks + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_clks", + "MetricExpr": "cpu@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ / tma_info_thread_clks + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_0", "MetricThreshold": "tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1329,7 +1354,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the CPU executed total of 1 uop per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "EXE_ACTIVITY.1_PORTS_UTIL / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.1_PORTS_UTIL / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issueL1;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_1", "MetricThreshold": "tma_ports_utilized_1 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1338,7 +1363,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 2 uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "EXE_ACTIVITY.2_PORTS_UTIL / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.2_PORTS_UTIL / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issue2P;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_2", "MetricThreshold": "tma_ports_utilized_2 > 0.15 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1347,7 +1372,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 3 or more uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "UOPS_EXECUTED.CYCLES_GE_3 / tma_info_clks", + "MetricExpr": "UOPS_EXECUTED.CYCLES_GE_3 / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_3m", "MetricThreshold": "tma_ports_utilized_3m > 0.7 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1356,16 +1381,18 @@ }, { "BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired", - "MetricExpr": "topdown\\-retiring / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_slots", - "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "topdown\\-retiring / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_thread_slots", + "MetricGroup": "Default;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired. Ideally; all pipeline slots would be attributed to the Retiring category. Retiring of 100% would indicate the maximum Pipeline_Width throughput was achieved. Maximizing Retiring typically increases the Instructions-per-cycle (see IPC metric). Note that a high Retiring value does not necessary mean there is no room for more performance. For example; Heavy-operations or Microcode Assists are categorized under Retiring. They often indicate suboptimal performance and can often be optimized or avoided. Sample with: UOPS_RETIRED.SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU issue-pipeline was stalled due to serializing operations", - "MetricExpr": "RESOURCE_STALLS.SCOREBOARD / tma_info_clks", + "MetricExpr": "RESOURCE_STALLS.SCOREBOARD / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL5;tma_L5_group;tma_issueSO;tma_ports_utilized_0_group", "MetricName": "tma_serializing_operation", "MetricThreshold": "tma_serializing_operation > 0.1 & (tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)))", @@ -1374,7 +1401,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to PAUSE Instructions", - "MetricExpr": "140 * MISC_RETIRED.PAUSE_INST / tma_info_clks", + "MetricExpr": "140 * MISC_RETIRED.PAUSE_INST / tma_info_thread_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_serializing_operation_group", "MetricName": "tma_slow_pause", "MetricThreshold": "tma_slow_pause > 0.05 & (tma_serializing_operation > 0.1 & (tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))))", @@ -1383,7 +1410,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles handling memory load split accesses - load that cross 64-byte cache line boundary", - "MetricExpr": "tma_info_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_split_loads", "MetricThreshold": "tma_split_loads > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1392,7 +1419,7 @@ }, { "BriefDescription": "This metric represents rate of split store accesses", - "MetricExpr": "MEM_INST_RETIRED.SPLIT_STORES / tma_info_core_clks", + "MetricExpr": "MEM_INST_RETIRED.SPLIT_STORES / tma_info_core_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_issueSpSt;tma_store_bound_group", "MetricName": "tma_split_stores", "MetricThreshold": "tma_split_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1401,16 +1428,16 @@ }, { "BriefDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors)", - "MetricExpr": "L1D_PEND_MISS.L2_STALL / tma_info_clks", + "MetricExpr": "L1D_PEND_MISS.L2_STALL / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueBW;tma_l3_bound_group", "MetricName": "tma_sq_full", "MetricThreshold": "tma_sq_full > 0.3 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_mem_bandwidth", + "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_mem_bandwidth", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write", - "MetricExpr": "EXE_ACTIVITY.BOUND_ON_STORES / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.BOUND_ON_STORES / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1419,7 +1446,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of cycles when the memory subsystem had loads blocked since they could not forward data from earlier (in program order) overlapping stores", - "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_clks", + "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_store_fwd_blk", "MetricThreshold": "tma_store_fwd_blk > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1428,7 +1455,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles the CPU spent handling L1D store misses", - "MetricExpr": "(L2_RQSTS.RFO_HIT * 10 * (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) + (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_clks", + "MetricExpr": "(L2_RQSTS.RFO_HIT * 10 * (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) + (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_thread_clks", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_issueSL;tma_store_bound_group", "MetricName": "tma_store_latency", "MetricThreshold": "tma_store_latency > 0.1 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1437,7 +1464,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Store operations", - "MetricExpr": "(UOPS_DISPATCHED.PORT_4_9 + UOPS_DISPATCHED.PORT_7_8) / (4 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED.PORT_4_9 + UOPS_DISPATCHED.PORT_7_8) / (4 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_store_op_utilization", "MetricThreshold": "tma_store_op_utilization > 0.6", @@ -1454,7 +1481,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles where the STLB was missed by store accesses, performing a hardware page walk", - "MetricExpr": "DTLB_STORE_MISSES.WALK_ACTIVE / tma_info_core_clks", + "MetricExpr": "DTLB_STORE_MISSES.WALK_ACTIVE / tma_info_core_core_clks", "MetricGroup": "MemoryTLB;TopdownL5;tma_L5_group;tma_dtlb_store_group", "MetricName": "tma_store_stlb_miss", "MetricThreshold": "tma_store_stlb_miss > 0.05 & (tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1462,7 +1489,7 @@ }, { "BriefDescription": "This metric estimates how often CPU was stalled due to Streaming store memory accesses; Streaming store optimize out a read request required by RFO stores", - "MetricExpr": "9 * OCR.STREAMING_WR.ANY_RESPONSE / tma_info_clks", + "MetricExpr": "9 * OCR.STREAMING_WR.ANY_RESPONSE / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueSmSt;tma_store_bound_group", "MetricName": "tma_streaming_stores", "MetricThreshold": "tma_streaming_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1471,7 +1498,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to new branch address clears", - "MetricExpr": "10 * BACLEARS.ANY / tma_info_clks", + "MetricExpr": "10 * BACLEARS.ANY / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;TopdownL4;tma_L4_group;tma_branch_resteers_group", "MetricName": "tma_unknown_branches", "MetricThreshold": "tma_unknown_branches > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", diff --git a/tools/perf/pmu-events/arch/x86/icelake/metricgroups.json b/tools/perf/pmu-events/arch/x86/icelake/metricgroups.json new file mode 100644 index 000000000000..a151ba9cccb0 --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/icelake/metricgroups.json @@ -0,0 +1,113 @@ +{ + "Backend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Bad": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BadSpec": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BigFoot": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BrMispredicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Branches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CacheMisses": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CodeGen": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Compute": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Cor": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSBmiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DataSharing": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Fed": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Flops": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpScalar": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Frontend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "HPC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IcMiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "InsType": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "L2Evicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "LSD": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MachineClears": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBound": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryTLB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_BW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_Lat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MicroSeq": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "OS": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Offcore": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PGO": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Pipeline": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PortsUtil": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Prefetches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Ret": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Retire": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SMT": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Server": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Snoop": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SoC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL1": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL2": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL3mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "TopdownL5": "Metrics for top-down breakdown at level 5", + "TopdownL6": "Metrics for top-down breakdown at level 6", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_L5_group": "Metrics for top-down breakdown at level 5", + "tma_L6_group": "Metrics for top-down breakdown at level 6", + "tma_alu_op_utilization_group": "Metrics contributing to tma_alu_op_utilization category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_branch_resteers_group": "Metrics contributing to tma_branch_resteers category", + "tma_core_bound_group": "Metrics contributing to tma_core_bound category", + "tma_dram_bound_group": "Metrics contributing to tma_dram_bound category", + "tma_dtlb_load_group": "Metrics contributing to tma_dtlb_load category", + "tma_dtlb_store_group": "Metrics contributing to tma_dtlb_store category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_fp_arith_group": "Metrics contributing to tma_fp_arith category", + "tma_fp_vector_group": "Metrics contributing to tma_fp_vector category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_heavy_operations_group": "Metrics contributing to tma_heavy_operations category", + "tma_issue2P": "Metrics related by the issue $issue2P", + "tma_issueBC": "Metrics related by the issue $issueBC", + "tma_issueBM": "Metrics related by the issue $issueBM", + "tma_issueBW": "Metrics related by the issue $issueBW", + "tma_issueD0": "Metrics related by the issue $issueD0", + "tma_issueFB": "Metrics related by the issue $issueFB", + "tma_issueFL": "Metrics related by the issue $issueFL", + "tma_issueL1": "Metrics related by the issue $issueL1", + "tma_issueLat": "Metrics related by the issue $issueLat", + "tma_issueMC": "Metrics related by the issue $issueMC", + "tma_issueMS": "Metrics related by the issue $issueMS", + "tma_issueMV": "Metrics related by the issue $issueMV", + "tma_issueRFO": "Metrics related by the issue $issueRFO", + "tma_issueSL": "Metrics related by the issue $issueSL", + "tma_issueSO": "Metrics related by the issue $issueSO", + "tma_issueSmSt": "Metrics related by the issue $issueSmSt", + "tma_issueSpSt": "Metrics related by the issue $issueSpSt", + "tma_issueSyncxn": "Metrics related by the issue $issueSyncxn", + "tma_issueTLB": "Metrics related by the issue $issueTLB", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_l3_bound_group": "Metrics contributing to tma_l3_bound category", + "tma_light_operations_group": "Metrics contributing to tma_light_operations category", + "tma_load_op_utilization_group": "Metrics contributing to tma_load_op_utilization category", + "tma_mem_latency_group": "Metrics contributing to tma_mem_latency category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_microcode_sequencer_group": "Metrics contributing to tma_microcode_sequencer category", + "tma_mite_group": "Metrics contributing to tma_mite category", + "tma_ports_utilization_group": "Metrics contributing to tma_ports_utilization category", + "tma_ports_utilized_0_group": "Metrics contributing to tma_ports_utilized_0 category", + "tma_ports_utilized_3m_group": "Metrics contributing to tma_ports_utilized_3m category", + "tma_retiring_group": "Metrics contributing to tma_retiring category", + "tma_serializing_operation_group": "Metrics contributing to tma_serializing_operation category", + "tma_store_bound_group": "Metrics contributing to tma_store_bound category", + "tma_store_op_utilization_group": "Metrics contributing to tma_store_op_utilization category" +} diff --git a/tools/perf/pmu-events/arch/x86/icelakex/icx-metrics.json b/tools/perf/pmu-events/arch/x86/icelakex/icx-metrics.json index 0f9b174dfc22..6f25b5b7aaf6 100644 --- a/tools/perf/pmu-events/arch/x86/icelakex/icx-metrics.json +++ b/tools/perf/pmu-events/arch/x86/icelakex/icx-metrics.json @@ -29,11 +29,244 @@ }, { "BriefDescription": "Uncore frequency per die [GHZ]", - "MetricExpr": "tma_info_socket_clks / #num_dies / duration_time / 1e9", + "MetricExpr": "tma_info_system_socket_clks / #num_dies / duration_time / 1e9", "MetricGroup": "SoC", "MetricName": "UNCORE_FREQ" }, { + "BriefDescription": "Cycles per instruction retired; indicating how much time each executed instruction took; in units of cycles.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD / INST_RETIRED.ANY", + "MetricName": "cpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "CPU operating frequency (in GHz)", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD / CPU_CLK_UNHALTED.REF_TSC * #SYSTEM_TSC_FREQ / 1e9", + "MetricName": "cpu_operating_frequency", + "ScaleUnit": "1GHz" + }, + { + "BriefDescription": "Percentage of time spent in the active CPU power state C0", + "MetricExpr": "tma_info_system_cpu_utilization", + "MetricName": "cpu_utilization", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for 2 megabyte page sizes) caused by demand data loads to the total number of completed instructions", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_COMPLETED_2M_4M / INST_RETIRED.ANY", + "MetricName": "dtlb_2nd_level_2mb_large_page_load_mpi", + "PublicDescription": "Ratio of number of completed page walks (for 2 megabyte page sizes) caused by demand data loads to the total number of completed instructions. This implies it missed in the Data Translation Lookaside Buffer (DTLB) and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data loads to the total number of completed instructions", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "dtlb_2nd_level_load_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data loads to the total number of completed instructions. This implies it missed in the DTLB and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data stores to the total number of completed instructions", + "MetricExpr": "DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "dtlb_2nd_level_store_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data stores to the total number of completed instructions. This implies it missed in the DTLB and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Bandwidth of IO writes that are initiated by end device controllers that are writing memory to the CPU.", + "MetricExpr": "(UNC_CHA_TOR_INSERTS.IO_HIT_ITOM + UNC_CHA_TOR_INSERTS.IO_MISS_ITOM + UNC_CHA_TOR_INSERTS.IO_HIT_ITOMCACHENEAR + UNC_CHA_TOR_INSERTS.IO_MISS_ITOMCACHENEAR) * 64 / 1e6 / duration_time", + "MetricName": "io_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for 2 megabyte and 4 megabyte page sizes) caused by a code fetch to the total number of completed instructions", + "MetricExpr": "ITLB_MISSES.WALK_COMPLETED_2M_4M / INST_RETIRED.ANY", + "MetricName": "itlb_2nd_level_large_page_mpi", + "PublicDescription": "Ratio of number of completed page walks (for 2 megabyte and 4 megabyte page sizes) caused by a code fetch to the total number of completed instructions. This implies it missed in the Instruction Translation Lookaside Buffer (ITLB) and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by a code fetch to the total number of completed instructions", + "MetricExpr": "ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "itlb_2nd_level_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by a code fetch to the total number of completed instructions. This implies it missed in the ITLB (Instruction TLB) and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read requests missing in L1 instruction cache (includes prefetches) to the total number of completed instructions", + "MetricExpr": "L2_RQSTS.ALL_CODE_RD / INST_RETIRED.ANY", + "MetricName": "l1_i_code_read_misses_with_prefetches_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of demand load requests hitting in L1 data cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_RETIRED.L1_HIT / INST_RETIRED.ANY", + "MetricName": "l1d_demand_data_read_hits_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of requests missing L1 data cache (includes data+rfo w/ prefetches) to the total number of completed instructions", + "MetricExpr": "L1D.REPLACEMENT / INST_RETIRED.ANY", + "MetricName": "l1d_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read request missing L2 cache to the total number of completed instructions", + "MetricExpr": "L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", + "MetricName": "l2_demand_code_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed demand load requests hitting in L2 cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_RETIRED.L2_HIT / INST_RETIRED.ANY", + "MetricName": "l2_demand_data_read_hits_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed data read request missing L2 cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_RETIRED.L2_MISS / INST_RETIRED.ANY", + "MetricName": "l2_demand_data_read_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of requests missing L2 cache (includes code+data+rfo w/ prefetches) to the total number of completed instructions", + "MetricExpr": "L2_LINES_IN.ALL / INST_RETIRED.ANY", + "MetricName": "l2_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read requests missing last level core cache (includes demand w/ prefetches) to the total number of completed instructions", + "MetricExpr": "(UNC_CHA_TOR_INSERTS.IA_MISS_CRD + UNC_CHA_TOR_INSERTS.IA_MISS_CRD_PREF) / INST_RETIRED.ANY", + "MetricName": "llc_code_read_mpi_demand_plus_prefetch", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand data read miss (read memory access) in nano seconds", + "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD / UNC_CHA_TOR_INSERTS.IA_MISS_DRD) / (UNC_CHA_CLOCKTICKS / (source_count(UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD) * #num_packages)) * duration_time", + "MetricName": "llc_demand_data_read_miss_latency", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand data read miss (read memory access) addressed to local memory in nano seconds", + "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_LOCAL / UNC_CHA_TOR_INSERTS.IA_MISS_DRD_LOCAL) / (UNC_CHA_CLOCKTICKS / (source_count(UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_LOCAL) * #num_packages)) * duration_time", + "MetricName": "llc_demand_data_read_miss_latency_for_local_requests", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand data read miss (read memory access) addressed to remote memory in nano seconds", + "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE / UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE) / (UNC_CHA_CLOCKTICKS / (source_count(UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE) * #num_packages)) * duration_time", + "MetricName": "llc_demand_data_read_miss_latency_for_remote_requests", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand data read miss (read memory access) addressed to Intel(R) Optane(TM) Persistent Memory(PMEM) in nano seconds", + "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_PMM / UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PMM) / (UNC_CHA_CLOCKTICKS / (source_count(UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_PMM) * #num_packages)) * duration_time", + "MetricName": "llc_demand_data_read_miss_to_pmem_latency", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Bandwidth (MB/sec) of read requests that miss the last level cache (LLC) and go to local memory.", + "MetricExpr": "UNC_CHA_REQUESTS.READS_LOCAL * 64 / 1e6 / duration_time", + "MetricName": "llc_miss_local_memory_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Bandwidth (MB/sec) of write requests that miss the last level cache (LLC) and go to local memory.", + "MetricExpr": "UNC_CHA_REQUESTS.WRITES_LOCAL * 64 / 1e6 / duration_time", + "MetricName": "llc_miss_local_memory_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Bandwidth (MB/sec) of read requests that miss the last level cache (LLC) and go to remote memory.", + "MetricExpr": "UNC_CHA_REQUESTS.READS_REMOTE * 64 / 1e6 / duration_time", + "MetricName": "llc_miss_remote_memory_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Bandwidth (MB/sec) of write requests that miss the last level cache (LLC) and go to remote memory.", + "MetricExpr": "UNC_CHA_REQUESTS.WRITES_REMOTE * 64 / 1e6 / duration_time", + "MetricName": "llc_miss_remote_memory_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "The ratio of number of completed memory load instructions to the total number completed instructions", + "MetricExpr": "MEM_INST_RETIRED.ALL_LOADS / INST_RETIRED.ANY", + "MetricName": "loads_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "DDR memory read bandwidth (MB/sec)", + "MetricExpr": "UNC_M_CAS_COUNT.RD * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "DDR memory bandwidth (MB/sec)", + "MetricExpr": "(UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_total", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "DDR memory write bandwidth (MB/sec)", + "MetricExpr": "UNC_M_CAS_COUNT.WR * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Memory write bandwidth (MB/sec) caused by directory updates; includes DDR and Intel(R) Optane(TM) Persistent Memory(PMEM).", + "MetricExpr": "(UNC_CHA_DIR_UPDATE.HA + UNC_CHA_DIR_UPDATE.TOR + UNC_M2M_DIRECTORY_UPDATE.ANY) * 64 / 1e6 / duration_time", + "MetricName": "memory_extra_write_bw_due_to_directory_updates", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Memory read that miss the last level cache (LLC) addressed to local DRAM as a percentage of total memory read accesses, does not include LLC prefetches.", + "MetricExpr": "(UNC_CHA_TOR_INSERTS.IA_MISS_DRD_LOCAL + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PREF_LOCAL) / (UNC_CHA_TOR_INSERTS.IA_MISS_DRD_LOCAL + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PREF_LOCAL + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PREF_REMOTE)", + "MetricName": "numa_reads_addressed_to_local_dram", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Memory reads that miss the last level cache (LLC) addressed to remote DRAM as a percentage of total memory read accesses, does not include LLC prefetches.", + "MetricExpr": "(UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PREF_REMOTE) / (UNC_CHA_TOR_INSERTS.IA_MISS_DRD_LOCAL + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PREF_LOCAL + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PREF_REMOTE)", + "MetricName": "numa_reads_addressed_to_remote_dram", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from decoded instruction cache (decoded stream buffer or DSB) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS + LSD.UOPS)", + "MetricName": "percent_uops_delivered_from_decoded_icache", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from legacy decode pipeline (Micro-instruction Translation Engine or MITE) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.MITE_UOPS / (IDQ.DSB_UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS + LSD.UOPS)", + "MetricName": "percent_uops_delivered_from_legacy_decode_pipeline", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from microcode sequencer (MS) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.MS_UOPS / (IDQ.DSB_UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS + LSD.UOPS)", + "MetricName": "percent_uops_delivered_from_microcode_sequencer", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Intel(R) Optane(TM) Persistent Memory(PMEM) memory read bandwidth (MB/sec)", + "MetricExpr": "UNC_M_PMM_RPQ_INSERTS * 64 / 1e6 / duration_time", + "MetricName": "pmem_memory_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Intel(R) Optane(TM) Persistent Memory(PMEM) memory bandwidth (MB/sec)", + "MetricExpr": "(UNC_M_PMM_RPQ_INSERTS + UNC_M_PMM_WPQ_INSERTS) * 64 / 1e6 / duration_time", + "MetricName": "pmem_memory_bandwidth_total", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Intel(R) Optane(TM) Persistent Memory(PMEM) memory write bandwidth (MB/sec)", + "MetricExpr": "UNC_M_PMM_WPQ_INSERTS * 64 / 1e6 / duration_time", + "MetricName": "pmem_memory_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { "BriefDescription": "Percentage of cycles spent in System Management Interrupts.", "MetricExpr": "((msr@aperf@ - cycles) / msr@aperf@ if msr@smi@ > 0 else 0)", "MetricGroup": "smi", @@ -49,8 +282,14 @@ "ScaleUnit": "1SMI#" }, { + "BriefDescription": "The ratio of number of completed memory store instructions to the total number completed instructions", + "MetricExpr": "MEM_INST_RETIRED.ALL_STORES / INST_RETIRED.ANY", + "MetricName": "stores_per_instr", + "ScaleUnit": "1per_instr" + }, + { "BriefDescription": "This metric estimates how often memory load accesses were aliased by preceding stores (in program order) with a 4K address offset", - "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_clks", + "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_4k_aliasing", "MetricThreshold": "tma_4k_aliasing > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -59,7 +298,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution ports for ALU operations.", - "MetricExpr": "(UOPS_DISPATCHED.PORT_0 + UOPS_DISPATCHED.PORT_1 + UOPS_DISPATCHED.PORT_5 + UOPS_DISPATCHED.PORT_6) / (4 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED.PORT_0 + UOPS_DISPATCHED.PORT_1 + UOPS_DISPATCHED.PORT_5 + UOPS_DISPATCHED.PORT_6) / (4 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_alu_op_utilization", "MetricThreshold": "tma_alu_op_utilization > 0.6", @@ -67,7 +306,7 @@ }, { "BriefDescription": "This metric estimates fraction of slots the CPU retired uops delivered by the Microcode_Sequencer as a result of Assists", - "MetricExpr": "100 * ASSISTS.ANY / tma_info_slots", + "MetricExpr": "100 * ASSISTS.ANY / tma_info_thread_slots", "MetricGroup": "TopdownL4;tma_L4_group;tma_microcode_sequencer_group", "MetricName": "tma_assists", "MetricThreshold": "tma_assists > 0.1 & (tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1)", @@ -76,25 +315,29 @@ }, { "BriefDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend", - "MetricExpr": "topdown\\-be\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 5 * cpu@INT_MISC.RECOVERY_CYCLES\\,cmask\\=1\\,edge@ / tma_info_slots", - "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "topdown\\-be\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 5 * cpu@INT_MISC.RECOVERY_CYCLES\\,cmask\\=1\\,edge@ / tma_info_thread_slots", + "MetricGroup": "Default;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend. Backend is the portion of the processor core where the out-of-order scheduler dispatches ready uops into their respective execution units; and once completed these uops get retired according to program order. For example; stalls due to data-cache misses or stalls due to the divider unit being overloaded are both categorized under Backend Bound. Backend Bound is further divided into two main categories: Memory Bound and Core Bound. Sample with: TOPDOWN.BACKEND_BOUND_SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations", + "DefaultMetricgroupName": "TopdownL1", "MetricExpr": "max(1 - (tma_frontend_bound + tma_backend_bound + tma_retiring), 0)", - "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", + "MetricGroup": "Default;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots wasted due to incorrect speculations. This include slots used to issue uops that do not eventually get retired and slots for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work due to miss-predicted branches are categorized under Bad Speculation category. Incorrect data speculation followed by Memory Ordering Nukes is another example.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring branch instructions.", - "MetricExpr": "tma_light_operations * BR_INST_RETIRED.ALL_BRANCHES / (tma_retiring * tma_info_slots)", + "MetricExpr": "tma_light_operations * BR_INST_RETIRED.ALL_BRANCHES / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_branch_instructions", "MetricThreshold": "tma_branch_instructions > 0.1 & tma_light_operations > 0.6", @@ -106,12 +349,13 @@ "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15", - "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_branch_misprediction_cost, tma_info_mispredictions, tma_mispredicts_resteers", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_info_bottleneck_mispredictions, tma_mispredicts_resteers", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers", - "MetricExpr": "INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks + tma_unknown_branches", + "MetricExpr": "INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks + tma_unknown_branches", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteers", "MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -129,7 +373,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Machine Clears", - "MetricExpr": "(1 - BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT)) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks", + "MetricExpr": "(1 - BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT)) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks", "MetricGroup": "BadSpec;MachineClears;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueMC", "MetricName": "tma_clears_resteers", "MetricThreshold": "tma_clears_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", @@ -139,7 +383,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to contested accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(44 * tma_info_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_HITM * (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM / (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM + OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) + 43.5 * tma_info_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "(44 * tma_info_system_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_HITM * (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM / (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM + OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) + 43.5 * tma_info_system_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_contested_accesses", "MetricThreshold": "tma_contested_accesses > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -152,13 +396,14 @@ "MetricGroup": "Backend;Compute;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck. Shortage in hardware compute resources; or dependencies in software's instructions are both categorized under Core Bound. Hence it may indicate the machine ran out of an out-of-order resource; certain execution units are overloaded or dependencies in program's data- or instruction-flow are limiting the performance (e.g. FP-chained long-latency arithmetic operations).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to data-sharing accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "43.5 * tma_info_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_L3_HIT_RETIRED.XSNP_HITM * (1 - OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM / (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM + OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "43.5 * tma_info_system_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_L3_HIT_RETIRED.XSNP_HITM * (1 - OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM / (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM + OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_data_sharing", "MetricThreshold": "tma_data_sharing > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -167,16 +412,16 @@ }, { "BriefDescription": "This metric represents fraction of cycles where decoder-0 was the only active decoder", - "MetricExpr": "(cpu@INST_DECODED.DECODERS\\,cmask\\=1@ - cpu@INST_DECODED.DECODERS\\,cmask\\=2@) / tma_info_core_clks / 2", + "MetricExpr": "(cpu@INST_DECODED.DECODERS\\,cmask\\=1@ - cpu@INST_DECODED.DECODERS\\,cmask\\=2@) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL4;tma_L4_group;tma_issueD0;tma_mite_group", "MetricName": "tma_decoder0_alone", - "MetricThreshold": "tma_decoder0_alone > 0.1 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 5 > 0.35))", + "MetricThreshold": "tma_decoder0_alone > 0.1 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 5 > 0.35))", "PublicDescription": "This metric represents fraction of cycles where decoder-0 was the only active decoder. Related metrics: tma_few_uops_instructions", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles where the Divider unit was active", - "MetricExpr": "ARITH.DIVIDER_ACTIVE / tma_info_clks", + "MetricExpr": "ARITH.DIVIDER_ACTIVE / tma_info_thread_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_divider", "MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -186,7 +431,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks - tma_l2_bound - tma_pmm_bound if #has_pmem > 0 else CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks - tma_l2_bound)", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_thread_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks - tma_l2_bound - tma_pmm_bound if #has_pmem > 0 else CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_thread_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks - tma_l2_bound)", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -195,43 +440,43 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline", - "MetricExpr": "(IDQ.DSB_CYCLES_ANY - IDQ.DSB_CYCLES_OK) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.DSB_CYCLES_ANY - IDQ.DSB_CYCLES_OK) / tma_info_core_core_clks / 2", "MetricGroup": "DSB;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_dsb", - "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 5 > 0.35)", + "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 5 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline. For example; inefficient utilization of the DSB cache structure or bank conflict when reading from it; are categorized here.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_dsb_switches", "MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb, tma_lcp", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses", - "MetricExpr": "min(7 * cpu@DTLB_LOAD_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_LOAD_MISSES.WALK_ACTIVE, max(CYCLE_ACTIVITY.CYCLES_MEM_ANY - CYCLE_ACTIVITY.CYCLES_L1D_MISS, 0)) / tma_info_clks", + "MetricExpr": "min(7 * cpu@DTLB_LOAD_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_LOAD_MISSES.WALK_ACTIVE, max(CYCLE_ACTIVITY.CYCLES_MEM_ANY - CYCLE_ACTIVITY.CYCLES_L1D_MISS, 0)) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group", "MetricName": "tma_dtlb_load", "MetricThreshold": "tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses. TLBs (Translation Look-aside Buffers) are processor caches for recently used entries out of the Page Tables that are used to map virtual- to physical-addresses by the operating system. This metric approximates the potential delay of demand loads missing the first-level data TLB (assuming worst case scenario with back to back misses to different pages). This includes hitting in the second-level TLB (STLB) as well as performing a hardware page walk on an STLB miss. Sample with: MEM_INST_RETIRED.STLB_MISS_LOADS_PS. Related metrics: tma_dtlb_store, tma_info_memory_data_tlbs", + "PublicDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses. TLBs (Translation Look-aside Buffers) are processor caches for recently used entries out of the Page Tables that are used to map virtual- to physical-addresses by the operating system. This metric approximates the potential delay of demand loads missing the first-level data TLB (assuming worst case scenario with back to back misses to different pages). This includes hitting in the second-level TLB (STLB) as well as performing a hardware page walk on an STLB miss. Sample with: MEM_INST_RETIRED.STLB_MISS_LOADS_PS. Related metrics: tma_dtlb_store, tma_info_bottleneck_memory_data_tlbs", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses", - "MetricExpr": "(7 * cpu@DTLB_STORE_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_STORE_MISSES.WALK_ACTIVE) / tma_info_core_clks", + "MetricExpr": "(7 * cpu@DTLB_STORE_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_STORE_MISSES.WALK_ACTIVE) / tma_info_core_core_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_store_bound_group", "MetricName": "tma_dtlb_store", "MetricThreshold": "tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses. As with ordinary data caching; focus on improving data locality and reducing working-set size to reduce DTLB overhead. Additionally; consider using profile-guided optimization (PGO) to collocate frequently-used data on the same page. Try using larger page sizes for large amounts of frequently-used data. Sample with: MEM_INST_RETIRED.STLB_MISS_STORES_PS. Related metrics: tma_dtlb_load, tma_info_memory_data_tlbs", + "PublicDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses. As with ordinary data caching; focus on improving data locality and reducing working-set size to reduce DTLB overhead. Additionally; consider using profile-guided optimization (PGO) to collocate frequently-used data on the same page. Try using larger page sizes for large amounts of frequently-used data. Sample with: MEM_INST_RETIRED.STLB_MISS_STORES_PS. Related metrics: tma_dtlb_load, tma_info_bottleneck_memory_data_tlbs", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates how often CPU was handling synchronizations due to False Sharing", - "MetricExpr": "48 * tma_info_average_frequency * OCR.DEMAND_RFO.L3_HIT.SNOOP_HITM / tma_info_clks", + "MetricExpr": "48 * tma_info_system_average_frequency * OCR.DEMAND_RFO.L3_HIT.SNOOP_HITM / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_store_bound_group", "MetricName": "tma_false_sharing", "MetricThreshold": "tma_false_sharing > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -240,11 +485,11 @@ }, { "BriefDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed", - "MetricExpr": "L1D_PEND_MISS.FB_FULL / tma_info_clks", + "MetricExpr": "L1D_PEND_MISS.FB_FULL / tma_info_thread_clks", "MetricGroup": "MemoryBW;TopdownL4;tma_L4_group;tma_issueBW;tma_issueSL;tma_issueSmSt;tma_l1_bound_group", "MetricName": "tma_fb_full", "MetricThreshold": "tma_fb_full > 0.3", - "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", + "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", "ScaleUnit": "100%" }, { @@ -252,16 +497,18 @@ "MetricExpr": "max(0, tma_frontend_bound - tma_fetch_latency)", "MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", "MetricName": "tma_fetch_bandwidth", - "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 5 > 0.35", - "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb, tma_lcp", + "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 5 > 0.35", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues", - "MetricExpr": "(5 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE - INT_MISC.UOP_DROPPING) / tma_info_slots", + "MetricExpr": "(5 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE - INT_MISC.UOP_DROPPING) / tma_info_thread_slots", "MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: FRONTEND_RETIRED.LATENCY_GE_16_PS;FRONTEND_RETIRED.LATENCY_GE_8_PS", "ScaleUnit": "100%" }, @@ -276,6 +523,7 @@ }, { "BriefDescription": "This metric represents overall arithmetic floating-point (FP) operations fraction the CPU has executed (retired)", + "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "tma_x87_use + tma_fp_scalar + tma_fp_vector", "MetricGroup": "HPC;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_fp_arith", @@ -285,7 +533,7 @@ }, { "BriefDescription": "This metric approximates arithmetic floating-point (FP) scalar uops fraction the CPU has retired", - "MetricExpr": "cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ / (tma_retiring * tma_info_slots)", + "MetricExpr": "cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL4;tma_L4_group;tma_fp_arith_group;tma_issue2P", "MetricName": "tma_fp_scalar", "MetricThreshold": "tma_fp_scalar > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6)", @@ -294,7 +542,7 @@ }, { "BriefDescription": "This metric approximates arithmetic floating-point (FP) vector uops fraction the CPU has retired aggregated across all vector widths", - "MetricExpr": "cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@ / (tma_retiring * tma_info_slots)", + "MetricExpr": "cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@ / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL4;tma_L4_group;tma_fp_arith_group;tma_issue2P", "MetricName": "tma_fp_vector", "MetricThreshold": "tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6)", @@ -303,7 +551,7 @@ }, { "BriefDescription": "This metric approximates arithmetic FP vector uops fraction the CPU has retired for 128-bit wide vectors", - "MetricExpr": "(FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL5;tma_L5_group;tma_fp_vector_group;tma_issue2P", "MetricName": "tma_fp_vector_128b", "MetricThreshold": "tma_fp_vector_128b > 0.1 & (tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6))", @@ -312,7 +560,7 @@ }, { "BriefDescription": "This metric approximates arithmetic FP vector uops fraction the CPU has retired for 256-bit wide vectors", - "MetricExpr": "(FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL5;tma_L5_group;tma_fp_vector_group;tma_issue2P", "MetricName": "tma_fp_vector_256b", "MetricThreshold": "tma_fp_vector_256b > 0.1 & (tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6))", @@ -321,7 +569,7 @@ }, { "BriefDescription": "This metric approximates arithmetic FP vector uops fraction the CPU has retired for 512-bit wide vectors", - "MetricExpr": "(FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL5;tma_L5_group;tma_fp_vector_group;tma_issue2P", "MetricName": "tma_fp_vector_512b", "MetricThreshold": "tma_fp_vector_512b > 0.1 & (tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6))", @@ -330,10 +578,12 @@ }, { "BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend", - "MetricExpr": "topdown\\-fe\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) - INT_MISC.UOP_DROPPING / tma_info_slots", - "MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "topdown\\-fe\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) - INT_MISC.UOP_DROPPING / tma_info_thread_slots", + "MetricGroup": "Default;PGO;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound. Sample with: FRONTEND_RETIRED.LATENCY_GE_4_PS", "ScaleUnit": "100%" }, @@ -343,12 +593,13 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_heavy_operations", "MetricThreshold": "tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to instruction cache misses", - "MetricExpr": "ICACHE_16B.IFDATA_STALL / tma_info_clks", + "MetricExpr": "ICACHE_16B.IFDATA_STALL / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;IcMiss;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_icache_misses", "MetricThreshold": "tma_icache_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -356,726 +607,741 @@ "ScaleUnit": "100%" }, { - "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", - "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time", - "MetricGroup": "Power;Summary", - "MetricName": "tma_info_average_frequency" + "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_thread_slots / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BrMispredicts;tma_issueBM", + "MetricName": "tma_info_bad_spec_branch_misprediction_cost", + "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_info_bottleneck_mispredictions, tma_mispredicts_resteers" + }, + { + "BriefDescription": "Instructions per retired mispredicts for conditional non-taken branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_NTAKEN", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_cond_ntaken", + "MetricThreshold": "tma_info_bad_spec_ipmisp_cond_ntaken < 200" + }, + { + "BriefDescription": "Instructions per retired mispredicts for conditional taken branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_TAKEN", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_cond_taken", + "MetricThreshold": "tma_info_bad_spec_ipmisp_cond_taken < 200" + }, + { + "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.INDIRECT", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_indirect", + "MetricThreshold": "tma_info_bad_spec_ipmisp_indirect < 1e3" + }, + { + "BriefDescription": "Instructions per retired mispredicts for return branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.RET", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_ret", + "MetricThreshold": "tma_info_bad_spec_ipmisp_ret < 500" + }, + { + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_core_ipmispredict", + "MetricGroup": "Bad;BadSpec;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmispredict", + "MetricThreshold": "tma_info_bad_spec_ipmispredict < 200" + }, + { + "BriefDescription": "Probability of Core Bound bottleneck hidden by SMT-profiling artifacts", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(100 * (1 - tma_core_bound / tma_ports_utilization if tma_core_bound < tma_ports_utilization else 1) if tma_info_system_smt_2t_utilization > 0.5 else 0)", + "MetricGroup": "Cor;SMT", + "MetricName": "tma_info_botlnk_l0_core_bound_likely", + "MetricThreshold": "tma_info_botlnk_l0_core_bound_likely > 0.5" + }, + { + "BriefDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_fetch_latency * tma_dsb_switches / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches) + tma_fetch_bandwidth * tma_mite / (tma_dsb + tma_mite))", + "MetricGroup": "DSBmiss;Fed;tma_issueFB", + "MetricName": "tma_info_botlnk_l2_dsb_misses", + "MetricThreshold": "tma_info_botlnk_l2_dsb_misses > 10", + "PublicDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp" + }, + { + "BriefDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck", + "MetricExpr": "100 * (tma_fetch_latency * tma_icache_misses / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", + "MetricGroup": "Fed;FetchLat;IcMiss;tma_issueFL", + "MetricName": "tma_info_botlnk_l2_ic_misses", + "MetricThreshold": "tma_info_botlnk_l2_ic_misses > 5", + "PublicDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck. Related metrics: " }, { "BriefDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses)", "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "100 * tma_fetch_latency * (tma_itlb_misses + tma_icache_misses + tma_unknown_branches) / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)", "MetricGroup": "BigFoot;Fed;Frontend;IcMiss;MemoryTLB;tma_issueBC", - "MetricName": "tma_info_big_code", - "MetricThreshold": "tma_info_big_code > 20", - "PublicDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses). Related metrics: tma_info_branching_overhead" + "MetricName": "tma_info_bottleneck_big_code", + "MetricThreshold": "tma_info_bottleneck_big_code > 20", + "PublicDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses). Related metrics: tma_info_bottleneck_branching_overhead" }, { - "BriefDescription": "Branch instructions per taken branch.", - "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", - "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_bptkbranch" + "BriefDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls)", + "MetricExpr": "100 * ((BR_INST_RETIRED.COND + 3 * BR_INST_RETIRED.NEAR_CALL + (BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL)) / tma_info_thread_slots)", + "MetricGroup": "Ret;tma_issueBC", + "MetricName": "tma_info_bottleneck_branching_overhead", + "MetricThreshold": "tma_info_bottleneck_branching_overhead > 10", + "PublicDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls). Related metrics: tma_info_bottleneck_big_code" }, { - "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", - "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_slots / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_branch_misprediction_cost", - "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_info_mispredictions, tma_mispredicts_resteers" + "BriefDescription": "Total pipeline cost of instruction fetch bandwidth related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_frontend_bound - tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) - tma_info_bottleneck_big_code", + "MetricGroup": "Fed;FetchBW;Frontend", + "MetricName": "tma_info_bottleneck_instruction_fetch_bw", + "MetricThreshold": "tma_info_bottleneck_instruction_fetch_bw > 20" }, { - "BriefDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls)", - "MetricExpr": "100 * ((BR_INST_RETIRED.COND + 3 * BR_INST_RETIRED.NEAR_CALL + (BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL)) / tma_info_slots)", - "MetricGroup": "Ret;tma_issueBC", - "MetricName": "tma_info_branching_overhead", - "MetricThreshold": "tma_info_branching_overhead > 10", - "PublicDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls). Related metrics: tma_info_big_code" + "BriefDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks", + "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_mem_bandwidth / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_sq_full / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full))) + tma_l1_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_fb_full / (tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk))", + "MetricGroup": "Mem;MemoryBW;Offcore;tma_issueBW", + "MetricName": "tma_info_bottleneck_memory_bandwidth", + "MetricThreshold": "tma_info_bottleneck_memory_bandwidth > 20", + "PublicDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks. Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full" }, { - "BriefDescription": "Fraction of branches that are CALL or RET", - "MetricExpr": "(BR_INST_RETIRED.NEAR_CALL + BR_INST_RETIRED.NEAR_RETURN) / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_callret" + "BriefDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_l1_bound / max(tma_memory_bound, tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_dtlb_load / max(tma_l1_bound, tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk)) + tma_store_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_dtlb_store / (tma_dtlb_store + tma_false_sharing + tma_split_stores + tma_store_latency + tma_streaming_stores)))", + "MetricGroup": "Mem;MemoryTLB;Offcore;tma_issueTLB", + "MetricName": "tma_info_bottleneck_memory_data_tlbs", + "MetricThreshold": "tma_info_bottleneck_memory_data_tlbs > 20", + "PublicDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs). Related metrics: tma_dtlb_load, tma_dtlb_store" }, { - "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "Pipeline", - "MetricName": "tma_info_clks" + "BriefDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_mem_latency / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_l3_hit_latency / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full)) + tma_l2_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound))", + "MetricGroup": "Mem;MemoryLat;Offcore;tma_issueLat", + "MetricName": "tma_info_bottleneck_memory_latency", + "MetricThreshold": "tma_info_bottleneck_memory_latency > 20", + "PublicDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches). Related metrics: tma_l3_hit_latency, tma_mem_latency" }, { - "BriefDescription": "STLB (2nd level TLB) code speculative misses per kilo instruction (misses of any page-size that complete the page walk)", - "MetricExpr": "1e3 * ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", - "MetricGroup": "Fed;MemoryTLB", - "MetricName": "tma_info_code_stlb_mpki" + "BriefDescription": "Total pipeline cost of Branch Misprediction related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", + "MetricGroup": "Bad;BadSpec;BrMispredicts;tma_issueBM", + "MetricName": "tma_info_bottleneck_mispredictions", + "MetricThreshold": "tma_info_bottleneck_mispredictions > 20", + "PublicDescription": "Total pipeline cost of Branch Misprediction related bottlenecks. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers" + }, + { + "BriefDescription": "Fraction of branches that are CALL or RET", + "MetricExpr": "(BR_INST_RETIRED.NEAR_CALL + BR_INST_RETIRED.NEAR_RETURN) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_callret" }, { "BriefDescription": "Fraction of branches that are non-taken conditionals", "MetricExpr": "BR_INST_RETIRED.COND_NTAKEN / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Bad;Branches;CodeGen;PGO", - "MetricName": "tma_info_cond_nt" + "MetricName": "tma_info_branches_cond_nt" }, { "BriefDescription": "Fraction of branches that are taken conditionals", "MetricExpr": "BR_INST_RETIRED.COND_TAKEN / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Bad;Branches;CodeGen;PGO", - "MetricName": "tma_info_cond_tk" + "MetricName": "tma_info_branches_cond_tk" }, { - "BriefDescription": "Probability of Core Bound bottleneck hidden by SMT-profiling artifacts", - "MetricExpr": "(100 * (1 - tma_core_bound / tma_ports_utilization if tma_core_bound < tma_ports_utilization else 1) if tma_info_smt_2t_utilization > 0.5 else 0)", - "MetricGroup": "Cor;SMT", - "MetricName": "tma_info_core_bound_likely", - "MetricThreshold": "tma_info_core_bound_likely > 0.5" + "BriefDescription": "Fraction of branches that are unconditional (direct or indirect) jumps", + "MetricExpr": "(BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_jump" + }, + { + "BriefDescription": "Fraction of branches of other types (not individually covered by other metrics in Info.Branches group)", + "MetricExpr": "1 - (tma_info_branches_cond_nt + tma_info_branches_cond_tk + tma_info_branches_callret + tma_info_branches_jump)", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_other_branches" }, { "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", "MetricExpr": "CPU_CLK_UNHALTED.DISTRIBUTED", "MetricGroup": "SMT", - "MetricName": "tma_info_core_clks" + "MetricName": "tma_info_core_core_clks" }, { "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks", "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_coreipc" + "MetricName": "tma_info_core_coreipc" }, { - "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricExpr": "1 / tma_info_ipc", - "MetricGroup": "Mem;Pipeline", - "MetricName": "tma_info_cpi" + "BriefDescription": "Floating Point Operations Per Cycle", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * cpu@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE\\,umask\\=0x18@ + 8 * cpu@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE\\,umask\\=0x60@ + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / tma_info_core_core_clks", + "MetricGroup": "Flops;Ret", + "MetricName": "tma_info_core_flopc" }, { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": "HPC;Summary", - "MetricName": "tma_info_cpu_utilization" + "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@) / (2 * tma_info_core_core_clks)", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_core_fp_arith_utilization", + "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." }, { - "BriefDescription": "Average Parallel L2 cache miss data reads", - "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", - "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_data_l2_mlp" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", + "MetricExpr": "UOPS_EXECUTED.THREAD / (UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1)", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", + "MetricName": "tma_info_core_ilp" }, { - "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time", - "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", - "MetricName": "tma_info_dram_bw_use", - "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_info_memory_bandwidth, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear)", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BadSpec;BrMispredicts;TopdownL1;tma_L1_group", + "MetricName": "tma_info_core_ipmispredict", + "MetricgroupNoGroup": "TopdownL1" }, { "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", "MetricExpr": "IDQ.DSB_UOPS / UOPS_ISSUED.ANY", "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", - "MetricName": "tma_info_dsb_coverage", - "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 5 > 0.35", - "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_misses, tma_info_iptb, tma_lcp" - }, - { - "BriefDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck", - "MetricExpr": "100 * (tma_fetch_latency * tma_dsb_switches / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches) + tma_fetch_bandwidth * tma_mite / (tma_dsb + tma_mite))", - "MetricGroup": "DSBmiss;Fed;tma_issueFB", - "MetricName": "tma_info_dsb_misses", - "MetricThreshold": "tma_info_dsb_misses > 10", - "PublicDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb, tma_lcp" + "MetricName": "tma_info_frontend_dsb_coverage", + "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 5 > 0.35", + "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_inst_mix_iptb, tma_lcp" }, { "BriefDescription": "Average number of cycles of a switch from the DSB fetch-unit to MITE fetch unit - see DSB_Switches tree node for details.", "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / cpu@DSB2MITE_SWITCHES.PENALTY_CYCLES\\,cmask\\=1\\,edge@", "MetricGroup": "DSBmiss", - "MetricName": "tma_info_dsb_switch_cost" - }, - { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", - "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", - "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", - "MetricName": "tma_info_execute" - }, - { - "BriefDescription": "The ratio of Executed- by Issued-Uops", - "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", - "MetricGroup": "Cor;Pipeline", - "MetricName": "tma_info_execute_per_issue", - "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." - }, - { - "BriefDescription": "Fill Buffer (FB) hits per kilo instructions for retired demand loads (L1D misses that merge into ongoing miss-handling entries)", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.FB_HIT / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_fb_hpki" + "MetricName": "tma_info_frontend_dsb_switch_cost" }, { "BriefDescription": "Average number of Uops issued by front-end when it issued something", "MetricExpr": "UOPS_ISSUED.ANY / cpu@UOPS_ISSUED.ANY\\,cmask\\=1@", "MetricGroup": "Fed;FetchBW", - "MetricName": "tma_info_fetch_upc" - }, - { - "BriefDescription": "Floating Point Operations Per Cycle", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * cpu@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE\\,umask\\=0x18@ + 8 * cpu@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE\\,umask\\=0x60@ + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / tma_info_core_clks", - "MetricGroup": "Flops;Ret", - "MetricName": "tma_info_flopc" + "MetricName": "tma_info_frontend_fetch_upc" }, { - "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@) / (2 * tma_info_core_clks)", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_fp_arith_utilization", - "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." + "BriefDescription": "Average Latency for L1 instruction cache misses", + "MetricExpr": "ICACHE_16B.IFDATA_STALL / cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@", + "MetricGroup": "Fed;FetchLat;IcMiss", + "MetricName": "tma_info_frontend_icache_miss_latency" }, { - "BriefDescription": "Giga Floating Point Operations Per Second", - "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * cpu@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE\\,umask\\=0x18@ + 8 * cpu@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE\\,umask\\=0x60@ + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / 1e9 / duration_time", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_gflops", - "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + "BriefDescription": "Instructions per non-speculative DSB miss (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / FRONTEND_RETIRED.ANY_DSB_MISS", + "MetricGroup": "DSBmiss;Fed", + "MetricName": "tma_info_frontend_ipdsb_miss_ret", + "MetricThreshold": "tma_info_frontend_ipdsb_miss_ret < 50" }, { - "BriefDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck", - "MetricExpr": "100 * (tma_fetch_latency * tma_icache_misses / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", - "MetricGroup": "Fed;FetchLat;IcMiss;tma_issueFL", - "MetricName": "tma_info_ic_misses", - "MetricThreshold": "tma_info_ic_misses > 5", - "PublicDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck. Related metrics: " + "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_inst_mix_instructions / BACLEARS.ANY", + "MetricGroup": "Fed", + "MetricName": "tma_info_frontend_ipunknown_branch" }, { - "BriefDescription": "Average Latency for L1 instruction cache misses", - "MetricExpr": "ICACHE_16B.IFDATA_STALL / cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@", - "MetricGroup": "Fed;FetchLat;IcMiss", - "MetricName": "tma_info_icache_miss_latency" + "BriefDescription": "L2 cache true code cacheline misses per kilo instruction", + "MetricExpr": "1e3 * FRONTEND_RETIRED.L2_MISS / INST_RETIRED.ANY", + "MetricGroup": "IcMiss", + "MetricName": "tma_info_frontend_l2mpki_code" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", - "MetricExpr": "UOPS_EXECUTED.THREAD / (UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1)", - "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", - "MetricName": "tma_info_ilp" + "BriefDescription": "L2 cache speculative code cacheline misses per kilo instruction", + "MetricExpr": "1e3 * L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", + "MetricGroup": "IcMiss", + "MetricName": "tma_info_frontend_l2mpki_code_all" }, { - "BriefDescription": "Total pipeline cost of instruction fetch bandwidth related bottlenecks", - "MetricExpr": "100 * (tma_frontend_bound - tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) - tma_info_big_code", - "MetricGroup": "Fed;FetchBW;Frontend", - "MetricName": "tma_info_instruction_fetch_bw", - "MetricThreshold": "tma_info_instruction_fetch_bw > 20" + "BriefDescription": "Branch instructions per taken branch.", + "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;PGO", + "MetricName": "tma_info_inst_mix_bptkbranch" }, { "BriefDescription": "Total number of retired Instructions", "MetricExpr": "INST_RETIRED.ANY", "MetricGroup": "Summary;TmaL1;tma_L1_group", - "MetricName": "tma_info_instructions", + "MetricName": "tma_info_inst_mix_instructions", "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" }, { - "BriefDescription": "Average IO (network or disk) Bandwidth Use for Reads [GB / sec]", - "MetricExpr": "(UNC_CHA_TOR_INSERTS.IO_HIT_ITOM + UNC_CHA_TOR_INSERTS.IO_MISS_ITOM + UNC_CHA_TOR_INSERTS.IO_HIT_ITOMCACHENEAR + UNC_CHA_TOR_INSERTS.IO_MISS_ITOMCACHENEAR) * 64 / 1e9 / duration_time", - "MetricGroup": "IoBW;Mem;Server;SoC", - "MetricName": "tma_info_io_read_bw" - }, - { - "BriefDescription": "Average IO (network or disk) Bandwidth Use for Writes [GB / sec]", - "MetricExpr": "UNC_CHA_TOR_INSERTS.IO_PCIRDCUR * 64 / 1e9 / duration_time", - "MetricGroup": "IoBW;Mem;Server;SoC", - "MetricName": "tma_info_io_write_bw" - }, - { "BriefDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_iparith", - "MetricThreshold": "tma_info_iparith < 10", + "MetricName": "tma_info_inst_mix_iparith", + "MetricThreshold": "tma_info_inst_mix_iparith < 10", "PublicDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate). May undercount due to FMA double counting. Approximated prior to BDW." }, { "BriefDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx128", - "MetricThreshold": "tma_info_iparith_avx128 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx128", + "MetricThreshold": "tma_info_inst_mix_iparith_avx128 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx256", - "MetricThreshold": "tma_info_iparith_avx256 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx256", + "MetricThreshold": "tma_info_inst_mix_iparith_avx256 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic AVX 512-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx512", - "MetricThreshold": "tma_info_iparith_avx512 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx512", + "MetricThreshold": "tma_info_inst_mix_iparith_avx512 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX 512-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_dp", - "MetricThreshold": "tma_info_iparith_scalar_dp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_dp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_dp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_SINGLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_sp", - "MetricThreshold": "tma_info_iparith_scalar_sp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_sp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_sp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per Branch (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Branches;Fed;InsType", - "MetricName": "tma_info_ipbranch", - "MetricThreshold": "tma_info_ipbranch < 8" - }, - { - "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": "Ret;Summary", - "MetricName": "tma_info_ipc" + "MetricName": "tma_info_inst_mix_ipbranch", + "MetricThreshold": "tma_info_inst_mix_ipbranch < 8" }, { "BriefDescription": "Instructions per (near) call (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_CALL", "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_ipcall", - "MetricThreshold": "tma_info_ipcall < 200" - }, - { - "BriefDescription": "Instructions per non-speculative DSB miss (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / FRONTEND_RETIRED.ANY_DSB_MISS", - "MetricGroup": "DSBmiss;Fed", - "MetricName": "tma_info_ipdsb_miss_ret", - "MetricThreshold": "tma_info_ipdsb_miss_ret < 50" - }, - { - "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", - "MetricGroup": "Branches;OS", - "MetricName": "tma_info_ipfarbranch", - "MetricThreshold": "tma_info_ipfarbranch < 1e6" + "MetricName": "tma_info_inst_mix_ipcall", + "MetricThreshold": "tma_info_inst_mix_ipcall < 200" }, { "BriefDescription": "Instructions per Floating Point (FP) Operation (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * cpu@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE\\,umask\\=0x18@ + 8 * cpu@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE\\,umask\\=0x60@ + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_ipflop", - "MetricThreshold": "tma_info_ipflop < 10" + "MetricName": "tma_info_inst_mix_ipflop", + "MetricThreshold": "tma_info_inst_mix_ipflop < 10" }, { "BriefDescription": "Instructions per Load (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_LOADS", "MetricGroup": "InsType", - "MetricName": "tma_info_ipload", - "MetricThreshold": "tma_info_ipload < 3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for conditional non-taken branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_NTAKEN", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_cond_ntaken", - "MetricThreshold": "tma_info_ipmisp_cond_ntaken < 200" - }, - { - "BriefDescription": "Instructions per retired mispredicts for conditional taken branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_TAKEN", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_cond_taken", - "MetricThreshold": "tma_info_ipmisp_cond_taken < 200" - }, - { - "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.INDIRECT", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_indirect", - "MetricThreshold": "tma_info_ipmisp_indirect < 1e3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for return branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.RET", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_ret", - "MetricThreshold": "tma_info_ipmisp_ret < 500" - }, - { - "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BadSpec;BrMispredicts", - "MetricName": "tma_info_ipmispredict", - "MetricThreshold": "tma_info_ipmispredict < 200" + "MetricName": "tma_info_inst_mix_ipload", + "MetricThreshold": "tma_info_inst_mix_ipload < 3" }, { "BriefDescription": "Instructions per Store (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_STORES", "MetricGroup": "InsType", - "MetricName": "tma_info_ipstore", - "MetricThreshold": "tma_info_ipstore < 8" + "MetricName": "tma_info_inst_mix_ipstore", + "MetricThreshold": "tma_info_inst_mix_ipstore < 8" }, { "BriefDescription": "Instructions per Software prefetch instruction (of any type: NTA/T0/T1/T2/Prefetch) (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / cpu@SW_PREFETCH_ACCESS.T0\\,umask\\=0xF@", "MetricGroup": "Prefetches", - "MetricName": "tma_info_ipswpf", - "MetricThreshold": "tma_info_ipswpf < 100" + "MetricName": "tma_info_inst_mix_ipswpf", + "MetricThreshold": "tma_info_inst_mix_ipswpf < 100" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW;Frontend;PGO;tma_issueFB", - "MetricName": "tma_info_iptb", - "MetricThreshold": "tma_info_iptb < 11", - "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_lcp" + "MetricName": "tma_info_inst_mix_iptb", + "MetricThreshold": "tma_info_inst_mix_iptb < 11", + "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_lcp" }, { - "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", - "MetricExpr": "tma_info_instructions / BACLEARS.ANY", - "MetricGroup": "Fed", - "MetricName": "tma_info_ipunknown_branch" + "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_core_l1d_cache_fill_bw" }, { - "BriefDescription": "Fraction of branches that are unconditional (direct or indirect) jumps", - "MetricExpr": "(BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL) / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_jump" + "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_core_l2_cache_fill_bw" }, { - "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_cpi" + "BriefDescription": "Rate of non silent evictions from the L2 cache per Kilo instruction", + "MetricExpr": "1e3 * L2_LINES_OUT.NON_SILENT / tma_info_inst_mix_instructions", + "MetricGroup": "L2Evicts;Mem;Server", + "MetricName": "tma_info_memory_core_l2_evictions_nonsilent_pki" }, { - "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_utilization", - "MetricThreshold": "tma_info_kernel_utilization > 0.05" + "BriefDescription": "Rate of silent evictions from the L2 cache per Kilo instruction where the evicted lines are dropped (no writeback to L3 or memory)", + "MetricExpr": "1e3 * L2_LINES_OUT.SILENT / tma_info_inst_mix_instructions", + "MetricGroup": "L2Evicts;Mem;Server", + "MetricName": "tma_info_memory_core_l2_evictions_silent_pki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw" + "BriefDescription": "Average per-core data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * OFFCORE_REQUESTS.ALL_REQUESTS / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_core_l3_cache_access_bw" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "tma_info_l1d_cache_fill_bw", + "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw_1t" + "MetricName": "tma_info_memory_core_l3_cache_fill_bw" + }, + { + "BriefDescription": "Fill Buffer (FB) hits per kilo instructions for retired demand loads (L1D misses that merge into ongoing miss-handling entries)", + "MetricExpr": "1e3 * MEM_LOAD_RETIRED.FB_HIT / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_fb_hpki" }, { "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L1_MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki" + "MetricName": "tma_info_memory_l1mpki" }, { "BriefDescription": "L1 cache true misses per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.ALL_DEMAND_DATA_RD / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki_load" - }, - { - "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw" - }, - { - "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "tma_info_l2_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw_1t" - }, - { - "BriefDescription": "Rate of non silent evictions from the L2 cache per Kilo instruction", - "MetricExpr": "1e3 * L2_LINES_OUT.NON_SILENT / tma_info_instructions", - "MetricGroup": "L2Evicts;Mem;Server", - "MetricName": "tma_info_l2_evictions_nonsilent_pki" - }, - { - "BriefDescription": "Rate of silent evictions from the L2 cache per Kilo instruction where the evicted lines are dropped (no writeback to L3 or memory)", - "MetricExpr": "1e3 * L2_LINES_OUT.SILENT / tma_info_instructions", - "MetricGroup": "L2Evicts;Mem;Server", - "MetricName": "tma_info_l2_evictions_silent_pki" + "MetricName": "tma_info_memory_l1mpki_load" }, { "BriefDescription": "L2 cache hits per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_HIT / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_load" + "MetricName": "tma_info_memory_l2hpki_load" }, { "BriefDescription": "L2 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L2_MISS / INST_RETIRED.ANY", "MetricGroup": "Backend;CacheMisses;Mem", - "MetricName": "tma_info_l2mpki" + "MetricName": "tma_info_memory_l2mpki" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all request types (including speculative)", - "MetricExpr": "1e3 * (OFFCORE_REQUESTS.ALL_DATA_RD - OFFCORE_REQUESTS.DEMAND_DATA_RD + L2_RQSTS.ALL_DEMAND_MISS + L2_RQSTS.SWPF_MISS) / tma_info_instructions", + "MetricExpr": "1e3 * (OFFCORE_REQUESTS.ALL_DATA_RD - OFFCORE_REQUESTS.DEMAND_DATA_RD + L2_RQSTS.ALL_DEMAND_MISS + L2_RQSTS.SWPF_MISS) / tma_info_inst_mix_instructions", "MetricGroup": "CacheMisses;Mem;Offcore", - "MetricName": "tma_info_l2mpki_all" - }, - { - "BriefDescription": "L2 cache true code cacheline misses per kilo instruction", - "MetricExpr": "1e3 * FRONTEND_RETIRED.L2_MISS / INST_RETIRED.ANY", - "MetricGroup": "IcMiss", - "MetricName": "tma_info_l2mpki_code" - }, - { - "BriefDescription": "L2 cache speculative code cacheline misses per kilo instruction", - "MetricExpr": "1e3 * L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", - "MetricGroup": "IcMiss", - "MetricName": "tma_info_l2mpki_code_all" + "MetricName": "tma_info_memory_l2mpki_all" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2mpki_load" - }, - { - "BriefDescription": "Average per-core data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * OFFCORE_REQUESTS.ALL_REQUESTS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw" + "MetricName": "tma_info_memory_l2mpki_load" }, { - "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_access_bw", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw_1t" + "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L3_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l3mpki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw" + "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", + "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_RETIRED.L1_MISS + MEM_LOAD_RETIRED.FB_HIT)", + "MetricGroup": "Mem;MemoryBound;MemoryLat", + "MetricName": "tma_info_memory_load_miss_real_latency" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw_1t" + "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", + "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", + "MetricGroup": "Mem;MemoryBW;MemoryBound", + "MetricName": "tma_info_memory_mlp", + "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" }, { - "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L3_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l3mpki" + "BriefDescription": "Average Parallel L2 cache miss data reads", + "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", + "MetricGroup": "Memory_BW;Offcore", + "MetricName": "tma_info_memory_oro_data_l2_mlp" }, { "BriefDescription": "Average Latency for L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS.DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l2_miss_latency" + "MetricName": "tma_info_memory_oro_load_l2_miss_latency" }, { "BriefDescription": "Average Parallel L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / cpu@OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD\\,cmask\\=1@", "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_load_l2_mlp" + "MetricName": "tma_info_memory_oro_load_l2_mlp" }, { "BriefDescription": "Average Latency for L3 cache miss demand Loads", "MetricExpr": "cpu@OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD\\,umask\\=0x10@ / OFFCORE_REQUESTS.L3_MISS_DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l3_miss_latency" + "MetricName": "tma_info_memory_oro_load_l3_miss_latency" }, { - "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", - "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_RETIRED.L1_MISS + MEM_LOAD_RETIRED.FB_HIT)", - "MetricGroup": "Mem;MemoryBound;MemoryLat", - "MetricName": "tma_info_load_miss_real_latency" + "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l1d_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l1d_cache_fill_bw_1t" + }, + { + "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l2_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l2_cache_fill_bw_1t" + }, + { + "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_access_bw", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_thread_l3_cache_access_bw_1t" + }, + { + "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l3_cache_fill_bw_1t" + }, + { + "BriefDescription": "STLB (2nd level TLB) code speculative misses per kilo instruction (misses of any page-size that complete the page walk)", + "MetricExpr": "1e3 * ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricGroup": "Fed;MemoryTLB", + "MetricName": "tma_info_memory_tlb_code_stlb_mpki" }, { "BriefDescription": "STLB (2nd level TLB) data load speculative misses per kilo instruction (misses of any page-size that complete the page walk)", "MetricExpr": "1e3 * DTLB_LOAD_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_load_stlb_mpki" + "MetricName": "tma_info_memory_tlb_load_stlb_mpki" + }, + { + "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", + "MetricExpr": "(ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING) / (2 * tma_info_core_core_clks)", + "MetricGroup": "Mem;MemoryTLB", + "MetricName": "tma_info_memory_tlb_page_walks_utilization", + "MetricThreshold": "tma_info_memory_tlb_page_walks_utilization > 0.5" + }, + { + "BriefDescription": "STLB (2nd level TLB) data store speculative misses per kilo instruction (misses of any page-size that complete the page walk)", + "MetricExpr": "1e3 * DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricGroup": "Mem;MemoryTLB", + "MetricName": "tma_info_memory_tlb_store_stlb_mpki" + }, + { + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", + "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", + "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", + "MetricName": "tma_info_pipeline_execute" + }, + { + "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "tma_retiring * tma_info_thread_slots / cpu@UOPS_RETIRED.SLOTS\\,cmask\\=1@", + "MetricGroup": "Pipeline;Ret", + "MetricName": "tma_info_pipeline_retire" + }, + { + "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time", + "MetricGroup": "Power;Summary", + "MetricName": "tma_info_system_average_frequency" + }, + { + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricGroup": "HPC;Summary", + "MetricName": "tma_info_system_cpu_utilization" + }, + { + "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", + "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time", + "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", + "MetricName": "tma_info_system_dram_bw_use", + "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_mem_bandwidth, tma_sq_full" + }, + { + "BriefDescription": "Giga Floating Point Operations Per Second", + "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * cpu@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE\\,umask\\=0x18@ + 8 * cpu@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE\\,umask\\=0x60@ + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / 1e9 / duration_time", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_system_gflops", + "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + }, + { + "BriefDescription": "Average IO (network or disk) Bandwidth Use for Reads [GB / sec]", + "MetricExpr": "(UNC_CHA_TOR_INSERTS.IO_HIT_ITOM + UNC_CHA_TOR_INSERTS.IO_MISS_ITOM + UNC_CHA_TOR_INSERTS.IO_HIT_ITOMCACHENEAR + UNC_CHA_TOR_INSERTS.IO_MISS_ITOMCACHENEAR) * 64 / 1e9 / duration_time", + "MetricGroup": "IoBW;Mem;Server;SoC", + "MetricName": "tma_info_system_io_read_bw" + }, + { + "BriefDescription": "Average IO (network or disk) Bandwidth Use for Writes [GB / sec]", + "MetricExpr": "UNC_CHA_TOR_INSERTS.IO_PCIRDCUR * 64 / 1e9 / duration_time", + "MetricGroup": "IoBW;Mem;Server;SoC", + "MetricName": "tma_info_system_io_write_bw" + }, + { + "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", + "MetricGroup": "Branches;OS", + "MetricName": "tma_info_system_ipfarbranch", + "MetricThreshold": "tma_info_system_ipfarbranch < 1e6" + }, + { + "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_cpi" + }, + { + "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_utilization", + "MetricThreshold": "tma_info_system_kernel_utilization > 0.05" }, { "BriefDescription": "Average latency of data read request to external DRAM memory [in nanoseconds]", "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_DDR / UNC_CHA_TOR_INSERTS.IA_MISS_DRD_DDR) / cha_0@event\\=0x0@", "MetricGroup": "Mem;MemoryLat;Server;SoC", - "MetricName": "tma_info_mem_dram_read_latency", + "MetricName": "tma_info_system_mem_dram_read_latency", "PublicDescription": "Average latency of data read request to external DRAM memory [in nanoseconds]. Accounts for demand loads and L1/L2 data-read prefetches" }, { "BriefDescription": "Average number of parallel data read requests to external memory", "MetricExpr": "UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD / UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD@thresh\\=1@", "MetricGroup": "Mem;MemoryBW;SoC", - "MetricName": "tma_info_mem_parallel_reads", + "MetricName": "tma_info_system_mem_parallel_reads", "PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches" }, { "BriefDescription": "Average latency of data read request to external 3D X-Point memory [in nanoseconds]", "MetricExpr": "(1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_PMM / UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PMM) / cha_0@event\\=0x0@ if #has_pmem > 0 else 0)", "MetricGroup": "Mem;MemoryLat;Server;SoC", - "MetricName": "tma_info_mem_pmm_read_latency", + "MetricName": "tma_info_system_mem_pmm_read_latency", "PublicDescription": "Average latency of data read request to external 3D X-Point memory [in nanoseconds]. Accounts for demand loads and L1/L2 data-read prefetches" }, { "BriefDescription": "Average latency of data read request to external memory (in nanoseconds)", - "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD / UNC_CHA_TOR_INSERTS.IA_MISS_DRD) / (tma_info_socket_clks / duration_time)", + "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD / UNC_CHA_TOR_INSERTS.IA_MISS_DRD) / (tma_info_system_socket_clks / duration_time)", "MetricGroup": "Mem;MemoryLat;SoC", - "MetricName": "tma_info_mem_read_latency", + "MetricName": "tma_info_system_mem_read_latency", "PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)" }, { - "BriefDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks", - "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_mem_bandwidth / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_sq_full / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full))) + tma_l1_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_fb_full / (tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk))", - "MetricGroup": "Mem;MemoryBW;Offcore;tma_issueBW", - "MetricName": "tma_info_memory_bandwidth", - "MetricThreshold": "tma_info_memory_bandwidth > 20", - "PublicDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks. Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_mem_bandwidth, tma_sq_full" - }, - { - "BriefDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs)", - "MetricExpr": "100 * tma_memory_bound * (tma_l1_bound / max(tma_memory_bound, tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_dtlb_load / max(tma_l1_bound, tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk)) + tma_store_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_dtlb_store / (tma_dtlb_store + tma_false_sharing + tma_split_stores + tma_store_latency + tma_streaming_stores)))", - "MetricGroup": "Mem;MemoryTLB;Offcore;tma_issueTLB", - "MetricName": "tma_info_memory_data_tlbs", - "MetricThreshold": "tma_info_memory_data_tlbs > 20", - "PublicDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs). Related metrics: tma_dtlb_load, tma_dtlb_store" - }, - { - "BriefDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches)", - "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_mem_latency / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_l3_hit_latency / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full)) + tma_l2_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound))", - "MetricGroup": "Mem;MemoryLat;Offcore;tma_issueLat", - "MetricName": "tma_info_memory_latency", - "MetricThreshold": "tma_info_memory_latency > 20", - "PublicDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches). Related metrics: tma_l3_hit_latency, tma_mem_latency" - }, - { - "BriefDescription": "Total pipeline cost of Branch Misprediction related bottlenecks", - "MetricExpr": "100 * (tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", - "MetricGroup": "Bad;BadSpec;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_mispredictions", - "MetricThreshold": "tma_info_mispredictions > 20", - "PublicDescription": "Total pipeline cost of Branch Misprediction related bottlenecks. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost, tma_mispredicts_resteers" - }, - { - "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", - "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", - "MetricGroup": "Mem;MemoryBW;MemoryBound", - "MetricName": "tma_info_mlp", - "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" - }, - { - "BriefDescription": "Fraction of branches of other types (not individually covered by other metrics in Info.Branches group)", - "MetricExpr": "1 - (tma_info_cond_nt + tma_info_cond_tk + tma_info_callret + tma_info_jump)", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_other_branches" - }, - { - "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", - "MetricExpr": "(ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING) / (2 * tma_info_core_clks)", - "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_page_walks_utilization", - "MetricThreshold": "tma_info_page_walks_utilization > 0.5" - }, - { "BriefDescription": "Average 3DXP Memory Bandwidth Use for reads [GB / sec]", "MetricExpr": "(64 * UNC_M_PMM_RPQ_INSERTS / 1e9 / duration_time if #has_pmem > 0 else 0)", "MetricGroup": "Mem;MemoryBW;Server;SoC", - "MetricName": "tma_info_pmm_read_bw" + "MetricName": "tma_info_system_pmm_read_bw" }, { "BriefDescription": "Average 3DXP Memory Bandwidth Use for Writes [GB / sec]", "MetricExpr": "(64 * UNC_M_PMM_WPQ_INSERTS / 1e9 / duration_time if #has_pmem > 0 else 0)", "MetricGroup": "Mem;MemoryBW;Server;SoC", - "MetricName": "tma_info_pmm_write_bw" + "MetricName": "tma_info_system_pmm_write_bw" }, { "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for baseline license level 0", - "MetricExpr": "CORE_POWER.LVL0_TURBO_LICENSE / tma_info_core_clks", + "MetricExpr": "CORE_POWER.LVL0_TURBO_LICENSE / tma_info_core_core_clks", "MetricGroup": "Power", - "MetricName": "tma_info_power_license0_utilization", + "MetricName": "tma_info_system_power_license0_utilization", "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for baseline license level 0. This includes non-AVX codes, SSE, AVX 128-bit, and low-current AVX 256-bit codes." }, { "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 1", - "MetricExpr": "CORE_POWER.LVL1_TURBO_LICENSE / tma_info_core_clks", + "MetricExpr": "CORE_POWER.LVL1_TURBO_LICENSE / tma_info_core_core_clks", "MetricGroup": "Power", - "MetricName": "tma_info_power_license1_utilization", - "MetricThreshold": "tma_info_power_license1_utilization > 0.5", + "MetricName": "tma_info_system_power_license1_utilization", + "MetricThreshold": "tma_info_system_power_license1_utilization > 0.5", "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 1. This includes high current AVX 256-bit instructions as well as low current AVX 512-bit instructions." }, { "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 2 (introduced in SKX)", - "MetricExpr": "CORE_POWER.LVL2_TURBO_LICENSE / tma_info_core_clks", + "MetricExpr": "CORE_POWER.LVL2_TURBO_LICENSE / tma_info_core_core_clks", "MetricGroup": "Power", - "MetricName": "tma_info_power_license2_utilization", - "MetricThreshold": "tma_info_power_license2_utilization > 0.5", + "MetricName": "tma_info_system_power_license2_utilization", + "MetricThreshold": "tma_info_system_power_license2_utilization > 0.5", "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 2 (introduced in SKX). This includes high current AVX 512-bit instructions." }, { - "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", - "MetricExpr": "tma_retiring * tma_info_slots / cpu@UOPS_RETIRED.SLOTS\\,cmask\\=1@", - "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_retire" - }, - { - "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", - "MetricExpr": "TOPDOWN.SLOTS", - "MetricGroup": "TmaL1;tma_L1_group", - "MetricName": "tma_info_slots" - }, - { - "BriefDescription": "Fraction of Physical Core issue-slots utilized by this Logical Processor", - "MetricExpr": "(tma_info_slots / (TOPDOWN.SLOTS / 2) if #SMT_on else 1)", - "MetricGroup": "SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_slots_utilization" - }, - { "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_DISTRIBUTED if #SMT_on else 0)", "MetricGroup": "SMT", - "MetricName": "tma_info_smt_2t_utilization" + "MetricName": "tma_info_system_smt_2t_utilization" }, { "BriefDescription": "Socket actual clocks when any core is active on that socket", "MetricExpr": "cha_0@event\\=0x0@", "MetricGroup": "SoC", - "MetricName": "tma_info_socket_clks" - }, - { - "BriefDescription": "STLB (2nd level TLB) data store speculative misses per kilo instruction (misses of any page-size that complete the page walk)", - "MetricExpr": "1e3 * DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", - "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_store_stlb_mpki" + "MetricName": "tma_info_system_socket_clks" }, { "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC", "MetricGroup": "Power", - "MetricName": "tma_info_turbo_utilization" + "MetricName": "tma_info_system_turbo_utilization" + }, + { + "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "Pipeline", + "MetricName": "tma_info_thread_clks" + }, + { + "BriefDescription": "Cycles Per Instruction (per Logical Processor)", + "MetricExpr": "1 / tma_info_thread_ipc", + "MetricGroup": "Mem;Pipeline", + "MetricName": "tma_info_thread_cpi" + }, + { + "BriefDescription": "The ratio of Executed- by Issued-Uops", + "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", + "MetricGroup": "Cor;Pipeline", + "MetricName": "tma_info_thread_execute_per_issue", + "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." + }, + { + "BriefDescription": "Instructions Per Cycle (per Logical Processor)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks", + "MetricGroup": "Ret;Summary", + "MetricName": "tma_info_thread_ipc" + }, + { + "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", + "MetricExpr": "TOPDOWN.SLOTS", + "MetricGroup": "TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots" + }, + { + "BriefDescription": "Fraction of Physical Core issue-slots utilized by this Logical Processor", + "MetricExpr": "(tma_info_thread_slots / (TOPDOWN.SLOTS / 2) if #SMT_on else 1)", + "MetricGroup": "SMT;TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots_utilization" }, { "BriefDescription": "Uops Per Instruction", - "MetricExpr": "tma_retiring * tma_info_slots / INST_RETIRED.ANY", + "MetricExpr": "tma_retiring * tma_info_thread_slots / INST_RETIRED.ANY", "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_uoppi", - "MetricThreshold": "tma_info_uoppi > 1.05" + "MetricName": "tma_info_thread_uoppi", + "MetricThreshold": "tma_info_thread_uoppi > 1.05" }, { "BriefDescription": "Instruction per taken branch", - "MetricExpr": "tma_retiring * tma_info_slots / BR_INST_RETIRED.NEAR_TAKEN", + "MetricExpr": "tma_retiring * tma_info_thread_slots / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW", - "MetricName": "tma_info_uptb", - "MetricThreshold": "tma_info_uptb < 7.5" + "MetricName": "tma_info_thread_uptb", + "MetricThreshold": "tma_info_thread_uptb < 7.5" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses", - "MetricExpr": "ICACHE_64B.IFTAG_STALL / tma_info_clks", + "MetricExpr": "ICACHE_64B.IFTAG_STALL / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -1084,7 +1350,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled without loads missing the L1 data cache", - "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_clks, 0)", + "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_thread_clks, 0)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_issueL1;tma_issueMC;tma_memory_bound_group", "MetricName": "tma_l1_bound", "MetricThreshold": "tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1094,7 +1360,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled due to L2 cache accesses by loads", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) / (MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + L1D_PEND_MISS.FB_FULL_PERIODS) * ((CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks)", + "MetricExpr": "MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) / (MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + L1D_PEND_MISS.FB_FULL_PERIODS) * ((CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l2_bound", "MetricThreshold": "tma_l2_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1103,7 +1369,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L2_MISS - CYCLE_ACTIVITY.STALLS_L3_MISS) / tma_info_clks", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L2_MISS - CYCLE_ACTIVITY.STALLS_L3_MISS) / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1112,20 +1378,20 @@ }, { "BriefDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited)", - "MetricExpr": "19 * tma_info_average_frequency * MEM_LOAD_RETIRED.L3_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "19 * tma_info_system_average_frequency * MEM_LOAD_RETIRED.L3_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "MemoryLat;TopdownL4;tma_L4_group;tma_issueLat;tma_l3_bound_group", "MetricName": "tma_l3_hit_latency", "MetricThreshold": "tma_l3_hit_latency > 0.1 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited). Avoiding private cache misses (i.e. L2 misses/L3 hits) will improve the latency; reduce contention with sibling physical cores and increase performance. Note the value of this node may overlap with its siblings. Sample with: MEM_LOAD_RETIRED.L3_HIT_PS. Related metrics: tma_info_memory_latency, tma_mem_latency", + "PublicDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited). Avoiding private cache misses (i.e. L2 misses/L3 hits) will improve the latency; reduce contention with sibling physical cores and increase performance. Note the value of this node may overlap with its siblings. Sample with: MEM_LOAD_RETIRED.L3_HIT_PS. Related metrics: tma_info_bottleneck_memory_latency, tma_mem_latency", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)", - "MetricExpr": "ILD_STALL.LCP / tma_info_clks", + "MetricExpr": "ILD_STALL.LCP / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_lcp", "MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb", + "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb", "ScaleUnit": "100%" }, { @@ -1134,12 +1400,13 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_light_operations", "MetricThreshold": "tma_light_operations > 0.6", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation). This correlates with total number of instructions used by the program. A uops-per-instruction (see UopPI metric) ratio of 1 or less should be expected for decently optimized software running on Intel Core/Xeon products. While this often indicates efficient X86 instructions were executed; high value does not necessarily mean better performance cannot be achieved. Sample with: INST_RETIRED.PREC_DIST", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Load operations", - "MetricExpr": "UOPS_DISPATCHED.PORT_2_3 / (2 * tma_info_core_clks)", + "MetricExpr": "UOPS_DISPATCHED.PORT_2_3 / (2 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_load_op_utilization", "MetricThreshold": "tma_load_op_utilization > 0.6", @@ -1156,7 +1423,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles where the Second-level TLB (STLB) was missed by load accesses, performing a hardware page walk", - "MetricExpr": "DTLB_LOAD_MISSES.WALK_ACTIVE / tma_info_clks", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_ACTIVE / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL5;tma_L5_group;tma_dtlb_load_group", "MetricName": "tma_load_stlb_miss", "MetricThreshold": "tma_load_stlb_miss > 0.05 & (tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1164,7 +1431,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from local memory", - "MetricExpr": "43.5 * tma_info_average_frequency * MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "43.5 * tma_info_system_average_frequency * MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Server;TopdownL5;tma_L5_group;tma_mem_latency_group", "MetricName": "tma_local_dram", "MetricThreshold": "tma_local_dram > 0.1 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1174,7 +1441,7 @@ { "BriefDescription": "This metric represents fraction of cycles the CPU spent handling cache misses due to lock operations", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(16 * max(0, MEM_INST_RETIRED.LOCK_LOADS - L2_RQSTS.ALL_RFO) + MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES * (10 * L2_RQSTS.RFO_HIT + min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO))) / tma_info_clks", + "MetricExpr": "(16 * max(0, MEM_INST_RETIRED.LOCK_LOADS - L2_RQSTS.ALL_RFO) + MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES * (10 * L2_RQSTS.RFO_HIT + min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO))) / tma_info_thread_clks", "MetricGroup": "Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_l1_bound_group", "MetricName": "tma_lock_latency", "MetricThreshold": "tma_lock_latency > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1187,25 +1454,26 @@ "MetricGroup": "BadSpec;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.1 & tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears. These slots are either wasted by uops fetched prior to the clear; or stalls the out-of-order portion of the machine needs to recover its state after the clear. For example; this can happen due to memory ordering Nukes (e.g. Memory Disambiguation) or Self-Modifying-Code (SMC) nukes. Sample with: MACHINE_CLEARS.COUNT. Related metrics: tma_clears_resteers, tma_contested_accesses, tma_data_sharing, tma_false_sharing, tma_l1_bound, tma_microcode_sequencer, tma_ms_switches, tma_remote_cache", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_clks", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW", "MetricName": "tma_mem_bandwidth", "MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_sq_full", + "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_sq_full", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat", "MetricName": "tma_mem_latency", "MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM). This metric does not aggregate requests from other Logical Processors/Physical Cores/sockets (see Uncore counters for that). Related metrics: tma_info_memory_latency, tma_l3_hit_latency", + "PublicDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM). This metric does not aggregate requests from other Logical Processors/Physical Cores/sockets (see Uncore counters for that). Related metrics: tma_info_bottleneck_memory_latency, tma_l3_hit_latency", "ScaleUnit": "100%" }, { @@ -1214,11 +1482,13 @@ "MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck. Memory Bound estimates fraction of slots where pipeline is likely stalled due to demand load or store instructions. This accounts mainly for (1) non-completed in-flight memory demand loads which coincides with execution units starvation; in addition to (2) cases where stores could impose backpressure on the pipeline when many of them get buffered at the same time (less common out of the two).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring memory operations -- uops for memory load or store accesses.", + "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "tma_light_operations * MEM_INST_RETIRED.ANY / INST_RETIRED.ANY", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_memory_operations", @@ -1227,7 +1497,7 @@ }, { "BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit", - "MetricExpr": "tma_retiring * tma_info_slots / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_slots", + "MetricExpr": "tma_retiring * tma_info_thread_slots / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_thread_slots", "MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS", "MetricName": "tma_microcode_sequencer", "MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1", @@ -1236,28 +1506,28 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage", - "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks", + "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks", "MetricGroup": "BadSpec;BrMispredicts;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueBM", "MetricName": "tma_mispredicts_resteers", "MetricThreshold": "tma_mispredicts_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost, tma_info_mispredictions", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost, tma_info_bottleneck_mispredictions", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline)", - "MetricExpr": "(IDQ.MITE_CYCLES_ANY - IDQ.MITE_CYCLES_OK) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.MITE_CYCLES_ANY - IDQ.MITE_CYCLES_OK) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_mite", - "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 5 > 0.35)", + "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 5 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline). This pipeline is used for code that was not pre-cached in the DSB or LSD. For example; inefficiencies due to asymmetric decoders; use of long immediate or LCP can manifest as MITE fetch bandwidth bottleneck. Sample with: FRONTEND_RETIRED.ANY_DSB_MISS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles where (only) 4 uops were delivered by the MITE pipeline", - "MetricExpr": "(cpu@IDQ.MITE_UOPS\\,cmask\\=4@ - cpu@IDQ.MITE_UOPS\\,cmask\\=5@) / tma_info_clks", + "MetricExpr": "(cpu@IDQ.MITE_UOPS\\,cmask\\=4@ - cpu@IDQ.MITE_UOPS\\,cmask\\=5@) / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchBW;TopdownL4;tma_L4_group;tma_mite_group", "MetricName": "tma_mite_4wide", - "MetricThreshold": "tma_mite_4wide > 0.05 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 5 > 0.35))", + "MetricThreshold": "tma_mite_4wide > 0.05 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 5 > 0.35))", "ScaleUnit": "100%" }, { @@ -1271,7 +1541,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)", - "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_clks", + "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_thread_clks", "MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO", "MetricName": "tma_ms_switches", "MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -1280,7 +1550,7 @@ }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring NOP (no op) instructions", - "MetricExpr": "tma_light_operations * INST_RETIRED.NOP / (tma_retiring * tma_info_slots)", + "MetricExpr": "tma_light_operations * INST_RETIRED.NOP / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_nop_instructions", "MetricThreshold": "tma_nop_instructions > 0.1 & tma_light_operations > 0.6", @@ -1289,6 +1559,7 @@ }, { "BriefDescription": "This metric represents the remaining light uops fraction the CPU has executed - remaining means not covered by other sibling nodes", + "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "max(0, tma_light_operations - (tma_fp_arith + tma_memory_operations + tma_branch_instructions + tma_nop_instructions))", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_other_light_ops", @@ -1298,7 +1569,7 @@ }, { "BriefDescription": "This metric roughly estimates (based on idle latencies) how often the CPU was stalled on accesses to external 3D-Xpoint (Crystal Ridge, a.k.a", - "MetricExpr": "(((1 - ((19 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + 10 * (MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS))) / (19 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + 10 * (MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + (25 * (MEM_LOAD_RETIRED.LOCAL_PMM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) if #has_pmem > 0 else 0) + 33 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_PMM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) if #has_pmem > 0 else 0))) if #has_pmem > 0 else 0)) * (CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks - tma_l2_bound) if 1e6 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_PMM + MEM_LOAD_RETIRED.LOCAL_PMM) > MEM_LOAD_RETIRED.L1_MISS else 0) if #has_pmem > 0 else 0)", + "MetricExpr": "(((1 - ((19 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + 10 * (MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS))) / (19 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + 10 * (MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + (25 * (MEM_LOAD_RETIRED.LOCAL_PMM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) if #has_pmem > 0 else 0) + 33 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_PMM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) if #has_pmem > 0 else 0))) if #has_pmem > 0 else 0)) * (CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_thread_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks - tma_l2_bound) if 1e6 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_PMM + MEM_LOAD_RETIRED.LOCAL_PMM) > MEM_LOAD_RETIRED.L1_MISS else 0) if #has_pmem > 0 else 0)", "MetricGroup": "MemoryBound;Server;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_pmm_bound", "MetricThreshold": "tma_pmm_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1307,7 +1578,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 0 ([SNB+] ALU; [HSW+] ALU and 2nd branch)", - "MetricExpr": "UOPS_DISPATCHED.PORT_0 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_0 / tma_info_core_core_clks", "MetricGroup": "Compute;TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_0", "MetricThreshold": "tma_port_0 > 0.6", @@ -1316,7 +1587,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 1 (ALU)", - "MetricExpr": "UOPS_DISPATCHED.PORT_1 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_1 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_1", "MetricThreshold": "tma_port_1 > 0.6", @@ -1325,7 +1596,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 5 ([SNB+] Branches and ALU; [HSW+] ALU)", - "MetricExpr": "UOPS_DISPATCHED.PORT_5 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_5 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_5", "MetricThreshold": "tma_port_5 > 0.6", @@ -1334,7 +1605,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 6 ([HSW+]Primary Branch and simple ALU)", - "MetricExpr": "UOPS_DISPATCHED.PORT_6 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_6 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_6", "MetricThreshold": "tma_port_6 > 0.6", @@ -1343,7 +1614,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)", - "MetricExpr": "((cpu@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) + (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL)) / tma_info_clks if ARITH.DIVIDER_ACTIVE < CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY else (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_clks)", + "MetricExpr": "((cpu@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) + (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL)) / tma_info_thread_clks if ARITH.DIVIDER_ACTIVE < CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY else (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_thread_clks)", "MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_ports_utilization", "MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -1352,7 +1623,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed no uops on any execution port (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "cpu@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ / tma_info_clks + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_clks", + "MetricExpr": "cpu@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ / tma_info_thread_clks + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_0", "MetricThreshold": "tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1361,7 +1632,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the CPU executed total of 1 uop per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "EXE_ACTIVITY.1_PORTS_UTIL / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.1_PORTS_UTIL / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issueL1;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_1", "MetricThreshold": "tma_ports_utilized_1 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1370,7 +1641,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 2 uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "EXE_ACTIVITY.2_PORTS_UTIL / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.2_PORTS_UTIL / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issue2P;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_2", "MetricThreshold": "tma_ports_utilized_2 > 0.15 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1379,7 +1650,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 3 or more uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "UOPS_EXECUTED.CYCLES_GE_3 / tma_info_clks", + "MetricExpr": "UOPS_EXECUTED.CYCLES_GE_3 / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_3m", "MetricThreshold": "tma_ports_utilized_3m > 0.7 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1388,7 +1659,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from remote cache in other sockets including synchronizations issues", - "MetricExpr": "(97 * tma_info_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM + 97 * tma_info_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "(97 * tma_info_system_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM + 97 * tma_info_system_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Offcore;Server;Snoop;TopdownL5;tma_L5_group;tma_issueSyncxn;tma_mem_latency_group", "MetricName": "tma_remote_cache", "MetricThreshold": "tma_remote_cache > 0.05 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1397,7 +1668,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from remote memory", - "MetricExpr": "108 * tma_info_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "108 * tma_info_system_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Server;Snoop;TopdownL5;tma_L5_group;tma_mem_latency_group", "MetricName": "tma_remote_dram", "MetricThreshold": "tma_remote_dram > 0.1 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1406,16 +1677,18 @@ }, { "BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired", - "MetricExpr": "topdown\\-retiring / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_slots", - "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "topdown\\-retiring / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_thread_slots", + "MetricGroup": "Default;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired. Ideally; all pipeline slots would be attributed to the Retiring category. Retiring of 100% would indicate the maximum Pipeline_Width throughput was achieved. Maximizing Retiring typically increases the Instructions-per-cycle (see IPC metric). Note that a high Retiring value does not necessary mean there is no room for more performance. For example; Heavy-operations or Microcode Assists are categorized under Retiring. They often indicate suboptimal performance and can often be optimized or avoided. Sample with: UOPS_RETIRED.SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU issue-pipeline was stalled due to serializing operations", - "MetricExpr": "RESOURCE_STALLS.SCOREBOARD / tma_info_clks", + "MetricExpr": "RESOURCE_STALLS.SCOREBOARD / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL5;tma_L5_group;tma_issueSO;tma_ports_utilized_0_group", "MetricName": "tma_serializing_operation", "MetricThreshold": "tma_serializing_operation > 0.1 & (tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)))", @@ -1424,7 +1697,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to PAUSE Instructions", - "MetricExpr": "37 * MISC_RETIRED.PAUSE_INST / tma_info_clks", + "MetricExpr": "37 * MISC_RETIRED.PAUSE_INST / tma_info_thread_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_serializing_operation_group", "MetricName": "tma_slow_pause", "MetricThreshold": "tma_slow_pause > 0.05 & (tma_serializing_operation > 0.1 & (tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))))", @@ -1433,7 +1706,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles handling memory load split accesses - load that cross 64-byte cache line boundary", - "MetricExpr": "tma_info_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_split_loads", "MetricThreshold": "tma_split_loads > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1442,7 +1715,7 @@ }, { "BriefDescription": "This metric represents rate of split store accesses", - "MetricExpr": "MEM_INST_RETIRED.SPLIT_STORES / tma_info_core_clks", + "MetricExpr": "MEM_INST_RETIRED.SPLIT_STORES / tma_info_core_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_issueSpSt;tma_store_bound_group", "MetricName": "tma_split_stores", "MetricThreshold": "tma_split_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1451,16 +1724,16 @@ }, { "BriefDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors)", - "MetricExpr": "L1D_PEND_MISS.L2_STALL / tma_info_clks", + "MetricExpr": "L1D_PEND_MISS.L2_STALL / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueBW;tma_l3_bound_group", "MetricName": "tma_sq_full", "MetricThreshold": "tma_sq_full > 0.3 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_mem_bandwidth", + "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_mem_bandwidth", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write", - "MetricExpr": "EXE_ACTIVITY.BOUND_ON_STORES / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.BOUND_ON_STORES / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1469,7 +1742,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of cycles when the memory subsystem had loads blocked since they could not forward data from earlier (in program order) overlapping stores", - "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_clks", + "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_store_fwd_blk", "MetricThreshold": "tma_store_fwd_blk > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1478,7 +1751,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles the CPU spent handling L1D store misses", - "MetricExpr": "(L2_RQSTS.RFO_HIT * 10 * (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) + (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_clks", + "MetricExpr": "(L2_RQSTS.RFO_HIT * 10 * (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) + (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_thread_clks", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_issueSL;tma_store_bound_group", "MetricName": "tma_store_latency", "MetricThreshold": "tma_store_latency > 0.1 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1487,7 +1760,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Store operations", - "MetricExpr": "(UOPS_DISPATCHED.PORT_4_9 + UOPS_DISPATCHED.PORT_7_8) / (4 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED.PORT_4_9 + UOPS_DISPATCHED.PORT_7_8) / (4 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_store_op_utilization", "MetricThreshold": "tma_store_op_utilization > 0.6", @@ -1504,7 +1777,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles where the STLB was missed by store accesses, performing a hardware page walk", - "MetricExpr": "DTLB_STORE_MISSES.WALK_ACTIVE / tma_info_core_clks", + "MetricExpr": "DTLB_STORE_MISSES.WALK_ACTIVE / tma_info_core_core_clks", "MetricGroup": "MemoryTLB;TopdownL5;tma_L5_group;tma_dtlb_store_group", "MetricName": "tma_store_stlb_miss", "MetricThreshold": "tma_store_stlb_miss > 0.05 & (tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1512,7 +1785,7 @@ }, { "BriefDescription": "This metric estimates how often CPU was stalled due to Streaming store memory accesses; Streaming store optimize out a read request required by RFO stores", - "MetricExpr": "9 * OCR.STREAMING_WR.ANY_RESPONSE / tma_info_clks", + "MetricExpr": "9 * OCR.STREAMING_WR.ANY_RESPONSE / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueSmSt;tma_store_bound_group", "MetricName": "tma_streaming_stores", "MetricThreshold": "tma_streaming_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1521,7 +1794,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to new branch address clears", - "MetricExpr": "10 * BACLEARS.ANY / tma_info_clks", + "MetricExpr": "10 * BACLEARS.ANY / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;TopdownL4;tma_L4_group;tma_branch_resteers_group", "MetricName": "tma_unknown_branches", "MetricThreshold": "tma_unknown_branches > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", @@ -1564,5 +1837,17 @@ "MetricGroup": "transaction", "MetricName": "tsx_transactional_cycles", "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uncore operating frequency in GHz", + "MetricExpr": "UNC_CHA_CLOCKTICKS / (source_count(UNC_CHA_CLOCKTICKS) * #num_packages) / 1e9 / duration_time", + "MetricName": "uncore_frequency", + "ScaleUnit": "1GHz" + }, + { + "BriefDescription": "Intel(R) Ultra Path Interconnect (UPI) data transmit bandwidth (MB/sec)", + "MetricExpr": "UNC_UPI_TxL_FLITS.ALL_DATA * 7.111111111111111 / 1e6 / duration_time", + "MetricName": "upi_data_transmit_bw", + "ScaleUnit": "1MB/s" } ] diff --git a/tools/perf/pmu-events/arch/x86/icelakex/metricgroups.json b/tools/perf/pmu-events/arch/x86/icelakex/metricgroups.json new file mode 100644 index 000000000000..bc6a9a4d27a9 --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/icelakex/metricgroups.json @@ -0,0 +1,114 @@ +{ + "Backend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Bad": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BadSpec": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BigFoot": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BrMispredicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Branches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CacheMisses": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CodeGen": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Compute": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Cor": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSBmiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DataSharing": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Fed": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Flops": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpScalar": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Frontend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "HPC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IcMiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "InsType": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IoBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "L2Evicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "LSD": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MachineClears": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBound": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryTLB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_BW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_Lat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MicroSeq": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "OS": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Offcore": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PGO": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Pipeline": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PortsUtil": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Prefetches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Ret": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Retire": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SMT": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Server": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Snoop": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SoC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL1": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL2": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL3mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "TopdownL5": "Metrics for top-down breakdown at level 5", + "TopdownL6": "Metrics for top-down breakdown at level 6", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_L5_group": "Metrics for top-down breakdown at level 5", + "tma_L6_group": "Metrics for top-down breakdown at level 6", + "tma_alu_op_utilization_group": "Metrics contributing to tma_alu_op_utilization category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_branch_resteers_group": "Metrics contributing to tma_branch_resteers category", + "tma_core_bound_group": "Metrics contributing to tma_core_bound category", + "tma_dram_bound_group": "Metrics contributing to tma_dram_bound category", + "tma_dtlb_load_group": "Metrics contributing to tma_dtlb_load category", + "tma_dtlb_store_group": "Metrics contributing to tma_dtlb_store category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_fp_arith_group": "Metrics contributing to tma_fp_arith category", + "tma_fp_vector_group": "Metrics contributing to tma_fp_vector category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_heavy_operations_group": "Metrics contributing to tma_heavy_operations category", + "tma_issue2P": "Metrics related by the issue $issue2P", + "tma_issueBC": "Metrics related by the issue $issueBC", + "tma_issueBM": "Metrics related by the issue $issueBM", + "tma_issueBW": "Metrics related by the issue $issueBW", + "tma_issueD0": "Metrics related by the issue $issueD0", + "tma_issueFB": "Metrics related by the issue $issueFB", + "tma_issueFL": "Metrics related by the issue $issueFL", + "tma_issueL1": "Metrics related by the issue $issueL1", + "tma_issueLat": "Metrics related by the issue $issueLat", + "tma_issueMC": "Metrics related by the issue $issueMC", + "tma_issueMS": "Metrics related by the issue $issueMS", + "tma_issueMV": "Metrics related by the issue $issueMV", + "tma_issueRFO": "Metrics related by the issue $issueRFO", + "tma_issueSL": "Metrics related by the issue $issueSL", + "tma_issueSO": "Metrics related by the issue $issueSO", + "tma_issueSmSt": "Metrics related by the issue $issueSmSt", + "tma_issueSpSt": "Metrics related by the issue $issueSpSt", + "tma_issueSyncxn": "Metrics related by the issue $issueSyncxn", + "tma_issueTLB": "Metrics related by the issue $issueTLB", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_l3_bound_group": "Metrics contributing to tma_l3_bound category", + "tma_light_operations_group": "Metrics contributing to tma_light_operations category", + "tma_load_op_utilization_group": "Metrics contributing to tma_load_op_utilization category", + "tma_mem_latency_group": "Metrics contributing to tma_mem_latency category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_microcode_sequencer_group": "Metrics contributing to tma_microcode_sequencer category", + "tma_mite_group": "Metrics contributing to tma_mite category", + "tma_ports_utilization_group": "Metrics contributing to tma_ports_utilization category", + "tma_ports_utilized_0_group": "Metrics contributing to tma_ports_utilized_0 category", + "tma_ports_utilized_3m_group": "Metrics contributing to tma_ports_utilized_3m category", + "tma_retiring_group": "Metrics contributing to tma_retiring category", + "tma_serializing_operation_group": "Metrics contributing to tma_serializing_operation category", + "tma_store_bound_group": "Metrics contributing to tma_store_bound category", + "tma_store_op_utilization_group": "Metrics contributing to tma_store_op_utilization category" +} diff --git a/tools/perf/pmu-events/arch/x86/ivybridge/ivb-metrics.json b/tools/perf/pmu-events/arch/x86/ivybridge/ivb-metrics.json index 5247f69c13b6..33fe555252b2 100644 --- a/tools/perf/pmu-events/arch/x86/ivybridge/ivb-metrics.json +++ b/tools/perf/pmu-events/arch/x86/ivybridge/ivb-metrics.json @@ -50,7 +50,7 @@ }, { "BriefDescription": "Uncore frequency per die [GHZ]", - "MetricExpr": "tma_info_socket_clks / #num_dies / duration_time / 1e9", + "MetricExpr": "tma_info_system_socket_clks / #num_dies / duration_time / 1e9", "MetricGroup": "SoC", "MetricName": "UNCORE_FREQ" }, @@ -71,7 +71,7 @@ }, { "BriefDescription": "This metric estimates how often memory load accesses were aliased by preceding stores (in program order) with a 4K address offset", - "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_clks", + "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_4k_aliasing", "MetricThreshold": "tma_4k_aliasing > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -81,7 +81,7 @@ { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution ports for ALU operations.", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5) / (3 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5) / (3 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_alu_op_utilization", "MetricThreshold": "tma_alu_op_utilization > 0.6", @@ -89,7 +89,7 @@ }, { "BriefDescription": "This metric estimates fraction of slots the CPU retired uops delivered by the Microcode_Sequencer as a result of Assists", - "MetricExpr": "100 * OTHER_ASSISTS.ANY_WB_ASSIST / tma_info_slots", + "MetricExpr": "100 * OTHER_ASSISTS.ANY_WB_ASSIST / tma_info_thread_slots", "MetricGroup": "TopdownL4;tma_L4_group;tma_microcode_sequencer_group", "MetricName": "tma_assists", "MetricThreshold": "tma_assists > 0.1 & (tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1)", @@ -103,15 +103,17 @@ "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend. Backend is the portion of the processor core where the out-of-order scheduler dispatches ready uops into their respective execution units; and once completed these uops get retired according to program order. For example; stalls due to data-cache misses or stalls due to the divider unit being overloaded are both categorized under Backend Bound. Backend Bound is further divided into two main categories: Memory Bound and Core Bound.", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations", - "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_slots", + "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots wasted due to incorrect speculations. This include slots used to issue uops that do not eventually get retired and slots for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work due to miss-predicted branches are categorized under Bad Speculation category. Incorrect data speculation followed by Memory Ordering Nukes is another example.", "ScaleUnit": "100%" }, @@ -122,12 +124,13 @@ "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15", - "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_branch_misprediction_cost, tma_mispredicts_resteers", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers", - "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_clks", + "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteers", "MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -147,7 +150,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to contested accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(60 * (MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.LLC_MISS))) + 43 * (MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.LLC_MISS)))) / tma_info_clks", + "MetricExpr": "(60 * (MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.LLC_MISS))) + 43 * (MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.LLC_MISS)))) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_contested_accesses", "MetricThreshold": "tma_contested_accesses > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -161,13 +164,14 @@ "MetricGroup": "Backend;Compute;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck. Shortage in hardware compute resources; or dependencies in software's instructions are both categorized under Core Bound. Hence it may indicate the machine ran out of an out-of-order resource; certain execution units are overloaded or dependencies in program's data- or instruction-flow are limiting the performance (e.g. FP-chained long-latency arithmetic operations).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to data-sharing accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "43 * (MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.LLC_MISS))) / tma_info_clks", + "MetricExpr": "43 * (MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.LLC_MISS))) / tma_info_thread_clks", "MetricGroup": "Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_data_sharing", "MetricThreshold": "tma_data_sharing > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -176,7 +180,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the Divider unit was active", - "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_clks", + "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_core_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_divider", "MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -186,7 +190,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_RETIRED.LLC_MISS)) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_clks", + "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_RETIRED.LLC_MISS)) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -195,25 +199,25 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline", - "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSB;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_dsb", - "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline. For example; inefficient utilization of the DSB cache structure or bank conflict when reading from it; are categorized here.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_dsb_switches", "MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb, tma_lcp", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses", - "MetricExpr": "(7 * DTLB_LOAD_MISSES.STLB_HIT + DTLB_LOAD_MISSES.WALK_DURATION) / tma_info_clks", + "MetricExpr": "(7 * DTLB_LOAD_MISSES.STLB_HIT + DTLB_LOAD_MISSES.WALK_DURATION) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group", "MetricName": "tma_dtlb_load", "MetricThreshold": "tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -222,7 +226,7 @@ }, { "BriefDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses", - "MetricExpr": "(7 * DTLB_STORE_MISSES.STLB_HIT + DTLB_STORE_MISSES.WALK_DURATION) / tma_info_clks", + "MetricExpr": "(7 * DTLB_STORE_MISSES.STLB_HIT + DTLB_STORE_MISSES.WALK_DURATION) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_store_bound_group", "MetricName": "tma_dtlb_store", "MetricThreshold": "tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -231,7 +235,7 @@ }, { "BriefDescription": "This metric roughly estimates how often CPU was handling synchronizations due to False Sharing", - "MetricExpr": "60 * OFFCORE_RESPONSE.DEMAND_RFO.LLC_HIT.HITM_OTHER_CORE / tma_info_clks", + "MetricExpr": "60 * OFFCORE_RESPONSE.DEMAND_RFO.LLC_HIT.HITM_OTHER_CORE / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_store_bound_group", "MetricName": "tma_false_sharing", "MetricThreshold": "tma_false_sharing > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -241,11 +245,11 @@ { "BriefDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "tma_info_load_miss_real_latency * cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@ / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@ / tma_info_thread_clks", "MetricGroup": "MemoryBW;TopdownL4;tma_L4_group;tma_issueBW;tma_issueSL;tma_issueSmSt;tma_l1_bound_group", "MetricName": "tma_fb_full", "MetricThreshold": "tma_fb_full > 0.3", - "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", + "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", "ScaleUnit": "100%" }, { @@ -253,16 +257,18 @@ "MetricExpr": "tma_frontend_bound - tma_fetch_latency", "MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", "MetricName": "tma_fetch_bandwidth", - "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_info_iptb, tma_lcp", + "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues", - "MetricExpr": "4 * min(CPU_CLK_UNHALTED.THREAD, IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE) / tma_info_slots", + "MetricExpr": "4 * min(CPU_CLK_UNHALTED.THREAD, IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE) / tma_info_thread_slots", "MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: RS_EVENTS.EMPTY_END", "ScaleUnit": "100%" }, @@ -295,10 +301,11 @@ }, { "BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend", - "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_slots", + "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_thread_slots", "MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound.", "ScaleUnit": "100%" }, @@ -308,363 +315,364 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_heavy_operations", "MetricThreshold": "tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to instruction cache misses.", - "MetricExpr": "ICACHE.IFETCH_STALL / tma_info_clks - tma_itlb_misses", + "MetricExpr": "ICACHE.IFETCH_STALL / tma_info_thread_clks - tma_itlb_misses", "MetricGroup": "BigFoot;FetchLat;IcMiss;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_icache_misses", "MetricThreshold": "tma_icache_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", "ScaleUnit": "100%" }, { - "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", - "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time", - "MetricGroup": "Power;Summary", - "MetricName": "tma_info_average_frequency" - }, - { - "BriefDescription": "Branch instructions per taken branch.", - "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", - "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_bptkbranch" + "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", + "MetricExpr": "tma_info_inst_mix_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_indirect", + "MetricThreshold": "tma_info_bad_spec_ipmisp_indirect < 1e3" }, { - "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "Pipeline", - "MetricName": "tma_info_clks" + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BadSpec;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmispredict", + "MetricThreshold": "tma_info_bad_spec_ipmispredict < 200" }, { "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", - "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_clks))", + "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_thread_clks))", "MetricGroup": "SMT", - "MetricName": "tma_info_core_clks" + "MetricName": "tma_info_core_core_clks" }, { "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks", "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_coreipc" + "MetricName": "tma_info_core_coreipc" }, { - "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricExpr": "1 / tma_info_ipc", - "MetricGroup": "Mem;Pipeline", - "MetricName": "tma_info_cpi" - }, - { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": "HPC;Summary", - "MetricName": "tma_info_cpu_utilization" - }, - { - "BriefDescription": "Average Parallel L2 cache miss data reads", - "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", - "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_data_l2_mlp" + "BriefDescription": "Floating Point Operations Per Cycle", + "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / tma_info_core_core_clks", + "MetricGroup": "Flops;Ret", + "MetricName": "tma_info_core_flopc" }, { - "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricExpr": "64 * (UNC_ARB_TRK_REQUESTS.ALL + UNC_ARB_COH_TRK_REQUESTS.ALL) / 1e6 / duration_time / 1e3", - "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", - "MetricName": "tma_info_dram_bw_use", - "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", + "MetricExpr": "UOPS_EXECUTED.THREAD / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC)", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", + "MetricName": "tma_info_core_ilp" }, { "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", "MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + LSD.UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)", "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", - "MetricName": "tma_info_dsb_coverage", - "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_iptb, tma_lcp" - }, - { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", - "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", - "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", - "MetricName": "tma_info_execute" - }, - { - "BriefDescription": "The ratio of Executed- by Issued-Uops", - "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", - "MetricGroup": "Cor;Pipeline", - "MetricName": "tma_info_execute_per_issue", - "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." - }, - { - "BriefDescription": "Floating Point Operations Per Cycle", - "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / tma_info_core_clks", - "MetricGroup": "Flops;Ret", - "MetricName": "tma_info_flopc" + "MetricName": "tma_info_frontend_dsb_coverage", + "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 4 > 0.35", + "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_inst_mix_iptb, tma_lcp" }, { - "BriefDescription": "Giga Floating Point Operations Per Second", - "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / 1e9 / duration_time", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_gflops", - "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_inst_mix_instructions / BACLEARS.ANY", + "MetricGroup": "Fed", + "MetricName": "tma_info_frontend_ipunknown_branch" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", - "MetricExpr": "UOPS_EXECUTED.THREAD / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC)", - "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", - "MetricName": "tma_info_ilp" + "BriefDescription": "Branch instructions per taken branch.", + "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;PGO", + "MetricName": "tma_info_inst_mix_bptkbranch" }, { "BriefDescription": "Total number of retired Instructions", "MetricExpr": "INST_RETIRED.ANY", "MetricGroup": "Summary;TmaL1;tma_L1_group", - "MetricName": "tma_info_instructions", + "MetricName": "tma_info_inst_mix_instructions", "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" }, { "BriefDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate)", "MetricExpr": "1 / (tma_fp_scalar + tma_fp_vector)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_iparith", - "MetricThreshold": "tma_info_iparith < 10", + "MetricName": "tma_info_inst_mix_iparith", + "MetricThreshold": "tma_info_inst_mix_iparith < 10", "PublicDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate). May undercount due to FMA double counting. Approximated prior to BDW." }, { "BriefDescription": "Instructions per Branch (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Branches;Fed;InsType", - "MetricName": "tma_info_ipbranch", - "MetricThreshold": "tma_info_ipbranch < 8" - }, - { - "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": "Ret;Summary", - "MetricName": "tma_info_ipc" + "MetricName": "tma_info_inst_mix_ipbranch", + "MetricThreshold": "tma_info_inst_mix_ipbranch < 8" }, { "BriefDescription": "Instructions per (near) call (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_CALL", "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_ipcall", - "MetricThreshold": "tma_info_ipcall < 200" - }, - { - "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", - "MetricGroup": "Branches;OS", - "MetricName": "tma_info_ipfarbranch", - "MetricThreshold": "tma_info_ipfarbranch < 1e6" + "MetricName": "tma_info_inst_mix_ipcall", + "MetricThreshold": "tma_info_inst_mix_ipcall < 200" }, { "BriefDescription": "Instructions per Load (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_LOADS", "MetricGroup": "InsType", - "MetricName": "tma_info_ipload", - "MetricThreshold": "tma_info_ipload < 3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", - "MetricExpr": "tma_info_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_indirect", - "MetricThreshold": "tma_info_ipmisp_indirect < 1e3" - }, - { - "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BadSpec;BrMispredicts", - "MetricName": "tma_info_ipmispredict", - "MetricThreshold": "tma_info_ipmispredict < 200" + "MetricName": "tma_info_inst_mix_ipload", + "MetricThreshold": "tma_info_inst_mix_ipload < 3" }, { "BriefDescription": "Instructions per Store (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_STORES", "MetricGroup": "InsType", - "MetricName": "tma_info_ipstore", - "MetricThreshold": "tma_info_ipstore < 8" + "MetricName": "tma_info_inst_mix_ipstore", + "MetricThreshold": "tma_info_inst_mix_ipstore < 8" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW;Frontend;PGO;tma_issueFB", - "MetricName": "tma_info_iptb", - "MetricThreshold": "tma_info_iptb < 9", - "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_lcp" - }, - { - "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", - "MetricExpr": "tma_info_instructions / BACLEARS.ANY", - "MetricGroup": "Fed", - "MetricName": "tma_info_ipunknown_branch" - }, - { - "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_cpi" - }, - { - "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_utilization", - "MetricThreshold": "tma_info_kernel_utilization > 0.05" + "MetricName": "tma_info_inst_mix_iptb", + "MetricThreshold": "tma_info_inst_mix_iptb < 9", + "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_lcp" }, { "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw" - }, - { - "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "tma_info_l1d_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw_1t" - }, - { - "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L1_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki" + "MetricName": "tma_info_memory_core_l1d_cache_fill_bw" }, { "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw" + "MetricName": "tma_info_memory_core_l2_cache_fill_bw" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "tma_info_l2_cache_fill_bw", + "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw_1t" + "MetricName": "tma_info_memory_core_l3_cache_fill_bw" + }, + { + "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L1_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l1mpki" }, { "BriefDescription": "L2 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L2_MISS / INST_RETIRED.ANY", "MetricGroup": "Backend;CacheMisses;Mem", - "MetricName": "tma_info_l2mpki" + "MetricName": "tma_info_memory_l2mpki" }, { - "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "0", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw_1t" + "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.LLC_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l3mpki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw" + "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_UOPS_RETIRED.L1_MISS + MEM_LOAD_UOPS_RETIRED.HIT_LFB)", + "MetricGroup": "Mem;MemoryBound;MemoryLat", + "MetricName": "tma_info_memory_load_miss_real_latency" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw_1t" + "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", + "MetricGroup": "Mem;MemoryBW;MemoryBound", + "MetricName": "tma_info_memory_mlp", + "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" }, { - "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.LLC_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l3mpki" + "BriefDescription": "Average Parallel L2 cache miss data reads", + "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", + "MetricGroup": "Memory_BW;Offcore", + "MetricName": "tma_info_memory_oro_data_l2_mlp" }, { "BriefDescription": "Average Latency for L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS.DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l2_miss_latency" + "MetricName": "tma_info_memory_oro_load_l2_miss_latency" }, { "BriefDescription": "Average Parallel L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_DATA_RD", "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_load_l2_mlp" + "MetricName": "tma_info_memory_oro_load_l2_mlp" }, { - "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_UOPS_RETIRED.L1_MISS + MEM_LOAD_UOPS_RETIRED.HIT_LFB)", - "MetricGroup": "Mem;MemoryBound;MemoryLat", - "MetricName": "tma_info_load_miss_real_latency" + "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l1d_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l1d_cache_fill_bw_1t" }, { - "BriefDescription": "Average number of parallel requests to external memory", - "MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_OCCUPANCY.CYCLES_WITH_ANY_REQUEST", - "MetricGroup": "Mem;SoC", - "MetricName": "tma_info_mem_parallel_requests", - "PublicDescription": "Average number of parallel requests to external memory. Accounts for all requests" + "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l2_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l2_cache_fill_bw_1t" }, { - "BriefDescription": "Average latency of all requests to external memory (in Uncore cycles)", - "MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_REQUESTS.ALL", - "MetricGroup": "Mem;SoC", - "MetricName": "tma_info_mem_request_latency" + "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "0", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_thread_l3_cache_access_bw_1t" }, { - "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", - "MetricGroup": "Mem;MemoryBW;MemoryBound", - "MetricName": "tma_info_mlp", - "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" + "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l3_cache_fill_bw_1t" }, { "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", - "MetricExpr": "(ITLB_MISSES.WALK_DURATION + DTLB_LOAD_MISSES.WALK_DURATION + DTLB_STORE_MISSES.WALK_DURATION) / tma_info_core_clks", + "MetricExpr": "(ITLB_MISSES.WALK_DURATION + DTLB_LOAD_MISSES.WALK_DURATION + DTLB_STORE_MISSES.WALK_DURATION) / tma_info_core_core_clks", "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_page_walks_utilization", - "MetricThreshold": "tma_info_page_walks_utilization > 0.5" + "MetricName": "tma_info_memory_tlb_page_walks_utilization", + "MetricThreshold": "tma_info_memory_tlb_page_walks_utilization > 0.5" + }, + { + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", + "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", + "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", + "MetricName": "tma_info_pipeline_execute" }, { "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@", "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_retire" + "MetricName": "tma_info_pipeline_retire" }, { - "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", - "MetricExpr": "4 * tma_info_core_clks", - "MetricGroup": "TmaL1;tma_L1_group", - "MetricName": "tma_info_slots" + "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time", + "MetricGroup": "Power;Summary", + "MetricName": "tma_info_system_average_frequency" + }, + { + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricGroup": "HPC;Summary", + "MetricName": "tma_info_system_cpu_utilization" + }, + { + "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", + "MetricExpr": "64 * (UNC_ARB_TRK_REQUESTS.ALL + UNC_ARB_COH_TRK_REQUESTS.ALL) / 1e6 / duration_time / 1e3", + "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", + "MetricName": "tma_info_system_dram_bw_use", + "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_mem_bandwidth, tma_sq_full" + }, + { + "BriefDescription": "Giga Floating Point Operations Per Second", + "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / 1e9 / duration_time", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_system_gflops", + "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + }, + { + "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", + "MetricGroup": "Branches;OS", + "MetricName": "tma_info_system_ipfarbranch", + "MetricThreshold": "tma_info_system_ipfarbranch < 1e6" + }, + { + "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_cpi" + }, + { + "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_utilization", + "MetricThreshold": "tma_info_system_kernel_utilization > 0.05" + }, + { + "BriefDescription": "Average number of parallel requests to external memory", + "MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_OCCUPANCY.CYCLES_WITH_ANY_REQUEST", + "MetricGroup": "Mem;SoC", + "MetricName": "tma_info_system_mem_parallel_requests", + "PublicDescription": "Average number of parallel requests to external memory. Accounts for all requests" + }, + { + "BriefDescription": "Average latency of all requests to external memory (in Uncore cycles)", + "MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_REQUESTS.ALL", + "MetricGroup": "Mem;SoC", + "MetricName": "tma_info_system_mem_request_latency" }, { "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / (CPU_CLK_UNHALTED.REF_XCLK_ANY / 2) if #SMT_on else 0)", "MetricGroup": "SMT", - "MetricName": "tma_info_smt_2t_utilization" + "MetricName": "tma_info_system_smt_2t_utilization" }, { "BriefDescription": "Socket actual clocks when any core is active on that socket", "MetricExpr": "UNC_CLOCK.SOCKET", "MetricGroup": "SoC", - "MetricName": "tma_info_socket_clks" + "MetricName": "tma_info_system_socket_clks" }, { "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC", "MetricGroup": "Power", - "MetricName": "tma_info_turbo_utilization" + "MetricName": "tma_info_system_turbo_utilization" + }, + { + "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "Pipeline", + "MetricName": "tma_info_thread_clks" + }, + { + "BriefDescription": "Cycles Per Instruction (per Logical Processor)", + "MetricExpr": "1 / tma_info_thread_ipc", + "MetricGroup": "Mem;Pipeline", + "MetricName": "tma_info_thread_cpi" + }, + { + "BriefDescription": "The ratio of Executed- by Issued-Uops", + "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", + "MetricGroup": "Cor;Pipeline", + "MetricName": "tma_info_thread_execute_per_issue", + "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." + }, + { + "BriefDescription": "Instructions Per Cycle (per Logical Processor)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks", + "MetricGroup": "Ret;Summary", + "MetricName": "tma_info_thread_ipc" + }, + { + "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", + "MetricExpr": "4 * tma_info_core_core_clks", + "MetricGroup": "TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots" }, { "BriefDescription": "Uops Per Instruction", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / INST_RETIRED.ANY", "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_uoppi", - "MetricThreshold": "tma_info_uoppi > 1.05" + "MetricName": "tma_info_thread_uoppi", + "MetricThreshold": "tma_info_thread_uoppi > 1.05" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW", - "MetricName": "tma_info_uptb", - "MetricThreshold": "tma_info_uptb < 6" + "MetricName": "tma_info_thread_uptb", + "MetricThreshold": "tma_info_thread_uptb < 6" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses", - "MetricExpr": "(12 * ITLB_MISSES.STLB_HIT + ITLB_MISSES.WALK_DURATION) / tma_info_clks", + "MetricExpr": "(12 * ITLB_MISSES.STLB_HIT + ITLB_MISSES.WALK_DURATION) / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -673,7 +681,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled without loads missing the L1 data cache", - "MetricExpr": "max((min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING) - CYCLE_ACTIVITY.STALLS_L1D_PENDING) / tma_info_clks, 0)", + "MetricExpr": "max((min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING) - CYCLE_ACTIVITY.STALLS_L1D_PENDING) / tma_info_thread_clks, 0)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_issueL1;tma_issueMC;tma_memory_bound_group", "MetricName": "tma_l1_bound", "MetricThreshold": "tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -682,7 +690,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled due to L2 cache accesses by loads", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L1D_PENDING - CYCLE_ACTIVITY.STALLS_L2_PENDING) / tma_info_clks", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L1D_PENDING - CYCLE_ACTIVITY.STALLS_L2_PENDING) / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l2_bound", "MetricThreshold": "tma_l2_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -692,7 +700,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_RETIRED.LLC_MISS) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_clks", + "MetricExpr": "MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_RETIRED.LLC_MISS) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -702,7 +710,7 @@ { "BriefDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "29 * (MEM_LOAD_UOPS_RETIRED.LLC_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.LLC_MISS))) / tma_info_clks", + "MetricExpr": "29 * (MEM_LOAD_UOPS_RETIRED.LLC_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_RETIRED.LLC_MISS))) / tma_info_thread_clks", "MetricGroup": "MemoryLat;TopdownL4;tma_L4_group;tma_issueLat;tma_l3_bound_group", "MetricName": "tma_l3_hit_latency", "MetricThreshold": "tma_l3_hit_latency > 0.1 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -711,11 +719,11 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)", - "MetricExpr": "ILD_STALL.LCP / tma_info_clks", + "MetricExpr": "ILD_STALL.LCP / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_lcp", "MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb", + "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb", "ScaleUnit": "100%" }, { @@ -724,13 +732,14 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_light_operations", "MetricThreshold": "tma_light_operations > 0.6", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation). This correlates with total number of instructions used by the program. A uops-per-instruction (see UopPI metric) ratio of 1 or less should be expected for decently optimized software running on Intel Core/Xeon products. While this often indicates efficient X86 instructions were executed; high value does not necessarily mean better performance cannot be achieved. Sample with: INST_RETIRED.PREC_DIST", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Load operations", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_load_op_utilization", "MetricThreshold": "tma_load_op_utilization > 0.6", @@ -740,7 +749,7 @@ { "BriefDescription": "This metric represents fraction of cycles the CPU spent handling cache misses due to lock operations", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO) / tma_info_clks", + "MetricExpr": "MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO) / tma_info_thread_clks", "MetricGroup": "Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_l1_bound_group", "MetricName": "tma_lock_latency", "MetricThreshold": "tma_lock_latency > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -754,21 +763,22 @@ "MetricGroup": "BadSpec;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.1 & tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears. These slots are either wasted by uops fetched prior to the clear; or stalls the out-of-order portion of the machine needs to recover its state after the clear. For example; this can happen due to memory ordering Nukes (e.g. Memory Disambiguation) or Self-Modifying-Code (SMC) nukes. Sample with: MACHINE_CLEARS.COUNT. Related metrics: tma_clears_resteers, tma_contested_accesses, tma_data_sharing, tma_false_sharing, tma_l1_bound, tma_microcode_sequencer, tma_ms_switches, tma_remote_cache", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=6@) / tma_info_clks", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=6@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW", "MetricName": "tma_mem_bandwidth", "MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_sq_full", + "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_sq_full", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat", "MetricName": "tma_mem_latency", "MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -778,16 +788,17 @@ { "BriefDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING) + RESOURCE_STALLS.SB) / (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", + "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING) + RESOURCE_STALLS.SB) / (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_thread_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", "MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck. Memory Bound estimates fraction of slots where pipeline is likely stalled due to demand load or store instructions. This accounts mainly for (1) non-completed in-flight memory demand loads which coincides with execution units starvation; in addition to (2) cases where stores could impose backpressure on the pipeline when many of them get buffered at the same time (less common out of the two).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_thread_slots", "MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS", "MetricName": "tma_microcode_sequencer", "MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1", @@ -796,16 +807,16 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline)", - "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_mite", - "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline). This pipeline is used for code that was not pre-cached in the DSB or LSD. For example; inefficiencies due to asymmetric decoders; use of long immediate or LCP can manifest as MITE fetch bandwidth bottleneck.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)", - "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_clks", + "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_thread_clks", "MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO", "MetricName": "tma_ms_switches", "MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -814,7 +825,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 0 ([SNB+] ALU; [HSW+] ALU and 2nd branch)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_core_clks", "MetricGroup": "Compute;TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_0", "MetricThreshold": "tma_port_0 > 0.6", @@ -823,7 +834,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 1 (ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_1", "MetricThreshold": "tma_port_1 > 0.6", @@ -832,7 +843,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 2 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_2", "MetricThreshold": "tma_port_2 > 0.6", @@ -841,7 +852,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 3 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_3", "MetricThreshold": "tma_port_3 > 0.6", @@ -859,7 +870,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 5 ([SNB+] Branches and ALU; [HSW+] ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_5", "MetricThreshold": "tma_port_5 > 0.6", @@ -869,7 +880,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING)) / tma_info_clks", + "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_thread_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING)) / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_ports_utilization", "MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -878,7 +889,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed no uops on any execution port (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,inv\\,cmask\\=1@ / 2 if #SMT_on else (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0)) / tma_info_core_clks)", + "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,inv\\,cmask\\=1@ / 2 if #SMT_on else (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0)) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_0", "MetricThreshold": "tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -887,7 +898,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the CPU executed total of 1 uop per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) / tma_info_core_clks)", + "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issueL1;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_1", "MetricThreshold": "tma_ports_utilized_1 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -896,7 +907,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 2 uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC - UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_clks)", + "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC - UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issue2P;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_2", "MetricThreshold": "tma_ports_utilized_2 > 0.15 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -905,7 +916,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 3 or more uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise).", - "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_clks", + "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_3m", "MetricThreshold": "tma_ports_utilized_3m > 0.7 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -913,17 +924,18 @@ }, { "BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired. Ideally; all pipeline slots would be attributed to the Retiring category. Retiring of 100% would indicate the maximum Pipeline_Width throughput was achieved. Maximizing Retiring typically increases the Instructions-per-cycle (see IPC metric). Note that a high Retiring value does not necessary mean there is no room for more performance. For example; Heavy-operations or Microcode Assists are categorized under Retiring. They often indicate suboptimal performance and can often be optimized or avoided. Sample with: UOPS_RETIRED.RETIRE_SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles handling memory load split accesses - load that cross 64-byte cache line boundary", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "13 * LD_BLOCKS.NO_SR / tma_info_clks", + "MetricExpr": "13 * LD_BLOCKS.NO_SR / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_split_loads", "MetricThreshold": "tma_split_loads > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -932,7 +944,7 @@ }, { "BriefDescription": "This metric represents rate of split store accesses", - "MetricExpr": "2 * MEM_UOPS_RETIRED.SPLIT_STORES / tma_info_core_clks", + "MetricExpr": "2 * MEM_UOPS_RETIRED.SPLIT_STORES / tma_info_core_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_issueSpSt;tma_store_bound_group", "MetricName": "tma_split_stores", "MetricThreshold": "tma_split_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -941,16 +953,16 @@ }, { "BriefDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors)", - "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_clks", + "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_core_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueBW;tma_l3_bound_group", "MetricName": "tma_sq_full", "MetricThreshold": "tma_sq_full > 0.3 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_mem_bandwidth", + "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_mem_bandwidth", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write", - "MetricExpr": "RESOURCE_STALLS.SB / tma_info_clks", + "MetricExpr": "RESOURCE_STALLS.SB / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -959,7 +971,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of cycles when the memory subsystem had loads blocked since they could not forward data from earlier (in program order) overlapping stores", - "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_clks", + "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_store_fwd_blk", "MetricThreshold": "tma_store_fwd_blk > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -969,7 +981,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU spent handling L1D store misses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(L2_RQSTS.RFO_HIT * 9 * (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) + (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_clks", + "MetricExpr": "(L2_RQSTS.RFO_HIT * 9 * (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) + (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_thread_clks", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_issueSL;tma_store_bound_group", "MetricName": "tma_store_latency", "MetricThreshold": "tma_store_latency > 0.1 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -978,7 +990,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Store operations", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_core_clks", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_store_op_utilization", "MetricThreshold": "tma_store_op_utilization > 0.6", diff --git a/tools/perf/pmu-events/arch/x86/ivybridge/metricgroups.json b/tools/perf/pmu-events/arch/x86/ivybridge/metricgroups.json new file mode 100644 index 000000000000..f6a0258e3241 --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/ivybridge/metricgroups.json @@ -0,0 +1,107 @@ +{ + "Backend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Bad": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BadSpec": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BigFoot": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BrMispredicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Branches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CacheMisses": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Compute": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Cor": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSBmiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DataSharing": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Fed": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Flops": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpScalar": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Frontend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "HPC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IcMiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "InsType": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "L2Evicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "LSD": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MachineClears": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBound": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryTLB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_BW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_Lat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MicroSeq": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "OS": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Offcore": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PGO": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Pipeline": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PortsUtil": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Ret": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Retire": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SMT": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Server": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Snoop": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SoC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL1": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL2": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL3mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "TopdownL5": "Metrics for top-down breakdown at level 5", + "TopdownL6": "Metrics for top-down breakdown at level 6", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_L5_group": "Metrics for top-down breakdown at level 5", + "tma_L6_group": "Metrics for top-down breakdown at level 6", + "tma_alu_op_utilization_group": "Metrics contributing to tma_alu_op_utilization category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_branch_resteers_group": "Metrics contributing to tma_branch_resteers category", + "tma_core_bound_group": "Metrics contributing to tma_core_bound category", + "tma_dram_bound_group": "Metrics contributing to tma_dram_bound category", + "tma_dtlb_load_group": "Metrics contributing to tma_dtlb_load category", + "tma_dtlb_store_group": "Metrics contributing to tma_dtlb_store category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_fp_arith_group": "Metrics contributing to tma_fp_arith category", + "tma_fp_vector_group": "Metrics contributing to tma_fp_vector category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_heavy_operations_group": "Metrics contributing to tma_heavy_operations category", + "tma_issue2P": "Metrics related by the issue $issue2P", + "tma_issueBM": "Metrics related by the issue $issueBM", + "tma_issueBW": "Metrics related by the issue $issueBW", + "tma_issueFB": "Metrics related by the issue $issueFB", + "tma_issueL1": "Metrics related by the issue $issueL1", + "tma_issueLat": "Metrics related by the issue $issueLat", + "tma_issueMC": "Metrics related by the issue $issueMC", + "tma_issueMS": "Metrics related by the issue $issueMS", + "tma_issueMV": "Metrics related by the issue $issueMV", + "tma_issueRFO": "Metrics related by the issue $issueRFO", + "tma_issueSL": "Metrics related by the issue $issueSL", + "tma_issueSO": "Metrics related by the issue $issueSO", + "tma_issueSmSt": "Metrics related by the issue $issueSmSt", + "tma_issueSpSt": "Metrics related by the issue $issueSpSt", + "tma_issueSyncxn": "Metrics related by the issue $issueSyncxn", + "tma_issueTLB": "Metrics related by the issue $issueTLB", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_l3_bound_group": "Metrics contributing to tma_l3_bound category", + "tma_light_operations_group": "Metrics contributing to tma_light_operations category", + "tma_load_op_utilization_group": "Metrics contributing to tma_load_op_utilization category", + "tma_mem_latency_group": "Metrics contributing to tma_mem_latency category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_microcode_sequencer_group": "Metrics contributing to tma_microcode_sequencer category", + "tma_ports_utilization_group": "Metrics contributing to tma_ports_utilization category", + "tma_ports_utilized_0_group": "Metrics contributing to tma_ports_utilized_0 category", + "tma_ports_utilized_3m_group": "Metrics contributing to tma_ports_utilized_3m category", + "tma_retiring_group": "Metrics contributing to tma_retiring category", + "tma_serializing_operation_group": "Metrics contributing to tma_serializing_operation category", + "tma_store_bound_group": "Metrics contributing to tma_store_bound category", + "tma_store_op_utilization_group": "Metrics contributing to tma_store_op_utilization category" +} diff --git a/tools/perf/pmu-events/arch/x86/ivytown/ivt-metrics.json b/tools/perf/pmu-events/arch/x86/ivytown/ivt-metrics.json index 89469b10fa30..f5e46a768fdd 100644 --- a/tools/perf/pmu-events/arch/x86/ivytown/ivt-metrics.json +++ b/tools/perf/pmu-events/arch/x86/ivytown/ivt-metrics.json @@ -50,7 +50,7 @@ }, { "BriefDescription": "Uncore frequency per die [GHZ]", - "MetricExpr": "tma_info_socket_clks / #num_dies / duration_time / 1e9", + "MetricExpr": "tma_info_system_socket_clks / #num_dies / duration_time / 1e9", "MetricGroup": "SoC", "MetricName": "UNCORE_FREQ" }, @@ -71,7 +71,7 @@ }, { "BriefDescription": "This metric estimates how often memory load accesses were aliased by preceding stores (in program order) with a 4K address offset", - "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_clks", + "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_4k_aliasing", "MetricThreshold": "tma_4k_aliasing > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -81,7 +81,7 @@ { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution ports for ALU operations.", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5) / (3 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5) / (3 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_alu_op_utilization", "MetricThreshold": "tma_alu_op_utilization > 0.6", @@ -89,7 +89,7 @@ }, { "BriefDescription": "This metric estimates fraction of slots the CPU retired uops delivered by the Microcode_Sequencer as a result of Assists", - "MetricExpr": "100 * OTHER_ASSISTS.ANY_WB_ASSIST / tma_info_slots", + "MetricExpr": "100 * OTHER_ASSISTS.ANY_WB_ASSIST / tma_info_thread_slots", "MetricGroup": "TopdownL4;tma_L4_group;tma_microcode_sequencer_group", "MetricName": "tma_assists", "MetricThreshold": "tma_assists > 0.1 & (tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1)", @@ -103,15 +103,17 @@ "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend. Backend is the portion of the processor core where the out-of-order scheduler dispatches ready uops into their respective execution units; and once completed these uops get retired according to program order. For example; stalls due to data-cache misses or stalls due to the divider unit being overloaded are both categorized under Backend Bound. Backend Bound is further divided into two main categories: Memory Bound and Core Bound.", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations", - "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_slots", + "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots wasted due to incorrect speculations. This include slots used to issue uops that do not eventually get retired and slots for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work due to miss-predicted branches are categorized under Bad Speculation category. Incorrect data speculation followed by Memory Ordering Nukes is another example.", "ScaleUnit": "100%" }, @@ -122,12 +124,13 @@ "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15", - "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_branch_misprediction_cost, tma_mispredicts_resteers", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers", - "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_clks", + "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteers", "MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -147,7 +150,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to contested accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(60 * (MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD))) + 43 * (MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD)))) / tma_info_clks", + "MetricExpr": "(60 * (MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD))) + 43 * (MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD)))) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_contested_accesses", "MetricThreshold": "tma_contested_accesses > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -161,13 +164,14 @@ "MetricGroup": "Backend;Compute;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck. Shortage in hardware compute resources; or dependencies in software's instructions are both categorized under Core Bound. Hence it may indicate the machine ran out of an out-of-order resource; certain execution units are overloaded or dependencies in program's data- or instruction-flow are limiting the performance (e.g. FP-chained long-latency arithmetic operations).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to data-sharing accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "43 * (MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD))) / tma_info_clks", + "MetricExpr": "43 * (MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD))) / tma_info_thread_clks", "MetricGroup": "Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_data_sharing", "MetricThreshold": "tma_data_sharing > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -176,7 +180,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the Divider unit was active", - "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_clks", + "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_core_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_divider", "MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -186,7 +190,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_RETIRED.LLC_MISS)) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_clks", + "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_RETIRED.LLC_MISS)) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -195,25 +199,25 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline", - "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSB;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_dsb", - "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline. For example; inefficient utilization of the DSB cache structure or bank conflict when reading from it; are categorized here.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_dsb_switches", "MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb, tma_lcp", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses", - "MetricExpr": "(7 * DTLB_LOAD_MISSES.STLB_HIT + DTLB_LOAD_MISSES.WALK_DURATION) / tma_info_clks", + "MetricExpr": "(7 * DTLB_LOAD_MISSES.STLB_HIT + DTLB_LOAD_MISSES.WALK_DURATION) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group", "MetricName": "tma_dtlb_load", "MetricThreshold": "tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -222,7 +226,7 @@ }, { "BriefDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses", - "MetricExpr": "(7 * DTLB_STORE_MISSES.STLB_HIT + DTLB_STORE_MISSES.WALK_DURATION) / tma_info_clks", + "MetricExpr": "(7 * DTLB_STORE_MISSES.STLB_HIT + DTLB_STORE_MISSES.WALK_DURATION) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_store_bound_group", "MetricName": "tma_dtlb_store", "MetricThreshold": "tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -231,7 +235,7 @@ }, { "BriefDescription": "This metric roughly estimates how often CPU was handling synchronizations due to False Sharing", - "MetricExpr": "(200 * OFFCORE_RESPONSE.DEMAND_RFO.LLC_MISS.REMOTE_HITM + 60 * OFFCORE_RESPONSE.DEMAND_RFO.LLC_HIT.HITM_OTHER_CORE) / tma_info_clks", + "MetricExpr": "(200 * OFFCORE_RESPONSE.DEMAND_RFO.LLC_MISS.REMOTE_HITM + 60 * OFFCORE_RESPONSE.DEMAND_RFO.LLC_HIT.HITM_OTHER_CORE) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_store_bound_group", "MetricName": "tma_false_sharing", "MetricThreshold": "tma_false_sharing > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -241,11 +245,11 @@ { "BriefDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "tma_info_load_miss_real_latency * cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@ / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@ / tma_info_thread_clks", "MetricGroup": "MemoryBW;TopdownL4;tma_L4_group;tma_issueBW;tma_issueSL;tma_issueSmSt;tma_l1_bound_group", "MetricName": "tma_fb_full", "MetricThreshold": "tma_fb_full > 0.3", - "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", + "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", "ScaleUnit": "100%" }, { @@ -253,16 +257,18 @@ "MetricExpr": "tma_frontend_bound - tma_fetch_latency", "MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", "MetricName": "tma_fetch_bandwidth", - "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_info_iptb, tma_lcp", + "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues", - "MetricExpr": "4 * min(CPU_CLK_UNHALTED.THREAD, IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE) / tma_info_slots", + "MetricExpr": "4 * min(CPU_CLK_UNHALTED.THREAD, IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE) / tma_info_thread_slots", "MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: RS_EVENTS.EMPTY_END", "ScaleUnit": "100%" }, @@ -295,10 +301,11 @@ }, { "BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend", - "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_slots", + "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_thread_slots", "MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound.", "ScaleUnit": "100%" }, @@ -308,364 +315,365 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_heavy_operations", "MetricThreshold": "tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to instruction cache misses.", - "MetricExpr": "ICACHE.IFETCH_STALL / tma_info_clks - tma_itlb_misses", + "MetricExpr": "ICACHE.IFETCH_STALL / tma_info_thread_clks - tma_itlb_misses", "MetricGroup": "BigFoot;FetchLat;IcMiss;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_icache_misses", "MetricThreshold": "tma_icache_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", "ScaleUnit": "100%" }, { - "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", - "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time", - "MetricGroup": "Power;Summary", - "MetricName": "tma_info_average_frequency" - }, - { - "BriefDescription": "Branch instructions per taken branch.", - "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", - "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_bptkbranch" + "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", + "MetricExpr": "tma_info_inst_mix_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_indirect", + "MetricThreshold": "tma_info_bad_spec_ipmisp_indirect < 1e3" }, { - "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "Pipeline", - "MetricName": "tma_info_clks" + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BadSpec;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmispredict", + "MetricThreshold": "tma_info_bad_spec_ipmispredict < 200" }, { "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", - "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_clks))", + "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_thread_clks))", "MetricGroup": "SMT", - "MetricName": "tma_info_core_clks" + "MetricName": "tma_info_core_core_clks" }, { "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks", "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_coreipc" + "MetricName": "tma_info_core_coreipc" }, { - "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricExpr": "1 / tma_info_ipc", - "MetricGroup": "Mem;Pipeline", - "MetricName": "tma_info_cpi" - }, - { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": "HPC;Summary", - "MetricName": "tma_info_cpu_utilization" - }, - { - "BriefDescription": "Average Parallel L2 cache miss data reads", - "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", - "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_data_l2_mlp" + "BriefDescription": "Floating Point Operations Per Cycle", + "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / tma_info_core_core_clks", + "MetricGroup": "Flops;Ret", + "MetricName": "tma_info_core_flopc" }, { - "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time", - "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", - "MetricName": "tma_info_dram_bw_use", - "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", + "MetricExpr": "UOPS_EXECUTED.THREAD / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC)", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", + "MetricName": "tma_info_core_ilp" }, { "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", "MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + LSD.UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)", "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", - "MetricName": "tma_info_dsb_coverage", - "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_iptb, tma_lcp" - }, - { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", - "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", - "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", - "MetricName": "tma_info_execute" - }, - { - "BriefDescription": "The ratio of Executed- by Issued-Uops", - "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", - "MetricGroup": "Cor;Pipeline", - "MetricName": "tma_info_execute_per_issue", - "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." - }, - { - "BriefDescription": "Floating Point Operations Per Cycle", - "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / tma_info_core_clks", - "MetricGroup": "Flops;Ret", - "MetricName": "tma_info_flopc" + "MetricName": "tma_info_frontend_dsb_coverage", + "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 4 > 0.35", + "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_inst_mix_iptb, tma_lcp" }, { - "BriefDescription": "Giga Floating Point Operations Per Second", - "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / 1e9 / duration_time", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_gflops", - "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_inst_mix_instructions / BACLEARS.ANY", + "MetricGroup": "Fed", + "MetricName": "tma_info_frontend_ipunknown_branch" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", - "MetricExpr": "UOPS_EXECUTED.THREAD / (cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC)", - "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", - "MetricName": "tma_info_ilp" + "BriefDescription": "Branch instructions per taken branch.", + "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;PGO", + "MetricName": "tma_info_inst_mix_bptkbranch" }, { "BriefDescription": "Total number of retired Instructions", "MetricExpr": "INST_RETIRED.ANY", "MetricGroup": "Summary;TmaL1;tma_L1_group", - "MetricName": "tma_info_instructions", + "MetricName": "tma_info_inst_mix_instructions", "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" }, { "BriefDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate)", "MetricExpr": "1 / (tma_fp_scalar + tma_fp_vector)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_iparith", - "MetricThreshold": "tma_info_iparith < 10", + "MetricName": "tma_info_inst_mix_iparith", + "MetricThreshold": "tma_info_inst_mix_iparith < 10", "PublicDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate). May undercount due to FMA double counting. Approximated prior to BDW." }, { "BriefDescription": "Instructions per Branch (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Branches;Fed;InsType", - "MetricName": "tma_info_ipbranch", - "MetricThreshold": "tma_info_ipbranch < 8" - }, - { - "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": "Ret;Summary", - "MetricName": "tma_info_ipc" + "MetricName": "tma_info_inst_mix_ipbranch", + "MetricThreshold": "tma_info_inst_mix_ipbranch < 8" }, { "BriefDescription": "Instructions per (near) call (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_CALL", "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_ipcall", - "MetricThreshold": "tma_info_ipcall < 200" - }, - { - "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", - "MetricGroup": "Branches;OS", - "MetricName": "tma_info_ipfarbranch", - "MetricThreshold": "tma_info_ipfarbranch < 1e6" + "MetricName": "tma_info_inst_mix_ipcall", + "MetricThreshold": "tma_info_inst_mix_ipcall < 200" }, { "BriefDescription": "Instructions per Load (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_LOADS", "MetricGroup": "InsType", - "MetricName": "tma_info_ipload", - "MetricThreshold": "tma_info_ipload < 3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", - "MetricExpr": "tma_info_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_indirect", - "MetricThreshold": "tma_info_ipmisp_indirect < 1e3" - }, - { - "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BadSpec;BrMispredicts", - "MetricName": "tma_info_ipmispredict", - "MetricThreshold": "tma_info_ipmispredict < 200" + "MetricName": "tma_info_inst_mix_ipload", + "MetricThreshold": "tma_info_inst_mix_ipload < 3" }, { "BriefDescription": "Instructions per Store (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_UOPS_RETIRED.ALL_STORES", "MetricGroup": "InsType", - "MetricName": "tma_info_ipstore", - "MetricThreshold": "tma_info_ipstore < 8" + "MetricName": "tma_info_inst_mix_ipstore", + "MetricThreshold": "tma_info_inst_mix_ipstore < 8" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW;Frontend;PGO;tma_issueFB", - "MetricName": "tma_info_iptb", - "MetricThreshold": "tma_info_iptb < 9", - "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_lcp" - }, - { - "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", - "MetricExpr": "tma_info_instructions / BACLEARS.ANY", - "MetricGroup": "Fed", - "MetricName": "tma_info_ipunknown_branch" - }, - { - "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_cpi" - }, - { - "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_utilization", - "MetricThreshold": "tma_info_kernel_utilization > 0.05" + "MetricName": "tma_info_inst_mix_iptb", + "MetricThreshold": "tma_info_inst_mix_iptb < 9", + "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_lcp" }, { "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw" - }, - { - "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "tma_info_l1d_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw_1t" - }, - { - "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L1_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki" + "MetricName": "tma_info_memory_core_l1d_cache_fill_bw" }, { "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw" + "MetricName": "tma_info_memory_core_l2_cache_fill_bw" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "tma_info_l2_cache_fill_bw", + "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw_1t" + "MetricName": "tma_info_memory_core_l3_cache_fill_bw" + }, + { + "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L1_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l1mpki" }, { "BriefDescription": "L2 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.L2_MISS / INST_RETIRED.ANY", "MetricGroup": "Backend;CacheMisses;Mem", - "MetricName": "tma_info_l2mpki" + "MetricName": "tma_info_memory_l2mpki" }, { - "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "0", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw_1t" + "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.LLC_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l3mpki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw" + "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_UOPS_RETIRED.L1_MISS + MEM_LOAD_UOPS_RETIRED.HIT_LFB)", + "MetricGroup": "Mem;MemoryBound;MemoryLat", + "MetricName": "tma_info_memory_load_miss_real_latency" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw_1t" + "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", + "MetricGroup": "Mem;MemoryBW;MemoryBound", + "MetricName": "tma_info_memory_mlp", + "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" }, { - "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_UOPS_RETIRED.LLC_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l3mpki" + "BriefDescription": "Average Parallel L2 cache miss data reads", + "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", + "MetricGroup": "Memory_BW;Offcore", + "MetricName": "tma_info_memory_oro_data_l2_mlp" }, { "BriefDescription": "Average Latency for L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS.DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l2_miss_latency" + "MetricName": "tma_info_memory_oro_load_l2_miss_latency" }, { "BriefDescription": "Average Parallel L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_DATA_RD", "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_load_l2_mlp" + "MetricName": "tma_info_memory_oro_load_l2_mlp" }, { - "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_UOPS_RETIRED.L1_MISS + MEM_LOAD_UOPS_RETIRED.HIT_LFB)", - "MetricGroup": "Mem;MemoryBound;MemoryLat", - "MetricName": "tma_info_load_miss_real_latency" + "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l1d_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l1d_cache_fill_bw_1t" }, { - "BriefDescription": "Average number of parallel data read requests to external memory", - "MetricExpr": "UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182\\,thresh\\=1@", - "MetricGroup": "Mem;MemoryBW;SoC", - "MetricName": "tma_info_mem_parallel_reads", - "PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches" + "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l2_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l2_cache_fill_bw_1t" }, { - "BriefDescription": "Average latency of data read request to external memory (in nanoseconds)", - "MetricExpr": "1e9 * (UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_INSERTS.MISS_OPCODE@filter_opc\\=0x182@) / (tma_info_socket_clks / duration_time)", - "MetricGroup": "Mem;MemoryLat;SoC", - "MetricName": "tma_info_mem_read_latency", - "PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)" + "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "0", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_thread_l3_cache_access_bw_1t" }, { - "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", - "MetricGroup": "Mem;MemoryBW;MemoryBound", - "MetricName": "tma_info_mlp", - "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" + "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l3_cache_fill_bw_1t" }, { "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", - "MetricExpr": "(ITLB_MISSES.WALK_DURATION + DTLB_LOAD_MISSES.WALK_DURATION + DTLB_STORE_MISSES.WALK_DURATION) / tma_info_core_clks", + "MetricExpr": "(ITLB_MISSES.WALK_DURATION + DTLB_LOAD_MISSES.WALK_DURATION + DTLB_STORE_MISSES.WALK_DURATION) / tma_info_core_core_clks", "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_page_walks_utilization", - "MetricThreshold": "tma_info_page_walks_utilization > 0.5" + "MetricName": "tma_info_memory_tlb_page_walks_utilization", + "MetricThreshold": "tma_info_memory_tlb_page_walks_utilization > 0.5" + }, + { + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", + "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", + "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", + "MetricName": "tma_info_pipeline_execute" }, { "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@", "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_retire" + "MetricName": "tma_info_pipeline_retire" }, { - "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", - "MetricExpr": "4 * tma_info_core_clks", - "MetricGroup": "TmaL1;tma_L1_group", - "MetricName": "tma_info_slots" + "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time", + "MetricGroup": "Power;Summary", + "MetricName": "tma_info_system_average_frequency" + }, + { + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricGroup": "HPC;Summary", + "MetricName": "tma_info_system_cpu_utilization" + }, + { + "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", + "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time", + "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", + "MetricName": "tma_info_system_dram_bw_use", + "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_mem_bandwidth, tma_sq_full" + }, + { + "BriefDescription": "Giga Floating Point Operations Per Second", + "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / 1e9 / duration_time", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_system_gflops", + "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + }, + { + "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", + "MetricGroup": "Branches;OS", + "MetricName": "tma_info_system_ipfarbranch", + "MetricThreshold": "tma_info_system_ipfarbranch < 1e6" + }, + { + "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_cpi" + }, + { + "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_utilization", + "MetricThreshold": "tma_info_system_kernel_utilization > 0.05" + }, + { + "BriefDescription": "Average number of parallel data read requests to external memory", + "MetricExpr": "UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182\\,thresh\\=1@", + "MetricGroup": "Mem;MemoryBW;SoC", + "MetricName": "tma_info_system_mem_parallel_reads", + "PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches" + }, + { + "BriefDescription": "Average latency of data read request to external memory (in nanoseconds)", + "MetricExpr": "1e9 * (UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_INSERTS.MISS_OPCODE@filter_opc\\=0x182@) / (tma_info_system_socket_clks / duration_time)", + "MetricGroup": "Mem;MemoryLat;SoC", + "MetricName": "tma_info_system_mem_read_latency", + "PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)" }, { "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / (CPU_CLK_UNHALTED.REF_XCLK_ANY / 2) if #SMT_on else 0)", "MetricGroup": "SMT", - "MetricName": "tma_info_smt_2t_utilization" + "MetricName": "tma_info_system_smt_2t_utilization" }, { "BriefDescription": "Socket actual clocks when any core is active on that socket", "MetricExpr": "cbox_0@event\\=0x0@", "MetricGroup": "SoC", - "MetricName": "tma_info_socket_clks" + "MetricName": "tma_info_system_socket_clks" }, { "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC", "MetricGroup": "Power", - "MetricName": "tma_info_turbo_utilization" + "MetricName": "tma_info_system_turbo_utilization" + }, + { + "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "Pipeline", + "MetricName": "tma_info_thread_clks" + }, + { + "BriefDescription": "Cycles Per Instruction (per Logical Processor)", + "MetricExpr": "1 / tma_info_thread_ipc", + "MetricGroup": "Mem;Pipeline", + "MetricName": "tma_info_thread_cpi" + }, + { + "BriefDescription": "The ratio of Executed- by Issued-Uops", + "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", + "MetricGroup": "Cor;Pipeline", + "MetricName": "tma_info_thread_execute_per_issue", + "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." + }, + { + "BriefDescription": "Instructions Per Cycle (per Logical Processor)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks", + "MetricGroup": "Ret;Summary", + "MetricName": "tma_info_thread_ipc" + }, + { + "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", + "MetricExpr": "4 * tma_info_core_core_clks", + "MetricGroup": "TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots" }, { "BriefDescription": "Uops Per Instruction", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / INST_RETIRED.ANY", "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_uoppi", - "MetricThreshold": "tma_info_uoppi > 1.05" + "MetricName": "tma_info_thread_uoppi", + "MetricThreshold": "tma_info_thread_uoppi > 1.05" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW", - "MetricName": "tma_info_uptb", - "MetricThreshold": "tma_info_uptb < 6" + "MetricName": "tma_info_thread_uptb", + "MetricThreshold": "tma_info_thread_uptb < 6" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses", - "MetricExpr": "(12 * ITLB_MISSES.STLB_HIT + ITLB_MISSES.WALK_DURATION) / tma_info_clks", + "MetricExpr": "(12 * ITLB_MISSES.STLB_HIT + ITLB_MISSES.WALK_DURATION) / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -674,7 +682,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled without loads missing the L1 data cache", - "MetricExpr": "max((min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING) - CYCLE_ACTIVITY.STALLS_L1D_PENDING) / tma_info_clks, 0)", + "MetricExpr": "max((min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING) - CYCLE_ACTIVITY.STALLS_L1D_PENDING) / tma_info_thread_clks, 0)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_issueL1;tma_issueMC;tma_memory_bound_group", "MetricName": "tma_l1_bound", "MetricThreshold": "tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -683,7 +691,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled due to L2 cache accesses by loads", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L1D_PENDING - CYCLE_ACTIVITY.STALLS_L2_PENDING) / tma_info_clks", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L1D_PENDING - CYCLE_ACTIVITY.STALLS_L2_PENDING) / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l2_bound", "MetricThreshold": "tma_l2_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -693,7 +701,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_RETIRED.LLC_MISS) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_clks", + "MetricExpr": "MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_RETIRED.LLC_MISS) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -703,7 +711,7 @@ { "BriefDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "41 * (MEM_LOAD_UOPS_RETIRED.LLC_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD))) / tma_info_clks", + "MetricExpr": "41 * (MEM_LOAD_UOPS_RETIRED.LLC_HIT * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD))) / tma_info_thread_clks", "MetricGroup": "MemoryLat;TopdownL4;tma_L4_group;tma_issueLat;tma_l3_bound_group", "MetricName": "tma_l3_hit_latency", "MetricThreshold": "tma_l3_hit_latency > 0.1 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -712,11 +720,11 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)", - "MetricExpr": "ILD_STALL.LCP / tma_info_clks", + "MetricExpr": "ILD_STALL.LCP / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_lcp", "MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb", + "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb", "ScaleUnit": "100%" }, { @@ -725,13 +733,14 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_light_operations", "MetricThreshold": "tma_light_operations > 0.6", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation). This correlates with total number of instructions used by the program. A uops-per-instruction (see UopPI metric) ratio of 1 or less should be expected for decently optimized software running on Intel Core/Xeon products. While this often indicates efficient X86 instructions were executed; high value does not necessarily mean better performance cannot be achieved. Sample with: INST_RETIRED.PREC_DIST", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Load operations", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_load_op_utilization", "MetricThreshold": "tma_load_op_utilization > 0.6", @@ -741,7 +750,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from local memory", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "200 * (MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD))) / tma_info_clks", + "MetricExpr": "200 * (MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD))) / tma_info_thread_clks", "MetricGroup": "Server;TopdownL5;tma_L5_group;tma_mem_latency_group", "MetricName": "tma_local_dram", "MetricThreshold": "tma_local_dram > 0.1 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -751,7 +760,7 @@ { "BriefDescription": "This metric represents fraction of cycles the CPU spent handling cache misses due to lock operations", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO) / tma_info_clks", + "MetricExpr": "MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO) / tma_info_thread_clks", "MetricGroup": "Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_l1_bound_group", "MetricName": "tma_lock_latency", "MetricThreshold": "tma_lock_latency > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -765,21 +774,22 @@ "MetricGroup": "BadSpec;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.1 & tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears. These slots are either wasted by uops fetched prior to the clear; or stalls the out-of-order portion of the machine needs to recover its state after the clear. For example; this can happen due to memory ordering Nukes (e.g. Memory Disambiguation) or Self-Modifying-Code (SMC) nukes. Sample with: MACHINE_CLEARS.COUNT. Related metrics: tma_clears_resteers, tma_contested_accesses, tma_data_sharing, tma_false_sharing, tma_l1_bound, tma_microcode_sequencer, tma_ms_switches, tma_remote_cache", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=6@) / tma_info_clks", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=6@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW", "MetricName": "tma_mem_bandwidth", "MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_sq_full", + "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_sq_full", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat", "MetricName": "tma_mem_latency", "MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -789,16 +799,17 @@ { "BriefDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING) + RESOURCE_STALLS.SB) / (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", + "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING) + RESOURCE_STALLS.SB) / (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_thread_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", "MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck. Memory Bound estimates fraction of slots where pipeline is likely stalled due to demand load or store instructions. This accounts mainly for (1) non-completed in-flight memory demand loads which coincides with execution units starvation; in addition to (2) cases where stores could impose backpressure on the pipeline when many of them get buffered at the same time (less common out of the two).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_thread_slots", "MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS", "MetricName": "tma_microcode_sequencer", "MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1", @@ -807,16 +818,16 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline)", - "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_mite", - "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline). This pipeline is used for code that was not pre-cached in the DSB or LSD. For example; inefficiencies due to asymmetric decoders; use of long immediate or LCP can manifest as MITE fetch bandwidth bottleneck.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)", - "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_clks", + "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_thread_clks", "MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO", "MetricName": "tma_ms_switches", "MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -825,7 +836,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 0 ([SNB+] ALU; [HSW+] ALU and 2nd branch)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_core_clks", "MetricGroup": "Compute;TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_0", "MetricThreshold": "tma_port_0 > 0.6", @@ -834,7 +845,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 1 (ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_1", "MetricThreshold": "tma_port_1 > 0.6", @@ -843,7 +854,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 2 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_2", "MetricThreshold": "tma_port_2 > 0.6", @@ -852,7 +863,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 3 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_3", "MetricThreshold": "tma_port_3 > 0.6", @@ -870,7 +881,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 5 ([SNB+] Branches and ALU; [HSW+] ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_5", "MetricThreshold": "tma_port_5 > 0.6", @@ -880,7 +891,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING)) / tma_info_clks", + "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) + UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - (UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC if tma_info_thread_ipc > 1.8 else UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_LDM_PENDING)) / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_ports_utilization", "MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -889,7 +900,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed no uops on any execution port (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,inv\\,cmask\\=1@ / 2 if #SMT_on else (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0)) / tma_info_core_clks)", + "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,inv\\,cmask\\=1@ / 2 if #SMT_on else (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_EXECUTE) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0)) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_0", "MetricThreshold": "tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -898,7 +909,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the CPU executed total of 1 uop per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) / tma_info_core_clks)", + "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=1@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_1_UOP_EXEC - UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issueL1;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_1", "MetricThreshold": "tma_ports_utilized_1 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -907,7 +918,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 2 uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC - UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_clks)", + "MetricExpr": "((cpu@UOPS_EXECUTED.CORE\\,cmask\\=2@ - cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@) / 2 if #SMT_on else (UOPS_EXECUTED.CYCLES_GE_2_UOPS_EXEC - UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_core_clks)", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issue2P;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_2", "MetricThreshold": "tma_ports_utilized_2 > 0.15 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -916,7 +927,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 3 or more uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise).", - "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_clks", + "MetricExpr": "(cpu@UOPS_EXECUTED.CORE\\,cmask\\=3@ / 2 if #SMT_on else UOPS_EXECUTED.CYCLES_GE_3_UOPS_EXEC) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_3m", "MetricThreshold": "tma_ports_utilized_3m > 0.7 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -925,7 +936,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from remote cache in other sockets including synchronizations issues", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(200 * (MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD))) + 180 * (MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD)))) / tma_info_clks", + "MetricExpr": "(200 * (MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD))) + 180 * (MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD)))) / tma_info_thread_clks", "MetricGroup": "Offcore;Server;Snoop;TopdownL5;tma_L5_group;tma_issueSyncxn;tma_mem_latency_group", "MetricName": "tma_remote_cache", "MetricThreshold": "tma_remote_cache > 0.05 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -935,7 +946,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from remote memory", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "310 * (MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD))) / tma_info_clks", + "MetricExpr": "310 * (MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_UOPS_RETIRED.HIT_LFB / (MEM_LOAD_UOPS_RETIRED.L2_HIT + MEM_LOAD_UOPS_RETIRED.LLC_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HIT + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_HITM + MEM_LOAD_UOPS_LLC_HIT_RETIRED.XSNP_MISS + MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_HITM + MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_FWD))) / tma_info_thread_clks", "MetricGroup": "Server;Snoop;TopdownL5;tma_L5_group;tma_mem_latency_group", "MetricName": "tma_remote_dram", "MetricThreshold": "tma_remote_dram > 0.1 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -944,17 +955,18 @@ }, { "BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired. Ideally; all pipeline slots would be attributed to the Retiring category. Retiring of 100% would indicate the maximum Pipeline_Width throughput was achieved. Maximizing Retiring typically increases the Instructions-per-cycle (see IPC metric). Note that a high Retiring value does not necessary mean there is no room for more performance. For example; Heavy-operations or Microcode Assists are categorized under Retiring. They often indicate suboptimal performance and can often be optimized or avoided. Sample with: UOPS_RETIRED.RETIRE_SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles handling memory load split accesses - load that cross 64-byte cache line boundary", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "13 * LD_BLOCKS.NO_SR / tma_info_clks", + "MetricExpr": "13 * LD_BLOCKS.NO_SR / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_split_loads", "MetricThreshold": "tma_split_loads > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -963,7 +975,7 @@ }, { "BriefDescription": "This metric represents rate of split store accesses", - "MetricExpr": "2 * MEM_UOPS_RETIRED.SPLIT_STORES / tma_info_core_clks", + "MetricExpr": "2 * MEM_UOPS_RETIRED.SPLIT_STORES / tma_info_core_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_issueSpSt;tma_store_bound_group", "MetricName": "tma_split_stores", "MetricThreshold": "tma_split_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -972,16 +984,16 @@ }, { "BriefDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors)", - "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_clks", + "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_core_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueBW;tma_l3_bound_group", "MetricName": "tma_sq_full", "MetricThreshold": "tma_sq_full > 0.3 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_mem_bandwidth", + "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_mem_bandwidth", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write", - "MetricExpr": "RESOURCE_STALLS.SB / tma_info_clks", + "MetricExpr": "RESOURCE_STALLS.SB / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -990,7 +1002,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of cycles when the memory subsystem had loads blocked since they could not forward data from earlier (in program order) overlapping stores", - "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_clks", + "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_store_fwd_blk", "MetricThreshold": "tma_store_fwd_blk > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1000,7 +1012,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU spent handling L1D store misses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(L2_RQSTS.RFO_HIT * 9 * (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) + (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_clks", + "MetricExpr": "(L2_RQSTS.RFO_HIT * 9 * (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) + (1 - MEM_UOPS_RETIRED.LOCK_LOADS / MEM_UOPS_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_thread_clks", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_issueSL;tma_store_bound_group", "MetricName": "tma_store_latency", "MetricThreshold": "tma_store_latency > 0.1 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1009,7 +1021,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Store operations", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_core_clks", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_store_op_utilization", "MetricThreshold": "tma_store_op_utilization > 0.6", diff --git a/tools/perf/pmu-events/arch/x86/ivytown/metricgroups.json b/tools/perf/pmu-events/arch/x86/ivytown/metricgroups.json new file mode 100644 index 000000000000..f6a0258e3241 --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/ivytown/metricgroups.json @@ -0,0 +1,107 @@ +{ + "Backend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Bad": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BadSpec": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BigFoot": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BrMispredicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Branches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CacheMisses": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Compute": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Cor": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSBmiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DataSharing": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Fed": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Flops": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpScalar": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Frontend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "HPC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IcMiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "InsType": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "L2Evicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "LSD": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MachineClears": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBound": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryTLB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_BW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_Lat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MicroSeq": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "OS": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Offcore": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PGO": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Pipeline": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PortsUtil": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Ret": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Retire": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SMT": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Server": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Snoop": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SoC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL1": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL2": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL3mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "TopdownL5": "Metrics for top-down breakdown at level 5", + "TopdownL6": "Metrics for top-down breakdown at level 6", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_L5_group": "Metrics for top-down breakdown at level 5", + "tma_L6_group": "Metrics for top-down breakdown at level 6", + "tma_alu_op_utilization_group": "Metrics contributing to tma_alu_op_utilization category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_branch_resteers_group": "Metrics contributing to tma_branch_resteers category", + "tma_core_bound_group": "Metrics contributing to tma_core_bound category", + "tma_dram_bound_group": "Metrics contributing to tma_dram_bound category", + "tma_dtlb_load_group": "Metrics contributing to tma_dtlb_load category", + "tma_dtlb_store_group": "Metrics contributing to tma_dtlb_store category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_fp_arith_group": "Metrics contributing to tma_fp_arith category", + "tma_fp_vector_group": "Metrics contributing to tma_fp_vector category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_heavy_operations_group": "Metrics contributing to tma_heavy_operations category", + "tma_issue2P": "Metrics related by the issue $issue2P", + "tma_issueBM": "Metrics related by the issue $issueBM", + "tma_issueBW": "Metrics related by the issue $issueBW", + "tma_issueFB": "Metrics related by the issue $issueFB", + "tma_issueL1": "Metrics related by the issue $issueL1", + "tma_issueLat": "Metrics related by the issue $issueLat", + "tma_issueMC": "Metrics related by the issue $issueMC", + "tma_issueMS": "Metrics related by the issue $issueMS", + "tma_issueMV": "Metrics related by the issue $issueMV", + "tma_issueRFO": "Metrics related by the issue $issueRFO", + "tma_issueSL": "Metrics related by the issue $issueSL", + "tma_issueSO": "Metrics related by the issue $issueSO", + "tma_issueSmSt": "Metrics related by the issue $issueSmSt", + "tma_issueSpSt": "Metrics related by the issue $issueSpSt", + "tma_issueSyncxn": "Metrics related by the issue $issueSyncxn", + "tma_issueTLB": "Metrics related by the issue $issueTLB", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_l3_bound_group": "Metrics contributing to tma_l3_bound category", + "tma_light_operations_group": "Metrics contributing to tma_light_operations category", + "tma_load_op_utilization_group": "Metrics contributing to tma_load_op_utilization category", + "tma_mem_latency_group": "Metrics contributing to tma_mem_latency category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_microcode_sequencer_group": "Metrics contributing to tma_microcode_sequencer category", + "tma_ports_utilization_group": "Metrics contributing to tma_ports_utilization category", + "tma_ports_utilized_0_group": "Metrics contributing to tma_ports_utilized_0 category", + "tma_ports_utilized_3m_group": "Metrics contributing to tma_ports_utilized_3m category", + "tma_retiring_group": "Metrics contributing to tma_retiring category", + "tma_serializing_operation_group": "Metrics contributing to tma_serializing_operation category", + "tma_store_bound_group": "Metrics contributing to tma_store_bound category", + "tma_store_op_utilization_group": "Metrics contributing to tma_store_op_utilization category" +} diff --git a/tools/perf/pmu-events/arch/x86/jaketown/jkt-metrics.json b/tools/perf/pmu-events/arch/x86/jaketown/jkt-metrics.json index e8f4e5c01c9f..35b1a3aa728d 100644 --- a/tools/perf/pmu-events/arch/x86/jaketown/jkt-metrics.json +++ b/tools/perf/pmu-events/arch/x86/jaketown/jkt-metrics.json @@ -50,7 +50,7 @@ }, { "BriefDescription": "Uncore frequency per die [GHZ]", - "MetricExpr": "tma_info_socket_clks / #num_dies / duration_time / 1e9", + "MetricExpr": "tma_info_system_socket_clks / #num_dies / duration_time / 1e9", "MetricGroup": "SoC", "MetricName": "UNCORE_FREQ" }, @@ -76,15 +76,17 @@ "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend. Backend is the portion of the processor core where the out-of-order scheduler dispatches ready uops into their respective execution units; and once completed these uops get retired according to program order. For example; stalls due to data-cache misses or stalls due to the divider unit being overloaded are both categorized under Backend Bound. Backend Bound is further divided into two main categories: Memory Bound and Core Bound.", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations", - "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_slots", + "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots wasted due to incorrect speculations. This include slots used to issue uops that do not eventually get retired and slots for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work due to miss-predicted branches are categorized under Bad Speculation category. Incorrect data speculation followed by Memory Ordering Nukes is another example.", "ScaleUnit": "100%" }, @@ -95,12 +97,13 @@ "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15", - "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_branch_misprediction_cost, tma_mispredicts_resteers", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers", - "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_clks", + "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteers", "MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -114,12 +117,13 @@ "MetricGroup": "Backend;Compute;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck. Shortage in hardware compute resources; or dependencies in software's instructions are both categorized under Core Bound. Hence it may indicate the machine ran out of an out-of-order resource; certain execution units are overloaded or dependencies in program's data- or instruction-flow are limiting the performance (e.g. FP-chained long-latency arithmetic operations).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles where the Divider unit was active", - "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_clks", + "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_core_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_divider", "MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -129,7 +133,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_RETIRED.LLC_MISS)) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_clks", + "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_RETIRED.LLC_MISS)) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -138,16 +142,16 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_dsb_switches", "MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_lcp", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses", - "MetricExpr": "(7 * DTLB_LOAD_MISSES.STLB_HIT + DTLB_LOAD_MISSES.WALK_DURATION) / tma_info_clks", + "MetricExpr": "(7 * DTLB_LOAD_MISSES.STLB_HIT + DTLB_LOAD_MISSES.WALK_DURATION) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group", "MetricName": "tma_dtlb_load", "MetricThreshold": "tma_dtlb_load > 0.1", @@ -159,16 +163,18 @@ "MetricExpr": "tma_frontend_bound - tma_fetch_latency", "MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", "MetricName": "tma_fetch_bandwidth", - "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_lcp", + "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_frontend_dsb_coverage, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues", - "MetricExpr": "4 * min(CPU_CLK_UNHALTED.THREAD, IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE) / tma_info_slots", + "MetricExpr": "4 * min(CPU_CLK_UNHALTED.THREAD, IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE) / tma_info_thread_slots", "MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: RS_EVENTS.EMPTY_END", "ScaleUnit": "100%" }, @@ -201,10 +207,11 @@ }, { "BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend", - "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_slots", + "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_thread_slots", "MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound.", "ScaleUnit": "100%" }, @@ -214,173 +221,174 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_heavy_operations", "MetricThreshold": "tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences.", "ScaleUnit": "100%" }, { - "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", - "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time", - "MetricGroup": "Power;Summary", - "MetricName": "tma_info_average_frequency" - }, - { - "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "Pipeline", - "MetricName": "tma_info_clks" - }, - { "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", - "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_clks))", + "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_thread_clks))", "MetricGroup": "SMT", - "MetricName": "tma_info_core_clks" + "MetricName": "tma_info_core_core_clks" }, { "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks", "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_coreipc" - }, - { - "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricExpr": "1 / tma_info_ipc", - "MetricGroup": "Mem;Pipeline", - "MetricName": "tma_info_cpi" + "MetricName": "tma_info_core_coreipc" }, { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": "HPC;Summary", - "MetricName": "tma_info_cpu_utilization" + "BriefDescription": "Floating Point Operations Per Cycle", + "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / tma_info_core_core_clks", + "MetricGroup": "Flops;Ret", + "MetricName": "tma_info_core_flopc" }, { - "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time", - "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", - "MetricName": "tma_info_dram_bw_use", - "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_mem_bandwidth" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", + "MetricExpr": "UOPS_DISPATCHED.THREAD / (cpu@UOPS_DISPATCHED.CORE\\,cmask\\=1@ / 2 if #SMT_on else cpu@UOPS_DISPATCHED.CORE\\,cmask\\=1@)", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", + "MetricName": "tma_info_core_ilp" }, { "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", "MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + LSD.UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)", "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", - "MetricName": "tma_info_dsb_coverage", - "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 4 > 0.35", + "MetricName": "tma_info_frontend_dsb_coverage", + "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 4 > 0.35", "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_lcp" }, { - "BriefDescription": "The ratio of Executed- by Issued-Uops", - "MetricExpr": "UOPS_DISPATCHED.THREAD / UOPS_ISSUED.ANY", - "MetricGroup": "Cor;Pipeline", - "MetricName": "tma_info_execute_per_issue", - "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." + "BriefDescription": "Total number of retired Instructions", + "MetricExpr": "INST_RETIRED.ANY", + "MetricGroup": "Summary;TmaL1;tma_L1_group", + "MetricName": "tma_info_inst_mix_instructions", + "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" }, { - "BriefDescription": "Floating Point Operations Per Cycle", - "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / tma_info_core_clks", - "MetricGroup": "Flops;Ret", - "MetricName": "tma_info_flopc" + "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@", + "MetricGroup": "Pipeline;Ret", + "MetricName": "tma_info_pipeline_retire" }, { - "BriefDescription": "Giga Floating Point Operations Per Second", - "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / 1e9 / duration_time", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_gflops", - "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time", + "MetricGroup": "Power;Summary", + "MetricName": "tma_info_system_average_frequency" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", - "MetricExpr": "UOPS_DISPATCHED.THREAD / (cpu@UOPS_DISPATCHED.CORE\\,cmask\\=1@ / 2 if #SMT_on else cpu@UOPS_DISPATCHED.CORE\\,cmask\\=1@)", - "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", - "MetricName": "tma_info_ilp" + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricGroup": "HPC;Summary", + "MetricName": "tma_info_system_cpu_utilization" }, { - "BriefDescription": "Total number of retired Instructions", - "MetricExpr": "INST_RETIRED.ANY", - "MetricGroup": "Summary;TmaL1;tma_L1_group", - "MetricName": "tma_info_instructions", - "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" + "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", + "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time", + "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", + "MetricName": "tma_info_system_dram_bw_use", + "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_mem_bandwidth" }, { - "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": "Ret;Summary", - "MetricName": "tma_info_ipc" + "BriefDescription": "Giga Floating Point Operations Per Second", + "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / 1e9 / duration_time", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_system_gflops", + "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." }, { "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", "MetricGroup": "Branches;OS", - "MetricName": "tma_info_ipfarbranch", - "MetricThreshold": "tma_info_ipfarbranch < 1e6" + "MetricName": "tma_info_system_ipfarbranch", + "MetricThreshold": "tma_info_system_ipfarbranch < 1e6" }, { "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", "MetricGroup": "OS", - "MetricName": "tma_info_kernel_cpi" + "MetricName": "tma_info_system_kernel_cpi" }, { "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", "MetricGroup": "OS", - "MetricName": "tma_info_kernel_utilization", - "MetricThreshold": "tma_info_kernel_utilization > 0.05" + "MetricName": "tma_info_system_kernel_utilization", + "MetricThreshold": "tma_info_system_kernel_utilization > 0.05" }, { "BriefDescription": "Average number of parallel data read requests to external memory", "MetricExpr": "UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182\\,thresh\\=1@", "MetricGroup": "Mem;MemoryBW;SoC", - "MetricName": "tma_info_mem_parallel_reads", + "MetricName": "tma_info_system_mem_parallel_reads", "PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches" }, { "BriefDescription": "Average latency of data read request to external memory (in nanoseconds)", - "MetricExpr": "1e9 * (UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_INSERTS.MISS_OPCODE@filter_opc\\=0x182@) / (tma_info_socket_clks / duration_time)", + "MetricExpr": "1e9 * (UNC_C_TOR_OCCUPANCY.MISS_OPCODE@filter_opc\\=0x182@ / UNC_C_TOR_INSERTS.MISS_OPCODE@filter_opc\\=0x182@) / (tma_info_system_socket_clks / duration_time)", "MetricGroup": "Mem;MemoryLat;SoC", - "MetricName": "tma_info_mem_read_latency", + "MetricName": "tma_info_system_mem_read_latency", "PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)" }, { - "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@", - "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_retire" - }, - { - "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", - "MetricExpr": "4 * tma_info_core_clks", - "MetricGroup": "TmaL1;tma_L1_group", - "MetricName": "tma_info_slots" - }, - { "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / (CPU_CLK_UNHALTED.REF_XCLK_ANY / 2) if #SMT_on else 0)", "MetricGroup": "SMT", - "MetricName": "tma_info_smt_2t_utilization" + "MetricName": "tma_info_system_smt_2t_utilization" }, { "BriefDescription": "Socket actual clocks when any core is active on that socket", "MetricExpr": "cbox_0@event\\=0x0@", "MetricGroup": "SoC", - "MetricName": "tma_info_socket_clks" + "MetricName": "tma_info_system_socket_clks" }, { "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC", "MetricGroup": "Power", - "MetricName": "tma_info_turbo_utilization" + "MetricName": "tma_info_system_turbo_utilization" + }, + { + "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "Pipeline", + "MetricName": "tma_info_thread_clks" + }, + { + "BriefDescription": "Cycles Per Instruction (per Logical Processor)", + "MetricExpr": "1 / tma_info_thread_ipc", + "MetricGroup": "Mem;Pipeline", + "MetricName": "tma_info_thread_cpi" + }, + { + "BriefDescription": "The ratio of Executed- by Issued-Uops", + "MetricExpr": "UOPS_DISPATCHED.THREAD / UOPS_ISSUED.ANY", + "MetricGroup": "Cor;Pipeline", + "MetricName": "tma_info_thread_execute_per_issue", + "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." + }, + { + "BriefDescription": "Instructions Per Cycle (per Logical Processor)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks", + "MetricGroup": "Ret;Summary", + "MetricName": "tma_info_thread_ipc" + }, + { + "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", + "MetricExpr": "4 * tma_info_core_core_clks", + "MetricGroup": "TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots" }, { "BriefDescription": "Uops Per Instruction", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / INST_RETIRED.ANY", "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_uoppi", - "MetricThreshold": "tma_info_uoppi > 1.05" + "MetricName": "tma_info_thread_uoppi", + "MetricThreshold": "tma_info_thread_uoppi > 1.05" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses", - "MetricExpr": "(12 * ITLB_MISSES.STLB_HIT + ITLB_MISSES.WALK_DURATION) / tma_info_clks", + "MetricExpr": "(12 * ITLB_MISSES.STLB_HIT + ITLB_MISSES.WALK_DURATION) / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -390,7 +398,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_RETIRED.LLC_MISS) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_clks", + "MetricExpr": "MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_RETIRED.LLC_MISS) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -399,11 +407,11 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)", - "MetricExpr": "ILD_STALL.LCP / tma_info_clks", + "MetricExpr": "ILD_STALL.LCP / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_lcp", "MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage", + "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage", "ScaleUnit": "100%" }, { @@ -412,6 +420,7 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_light_operations", "MetricThreshold": "tma_light_operations > 0.6", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation). This correlates with total number of instructions used by the program. A uops-per-instruction (see UopPI metric) ratio of 1 or less should be expected for decently optimized software running on Intel Core/Xeon products. While this often indicates efficient X86 instructions were executed; high value does not necessarily mean better performance cannot be achieved. Sample with: INST_RETIRED.PREC_DIST", "ScaleUnit": "100%" }, @@ -422,21 +431,22 @@ "MetricGroup": "BadSpec;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.1 & tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears. These slots are either wasted by uops fetched prior to the clear; or stalls the out-of-order portion of the machine needs to recover its state after the clear. For example; this can happen due to memory ordering Nukes (e.g. Memory Disambiguation) or Self-Modifying-Code (SMC) nukes. Sample with: MACHINE_CLEARS.COUNT. Related metrics: tma_clears_resteers, tma_l1_bound, tma_microcode_sequencer, tma_ms_switches, tma_remote_cache", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=6@) / tma_info_clks", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=6@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW", "MetricName": "tma_mem_bandwidth", "MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_info_dram_bw_use", + "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_info_system_dram_bw_use", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat", "MetricName": "tma_mem_latency", "MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -446,16 +456,17 @@ { "BriefDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_L1D_PENDING) + RESOURCE_STALLS.SB) / (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_DISPATCH) + cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=1@ - (cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=3@ if tma_info_ipc > 1.8 else cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", + "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_L1D_PENDING) + RESOURCE_STALLS.SB) / (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_DISPATCH) + cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=1@ - (cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=3@ if tma_info_thread_ipc > 1.8 else cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", "MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck. Memory Bound estimates fraction of slots where pipeline is likely stalled due to demand load or store instructions. This accounts mainly for (1) non-completed in-flight memory demand loads which coincides with execution units starvation; in addition to (2) cases where stores could impose backpressure on the pipeline when many of them get buffered at the same time (less common out of the two).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_thread_slots", "MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS", "MetricName": "tma_microcode_sequencer", "MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1", @@ -464,7 +475,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)", - "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_clks", + "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_thread_clks", "MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO", "MetricName": "tma_ms_switches", "MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -474,7 +485,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_DISPATCH) + cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=1@ - (cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=3@ if tma_info_ipc > 1.8 else cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_L1D_PENDING)) / tma_info_clks", + "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_DISPATCH) + cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=1@ - (cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=3@ if tma_info_thread_ipc > 1.8 else cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_L1D_PENDING)) / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_ports_utilization", "MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -483,16 +494,17 @@ }, { "BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired. Ideally; all pipeline slots would be attributed to the Retiring category. Retiring of 100% would indicate the maximum Pipeline_Width throughput was achieved. Maximizing Retiring typically increases the Instructions-per-cycle (see IPC metric). Note that a high Retiring value does not necessary mean there is no room for more performance. For example; Heavy-operations or Microcode Assists are categorized under Retiring. They often indicate suboptimal performance and can often be optimized or avoided. Sample with: UOPS_RETIRED.RETIRE_SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write", - "MetricExpr": "RESOURCE_STALLS.SB / tma_info_clks", + "MetricExpr": "RESOURCE_STALLS.SB / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", diff --git a/tools/perf/pmu-events/arch/x86/jaketown/metricgroups.json b/tools/perf/pmu-events/arch/x86/jaketown/metricgroups.json new file mode 100644 index 000000000000..bebb85945d62 --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/jaketown/metricgroups.json @@ -0,0 +1,100 @@ +{ + "Backend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Bad": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BadSpec": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BigFoot": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BrMispredicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Branches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CacheMisses": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Compute": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Cor": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSBmiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Fed": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Flops": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpScalar": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Frontend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "HPC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IcMiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "InsType": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "L2Evicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "LSD": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MachineClears": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBound": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryTLB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MicroSeq": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "OS": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Offcore": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PGO": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Pipeline": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PortsUtil": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Ret": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Retire": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SMT": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Server": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Snoop": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SoC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL1": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL2": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL3mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "TopdownL5": "Metrics for top-down breakdown at level 5", + "TopdownL6": "Metrics for top-down breakdown at level 6", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_L5_group": "Metrics for top-down breakdown at level 5", + "tma_L6_group": "Metrics for top-down breakdown at level 6", + "tma_alu_op_utilization_group": "Metrics contributing to tma_alu_op_utilization category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_branch_resteers_group": "Metrics contributing to tma_branch_resteers category", + "tma_core_bound_group": "Metrics contributing to tma_core_bound category", + "tma_dram_bound_group": "Metrics contributing to tma_dram_bound category", + "tma_dtlb_load_group": "Metrics contributing to tma_dtlb_load category", + "tma_dtlb_store_group": "Metrics contributing to tma_dtlb_store category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_fp_arith_group": "Metrics contributing to tma_fp_arith category", + "tma_fp_vector_group": "Metrics contributing to tma_fp_vector category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_heavy_operations_group": "Metrics contributing to tma_heavy_operations category", + "tma_issue2P": "Metrics related by the issue $issue2P", + "tma_issueBM": "Metrics related by the issue $issueBM", + "tma_issueBW": "Metrics related by the issue $issueBW", + "tma_issueFB": "Metrics related by the issue $issueFB", + "tma_issueL1": "Metrics related by the issue $issueL1", + "tma_issueLat": "Metrics related by the issue $issueLat", + "tma_issueMC": "Metrics related by the issue $issueMC", + "tma_issueMS": "Metrics related by the issue $issueMS", + "tma_issueMV": "Metrics related by the issue $issueMV", + "tma_issueRFO": "Metrics related by the issue $issueRFO", + "tma_issueSL": "Metrics related by the issue $issueSL", + "tma_issueSO": "Metrics related by the issue $issueSO", + "tma_issueSmSt": "Metrics related by the issue $issueSmSt", + "tma_issueSyncxn": "Metrics related by the issue $issueSyncxn", + "tma_issueTLB": "Metrics related by the issue $issueTLB", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_light_operations_group": "Metrics contributing to tma_light_operations category", + "tma_mem_latency_group": "Metrics contributing to tma_mem_latency category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_microcode_sequencer_group": "Metrics contributing to tma_microcode_sequencer category", + "tma_ports_utilization_group": "Metrics contributing to tma_ports_utilization category", + "tma_ports_utilized_0_group": "Metrics contributing to tma_ports_utilized_0 category", + "tma_retiring_group": "Metrics contributing to tma_retiring category", + "tma_serializing_operation_group": "Metrics contributing to tma_serializing_operation category", + "tma_store_bound_group": "Metrics contributing to tma_store_bound category", + "tma_store_op_utilization_group": "Metrics contributing to tma_store_op_utilization category" +} diff --git a/tools/perf/pmu-events/arch/x86/mapfile.csv b/tools/perf/pmu-events/arch/x86/mapfile.csv index 66c37a3cbf43..6543a68d4a17 100644 --- a/tools/perf/pmu-events/arch/x86/mapfile.csv +++ b/tools/perf/pmu-events/arch/x86/mapfile.csv @@ -1,19 +1,19 @@ Family-model,Version,Filename,EventType -GenuineIntel-6-(97|9A|B7|BA|BF),v1.20,alderlake,core -GenuineIntel-6-BE,v1.20,alderlaken,core +GenuineIntel-6-(97|9A|B7|BA|BF),v1.21,alderlake,core +GenuineIntel-6-BE,v1.21,alderlaken,core GenuineIntel-6-(1C|26|27|35|36),v4,bonnell,core -GenuineIntel-6-(3D|47),v27,broadwell,core -GenuineIntel-6-56,v9,broadwellde,core -GenuineIntel-6-4F,v20,broadwellx,core -GenuineIntel-6-55-[56789ABCDEF],v1.17,cascadelakex,core -GenuineIntel-6-9[6C],v1.03,elkhartlake,core +GenuineIntel-6-(3D|47),v28,broadwell,core +GenuineIntel-6-56,v10,broadwellde,core +GenuineIntel-6-4F,v21,broadwellx,core +GenuineIntel-6-55-[56789ABCDEF],v1.18,cascadelakex,core +GenuineIntel-6-9[6C],v1.04,elkhartlake,core GenuineIntel-6-5[CF],v13,goldmont,core GenuineIntel-6-7A,v1.01,goldmontplus,core GenuineIntel-6-B6,v1.00,grandridge,core GenuineIntel-6-A[DE],v1.01,graniterapids,core GenuineIntel-6-(3C|45|46),v33,haswell,core GenuineIntel-6-3F,v27,haswellx,core -GenuineIntel-6-(7D|7E|A7),v1.17,icelake,core +GenuineIntel-6-(7D|7E|A7),v1.18,icelake,core GenuineIntel-6-6[AC],v1.20,icelakex,core GenuineIntel-6-3A,v24,ivybridge,core GenuineIntel-6-3E,v23,ivytown,core @@ -23,13 +23,13 @@ GenuineIntel-6-A[AC],v1.01,meteorlake,core GenuineIntel-6-1[AEF],v3,nehalemep,core GenuineIntel-6-2E,v3,nehalemex,core GenuineIntel-6-2A,v19,sandybridge,core -GenuineIntel-6-(8F|CF),v1.12,sapphirerapids,core +GenuineIntel-6-(8F|CF),v1.13,sapphirerapids,core GenuineIntel-6-AF,v1.00,sierraforest,core GenuineIntel-6-(37|4A|4C|4D|5A),v15,silvermont,core -GenuineIntel-6-(4E|5E|8E|9E|A5|A6),v55,skylake,core -GenuineIntel-6-55-[01234],v1.29,skylakex,core -GenuineIntel-6-86,v1.20,snowridgex,core -GenuineIntel-6-8[CD],v1.10,tigerlake,core +GenuineIntel-6-(4E|5E|8E|9E|A5|A6),v56,skylake,core +GenuineIntel-6-55-[01234],v1.30,skylakex,core +GenuineIntel-6-86,v1.21,snowridgex,core +GenuineIntel-6-8[CD],v1.12,tigerlake,core GenuineIntel-6-2C,v4,westmereep-dp,core GenuineIntel-6-25,v3,westmereep-sp,core GenuineIntel-6-2F,v3,westmereex,core diff --git a/tools/perf/pmu-events/arch/x86/sandybridge/metricgroups.json b/tools/perf/pmu-events/arch/x86/sandybridge/metricgroups.json new file mode 100644 index 000000000000..bebb85945d62 --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/sandybridge/metricgroups.json @@ -0,0 +1,100 @@ +{ + "Backend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Bad": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BadSpec": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BigFoot": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BrMispredicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Branches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CacheMisses": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Compute": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Cor": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSBmiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Fed": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Flops": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpScalar": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Frontend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "HPC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IcMiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "InsType": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "L2Evicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "LSD": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MachineClears": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBound": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryTLB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MicroSeq": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "OS": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Offcore": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PGO": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Pipeline": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PortsUtil": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Ret": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Retire": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SMT": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Server": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Snoop": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SoC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL1": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL2": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL3mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "TopdownL5": "Metrics for top-down breakdown at level 5", + "TopdownL6": "Metrics for top-down breakdown at level 6", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_L5_group": "Metrics for top-down breakdown at level 5", + "tma_L6_group": "Metrics for top-down breakdown at level 6", + "tma_alu_op_utilization_group": "Metrics contributing to tma_alu_op_utilization category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_branch_resteers_group": "Metrics contributing to tma_branch_resteers category", + "tma_core_bound_group": "Metrics contributing to tma_core_bound category", + "tma_dram_bound_group": "Metrics contributing to tma_dram_bound category", + "tma_dtlb_load_group": "Metrics contributing to tma_dtlb_load category", + "tma_dtlb_store_group": "Metrics contributing to tma_dtlb_store category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_fp_arith_group": "Metrics contributing to tma_fp_arith category", + "tma_fp_vector_group": "Metrics contributing to tma_fp_vector category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_heavy_operations_group": "Metrics contributing to tma_heavy_operations category", + "tma_issue2P": "Metrics related by the issue $issue2P", + "tma_issueBM": "Metrics related by the issue $issueBM", + "tma_issueBW": "Metrics related by the issue $issueBW", + "tma_issueFB": "Metrics related by the issue $issueFB", + "tma_issueL1": "Metrics related by the issue $issueL1", + "tma_issueLat": "Metrics related by the issue $issueLat", + "tma_issueMC": "Metrics related by the issue $issueMC", + "tma_issueMS": "Metrics related by the issue $issueMS", + "tma_issueMV": "Metrics related by the issue $issueMV", + "tma_issueRFO": "Metrics related by the issue $issueRFO", + "tma_issueSL": "Metrics related by the issue $issueSL", + "tma_issueSO": "Metrics related by the issue $issueSO", + "tma_issueSmSt": "Metrics related by the issue $issueSmSt", + "tma_issueSyncxn": "Metrics related by the issue $issueSyncxn", + "tma_issueTLB": "Metrics related by the issue $issueTLB", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_light_operations_group": "Metrics contributing to tma_light_operations category", + "tma_mem_latency_group": "Metrics contributing to tma_mem_latency category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_microcode_sequencer_group": "Metrics contributing to tma_microcode_sequencer category", + "tma_ports_utilization_group": "Metrics contributing to tma_ports_utilization category", + "tma_ports_utilized_0_group": "Metrics contributing to tma_ports_utilized_0 category", + "tma_retiring_group": "Metrics contributing to tma_retiring category", + "tma_serializing_operation_group": "Metrics contributing to tma_serializing_operation category", + "tma_store_bound_group": "Metrics contributing to tma_store_bound category", + "tma_store_op_utilization_group": "Metrics contributing to tma_store_op_utilization category" +} diff --git a/tools/perf/pmu-events/arch/x86/sandybridge/snb-metrics.json b/tools/perf/pmu-events/arch/x86/sandybridge/snb-metrics.json index 4a99fe515f4b..8898b6fd0dea 100644 --- a/tools/perf/pmu-events/arch/x86/sandybridge/snb-metrics.json +++ b/tools/perf/pmu-events/arch/x86/sandybridge/snb-metrics.json @@ -50,7 +50,7 @@ }, { "BriefDescription": "Uncore frequency per die [GHZ]", - "MetricExpr": "tma_info_socket_clks / #num_dies / duration_time / 1e9", + "MetricExpr": "tma_info_system_socket_clks / #num_dies / duration_time / 1e9", "MetricGroup": "SoC", "MetricName": "UNCORE_FREQ" }, @@ -76,15 +76,17 @@ "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend. Backend is the portion of the processor core where the out-of-order scheduler dispatches ready uops into their respective execution units; and once completed these uops get retired according to program order. For example; stalls due to data-cache misses or stalls due to the divider unit being overloaded are both categorized under Backend Bound. Backend Bound is further divided into two main categories: Memory Bound and Core Bound.", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations", - "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_slots", + "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots wasted due to incorrect speculations. This include slots used to issue uops that do not eventually get retired and slots for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work due to miss-predicted branches are categorized under Bad Speculation category. Incorrect data speculation followed by Memory Ordering Nukes is another example.", "ScaleUnit": "100%" }, @@ -95,12 +97,13 @@ "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15", - "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_branch_misprediction_cost, tma_mispredicts_resteers", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers", - "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_clks", + "MetricExpr": "12 * (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT + BACLEARS.ANY) / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteers", "MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -114,12 +117,13 @@ "MetricGroup": "Backend;Compute;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck. Shortage in hardware compute resources; or dependencies in software's instructions are both categorized under Core Bound. Hence it may indicate the machine ran out of an out-of-order resource; certain execution units are overloaded or dependencies in program's data- or instruction-flow are limiting the performance (e.g. FP-chained long-latency arithmetic operations).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles where the Divider unit was active", - "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_clks", + "MetricExpr": "ARITH.FPU_DIV_ACTIVE / tma_info_core_core_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_divider", "MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -129,7 +133,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_MISC_RETIRED.LLC_MISS)) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_clks", + "MetricExpr": "(1 - MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_MISC_RETIRED.LLC_MISS)) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -138,16 +142,16 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_dsb_switches", "MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_lcp", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Related metrics: tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses", - "MetricExpr": "(7 * DTLB_LOAD_MISSES.STLB_HIT + DTLB_LOAD_MISSES.WALK_DURATION) / tma_info_clks", + "MetricExpr": "(7 * DTLB_LOAD_MISSES.STLB_HIT + DTLB_LOAD_MISSES.WALK_DURATION) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group", "MetricName": "tma_dtlb_load", "MetricThreshold": "tma_dtlb_load > 0.1", @@ -159,16 +163,18 @@ "MetricExpr": "tma_frontend_bound - tma_fetch_latency", "MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", "MetricName": "tma_fetch_bandwidth", - "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_lcp", + "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Related metrics: tma_dsb_switches, tma_info_frontend_dsb_coverage, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues", - "MetricExpr": "4 * min(CPU_CLK_UNHALTED.THREAD, IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE) / tma_info_slots", + "MetricExpr": "4 * min(CPU_CLK_UNHALTED.THREAD, IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE) / tma_info_thread_slots", "MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: RS_EVENTS.EMPTY_END", "ScaleUnit": "100%" }, @@ -201,10 +207,11 @@ }, { "BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend", - "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_slots", + "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_thread_slots", "MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound.", "ScaleUnit": "100%" }, @@ -214,172 +221,173 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_heavy_operations", "MetricThreshold": "tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences.", "ScaleUnit": "100%" }, { - "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", - "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time", - "MetricGroup": "Power;Summary", - "MetricName": "tma_info_average_frequency" - }, - { - "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "Pipeline", - "MetricName": "tma_info_clks" - }, - { "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", - "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_clks))", + "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_thread_clks))", "MetricGroup": "SMT", - "MetricName": "tma_info_core_clks" + "MetricName": "tma_info_core_core_clks" }, { "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks", "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_coreipc" - }, - { - "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricExpr": "1 / tma_info_ipc", - "MetricGroup": "Mem;Pipeline", - "MetricName": "tma_info_cpi" + "MetricName": "tma_info_core_coreipc" }, { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": "HPC;Summary", - "MetricName": "tma_info_cpu_utilization" + "BriefDescription": "Floating Point Operations Per Cycle", + "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / tma_info_core_core_clks", + "MetricGroup": "Flops;Ret", + "MetricName": "tma_info_core_flopc" }, { - "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricExpr": "64 * (UNC_ARB_TRK_REQUESTS.ALL + UNC_ARB_COH_TRK_REQUESTS.ALL) / 1e6 / duration_time / 1e3", - "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", - "MetricName": "tma_info_dram_bw_use", - "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_mem_bandwidth" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", + "MetricExpr": "UOPS_DISPATCHED.THREAD / (cpu@UOPS_DISPATCHED.CORE\\,cmask\\=1@ / 2 if #SMT_on else cpu@UOPS_DISPATCHED.CORE\\,cmask\\=1@)", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", + "MetricName": "tma_info_core_ilp" }, { "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", "MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + LSD.UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)", "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", - "MetricName": "tma_info_dsb_coverage", - "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 4 > 0.35", + "MetricName": "tma_info_frontend_dsb_coverage", + "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 4 > 0.35", "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_lcp" }, { - "BriefDescription": "The ratio of Executed- by Issued-Uops", - "MetricExpr": "UOPS_DISPATCHED.THREAD / UOPS_ISSUED.ANY", - "MetricGroup": "Cor;Pipeline", - "MetricName": "tma_info_execute_per_issue", - "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." + "BriefDescription": "Total number of retired Instructions", + "MetricExpr": "INST_RETIRED.ANY", + "MetricGroup": "Summary;TmaL1;tma_L1_group", + "MetricName": "tma_info_inst_mix_instructions", + "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" }, { - "BriefDescription": "Floating Point Operations Per Cycle", - "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / tma_info_core_clks", - "MetricGroup": "Flops;Ret", - "MetricName": "tma_info_flopc" + "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@", + "MetricGroup": "Pipeline;Ret", + "MetricName": "tma_info_pipeline_retire" }, { - "BriefDescription": "Giga Floating Point Operations Per Second", - "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / 1e9 / duration_time", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_gflops", - "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time", + "MetricGroup": "Power;Summary", + "MetricName": "tma_info_system_average_frequency" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", - "MetricExpr": "UOPS_DISPATCHED.THREAD / (cpu@UOPS_DISPATCHED.CORE\\,cmask\\=1@ / 2 if #SMT_on else cpu@UOPS_DISPATCHED.CORE\\,cmask\\=1@)", - "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", - "MetricName": "tma_info_ilp" + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricGroup": "HPC;Summary", + "MetricName": "tma_info_system_cpu_utilization" }, { - "BriefDescription": "Total number of retired Instructions", - "MetricExpr": "INST_RETIRED.ANY", - "MetricGroup": "Summary;TmaL1;tma_L1_group", - "MetricName": "tma_info_instructions", - "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" + "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", + "MetricExpr": "64 * (UNC_ARB_TRK_REQUESTS.ALL + UNC_ARB_COH_TRK_REQUESTS.ALL) / 1e6 / duration_time / 1e3", + "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", + "MetricName": "tma_info_system_dram_bw_use", + "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_mem_bandwidth" }, { - "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": "Ret;Summary", - "MetricName": "tma_info_ipc" + "BriefDescription": "Giga Floating Point Operations Per Second", + "MetricExpr": "(FP_COMP_OPS_EXE.SSE_SCALAR_SINGLE + FP_COMP_OPS_EXE.SSE_SCALAR_DOUBLE + 2 * FP_COMP_OPS_EXE.SSE_PACKED_DOUBLE + 4 * (FP_COMP_OPS_EXE.SSE_PACKED_SINGLE + SIMD_FP_256.PACKED_DOUBLE) + 8 * SIMD_FP_256.PACKED_SINGLE) / 1e9 / duration_time", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_system_gflops", + "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." }, { "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", "MetricGroup": "Branches;OS", - "MetricName": "tma_info_ipfarbranch", - "MetricThreshold": "tma_info_ipfarbranch < 1e6" + "MetricName": "tma_info_system_ipfarbranch", + "MetricThreshold": "tma_info_system_ipfarbranch < 1e6" }, { "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", "MetricGroup": "OS", - "MetricName": "tma_info_kernel_cpi" + "MetricName": "tma_info_system_kernel_cpi" }, { "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", "MetricGroup": "OS", - "MetricName": "tma_info_kernel_utilization", - "MetricThreshold": "tma_info_kernel_utilization > 0.05" + "MetricName": "tma_info_system_kernel_utilization", + "MetricThreshold": "tma_info_system_kernel_utilization > 0.05" }, { "BriefDescription": "Average number of parallel requests to external memory", "MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_OCCUPANCY.CYCLES_WITH_ANY_REQUEST", "MetricGroup": "Mem;SoC", - "MetricName": "tma_info_mem_parallel_requests", + "MetricName": "tma_info_system_mem_parallel_requests", "PublicDescription": "Average number of parallel requests to external memory. Accounts for all requests" }, { "BriefDescription": "Average latency of all requests to external memory (in Uncore cycles)", "MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_REQUESTS.ALL", "MetricGroup": "Mem;SoC", - "MetricName": "tma_info_mem_request_latency" - }, - { - "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@", - "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_retire" - }, - { - "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", - "MetricExpr": "4 * tma_info_core_clks", - "MetricGroup": "TmaL1;tma_L1_group", - "MetricName": "tma_info_slots" + "MetricName": "tma_info_system_mem_request_latency" }, { "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / (CPU_CLK_UNHALTED.REF_XCLK_ANY / 2) if #SMT_on else 0)", "MetricGroup": "SMT", - "MetricName": "tma_info_smt_2t_utilization" + "MetricName": "tma_info_system_smt_2t_utilization" }, { "BriefDescription": "Socket actual clocks when any core is active on that socket", "MetricExpr": "UNC_CLOCK.SOCKET", "MetricGroup": "SoC", - "MetricName": "tma_info_socket_clks" + "MetricName": "tma_info_system_socket_clks" }, { "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC", "MetricGroup": "Power", - "MetricName": "tma_info_turbo_utilization" + "MetricName": "tma_info_system_turbo_utilization" + }, + { + "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "Pipeline", + "MetricName": "tma_info_thread_clks" + }, + { + "BriefDescription": "Cycles Per Instruction (per Logical Processor)", + "MetricExpr": "1 / tma_info_thread_ipc", + "MetricGroup": "Mem;Pipeline", + "MetricName": "tma_info_thread_cpi" + }, + { + "BriefDescription": "The ratio of Executed- by Issued-Uops", + "MetricExpr": "UOPS_DISPATCHED.THREAD / UOPS_ISSUED.ANY", + "MetricGroup": "Cor;Pipeline", + "MetricName": "tma_info_thread_execute_per_issue", + "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." + }, + { + "BriefDescription": "Instructions Per Cycle (per Logical Processor)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks", + "MetricGroup": "Ret;Summary", + "MetricName": "tma_info_thread_ipc" + }, + { + "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", + "MetricExpr": "4 * tma_info_core_core_clks", + "MetricGroup": "TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots" }, { "BriefDescription": "Uops Per Instruction", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / INST_RETIRED.ANY", "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_uoppi", - "MetricThreshold": "tma_info_uoppi > 1.05" + "MetricName": "tma_info_thread_uoppi", + "MetricThreshold": "tma_info_thread_uoppi > 1.05" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses", - "MetricExpr": "(12 * ITLB_MISSES.STLB_HIT + ITLB_MISSES.WALK_DURATION) / tma_info_clks", + "MetricExpr": "(12 * ITLB_MISSES.STLB_HIT + ITLB_MISSES.WALK_DURATION) / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -389,7 +397,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core", "MetricConstraint": "NO_GROUP_EVENTS_SMT", - "MetricExpr": "MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_MISC_RETIRED.LLC_MISS) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_clks", + "MetricExpr": "MEM_LOAD_UOPS_RETIRED.LLC_HIT / (MEM_LOAD_UOPS_RETIRED.LLC_HIT + 7 * MEM_LOAD_UOPS_MISC_RETIRED.LLC_MISS) * CYCLE_ACTIVITY.STALLS_L2_PENDING / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -398,11 +406,11 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)", - "MetricExpr": "ILD_STALL.LCP / tma_info_clks", + "MetricExpr": "ILD_STALL.LCP / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_lcp", "MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage", + "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage", "ScaleUnit": "100%" }, { @@ -411,6 +419,7 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_light_operations", "MetricThreshold": "tma_light_operations > 0.6", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation). This correlates with total number of instructions used by the program. A uops-per-instruction (see UopPI metric) ratio of 1 or less should be expected for decently optimized software running on Intel Core/Xeon products. While this often indicates efficient X86 instructions were executed; high value does not necessarily mean better performance cannot be achieved. Sample with: INST_RETIRED.PREC_DIST", "ScaleUnit": "100%" }, @@ -421,21 +430,22 @@ "MetricGroup": "BadSpec;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.1 & tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears. These slots are either wasted by uops fetched prior to the clear; or stalls the out-of-order portion of the machine needs to recover its state after the clear. For example; this can happen due to memory ordering Nukes (e.g. Memory Disambiguation) or Self-Modifying-Code (SMC) nukes. Sample with: MACHINE_CLEARS.COUNT. Related metrics: tma_clears_resteers, tma_l1_bound, tma_microcode_sequencer, tma_ms_switches, tma_remote_cache", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=6@) / tma_info_clks", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=6@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW", "MetricName": "tma_mem_bandwidth", "MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_info_dram_bw_use", + "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_info_system_dram_bw_use", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat", "MetricName": "tma_mem_latency", "MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -445,16 +455,17 @@ { "BriefDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_L1D_PENDING) + RESOURCE_STALLS.SB) / (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_DISPATCH) + cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=1@ - (cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=3@ if tma_info_ipc > 1.8 else cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", + "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_L1D_PENDING) + RESOURCE_STALLS.SB) / (min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_DISPATCH) + cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=1@ - (cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=3@ if tma_info_thread_ipc > 1.8 else cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB) * tma_backend_bound", "MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck. Memory Bound estimates fraction of slots where pipeline is likely stalled due to demand load or store instructions. This accounts mainly for (1) non-completed in-flight memory demand loads which coincides with execution units starvation; in addition to (2) cases where stores could impose backpressure on the pipeline when many of them get buffered at the same time (less common out of the two).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_thread_slots", "MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS", "MetricName": "tma_microcode_sequencer", "MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1", @@ -463,7 +474,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)", - "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_clks", + "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_thread_clks", "MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO", "MetricName": "tma_ms_switches", "MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -473,7 +484,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_DISPATCH) + cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=1@ - (cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=3@ if tma_info_ipc > 1.8 else cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_L1D_PENDING)) / tma_info_clks", + "MetricExpr": "(min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.CYCLES_NO_DISPATCH) + cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=1@ - (cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=3@ if tma_info_thread_ipc > 1.8 else cpu@UOPS_DISPATCHED.THREAD\\,cmask\\=2@) - (RS_EVENTS.EMPTY_CYCLES if tma_fetch_latency > 0.1 else 0) + RESOURCE_STALLS.SB - RESOURCE_STALLS.SB - min(CPU_CLK_UNHALTED.THREAD, CYCLE_ACTIVITY.STALLS_L1D_PENDING)) / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_ports_utilization", "MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -482,16 +493,17 @@ }, { "BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired. Ideally; all pipeline slots would be attributed to the Retiring category. Retiring of 100% would indicate the maximum Pipeline_Width throughput was achieved. Maximizing Retiring typically increases the Instructions-per-cycle (see IPC metric). Note that a high Retiring value does not necessary mean there is no room for more performance. For example; Heavy-operations or Microcode Assists are categorized under Retiring. They often indicate suboptimal performance and can often be optimized or avoided. Sample with: UOPS_RETIRED.RETIRE_SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write", - "MetricExpr": "RESOURCE_STALLS.SB / tma_info_clks", + "MetricExpr": "RESOURCE_STALLS.SB / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", diff --git a/tools/perf/pmu-events/arch/x86/sapphirerapids/memory.json b/tools/perf/pmu-events/arch/x86/sapphirerapids/memory.json index b72a36999930..e8bf7c9c44e1 100644 --- a/tools/perf/pmu-events/arch/x86/sapphirerapids/memory.json +++ b/tools/perf/pmu-events/arch/x86/sapphirerapids/memory.json @@ -32,18 +32,20 @@ "UMask": "0x3" }, { - "BriefDescription": "MEMORY_ACTIVITY.STALLS_L2_MISS", + "BriefDescription": "Execution stalls while L2 cache miss demand cacheable load request is outstanding.", "CounterMask": "5", "EventCode": "0x47", "EventName": "MEMORY_ACTIVITY.STALLS_L2_MISS", + "PublicDescription": "Execution stalls while L2 cache miss demand cacheable load request is outstanding (will not count for uncacheable demand requests e.g. bus lock).", "SampleAfterValue": "1000003", "UMask": "0x5" }, { - "BriefDescription": "MEMORY_ACTIVITY.STALLS_L3_MISS", + "BriefDescription": "Execution stalls while L3 cache miss demand cacheable load request is outstanding.", "CounterMask": "9", "EventCode": "0x47", "EventName": "MEMORY_ACTIVITY.STALLS_L3_MISS", + "PublicDescription": "Execution stalls while L3 cache miss demand cacheable load request is outstanding (will not count for uncacheable demand requests e.g. bus lock).", "SampleAfterValue": "1000003", "UMask": "0x9" }, diff --git a/tools/perf/pmu-events/arch/x86/sapphirerapids/metricgroups.json b/tools/perf/pmu-events/arch/x86/sapphirerapids/metricgroups.json new file mode 100644 index 000000000000..e6f7934320bf --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/sapphirerapids/metricgroups.json @@ -0,0 +1,118 @@ +{ + "Backend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Bad": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BadSpec": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BigFoot": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BrMispredicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Branches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CacheMisses": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CodeGen": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Compute": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Cor": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSBmiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DataSharing": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Fed": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Flops": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpScalar": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Frontend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "HPC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IcMiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "InsType": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IntVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IoBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "L2Evicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "LSD": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MachineClears": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBound": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryTLB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_BW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_Lat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MicroSeq": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "OS": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Offcore": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PGO": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Pipeline": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PortsUtil": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Prefetches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Ret": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Retire": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SMT": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Server": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Snoop": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SoC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL1": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL2": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL3mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "TopdownL5": "Metrics for top-down breakdown at level 5", + "TopdownL6": "Metrics for top-down breakdown at level 6", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_L5_group": "Metrics for top-down breakdown at level 5", + "tma_L6_group": "Metrics for top-down breakdown at level 6", + "tma_alu_op_utilization_group": "Metrics contributing to tma_alu_op_utilization category", + "tma_assists_group": "Metrics contributing to tma_assists category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_branch_resteers_group": "Metrics contributing to tma_branch_resteers category", + "tma_core_bound_group": "Metrics contributing to tma_core_bound category", + "tma_dram_bound_group": "Metrics contributing to tma_dram_bound category", + "tma_dtlb_load_group": "Metrics contributing to tma_dtlb_load category", + "tma_dtlb_store_group": "Metrics contributing to tma_dtlb_store category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_fp_arith_group": "Metrics contributing to tma_fp_arith category", + "tma_fp_vector_group": "Metrics contributing to tma_fp_vector category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_heavy_operations_group": "Metrics contributing to tma_heavy_operations category", + "tma_int_operations_group": "Metrics contributing to tma_int_operations category", + "tma_issue2P": "Metrics related by the issue $issue2P", + "tma_issueBC": "Metrics related by the issue $issueBC", + "tma_issueBM": "Metrics related by the issue $issueBM", + "tma_issueBW": "Metrics related by the issue $issueBW", + "tma_issueD0": "Metrics related by the issue $issueD0", + "tma_issueFB": "Metrics related by the issue $issueFB", + "tma_issueFL": "Metrics related by the issue $issueFL", + "tma_issueL1": "Metrics related by the issue $issueL1", + "tma_issueLat": "Metrics related by the issue $issueLat", + "tma_issueMC": "Metrics related by the issue $issueMC", + "tma_issueMS": "Metrics related by the issue $issueMS", + "tma_issueMV": "Metrics related by the issue $issueMV", + "tma_issueRFO": "Metrics related by the issue $issueRFO", + "tma_issueSL": "Metrics related by the issue $issueSL", + "tma_issueSO": "Metrics related by the issue $issueSO", + "tma_issueSmSt": "Metrics related by the issue $issueSmSt", + "tma_issueSpSt": "Metrics related by the issue $issueSpSt", + "tma_issueSyncxn": "Metrics related by the issue $issueSyncxn", + "tma_issueTLB": "Metrics related by the issue $issueTLB", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_l3_bound_group": "Metrics contributing to tma_l3_bound category", + "tma_light_operations_group": "Metrics contributing to tma_light_operations category", + "tma_load_op_utilization_group": "Metrics contributing to tma_load_op_utilization category", + "tma_mem_bandwidth_group": "Metrics contributing to tma_mem_bandwidth category", + "tma_mem_latency_group": "Metrics contributing to tma_mem_latency category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_microcode_sequencer_group": "Metrics contributing to tma_microcode_sequencer category", + "tma_mite_group": "Metrics contributing to tma_mite category", + "tma_ports_utilization_group": "Metrics contributing to tma_ports_utilization category", + "tma_ports_utilized_0_group": "Metrics contributing to tma_ports_utilized_0 category", + "tma_ports_utilized_3m_group": "Metrics contributing to tma_ports_utilized_3m category", + "tma_retiring_group": "Metrics contributing to tma_retiring category", + "tma_serializing_operation_group": "Metrics contributing to tma_serializing_operation category", + "tma_store_bound_group": "Metrics contributing to tma_store_bound category", + "tma_store_op_utilization_group": "Metrics contributing to tma_store_op_utilization category" +} diff --git a/tools/perf/pmu-events/arch/x86/sapphirerapids/spr-metrics.json b/tools/perf/pmu-events/arch/x86/sapphirerapids/spr-metrics.json index 126300b7ae77..c732982f70b5 100644 --- a/tools/perf/pmu-events/arch/x86/sapphirerapids/spr-metrics.json +++ b/tools/perf/pmu-events/arch/x86/sapphirerapids/spr-metrics.json @@ -29,11 +29,262 @@ }, { "BriefDescription": "Uncore frequency per die [GHZ]", - "MetricExpr": "tma_info_socket_clks / #num_dies / duration_time / 1e9", + "MetricExpr": "tma_info_system_socket_clks / #num_dies / duration_time / 1e9", "MetricGroup": "SoC", "MetricName": "UNCORE_FREQ" }, { + "BriefDescription": "Cycles per instruction retired; indicating how much time each executed instruction took; in units of cycles.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD / INST_RETIRED.ANY", + "MetricName": "cpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "CPU operating frequency (in GHz)", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD / CPU_CLK_UNHALTED.REF_TSC * #SYSTEM_TSC_FREQ / 1e9", + "MetricName": "cpu_operating_frequency", + "ScaleUnit": "1GHz" + }, + { + "BriefDescription": "Percentage of time spent in the active CPU power state C0", + "MetricExpr": "tma_info_system_cpu_utilization", + "MetricName": "cpu_utilization", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for 2 megabyte page sizes) caused by demand data loads to the total number of completed instructions", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_COMPLETED_2M_4M / INST_RETIRED.ANY", + "MetricName": "dtlb_2nd_level_2mb_large_page_load_mpi", + "PublicDescription": "Ratio of number of completed page walks (for 2 megabyte page sizes) caused by demand data loads to the total number of completed instructions. This implies it missed in the Data Translation Lookaside Buffer (DTLB) and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data loads to the total number of completed instructions", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "dtlb_2nd_level_load_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data loads to the total number of completed instructions. This implies it missed in the DTLB and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data stores to the total number of completed instructions", + "MetricExpr": "DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "dtlb_2nd_level_store_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data stores to the total number of completed instructions. This implies it missed in the DTLB and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Bandwidth of IO reads that are initiated by end device controllers that are requesting memory from the CPU.", + "MetricExpr": "UNC_CHA_TOR_INSERTS.IO_PCIRDCUR * 64 / 1e6 / duration_time", + "MetricName": "io_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Bandwidth of IO writes that are initiated by end device controllers that are writing memory to the CPU.", + "MetricExpr": "(UNC_CHA_TOR_INSERTS.IO_ITOM + UNC_CHA_TOR_INSERTS.IO_ITOMCACHENEAR) * 64 / 1e6 / duration_time", + "MetricName": "io_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for 2 megabyte and 4 megabyte page sizes) caused by a code fetch to the total number of completed instructions", + "MetricExpr": "ITLB_MISSES.WALK_COMPLETED_2M_4M / INST_RETIRED.ANY", + "MetricName": "itlb_2nd_level_large_page_mpi", + "PublicDescription": "Ratio of number of completed page walks (for 2 megabyte and 4 megabyte page sizes) caused by a code fetch to the total number of completed instructions. This implies it missed in the Instruction Translation Lookaside Buffer (ITLB) and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by a code fetch to the total number of completed instructions", + "MetricExpr": "ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "itlb_2nd_level_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by a code fetch to the total number of completed instructions. This implies it missed in the ITLB (Instruction TLB) and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read requests missing in L1 instruction cache (includes prefetches) to the total number of completed instructions", + "MetricExpr": "L2_RQSTS.ALL_CODE_RD / INST_RETIRED.ANY", + "MetricName": "l1_i_code_read_misses_with_prefetches_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of demand load requests hitting in L1 data cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_RETIRED.L1_HIT / INST_RETIRED.ANY", + "MetricName": "l1d_demand_data_read_hits_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of requests missing L1 data cache (includes data+rfo w/ prefetches) to the total number of completed instructions", + "MetricExpr": "L1D.REPLACEMENT / INST_RETIRED.ANY", + "MetricName": "l1d_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read request missing L2 cache to the total number of completed instructions", + "MetricExpr": "L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", + "MetricName": "l2_demand_code_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed demand load requests hitting in L2 cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_RETIRED.L2_HIT / INST_RETIRED.ANY", + "MetricName": "l2_demand_data_read_hits_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed data read request missing L2 cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_RETIRED.L2_MISS / INST_RETIRED.ANY", + "MetricName": "l2_demand_data_read_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of requests missing L2 cache (includes code+data+rfo w/ prefetches) to the total number of completed instructions", + "MetricExpr": "L2_LINES_IN.ALL / INST_RETIRED.ANY", + "MetricName": "l2_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read requests missing last level core cache (includes demand w/ prefetches) to the total number of completed instructions", + "MetricExpr": "UNC_CHA_TOR_INSERTS.IA_MISS_CRD / INST_RETIRED.ANY", + "MetricName": "llc_code_read_mpi_demand_plus_prefetch", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of data read requests missing last level core cache (includes demand w/ prefetches) to the total number of completed instructions", + "MetricExpr": "(UNC_CHA_TOR_INSERTS.IA_MISS_LLCPREFDATA + UNC_CHA_TOR_INSERTS.IA_MISS_DRD + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PREF) / INST_RETIRED.ANY", + "MetricName": "llc_data_read_mpi_demand_plus_prefetch", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand data read miss (read memory access) in nano seconds", + "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD / UNC_CHA_TOR_INSERTS.IA_MISS_DRD) / (UNC_CHA_CLOCKTICKS / (source_count(UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD) * #num_packages)) * duration_time", + "MetricName": "llc_demand_data_read_miss_latency", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand data read miss (read memory access) addressed to local memory in nano seconds", + "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_LOCAL / UNC_CHA_TOR_INSERTS.IA_MISS_DRD_LOCAL) / (UNC_CHA_CLOCKTICKS / (source_count(UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_LOCAL) * #num_packages)) * duration_time", + "MetricName": "llc_demand_data_read_miss_latency_for_local_requests", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand data read miss (read memory access) addressed to remote memory in nano seconds", + "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE / UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE) / (UNC_CHA_CLOCKTICKS / (source_count(UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_REMOTE) * #num_packages)) * duration_time", + "MetricName": "llc_demand_data_read_miss_latency_for_remote_requests", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand data read miss (read memory access) addressed to DRAM in nano seconds", + "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_DDR / UNC_CHA_TOR_INSERTS.IA_MISS_DRD_DDR) / (UNC_CHA_CLOCKTICKS / (source_count(UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_DDR) * #num_packages)) * duration_time", + "MetricName": "llc_demand_data_read_miss_to_dram_latency", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand data read miss (read memory access) addressed to Intel(R) Optane(TM) Persistent Memory(PMEM) in nano seconds", + "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_PMM / UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PMM) / (UNC_CHA_CLOCKTICKS / (source_count(UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_PMM) * #num_packages)) * duration_time", + "MetricName": "llc_demand_data_read_miss_to_pmem_latency", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Bandwidth (MB/sec) of read requests that miss the last level cache (LLC) and go to local memory.", + "MetricExpr": "UNC_CHA_REQUESTS.READS_LOCAL * 64 / 1e6 / duration_time", + "MetricName": "llc_miss_local_memory_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Bandwidth (MB/sec) of write requests that miss the last level cache (LLC) and go to local memory.", + "MetricExpr": "UNC_CHA_REQUESTS.WRITES_LOCAL * 64 / 1e6 / duration_time", + "MetricName": "llc_miss_local_memory_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Bandwidth (MB/sec) of read requests that miss the last level cache (LLC) and go to remote memory.", + "MetricExpr": "UNC_CHA_REQUESTS.READS_REMOTE * 64 / 1e6 / duration_time", + "MetricName": "llc_miss_remote_memory_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Bandwidth (MB/sec) of write requests that miss the last level cache (LLC) and go to remote memory.", + "MetricExpr": "UNC_CHA_REQUESTS.WRITES_REMOTE * 64 / 1e6 / duration_time", + "MetricName": "llc_miss_remote_memory_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "The ratio of number of completed memory load instructions to the total number completed instructions", + "MetricExpr": "MEM_INST_RETIRED.ALL_LOADS / INST_RETIRED.ANY", + "MetricName": "loads_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "DDR memory read bandwidth (MB/sec)", + "MetricExpr": "UNC_M_CAS_COUNT.RD * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "DDR memory bandwidth (MB/sec)", + "MetricExpr": "(UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_total", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "DDR memory write bandwidth (MB/sec)", + "MetricExpr": "UNC_M_CAS_COUNT.WR * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Memory write bandwidth (MB/sec) caused by directory updates; includes DDR and Intel(R) Optane(TM) Persistent Memory(PMEM).", + "MetricExpr": "(UNC_CHA_DIR_UPDATE.HA + UNC_CHA_DIR_UPDATE.TOR + UNC_M2M_DIRECTORY_UPDATE.ANY) * 64 / 1e6 / duration_time", + "MetricName": "memory_extra_write_bw_due_to_directory_updates", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Memory read that miss the last level cache (LLC) addressed to local DRAM as a percentage of total memory read accesses, does not include LLC prefetches.", + "MetricExpr": "(UNC_CHA_TOR_INSERTS.IA_MISS_DRD_LOCAL + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PREF_LOCAL) / (UNC_CHA_TOR_INSERTS.IA_MISS_DRD_LOCAL + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PREF_LOCAL + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PREF_REMOTE)", + "MetricName": "numa_reads_addressed_to_local_dram", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Memory reads that miss the last level cache (LLC) addressed to remote DRAM as a percentage of total memory read accesses, does not include LLC prefetches.", + "MetricExpr": "(UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PREF_REMOTE) / (UNC_CHA_TOR_INSERTS.IA_MISS_DRD_LOCAL + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PREF_LOCAL + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_REMOTE + UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PREF_REMOTE)", + "MetricName": "numa_reads_addressed_to_remote_dram", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from decoded instruction cache (decoded stream buffer or DSB) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS + LSD.UOPS)", + "MetricName": "percent_uops_delivered_from_decoded_icache", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from legacy decode pipeline (Micro-instruction Translation Engine or MITE) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.MITE_UOPS / (IDQ.DSB_UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS + LSD.UOPS)", + "MetricName": "percent_uops_delivered_from_legacy_decode_pipeline", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from microcode sequencer (MS) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.MS_UOPS / (IDQ.DSB_UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS + LSD.UOPS)", + "MetricName": "percent_uops_delivered_from_microcode_sequencer", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Intel(R) Optane(TM) Persistent Memory(PMEM) memory read bandwidth (MB/sec)", + "MetricExpr": "UNC_M_PMM_RPQ_INSERTS * 64 / 1e6 / duration_time", + "MetricName": "pmem_memory_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Intel(R) Optane(TM) Persistent Memory(PMEM) memory bandwidth (MB/sec)", + "MetricExpr": "(UNC_M_PMM_RPQ_INSERTS + UNC_M_PMM_WPQ_INSERTS) * 64 / 1e6 / duration_time", + "MetricName": "pmem_memory_bandwidth_total", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Intel(R) Optane(TM) Persistent Memory(PMEM) memory write bandwidth (MB/sec)", + "MetricExpr": "UNC_M_PMM_WPQ_INSERTS * 64 / 1e6 / duration_time", + "MetricName": "pmem_memory_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { "BriefDescription": "Percentage of cycles spent in System Management Interrupts.", "MetricExpr": "((msr@aperf@ - cycles) / msr@aperf@ if msr@smi@ > 0 else 0)", "MetricGroup": "smi", @@ -49,8 +300,14 @@ "ScaleUnit": "1SMI#" }, { + "BriefDescription": "The ratio of number of completed memory store instructions to the total number completed instructions", + "MetricExpr": "MEM_INST_RETIRED.ALL_STORES / INST_RETIRED.ANY", + "MetricName": "stores_per_instr", + "ScaleUnit": "1per_instr" + }, + { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution ports for ALU operations.", - "MetricExpr": "(UOPS_DISPATCHED.PORT_0 + UOPS_DISPATCHED.PORT_1 + UOPS_DISPATCHED.PORT_5_11 + UOPS_DISPATCHED.PORT_6) / (5 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED.PORT_0 + UOPS_DISPATCHED.PORT_1 + UOPS_DISPATCHED.PORT_5_11 + UOPS_DISPATCHED.PORT_6) / (5 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_alu_op_utilization", "MetricThreshold": "tma_alu_op_utilization > 0.6", @@ -58,7 +315,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles where the Advanced Matrix Extensions (AMX) execution engine was busy with tile (arithmetic) operations", - "MetricExpr": "EXE.AMX_BUSY / tma_info_core_clks", + "MetricExpr": "EXE.AMX_BUSY / tma_info_core_core_clks", "MetricGroup": "Compute;HPC;Server;TopdownL5;tma_L5_group;tma_ports_utilized_0_group", "MetricName": "tma_amx_busy", "MetricThreshold": "tma_amx_busy > 0.5 & (tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)))", @@ -66,7 +323,7 @@ }, { "BriefDescription": "This metric estimates fraction of slots the CPU retired uops delivered by the Microcode_Sequencer as a result of Assists", - "MetricExpr": "100 * cpu@ASSISTS.ANY\\,umask\\=0x1B@ / tma_info_slots", + "MetricExpr": "100 * cpu@ASSISTS.ANY\\,umask\\=0x1B@ / tma_info_thread_slots", "MetricGroup": "TopdownL4;tma_L4_group;tma_microcode_sequencer_group", "MetricName": "tma_assists", "MetricThreshold": "tma_assists > 0.1 & (tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1)", @@ -75,7 +332,7 @@ }, { "BriefDescription": "This metric estimates fraction of slots the CPU retired uops as a result of handing SSE to AVX* or AVX* to SSE transition Assists.", - "MetricExpr": "63 * ASSISTS.SSE_AVX_MIX / tma_info_slots", + "MetricExpr": "63 * ASSISTS.SSE_AVX_MIX / tma_info_thread_slots", "MetricGroup": "HPC;TopdownL5;tma_L5_group;tma_assists_group", "MetricName": "tma_avx_assists", "MetricThreshold": "tma_avx_assists > 0.1", @@ -83,34 +340,40 @@ }, { "BriefDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend", - "MetricExpr": "topdown\\-be\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_slots", - "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "topdown\\-be\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_thread_slots", + "MetricGroup": "Default;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend. Backend is the portion of the processor core where the out-of-order scheduler dispatches ready uops into their respective execution units; and once completed these uops get retired according to program order. For example; stalls due to data-cache misses or stalls due to the divider unit being overloaded are both categorized under Backend Bound. Backend Bound is further divided into two main categories: Memory Bound and Core Bound. Sample with: TOPDOWN.BACKEND_BOUND_SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations", + "DefaultMetricgroupName": "TopdownL1", "MetricExpr": "max(1 - (tma_frontend_bound + tma_backend_bound + tma_retiring), 0)", - "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", + "MetricGroup": "Default;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots wasted due to incorrect speculations. This include slots used to issue uops that do not eventually get retired and slots for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work due to miss-predicted branches are categorized under Bad Speculation category. Incorrect data speculation followed by Memory Ordering Nukes is another example.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction", - "MetricExpr": "topdown\\-br\\-mispredict / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_slots", - "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", + "DefaultMetricgroupName": "TopdownL2", + "MetricExpr": "topdown\\-br\\-mispredict / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_thread_slots", + "MetricGroup": "BadSpec;BrMispredicts;Default;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15", - "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: TOPDOWN.BR_MISPREDICT_SLOTS. Related metrics: tma_info_branch_misprediction_cost, tma_info_mispredictions, tma_mispredicts_resteers", + "MetricgroupNoGroup": "TopdownL2;Default", + "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: TOPDOWN.BR_MISPREDICT_SLOTS. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_info_bottleneck_mispredictions, tma_mispredicts_resteers", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers", - "MetricExpr": "INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks + tma_unknown_branches", + "MetricExpr": "INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks + tma_unknown_branches", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteers", "MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -128,7 +391,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Machine Clears", - "MetricExpr": "(1 - tma_branch_mispredicts / tma_bad_speculation) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks", + "MetricExpr": "(1 - tma_branch_mispredicts / tma_bad_speculation) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks", "MetricGroup": "BadSpec;MachineClears;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueMC", "MetricName": "tma_clears_resteers", "MetricThreshold": "tma_clears_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", @@ -138,7 +401,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to contested accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(76 * tma_info_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_FWD * (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM / (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM + OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) + 75.5 * tma_info_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "(76 * tma_info_system_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_FWD * (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM / (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM + OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) + 75.5 * tma_info_system_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_contested_accesses", "MetricThreshold": "tma_contested_accesses > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -147,17 +410,19 @@ }, { "BriefDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck", + "DefaultMetricgroupName": "TopdownL2", "MetricExpr": "max(0, tma_backend_bound - tma_memory_bound)", - "MetricGroup": "Backend;Compute;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", + "MetricGroup": "Backend;Compute;Default;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2;Default", "PublicDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck. Shortage in hardware compute resources; or dependencies in software's instructions are both categorized under Core Bound. Hence it may indicate the machine ran out of an out-of-order resource; certain execution units are overloaded or dependencies in program's data- or instruction-flow are limiting the performance (e.g. FP-chained long-latency arithmetic operations).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to data-sharing accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "75.5 * tma_info_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_NO_FWD + MEM_LOAD_L3_HIT_RETIRED.XSNP_FWD * (1 - OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM / (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM + OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "75.5 * tma_info_system_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_NO_FWD + MEM_LOAD_L3_HIT_RETIRED.XSNP_FWD * (1 - OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM / (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM + OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_data_sharing", "MetricThreshold": "tma_data_sharing > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -166,16 +431,16 @@ }, { "BriefDescription": "This metric represents fraction of cycles where decoder-0 was the only active decoder", - "MetricExpr": "(cpu@INST_DECODED.DECODERS\\,cmask\\=1@ - cpu@INST_DECODED.DECODERS\\,cmask\\=2@) / tma_info_core_clks / 2", + "MetricExpr": "(cpu@INST_DECODED.DECODERS\\,cmask\\=1@ - cpu@INST_DECODED.DECODERS\\,cmask\\=2@) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL4;tma_L4_group;tma_issueD0;tma_mite_group", "MetricName": "tma_decoder0_alone", - "MetricThreshold": "tma_decoder0_alone > 0.1 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 6 > 0.35))", + "MetricThreshold": "tma_decoder0_alone > 0.1 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 6 > 0.35))", "PublicDescription": "This metric represents fraction of cycles where decoder-0 was the only active decoder. Related metrics: tma_few_uops_instructions", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles where the Divider unit was active", - "MetricExpr": "ARITH.DIV_ACTIVE / tma_info_clks", + "MetricExpr": "ARITH.DIV_ACTIVE / tma_info_thread_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_divider", "MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -185,7 +450,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(MEMORY_ACTIVITY.STALLS_L3_MISS / tma_info_clks - tma_pmm_bound if #has_pmem > 0 else MEMORY_ACTIVITY.STALLS_L3_MISS / tma_info_clks)", + "MetricExpr": "(MEMORY_ACTIVITY.STALLS_L3_MISS / tma_info_thread_clks - tma_pmm_bound if #has_pmem > 0 else MEMORY_ACTIVITY.STALLS_L3_MISS / tma_info_thread_clks)", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -194,43 +459,43 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline", - "MetricExpr": "(IDQ.DSB_CYCLES_ANY - IDQ.DSB_CYCLES_OK) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.DSB_CYCLES_ANY - IDQ.DSB_CYCLES_OK) / tma_info_core_core_clks / 2", "MetricGroup": "DSB;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_dsb", - "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 6 > 0.35)", + "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 6 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline. For example; inefficient utilization of the DSB cache structure or bank conflict when reading from it; are categorized here.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_dsb_switches", "MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb, tma_lcp", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses", - "MetricExpr": "min(7 * cpu@DTLB_LOAD_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_LOAD_MISSES.WALK_ACTIVE, max(CYCLE_ACTIVITY.CYCLES_MEM_ANY - MEMORY_ACTIVITY.CYCLES_L1D_MISS, 0)) / tma_info_clks", + "MetricExpr": "min(7 * cpu@DTLB_LOAD_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_LOAD_MISSES.WALK_ACTIVE, max(CYCLE_ACTIVITY.CYCLES_MEM_ANY - MEMORY_ACTIVITY.CYCLES_L1D_MISS, 0)) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group", "MetricName": "tma_dtlb_load", "MetricThreshold": "tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses. TLBs (Translation Look-aside Buffers) are processor caches for recently used entries out of the Page Tables that are used to map virtual- to physical-addresses by the operating system. This metric approximates the potential delay of demand loads missing the first-level data TLB (assuming worst case scenario with back to back misses to different pages). This includes hitting in the second-level TLB (STLB) as well as performing a hardware page walk on an STLB miss. Sample with: MEM_INST_RETIRED.STLB_MISS_LOADS_PS. Related metrics: tma_dtlb_store, tma_info_memory_data_tlbs", + "PublicDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses. TLBs (Translation Look-aside Buffers) are processor caches for recently used entries out of the Page Tables that are used to map virtual- to physical-addresses by the operating system. This metric approximates the potential delay of demand loads missing the first-level data TLB (assuming worst case scenario with back to back misses to different pages). This includes hitting in the second-level TLB (STLB) as well as performing a hardware page walk on an STLB miss. Sample with: MEM_INST_RETIRED.STLB_MISS_LOADS_PS. Related metrics: tma_dtlb_store, tma_info_bottleneck_memory_data_tlbs", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses", - "MetricExpr": "(7 * cpu@DTLB_STORE_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_STORE_MISSES.WALK_ACTIVE) / tma_info_core_clks", + "MetricExpr": "(7 * cpu@DTLB_STORE_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_STORE_MISSES.WALK_ACTIVE) / tma_info_core_core_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_store_bound_group", "MetricName": "tma_dtlb_store", "MetricThreshold": "tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses. As with ordinary data caching; focus on improving data locality and reducing working-set size to reduce DTLB overhead. Additionally; consider using profile-guided optimization (PGO) to collocate frequently-used data on the same page. Try using larger page sizes for large amounts of frequently-used data. Sample with: MEM_INST_RETIRED.STLB_MISS_STORES_PS. Related metrics: tma_dtlb_load, tma_info_memory_data_tlbs", + "PublicDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses. As with ordinary data caching; focus on improving data locality and reducing working-set size to reduce DTLB overhead. Additionally; consider using profile-guided optimization (PGO) to collocate frequently-used data on the same page. Try using larger page sizes for large amounts of frequently-used data. Sample with: MEM_INST_RETIRED.STLB_MISS_STORES_PS. Related metrics: tma_dtlb_load, tma_info_bottleneck_memory_data_tlbs", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates how often CPU was handling synchronizations due to False Sharing", - "MetricExpr": "80 * tma_info_average_frequency * OCR.DEMAND_RFO.L3_HIT.SNOOP_HITM / tma_info_clks", + "MetricExpr": "80 * tma_info_system_average_frequency * OCR.DEMAND_RFO.L3_HIT.SNOOP_HITM / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_store_bound_group", "MetricName": "tma_false_sharing", "MetricThreshold": "tma_false_sharing > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -239,28 +504,32 @@ }, { "BriefDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed", - "MetricExpr": "L1D_PEND_MISS.FB_FULL / tma_info_clks", + "MetricExpr": "L1D_PEND_MISS.FB_FULL / tma_info_thread_clks", "MetricGroup": "MemoryBW;TopdownL4;tma_L4_group;tma_issueBW;tma_issueSL;tma_issueSmSt;tma_l1_bound_group", "MetricName": "tma_fb_full", "MetricThreshold": "tma_fb_full > 0.3", - "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", + "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues", + "DefaultMetricgroupName": "TopdownL2", "MetricExpr": "max(0, tma_frontend_bound - tma_fetch_latency)", - "MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", + "MetricGroup": "Default;FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", "MetricName": "tma_fetch_bandwidth", - "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 6 > 0.35", - "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb, tma_lcp", + "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 6 > 0.35", + "MetricgroupNoGroup": "TopdownL2;Default", + "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues", - "MetricExpr": "topdown\\-fetch\\-lat / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) - INT_MISC.UOP_DROPPING / tma_info_slots", - "MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", + "DefaultMetricgroupName": "TopdownL2", + "MetricExpr": "topdown\\-fetch\\-lat / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) - INT_MISC.UOP_DROPPING / tma_info_thread_slots", + "MetricGroup": "Default;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL2;Default", "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: FRONTEND_RETIRED.LATENCY_GE_16_PS;FRONTEND_RETIRED.LATENCY_GE_8_PS", "ScaleUnit": "100%" }, @@ -275,7 +544,7 @@ }, { "BriefDescription": "This metric approximates arithmetic floating-point (FP) matrix uops fraction the CPU has retired (aggregated across all supported FP datatypes in AMX engine)", - "MetricExpr": "cpu@AMX_OPS_RETIRED.BF16\\,cmask\\=1@ / (tma_retiring * tma_info_slots)", + "MetricExpr": "cpu@AMX_OPS_RETIRED.BF16\\,cmask\\=1@ / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;HPC;Pipeline;Server;TopdownL4;tma_L4_group;tma_fp_arith_group", "MetricName": "tma_fp_amx", "MetricThreshold": "tma_fp_amx > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6)", @@ -284,6 +553,7 @@ }, { "BriefDescription": "This metric represents overall arithmetic floating-point (FP) operations fraction the CPU has executed (retired)", + "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "tma_x87_use + tma_fp_scalar + tma_fp_vector + tma_fp_amx", "MetricGroup": "HPC;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_fp_arith", @@ -293,7 +563,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of slots the CPU retired uops as a result of handing Floating Point (FP) Assists", - "MetricExpr": "30 * ASSISTS.FP / tma_info_slots", + "MetricExpr": "30 * ASSISTS.FP / tma_info_thread_slots", "MetricGroup": "HPC;TopdownL5;tma_L5_group;tma_assists_group", "MetricName": "tma_fp_assists", "MetricThreshold": "tma_fp_assists > 0.1", @@ -302,7 +572,7 @@ }, { "BriefDescription": "This metric approximates arithmetic floating-point (FP) scalar uops fraction the CPU has retired", - "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + FP_ARITH_INST_RETIRED2.SCALAR) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + FP_ARITH_INST_RETIRED2.SCALAR) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL4;tma_L4_group;tma_fp_arith_group;tma_issue2P", "MetricName": "tma_fp_scalar", "MetricThreshold": "tma_fp_scalar > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6)", @@ -311,7 +581,7 @@ }, { "BriefDescription": "This metric approximates arithmetic floating-point (FP) vector uops fraction the CPU has retired aggregated across all vector widths", - "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@ + FP_ARITH_INST_RETIRED2.VECTOR) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@ + FP_ARITH_INST_RETIRED2.VECTOR) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL4;tma_L4_group;tma_fp_arith_group;tma_issue2P", "MetricName": "tma_fp_vector", "MetricThreshold": "tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6)", @@ -320,7 +590,7 @@ }, { "BriefDescription": "This metric approximates arithmetic FP vector uops fraction the CPU has retired for 128-bit wide vectors", - "MetricExpr": "(FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED2.128B_PACKED_HALF) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED2.128B_PACKED_HALF) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL5;tma_L5_group;tma_fp_vector_group;tma_issue2P", "MetricName": "tma_fp_vector_128b", "MetricThreshold": "tma_fp_vector_128b > 0.1 & (tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6))", @@ -329,7 +599,7 @@ }, { "BriefDescription": "This metric approximates arithmetic FP vector uops fraction the CPU has retired for 256-bit wide vectors", - "MetricExpr": "(FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED2.256B_PACKED_HALF) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED2.256B_PACKED_HALF) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL5;tma_L5_group;tma_fp_vector_group;tma_issue2P", "MetricName": "tma_fp_vector_256b", "MetricThreshold": "tma_fp_vector_256b > 0.1 & (tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6))", @@ -338,7 +608,7 @@ }, { "BriefDescription": "This metric approximates arithmetic FP vector uops fraction the CPU has retired for 512-bit wide vectors", - "MetricExpr": "(FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE + FP_ARITH_INST_RETIRED2.512B_PACKED_HALF) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE + FP_ARITH_INST_RETIRED2.512B_PACKED_HALF) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL5;tma_L5_group;tma_fp_vector_group;tma_issue2P", "MetricName": "tma_fp_vector_512b", "MetricThreshold": "tma_fp_vector_512b > 0.1 & (tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6))", @@ -347,16 +617,18 @@ }, { "BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend", - "MetricExpr": "topdown\\-fe\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) - INT_MISC.UOP_DROPPING / tma_info_slots", - "MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "topdown\\-fe\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) - INT_MISC.UOP_DROPPING / tma_info_thread_slots", + "MetricGroup": "Default;PGO;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound. Sample with: FRONTEND_RETIRED.LATENCY_GE_4_PS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring fused instructions -- where one uop can represent multiple contiguous instructions", - "MetricExpr": "tma_light_operations * INST_RETIRED.MACRO_FUSED / (tma_retiring * tma_info_slots)", + "MetricExpr": "tma_light_operations * INST_RETIRED.MACRO_FUSED / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_fused_instructions", "MetricThreshold": "tma_fused_instructions > 0.1 & tma_light_operations > 0.6", @@ -365,16 +637,18 @@ }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences", - "MetricExpr": "topdown\\-heavy\\-ops / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_slots", - "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", + "DefaultMetricgroupName": "TopdownL2", + "MetricExpr": "topdown\\-heavy\\-ops / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_thread_slots", + "MetricGroup": "Default;Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_heavy_operations", "MetricThreshold": "tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL2;Default", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences. Sample with: UOPS_RETIRED.HEAVY", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to instruction cache misses", - "MetricExpr": "ICACHE_DATA.STALLS / tma_info_clks", + "MetricExpr": "ICACHE_DATA.STALLS / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;IcMiss;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_icache_misses", "MetricThreshold": "tma_icache_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -382,746 +656,754 @@ "ScaleUnit": "100%" }, { - "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", - "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time", - "MetricGroup": "Power;Summary", - "MetricName": "tma_info_average_frequency" + "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_thread_slots / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BrMispredicts;tma_issueBM", + "MetricName": "tma_info_bad_spec_branch_misprediction_cost", + "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_info_bottleneck_mispredictions, tma_mispredicts_resteers" + }, + { + "BriefDescription": "Instructions per retired mispredicts for conditional non-taken branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_NTAKEN", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_cond_ntaken", + "MetricThreshold": "tma_info_bad_spec_ipmisp_cond_ntaken < 200" + }, + { + "BriefDescription": "Instructions per retired mispredicts for conditional taken branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_TAKEN", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_cond_taken", + "MetricThreshold": "tma_info_bad_spec_ipmisp_cond_taken < 200" + }, + { + "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.INDIRECT", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_indirect", + "MetricThreshold": "tma_info_bad_spec_ipmisp_indirect < 1e3" + }, + { + "BriefDescription": "Instructions per retired mispredicts for return branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.RET", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_ret", + "MetricThreshold": "tma_info_bad_spec_ipmisp_ret < 500" + }, + { + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BadSpec;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmispredict", + "MetricThreshold": "tma_info_bad_spec_ipmispredict < 200" + }, + { + "BriefDescription": "Probability of Core Bound bottleneck hidden by SMT-profiling artifacts", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(100 * (1 - tma_core_bound / tma_ports_utilization if tma_core_bound < tma_ports_utilization else 1) if tma_info_system_smt_2t_utilization > 0.5 else 0)", + "MetricGroup": "Cor;SMT", + "MetricName": "tma_info_botlnk_l0_core_bound_likely", + "MetricThreshold": "tma_info_botlnk_l0_core_bound_likely > 0.5" + }, + { + "BriefDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_fetch_latency * tma_dsb_switches / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches) + tma_fetch_bandwidth * tma_mite / (tma_dsb + tma_mite))", + "MetricGroup": "DSBmiss;Fed;tma_issueFB", + "MetricName": "tma_info_botlnk_l2_dsb_misses", + "MetricThreshold": "tma_info_botlnk_l2_dsb_misses > 10", + "PublicDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp" + }, + { + "BriefDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck", + "MetricExpr": "100 * (tma_fetch_latency * tma_icache_misses / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", + "MetricGroup": "Fed;FetchLat;IcMiss;tma_issueFL", + "MetricName": "tma_info_botlnk_l2_ic_misses", + "MetricThreshold": "tma_info_botlnk_l2_ic_misses > 5", + "PublicDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck. Related metrics: " }, { "BriefDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses)", "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "100 * tma_fetch_latency * (tma_itlb_misses + tma_icache_misses + tma_unknown_branches) / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)", "MetricGroup": "BigFoot;Fed;Frontend;IcMiss;MemoryTLB;tma_issueBC", - "MetricName": "tma_info_big_code", - "MetricThreshold": "tma_info_big_code > 20", - "PublicDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses). Related metrics: tma_info_branching_overhead" + "MetricName": "tma_info_bottleneck_big_code", + "MetricThreshold": "tma_info_bottleneck_big_code > 20", + "PublicDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses). Related metrics: tma_info_bottleneck_branching_overhead" }, { - "BriefDescription": "Branch instructions per taken branch.", - "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", - "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_bptkbranch" + "BriefDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls)", + "MetricExpr": "100 * ((BR_INST_RETIRED.COND + 3 * BR_INST_RETIRED.NEAR_CALL + (BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL)) / tma_info_thread_slots)", + "MetricGroup": "Ret;tma_issueBC", + "MetricName": "tma_info_bottleneck_branching_overhead", + "MetricThreshold": "tma_info_bottleneck_branching_overhead > 10", + "PublicDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls). Related metrics: tma_info_bottleneck_big_code" }, { - "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", - "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_slots / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_branch_misprediction_cost", - "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_info_mispredictions, tma_mispredicts_resteers" + "BriefDescription": "Total pipeline cost of instruction fetch bandwidth related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_frontend_bound - tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) - tma_info_bottleneck_big_code", + "MetricGroup": "Fed;FetchBW;Frontend", + "MetricName": "tma_info_bottleneck_instruction_fetch_bw", + "MetricThreshold": "tma_info_bottleneck_instruction_fetch_bw > 20" }, { - "BriefDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls)", - "MetricExpr": "100 * ((BR_INST_RETIRED.COND + 3 * BR_INST_RETIRED.NEAR_CALL + (BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL)) / tma_info_slots)", - "MetricGroup": "Ret;tma_issueBC", - "MetricName": "tma_info_branching_overhead", - "MetricThreshold": "tma_info_branching_overhead > 10", - "PublicDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls). Related metrics: tma_info_big_code" + "BriefDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks", + "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_mem_bandwidth / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_sq_full / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full))) + tma_l1_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_fb_full / (tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk))", + "MetricGroup": "Mem;MemoryBW;Offcore;tma_issueBW", + "MetricName": "tma_info_bottleneck_memory_bandwidth", + "MetricThreshold": "tma_info_bottleneck_memory_bandwidth > 20", + "PublicDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks. Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full" }, { - "BriefDescription": "Fraction of branches that are CALL or RET", - "MetricExpr": "(BR_INST_RETIRED.NEAR_CALL + BR_INST_RETIRED.NEAR_RETURN) / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_callret" + "BriefDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_l1_bound / max(tma_memory_bound, tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_dtlb_load / max(tma_l1_bound, tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk)) + tma_store_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_dtlb_store / (tma_dtlb_store + tma_false_sharing + tma_split_stores + tma_store_latency + tma_streaming_stores)))", + "MetricGroup": "Mem;MemoryTLB;Offcore;tma_issueTLB", + "MetricName": "tma_info_bottleneck_memory_data_tlbs", + "MetricThreshold": "tma_info_bottleneck_memory_data_tlbs > 20", + "PublicDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs). Related metrics: tma_dtlb_load, tma_dtlb_store" }, { - "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "Pipeline", - "MetricName": "tma_info_clks" + "BriefDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_mem_latency / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_l3_hit_latency / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full)) + tma_l2_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound))", + "MetricGroup": "Mem;MemoryLat;Offcore;tma_issueLat", + "MetricName": "tma_info_bottleneck_memory_latency", + "MetricThreshold": "tma_info_bottleneck_memory_latency > 20", + "PublicDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches). Related metrics: tma_l3_hit_latency, tma_mem_latency" }, { - "BriefDescription": "STLB (2nd level TLB) code speculative misses per kilo instruction (misses of any page-size that complete the page walk)", - "MetricExpr": "1e3 * ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", - "MetricGroup": "Fed;MemoryTLB", - "MetricName": "tma_info_code_stlb_mpki" + "BriefDescription": "Total pipeline cost of Branch Misprediction related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", + "MetricGroup": "Bad;BadSpec;BrMispredicts;tma_issueBM", + "MetricName": "tma_info_bottleneck_mispredictions", + "MetricThreshold": "tma_info_bottleneck_mispredictions > 20", + "PublicDescription": "Total pipeline cost of Branch Misprediction related bottlenecks. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers" + }, + { + "BriefDescription": "Fraction of branches that are CALL or RET", + "MetricExpr": "(BR_INST_RETIRED.NEAR_CALL + BR_INST_RETIRED.NEAR_RETURN) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_callret" }, { "BriefDescription": "Fraction of branches that are non-taken conditionals", "MetricExpr": "BR_INST_RETIRED.COND_NTAKEN / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Bad;Branches;CodeGen;PGO", - "MetricName": "tma_info_cond_nt" + "MetricName": "tma_info_branches_cond_nt" }, { "BriefDescription": "Fraction of branches that are taken conditionals", "MetricExpr": "BR_INST_RETIRED.COND_TAKEN / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Bad;Branches;CodeGen;PGO", - "MetricName": "tma_info_cond_tk" + "MetricName": "tma_info_branches_cond_tk" }, { - "BriefDescription": "Probability of Core Bound bottleneck hidden by SMT-profiling artifacts", - "MetricExpr": "(100 * (1 - tma_core_bound / tma_ports_utilization if tma_core_bound < tma_ports_utilization else 1) if tma_info_smt_2t_utilization > 0.5 else 0)", - "MetricGroup": "Cor;SMT", - "MetricName": "tma_info_core_bound_likely", - "MetricThreshold": "tma_info_core_bound_likely > 0.5" + "BriefDescription": "Fraction of branches that are unconditional (direct or indirect) jumps", + "MetricExpr": "(BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_jump" + }, + { + "BriefDescription": "Fraction of branches of other types (not individually covered by other metrics in Info.Branches group)", + "MetricExpr": "1 - (tma_info_branches_cond_nt + tma_info_branches_cond_tk + tma_info_branches_callret + tma_info_branches_jump)", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_other_branches" }, { "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", "MetricExpr": "CPU_CLK_UNHALTED.DISTRIBUTED", "MetricGroup": "SMT", - "MetricName": "tma_info_core_clks" + "MetricName": "tma_info_core_core_clks" }, { "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks", "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_coreipc" - }, - { - "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricExpr": "1 / tma_info_ipc", - "MetricGroup": "Mem;Pipeline", - "MetricName": "tma_info_cpi" + "MetricName": "tma_info_core_coreipc" }, { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": "HPC;Summary", - "MetricName": "tma_info_cpu_utilization" + "BriefDescription": "Floating Point Operations Per Cycle", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + FP_ARITH_INST_RETIRED2.SCALAR_HALF + 2 * (FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED2.COMPLEX_SCALAR_HALF) + 4 * cpu@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE\\,umask\\=0x18@ + 8 * (FP_ARITH_INST_RETIRED2.128B_PACKED_HALF + cpu@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE\\,umask\\=0x60@) + 16 * (FP_ARITH_INST_RETIRED2.256B_PACKED_HALF + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) + 32 * FP_ARITH_INST_RETIRED2.512B_PACKED_HALF + 4 * AMX_OPS_RETIRED.BF16", + "MetricGroup": "Flops;Ret", + "MetricName": "tma_info_core_flopc" }, { - "BriefDescription": "Average Parallel L2 cache miss data reads", - "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", - "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_data_l2_mlp" + "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(FP_ARITH_DISPATCHED.PORT_0 + FP_ARITH_DISPATCHED.PORT_1 + FP_ARITH_DISPATCHED.PORT_5) / (2 * tma_info_core_core_clks)", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_core_fp_arith_utilization", + "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." }, { - "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time", - "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", - "MetricName": "tma_info_dram_bw_use", - "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_info_memory_bandwidth, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", + "MetricExpr": "UOPS_EXECUTED.THREAD / (UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1)", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", + "MetricName": "tma_info_core_ilp" }, { "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", "MetricExpr": "IDQ.DSB_UOPS / UOPS_ISSUED.ANY", "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", - "MetricName": "tma_info_dsb_coverage", - "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 6 > 0.35", - "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_misses, tma_info_iptb, tma_lcp" - }, - { - "BriefDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck", - "MetricExpr": "100 * (tma_fetch_latency * tma_dsb_switches / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches) + tma_fetch_bandwidth * tma_mite / (tma_dsb + tma_mite))", - "MetricGroup": "DSBmiss;Fed;tma_issueFB", - "MetricName": "tma_info_dsb_misses", - "MetricThreshold": "tma_info_dsb_misses > 10", - "PublicDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb, tma_lcp" + "MetricName": "tma_info_frontend_dsb_coverage", + "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 6 > 0.35", + "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_inst_mix_iptb, tma_lcp" }, { "BriefDescription": "Average number of cycles of a switch from the DSB fetch-unit to MITE fetch unit - see DSB_Switches tree node for details.", "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / cpu@DSB2MITE_SWITCHES.PENALTY_CYCLES\\,cmask\\=1\\,edge@", "MetricGroup": "DSBmiss", - "MetricName": "tma_info_dsb_switch_cost" - }, - { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", - "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", - "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", - "MetricName": "tma_info_execute" - }, - { - "BriefDescription": "The ratio of Executed- by Issued-Uops", - "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", - "MetricGroup": "Cor;Pipeline", - "MetricName": "tma_info_execute_per_issue", - "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." - }, - { - "BriefDescription": "Fill Buffer (FB) hits per kilo instructions for retired demand loads (L1D misses that merge into ongoing miss-handling entries)", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.FB_HIT / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_fb_hpki" + "MetricName": "tma_info_frontend_dsb_switch_cost" }, { "BriefDescription": "Average number of Uops issued by front-end when it issued something", "MetricExpr": "UOPS_ISSUED.ANY / cpu@UOPS_ISSUED.ANY\\,cmask\\=1@", "MetricGroup": "Fed;FetchBW", - "MetricName": "tma_info_fetch_upc" + "MetricName": "tma_info_frontend_fetch_upc" }, { - "BriefDescription": "Floating Point Operations Per Cycle", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + FP_ARITH_INST_RETIRED2.SCALAR_HALF + 2 * (FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED2.COMPLEX_SCALAR_HALF) + 4 * cpu@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE\\,umask\\=0x18@ + 8 * (FP_ARITH_INST_RETIRED2.128B_PACKED_HALF + cpu@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE\\,umask\\=0x60@) + 16 * (FP_ARITH_INST_RETIRED2.256B_PACKED_HALF + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) + 32 * FP_ARITH_INST_RETIRED2.512B_PACKED_HALF + 4 * AMX_OPS_RETIRED.BF16", - "MetricGroup": "Flops;Ret", - "MetricName": "tma_info_flopc" - }, - { - "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(FP_ARITH_DISPATCHED.PORT_0 + FP_ARITH_DISPATCHED.PORT_1 + FP_ARITH_DISPATCHED.PORT_5) / (2 * tma_info_core_clks)", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_fp_arith_utilization", - "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." + "BriefDescription": "Average Latency for L1 instruction cache misses", + "MetricExpr": "ICACHE_DATA.STALLS / cpu@ICACHE_DATA.STALLS\\,cmask\\=1\\,edge@", + "MetricGroup": "Fed;FetchLat;IcMiss", + "MetricName": "tma_info_frontend_icache_miss_latency" }, { - "BriefDescription": "Giga Floating Point Operations Per Second", - "MetricExpr": "tma_info_flopc / duration_time", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_gflops", - "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + "BriefDescription": "Instructions per non-speculative DSB miss (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / FRONTEND_RETIRED.ANY_DSB_MISS", + "MetricGroup": "DSBmiss;Fed", + "MetricName": "tma_info_frontend_ipdsb_miss_ret", + "MetricThreshold": "tma_info_frontend_ipdsb_miss_ret < 50" }, { - "BriefDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck", - "MetricExpr": "100 * (tma_fetch_latency * tma_icache_misses / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", - "MetricGroup": "Fed;FetchLat;IcMiss;tma_issueFL", - "MetricName": "tma_info_ic_misses", - "MetricThreshold": "tma_info_ic_misses > 5", - "PublicDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck. Related metrics: " + "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_inst_mix_instructions / BACLEARS.ANY", + "MetricGroup": "Fed", + "MetricName": "tma_info_frontend_ipunknown_branch" }, { - "BriefDescription": "Average Latency for L1 instruction cache misses", - "MetricExpr": "ICACHE_DATA.STALLS / cpu@ICACHE_DATA.STALLS\\,cmask\\=1\\,edge@", - "MetricGroup": "Fed;FetchLat;IcMiss", - "MetricName": "tma_info_icache_miss_latency" + "BriefDescription": "L2 cache true code cacheline misses per kilo instruction", + "MetricExpr": "1e3 * FRONTEND_RETIRED.L2_MISS / INST_RETIRED.ANY", + "MetricGroup": "IcMiss", + "MetricName": "tma_info_frontend_l2mpki_code" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", - "MetricExpr": "UOPS_EXECUTED.THREAD / (UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1)", - "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", - "MetricName": "tma_info_ilp" + "BriefDescription": "L2 cache speculative code cacheline misses per kilo instruction", + "MetricExpr": "1e3 * L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", + "MetricGroup": "IcMiss", + "MetricName": "tma_info_frontend_l2mpki_code_all" }, { - "BriefDescription": "Total pipeline cost of instruction fetch bandwidth related bottlenecks", - "MetricExpr": "100 * (tma_frontend_bound - tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) - tma_info_big_code", - "MetricGroup": "Fed;FetchBW;Frontend", - "MetricName": "tma_info_instruction_fetch_bw", - "MetricThreshold": "tma_info_instruction_fetch_bw > 20" + "BriefDescription": "Branch instructions per taken branch.", + "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;PGO", + "MetricName": "tma_info_inst_mix_bptkbranch" }, { "BriefDescription": "Total number of retired Instructions", "MetricExpr": "INST_RETIRED.ANY", "MetricGroup": "Summary;TmaL1;tma_L1_group", - "MetricName": "tma_info_instructions", + "MetricName": "tma_info_inst_mix_instructions", "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" }, { - "BriefDescription": "Average IO (network or disk) Bandwidth Use for Writes [GB / sec]", - "MetricExpr": "UNC_CHA_TOR_INSERTS.IO_PCIRDCUR * 64 / 1e9 / duration_time", - "MetricGroup": "IoBW;Mem;Server;SoC", - "MetricName": "tma_info_io_write_bw" - }, - { "BriefDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + FP_ARITH_INST_RETIRED2.SCALAR + (cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@ + FP_ARITH_INST_RETIRED2.VECTOR))", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_iparith", - "MetricThreshold": "tma_info_iparith < 10", + "MetricName": "tma_info_inst_mix_iparith", + "MetricThreshold": "tma_info_inst_mix_iparith < 10", "PublicDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate). May undercount due to FMA double counting. Approximated prior to BDW." }, { "BriefDescription": "Instructions per FP Arithmetic AMX operation (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / AMX_OPS_RETIRED.BF16", "MetricGroup": "Flops;FpVector;InsType;Server", - "MetricName": "tma_info_iparith_amx_f16", - "MetricThreshold": "tma_info_iparith_amx_f16 < 10", + "MetricName": "tma_info_inst_mix_iparith_amx_f16", + "MetricThreshold": "tma_info_inst_mix_iparith_amx_f16 < 10", "PublicDescription": "Instructions per FP Arithmetic AMX operation (lower number means higher occurrence rate). Operations factored per matrices' sizes of the AMX instructions." }, { "BriefDescription": "Instructions per Integer Arithmetic AMX operation (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / AMX_OPS_RETIRED.INT8", "MetricGroup": "InsType;IntVector;Server", - "MetricName": "tma_info_iparith_amx_int8", - "MetricThreshold": "tma_info_iparith_amx_int8 < 10", + "MetricName": "tma_info_inst_mix_iparith_amx_int8", + "MetricThreshold": "tma_info_inst_mix_iparith_amx_int8 < 10", "PublicDescription": "Instructions per Integer Arithmetic AMX operation (lower number means higher occurrence rate). Operations factored per matrices' sizes of the AMX instructions." }, { "BriefDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED2.128B_PACKED_HALF)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx128", - "MetricThreshold": "tma_info_iparith_avx128 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx128", + "MetricThreshold": "tma_info_inst_mix_iparith_avx128 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED2.256B_PACKED_HALF)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx256", - "MetricThreshold": "tma_info_iparith_avx256 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx256", + "MetricThreshold": "tma_info_inst_mix_iparith_avx256 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic AVX 512-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE + FP_ARITH_INST_RETIRED2.512B_PACKED_HALF)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx512", - "MetricThreshold": "tma_info_iparith_avx512 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx512", + "MetricThreshold": "tma_info_inst_mix_iparith_avx512 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX 512-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_dp", - "MetricThreshold": "tma_info_iparith_scalar_dp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_dp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_dp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_SINGLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_sp", - "MetricThreshold": "tma_info_iparith_scalar_sp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_sp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_sp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { - "BriefDescription": "Instructions per a microcode Assist invocation", - "MetricExpr": "INST_RETIRED.ANY / cpu@ASSISTS.ANY\\,umask\\=0x1B@", - "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_ipassist", - "MetricThreshold": "tma_info_ipassist < 100e3", - "PublicDescription": "Instructions per a microcode Assist invocation. See Assists tree node for details (lower number means higher occurrence rate)" - }, - { "BriefDescription": "Instructions per Branch (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Branches;Fed;InsType", - "MetricName": "tma_info_ipbranch", - "MetricThreshold": "tma_info_ipbranch < 8" - }, - { - "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": "Ret;Summary", - "MetricName": "tma_info_ipc" + "MetricName": "tma_info_inst_mix_ipbranch", + "MetricThreshold": "tma_info_inst_mix_ipbranch < 8" }, { "BriefDescription": "Instructions per (near) call (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_CALL", "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_ipcall", - "MetricThreshold": "tma_info_ipcall < 200" - }, - { - "BriefDescription": "Instructions per non-speculative DSB miss (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / FRONTEND_RETIRED.ANY_DSB_MISS", - "MetricGroup": "DSBmiss;Fed", - "MetricName": "tma_info_ipdsb_miss_ret", - "MetricThreshold": "tma_info_ipdsb_miss_ret < 50" - }, - { - "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", - "MetricGroup": "Branches;OS", - "MetricName": "tma_info_ipfarbranch", - "MetricThreshold": "tma_info_ipfarbranch < 1e6" + "MetricName": "tma_info_inst_mix_ipcall", + "MetricThreshold": "tma_info_inst_mix_ipcall < 200" }, { "BriefDescription": "Instructions per Floating Point (FP) Operation (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_flopc", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_flopc", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_ipflop", - "MetricThreshold": "tma_info_ipflop < 10" + "MetricName": "tma_info_inst_mix_ipflop", + "MetricThreshold": "tma_info_inst_mix_ipflop < 10" }, { "BriefDescription": "Instructions per Load (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_LOADS", "MetricGroup": "InsType", - "MetricName": "tma_info_ipload", - "MetricThreshold": "tma_info_ipload < 3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for conditional non-taken branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_NTAKEN", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_cond_ntaken", - "MetricThreshold": "tma_info_ipmisp_cond_ntaken < 200" - }, - { - "BriefDescription": "Instructions per retired mispredicts for conditional taken branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_TAKEN", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_cond_taken", - "MetricThreshold": "tma_info_ipmisp_cond_taken < 200" - }, - { - "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.INDIRECT", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_indirect", - "MetricThreshold": "tma_info_ipmisp_indirect < 1e3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for return branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.RET", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_ret", - "MetricThreshold": "tma_info_ipmisp_ret < 500" - }, - { - "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BadSpec;BrMispredicts", - "MetricName": "tma_info_ipmispredict", - "MetricThreshold": "tma_info_ipmispredict < 200" + "MetricName": "tma_info_inst_mix_ipload", + "MetricThreshold": "tma_info_inst_mix_ipload < 3" }, { "BriefDescription": "Instructions per Store (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_STORES", "MetricGroup": "InsType", - "MetricName": "tma_info_ipstore", - "MetricThreshold": "tma_info_ipstore < 8" + "MetricName": "tma_info_inst_mix_ipstore", + "MetricThreshold": "tma_info_inst_mix_ipstore < 8" }, { "BriefDescription": "Instructions per Software prefetch instruction (of any type: NTA/T0/T1/T2/Prefetch) (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / cpu@SW_PREFETCH_ACCESS.T0\\,umask\\=0xF@", "MetricGroup": "Prefetches", - "MetricName": "tma_info_ipswpf", - "MetricThreshold": "tma_info_ipswpf < 100" + "MetricName": "tma_info_inst_mix_ipswpf", + "MetricThreshold": "tma_info_inst_mix_ipswpf < 100" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW;Frontend;PGO;tma_issueFB", - "MetricName": "tma_info_iptb", - "MetricThreshold": "tma_info_iptb < 13", - "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_lcp" + "MetricName": "tma_info_inst_mix_iptb", + "MetricThreshold": "tma_info_inst_mix_iptb < 13", + "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_lcp" }, { - "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", - "MetricExpr": "tma_info_instructions / BACLEARS.ANY", - "MetricGroup": "Fed", - "MetricName": "tma_info_ipunknown_branch" + "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_core_l1d_cache_fill_bw" }, { - "BriefDescription": "Fraction of branches that are unconditional (direct or indirect) jumps", - "MetricExpr": "(BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL) / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_jump" + "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_core_l2_cache_fill_bw" }, { - "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_cpi" + "BriefDescription": "Rate of non silent evictions from the L2 cache per Kilo instruction", + "MetricExpr": "1e3 * L2_LINES_OUT.NON_SILENT / tma_info_inst_mix_instructions", + "MetricGroup": "L2Evicts;Mem;Server", + "MetricName": "tma_info_memory_core_l2_evictions_nonsilent_pki" }, { - "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_utilization", - "MetricThreshold": "tma_info_kernel_utilization > 0.05" + "BriefDescription": "Rate of silent evictions from the L2 cache per Kilo instruction where the evicted lines are dropped (no writeback to L3 or memory)", + "MetricExpr": "1e3 * L2_LINES_OUT.SILENT / tma_info_inst_mix_instructions", + "MetricGroup": "L2Evicts;Mem;Server", + "MetricName": "tma_info_memory_core_l2_evictions_silent_pki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw" + "BriefDescription": "Average per-core data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * OFFCORE_REQUESTS.ALL_REQUESTS / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_core_l3_cache_access_bw" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "tma_info_l1d_cache_fill_bw", + "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw_1t" + "MetricName": "tma_info_memory_core_l3_cache_fill_bw" + }, + { + "BriefDescription": "Fill Buffer (FB) hits per kilo instructions for retired demand loads (L1D misses that merge into ongoing miss-handling entries)", + "MetricExpr": "1e3 * MEM_LOAD_RETIRED.FB_HIT / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_fb_hpki" }, { "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L1_MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki" + "MetricName": "tma_info_memory_l1mpki" }, { "BriefDescription": "L1 cache true misses per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.ALL_DEMAND_DATA_RD / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki_load" - }, - { - "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw" - }, - { - "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "tma_info_l2_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw_1t" - }, - { - "BriefDescription": "Rate of non silent evictions from the L2 cache per Kilo instruction", - "MetricExpr": "1e3 * L2_LINES_OUT.NON_SILENT / tma_info_instructions", - "MetricGroup": "L2Evicts;Mem;Server", - "MetricName": "tma_info_l2_evictions_nonsilent_pki" - }, - { - "BriefDescription": "Rate of silent evictions from the L2 cache per Kilo instruction where the evicted lines are dropped (no writeback to L3 or memory)", - "MetricExpr": "1e3 * L2_LINES_OUT.SILENT / tma_info_instructions", - "MetricGroup": "L2Evicts;Mem;Server", - "MetricName": "tma_info_l2_evictions_silent_pki" + "MetricName": "tma_info_memory_l1mpki_load" }, { "BriefDescription": "L2 cache hits per kilo instruction for all request types (including speculative)", "MetricExpr": "1e3 * (L2_RQSTS.REFERENCES - L2_RQSTS.MISS) / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_all" + "MetricName": "tma_info_memory_l2hpki_all" }, { "BriefDescription": "L2 cache hits per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_HIT / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_load" + "MetricName": "tma_info_memory_l2hpki_load" }, { "BriefDescription": "L2 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L2_MISS / INST_RETIRED.ANY", "MetricGroup": "Backend;CacheMisses;Mem", - "MetricName": "tma_info_l2mpki" + "MetricName": "tma_info_memory_l2mpki" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all request types (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem;Offcore", - "MetricName": "tma_info_l2mpki_all" - }, - { - "BriefDescription": "L2 cache true code cacheline misses per kilo instruction", - "MetricExpr": "1e3 * FRONTEND_RETIRED.L2_MISS / INST_RETIRED.ANY", - "MetricGroup": "IcMiss", - "MetricName": "tma_info_l2mpki_code" - }, - { - "BriefDescription": "L2 cache speculative code cacheline misses per kilo instruction", - "MetricExpr": "1e3 * L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", - "MetricGroup": "IcMiss", - "MetricName": "tma_info_l2mpki_code_all" + "MetricName": "tma_info_memory_l2mpki_all" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2mpki_load" + "MetricName": "tma_info_memory_l2mpki_load" }, { - "BriefDescription": "Average per-core data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * OFFCORE_REQUESTS.ALL_REQUESTS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw" - }, - { - "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_access_bw", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw_1t" + "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L3_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l3mpki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw" + "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", + "MetricExpr": "L1D_PEND_MISS.PENDING / MEM_LOAD_COMPLETED.L1_MISS_ANY", + "MetricGroup": "Mem;MemoryBound;MemoryLat", + "MetricName": "tma_info_memory_load_miss_real_latency" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw_1t" + "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", + "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", + "MetricGroup": "Mem;MemoryBW;MemoryBound", + "MetricName": "tma_info_memory_mlp", + "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" }, { - "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L3_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l3mpki" + "BriefDescription": "Average Parallel L2 cache miss data reads", + "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", + "MetricGroup": "Memory_BW;Offcore", + "MetricName": "tma_info_memory_oro_data_l2_mlp" }, { "BriefDescription": "Average Latency for L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS.DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l2_miss_latency" + "MetricName": "tma_info_memory_oro_load_l2_miss_latency" }, { "BriefDescription": "Average Parallel L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / cpu@OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD\\,cmask\\=1@", "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_load_l2_mlp" + "MetricName": "tma_info_memory_oro_load_l2_mlp" }, { "BriefDescription": "Average Latency for L3 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.L3_MISS_DEMAND_DATA_RD / OFFCORE_REQUESTS.L3_MISS_DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l3_miss_latency" + "MetricName": "tma_info_memory_oro_load_l3_miss_latency" }, { - "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", - "MetricExpr": "L1D_PEND_MISS.PENDING / MEM_LOAD_COMPLETED.L1_MISS_ANY", - "MetricGroup": "Mem;MemoryBound;MemoryLat", - "MetricName": "tma_info_load_miss_real_latency" + "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l1d_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l1d_cache_fill_bw_1t" + }, + { + "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l2_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l2_cache_fill_bw_1t" + }, + { + "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_access_bw", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_thread_l3_cache_access_bw_1t" + }, + { + "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l3_cache_fill_bw_1t" + }, + { + "BriefDescription": "STLB (2nd level TLB) code speculative misses per kilo instruction (misses of any page-size that complete the page walk)", + "MetricExpr": "1e3 * ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricGroup": "Fed;MemoryTLB", + "MetricName": "tma_info_memory_tlb_code_stlb_mpki" }, { "BriefDescription": "STLB (2nd level TLB) data load speculative misses per kilo instruction (misses of any page-size that complete the page walk)", "MetricExpr": "1e3 * DTLB_LOAD_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_load_stlb_mpki" + "MetricName": "tma_info_memory_tlb_load_stlb_mpki" + }, + { + "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", + "MetricExpr": "(ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING) / (4 * tma_info_core_core_clks)", + "MetricGroup": "Mem;MemoryTLB", + "MetricName": "tma_info_memory_tlb_page_walks_utilization", + "MetricThreshold": "tma_info_memory_tlb_page_walks_utilization > 0.5" + }, + { + "BriefDescription": "STLB (2nd level TLB) data store speculative misses per kilo instruction (misses of any page-size that complete the page walk)", + "MetricExpr": "1e3 * DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricGroup": "Mem;MemoryTLB", + "MetricName": "tma_info_memory_tlb_store_stlb_mpki" + }, + { + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", + "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", + "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", + "MetricName": "tma_info_pipeline_execute" + }, + { + "BriefDescription": "Instructions per a microcode Assist invocation", + "MetricExpr": "INST_RETIRED.ANY / cpu@ASSISTS.ANY\\,umask\\=0x1B@", + "MetricGroup": "Pipeline;Ret;Retire", + "MetricName": "tma_info_pipeline_ipassist", + "MetricThreshold": "tma_info_pipeline_ipassist < 100e3", + "PublicDescription": "Instructions per a microcode Assist invocation. See Assists tree node for details (lower number means higher occurrence rate)" + }, + { + "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "tma_retiring * tma_info_thread_slots / cpu@UOPS_RETIRED.SLOTS\\,cmask\\=1@", + "MetricGroup": "Pipeline;Ret", + "MetricName": "tma_info_pipeline_retire" + }, + { + "BriefDescription": "Estimated fraction of retirement-cycles dealing with repeat instructions", + "MetricExpr": "INST_RETIRED.REP_ITERATION / cpu@UOPS_RETIRED.SLOTS\\,cmask\\=1@", + "MetricGroup": "Pipeline;Ret", + "MetricName": "tma_info_pipeline_strings_cycles", + "MetricThreshold": "tma_info_pipeline_strings_cycles > 0.1" + }, + { + "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time", + "MetricGroup": "Power;Summary", + "MetricName": "tma_info_system_average_frequency" + }, + { + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricGroup": "HPC;Summary", + "MetricName": "tma_info_system_cpu_utilization" + }, + { + "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", + "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time", + "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", + "MetricName": "tma_info_system_dram_bw_use", + "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_mem_bandwidth, tma_sq_full" + }, + { + "BriefDescription": "Giga Floating Point Operations Per Second", + "MetricExpr": "tma_info_core_flopc / duration_time", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_system_gflops", + "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + }, + { + "BriefDescription": "Average IO (network or disk) Bandwidth Use for Writes [GB / sec]", + "MetricExpr": "UNC_CHA_TOR_INSERTS.IO_PCIRDCUR * 64 / 1e9 / duration_time", + "MetricGroup": "IoBW;Mem;Server;SoC", + "MetricName": "tma_info_system_io_write_bw" + }, + { + "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", + "MetricGroup": "Branches;OS", + "MetricName": "tma_info_system_ipfarbranch", + "MetricThreshold": "tma_info_system_ipfarbranch < 1e6" + }, + { + "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_cpi" + }, + { + "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_utilization", + "MetricThreshold": "tma_info_system_kernel_utilization > 0.05" }, { "BriefDescription": "Average latency of data read request to external DRAM memory [in nanoseconds]", "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_DDR / UNC_CHA_TOR_INSERTS.IA_MISS_DRD_DDR) / uncore_cha_0@event\\=0x1@", "MetricGroup": "Mem;MemoryLat;Server;SoC", - "MetricName": "tma_info_mem_dram_read_latency", + "MetricName": "tma_info_system_mem_dram_read_latency", "PublicDescription": "Average latency of data read request to external DRAM memory [in nanoseconds]. Accounts for demand loads and L1/L2 data-read prefetches" }, { "BriefDescription": "Average number of parallel data read requests to external memory", "MetricExpr": "UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD / UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD@thresh\\=1@", "MetricGroup": "Mem;MemoryBW;SoC", - "MetricName": "tma_info_mem_parallel_reads", + "MetricName": "tma_info_system_mem_parallel_reads", "PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches" }, { "BriefDescription": "Average latency of data read request to external 3D X-Point memory [in nanoseconds]", "MetricExpr": "(1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD_PMM / UNC_CHA_TOR_INSERTS.IA_MISS_DRD_PMM) / uncore_cha_0@event\\=0x1@ if #has_pmem > 0 else 0)", "MetricGroup": "Mem;MemoryLat;Server;SoC", - "MetricName": "tma_info_mem_pmm_read_latency", + "MetricName": "tma_info_system_mem_pmm_read_latency", "PublicDescription": "Average latency of data read request to external 3D X-Point memory [in nanoseconds]. Accounts for demand loads and L1/L2 data-read prefetches" }, { "BriefDescription": "Average latency of data read request to external memory (in nanoseconds)", - "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD / UNC_CHA_TOR_INSERTS.IA_MISS_DRD) / (tma_info_socket_clks / duration_time)", + "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD / UNC_CHA_TOR_INSERTS.IA_MISS_DRD) / (tma_info_system_socket_clks / duration_time)", "MetricGroup": "Mem;MemoryLat;SoC", - "MetricName": "tma_info_mem_read_latency", + "MetricName": "tma_info_system_mem_read_latency", "PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)" }, { - "BriefDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks", - "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_mem_bandwidth / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_sq_full / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full))) + tma_l1_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_fb_full / (tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk))", - "MetricGroup": "Mem;MemoryBW;Offcore;tma_issueBW", - "MetricName": "tma_info_memory_bandwidth", - "MetricThreshold": "tma_info_memory_bandwidth > 20", - "PublicDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks. Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_mem_bandwidth, tma_sq_full" - }, - { - "BriefDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs)", - "MetricExpr": "100 * tma_memory_bound * (tma_l1_bound / max(tma_memory_bound, tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_dtlb_load / max(tma_l1_bound, tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk)) + tma_store_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_dtlb_store / (tma_dtlb_store + tma_false_sharing + tma_split_stores + tma_store_latency + tma_streaming_stores)))", - "MetricGroup": "Mem;MemoryTLB;Offcore;tma_issueTLB", - "MetricName": "tma_info_memory_data_tlbs", - "MetricThreshold": "tma_info_memory_data_tlbs > 20", - "PublicDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs). Related metrics: tma_dtlb_load, tma_dtlb_store" - }, - { - "BriefDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches)", - "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_mem_latency / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound) * (tma_l3_hit_latency / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full)) + tma_l2_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_pmm_bound + tma_store_bound))", - "MetricGroup": "Mem;MemoryLat;Offcore;tma_issueLat", - "MetricName": "tma_info_memory_latency", - "MetricThreshold": "tma_info_memory_latency > 20", - "PublicDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches). Related metrics: tma_l3_hit_latency, tma_mem_latency" - }, - { - "BriefDescription": "Total pipeline cost of Branch Misprediction related bottlenecks", - "MetricExpr": "100 * (tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", - "MetricGroup": "Bad;BadSpec;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_mispredictions", - "MetricThreshold": "tma_info_mispredictions > 20", - "PublicDescription": "Total pipeline cost of Branch Misprediction related bottlenecks. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost, tma_mispredicts_resteers" - }, - { - "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", - "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", - "MetricGroup": "Mem;MemoryBW;MemoryBound", - "MetricName": "tma_info_mlp", - "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" - }, - { - "BriefDescription": "Fraction of branches of other types (not individually covered by other metrics in Info.Branches group)", - "MetricExpr": "1 - (tma_info_cond_nt + tma_info_cond_tk + tma_info_callret + tma_info_jump)", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_other_branches" - }, - { - "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", - "MetricExpr": "(ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING) / (4 * tma_info_core_clks)", - "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_page_walks_utilization", - "MetricThreshold": "tma_info_page_walks_utilization > 0.5" - }, - { "BriefDescription": "Average 3DXP Memory Bandwidth Use for reads [GB / sec]", "MetricExpr": "(64 * UNC_M_PMM_RPQ_INSERTS / 1e9 / duration_time if #has_pmem > 0 else 0)", "MetricGroup": "Mem;MemoryBW;Server;SoC", - "MetricName": "tma_info_pmm_read_bw" + "MetricName": "tma_info_system_pmm_read_bw" }, { "BriefDescription": "Average 3DXP Memory Bandwidth Use for Writes [GB / sec]", "MetricExpr": "(64 * UNC_M_PMM_WPQ_INSERTS / 1e9 / duration_time if #has_pmem > 0 else 0)", "MetricGroup": "Mem;MemoryBW;Server;SoC", - "MetricName": "tma_info_pmm_write_bw" - }, - { - "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", - "MetricExpr": "tma_retiring * tma_info_slots / cpu@UOPS_RETIRED.SLOTS\\,cmask\\=1@", - "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_retire" - }, - { - "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", - "MetricExpr": "TOPDOWN.SLOTS", - "MetricGroup": "TmaL1;tma_L1_group", - "MetricName": "tma_info_slots" - }, - { - "BriefDescription": "Fraction of Physical Core issue-slots utilized by this Logical Processor", - "MetricExpr": "(tma_info_slots / (TOPDOWN.SLOTS / 2) if #SMT_on else 1)", - "MetricGroup": "SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_slots_utilization" + "MetricName": "tma_info_system_pmm_write_bw" }, { "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_DISTRIBUTED if #SMT_on else 0)", "MetricGroup": "SMT", - "MetricName": "tma_info_smt_2t_utilization" + "MetricName": "tma_info_system_smt_2t_utilization" }, { "BriefDescription": "Socket actual clocks when any core is active on that socket", "MetricExpr": "uncore_cha_0@event\\=0x1@", "MetricGroup": "SoC", - "MetricName": "tma_info_socket_clks" - }, - { - "BriefDescription": "STLB (2nd level TLB) data store speculative misses per kilo instruction (misses of any page-size that complete the page walk)", - "MetricExpr": "1e3 * DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", - "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_store_stlb_mpki" - }, - { - "BriefDescription": "Estimated fraction of retirement-cycles dealing with repeat instructions", - "MetricExpr": "INST_RETIRED.REP_ITERATION / cpu@UOPS_RETIRED.SLOTS\\,cmask\\=1@", - "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_strings_cycles", - "MetricThreshold": "tma_info_strings_cycles > 0.1" + "MetricName": "tma_info_system_socket_clks" }, { "BriefDescription": "Tera Integer (matrix) Operations Per Second", "MetricExpr": "8 * AMX_OPS_RETIRED.INT8 / 1e12 / duration_time", "MetricGroup": "Cor;HPC;IntVector;Server", - "MetricName": "tma_info_tiops" + "MetricName": "tma_info_system_tiops" }, { "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC", "MetricGroup": "Power", - "MetricName": "tma_info_turbo_utilization" - }, - { - "BriefDescription": "Uops Per Instruction", - "MetricExpr": "tma_retiring * tma_info_slots / INST_RETIRED.ANY", - "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_uoppi", - "MetricThreshold": "tma_info_uoppi > 1.05" + "MetricName": "tma_info_system_turbo_utilization" }, { "BriefDescription": "Cross-socket Ultra Path Interconnect (UPI) data transmit bandwidth for data only [MB / sec]", "MetricExpr": "UNC_UPI_TxL_FLITS.ALL_DATA * 64 / 9 / 1e6", "MetricGroup": "Server;SoC", - "MetricName": "tma_info_upi_data_transmit_bw" + "MetricName": "tma_info_system_upi_data_transmit_bw" + }, + { + "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "Pipeline", + "MetricName": "tma_info_thread_clks" + }, + { + "BriefDescription": "Cycles Per Instruction (per Logical Processor)", + "MetricExpr": "1 / tma_info_thread_ipc", + "MetricGroup": "Mem;Pipeline", + "MetricName": "tma_info_thread_cpi" + }, + { + "BriefDescription": "The ratio of Executed- by Issued-Uops", + "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", + "MetricGroup": "Cor;Pipeline", + "MetricName": "tma_info_thread_execute_per_issue", + "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." + }, + { + "BriefDescription": "Instructions Per Cycle (per Logical Processor)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks", + "MetricGroup": "Ret;Summary", + "MetricName": "tma_info_thread_ipc" + }, + { + "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", + "MetricExpr": "TOPDOWN.SLOTS", + "MetricGroup": "TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots" + }, + { + "BriefDescription": "Fraction of Physical Core issue-slots utilized by this Logical Processor", + "MetricExpr": "(tma_info_thread_slots / (TOPDOWN.SLOTS / 2) if #SMT_on else 1)", + "MetricGroup": "SMT;TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots_utilization" + }, + { + "BriefDescription": "Uops Per Instruction", + "MetricExpr": "tma_retiring * tma_info_thread_slots / INST_RETIRED.ANY", + "MetricGroup": "Pipeline;Ret;Retire", + "MetricName": "tma_info_thread_uoppi", + "MetricThreshold": "tma_info_thread_uoppi > 1.05" }, { "BriefDescription": "Instruction per taken branch", - "MetricExpr": "tma_retiring * tma_info_slots / BR_INST_RETIRED.NEAR_TAKEN", + "MetricExpr": "tma_retiring * tma_info_thread_slots / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW", - "MetricName": "tma_info_uptb", - "MetricThreshold": "tma_info_uptb < 9" + "MetricName": "tma_info_thread_uptb", + "MetricThreshold": "tma_info_thread_uptb < 9" }, { "BriefDescription": "This metric approximates arithmetic Integer (Int) matrix uops fraction the CPU has retired (aggregated across all supported Int datatypes in AMX engine)", - "MetricExpr": "cpu@AMX_OPS_RETIRED.INT8\\,cmask\\=1@ / (tma_retiring * tma_info_slots)", + "MetricExpr": "cpu@AMX_OPS_RETIRED.INT8\\,cmask\\=1@ / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;HPC;IntVector;Pipeline;Server;TopdownL4;tma_L4_group;tma_int_operations_group", "MetricName": "tma_int_amx", "MetricThreshold": "tma_int_amx > 0.1 & (tma_int_operations > 0.1 & tma_light_operations > 0.6)", @@ -1139,7 +1421,7 @@ }, { "BriefDescription": "This metric represents 128-bit vector Integer ADD/SUB/SAD or VNNI (Vector Neural Network Instructions) uops fraction the CPU has retired", - "MetricExpr": "(INT_VEC_RETIRED.ADD_128 + INT_VEC_RETIRED.VNNI_128) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(INT_VEC_RETIRED.ADD_128 + INT_VEC_RETIRED.VNNI_128) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;IntVector;Pipeline;TopdownL4;tma_L4_group;tma_int_operations_group;tma_issue2P", "MetricName": "tma_int_vector_128b", "MetricThreshold": "tma_int_vector_128b > 0.1 & (tma_int_operations > 0.1 & tma_light_operations > 0.6)", @@ -1148,7 +1430,7 @@ }, { "BriefDescription": "This metric represents 256-bit vector Integer ADD/SUB/SAD or VNNI (Vector Neural Network Instructions) uops fraction the CPU has retired", - "MetricExpr": "(INT_VEC_RETIRED.ADD_256 + INT_VEC_RETIRED.MUL_256 + INT_VEC_RETIRED.VNNI_256) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(INT_VEC_RETIRED.ADD_256 + INT_VEC_RETIRED.MUL_256 + INT_VEC_RETIRED.VNNI_256) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;IntVector;Pipeline;TopdownL4;tma_L4_group;tma_int_operations_group;tma_issue2P", "MetricName": "tma_int_vector_256b", "MetricThreshold": "tma_int_vector_256b > 0.1 & (tma_int_operations > 0.1 & tma_light_operations > 0.6)", @@ -1157,7 +1439,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses", - "MetricExpr": "ICACHE_TAG.STALLS / tma_info_clks", + "MetricExpr": "ICACHE_TAG.STALLS / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -1166,7 +1448,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled without loads missing the L1 data cache", - "MetricExpr": "max((EXE_ACTIVITY.BOUND_ON_LOADS - MEMORY_ACTIVITY.STALLS_L1D_MISS) / tma_info_clks, 0)", + "MetricExpr": "max((EXE_ACTIVITY.BOUND_ON_LOADS - MEMORY_ACTIVITY.STALLS_L1D_MISS) / tma_info_thread_clks, 0)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_issueL1;tma_issueMC;tma_memory_bound_group", "MetricName": "tma_l1_bound", "MetricThreshold": "tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1176,7 +1458,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled due to L2 cache accesses by loads", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(MEMORY_ACTIVITY.STALLS_L1D_MISS - MEMORY_ACTIVITY.STALLS_L2_MISS) / tma_info_clks", + "MetricExpr": "(MEMORY_ACTIVITY.STALLS_L1D_MISS - MEMORY_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l2_bound", "MetricThreshold": "tma_l2_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1185,7 +1467,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core", - "MetricExpr": "(MEMORY_ACTIVITY.STALLS_L2_MISS - MEMORY_ACTIVITY.STALLS_L3_MISS) / tma_info_clks", + "MetricExpr": "(MEMORY_ACTIVITY.STALLS_L2_MISS - MEMORY_ACTIVITY.STALLS_L3_MISS) / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1194,34 +1476,36 @@ }, { "BriefDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited)", - "MetricExpr": "33 * tma_info_average_frequency * MEM_LOAD_RETIRED.L3_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "33 * tma_info_system_average_frequency * MEM_LOAD_RETIRED.L3_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "MemoryLat;TopdownL4;tma_L4_group;tma_issueLat;tma_l3_bound_group", "MetricName": "tma_l3_hit_latency", "MetricThreshold": "tma_l3_hit_latency > 0.1 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited). Avoiding private cache misses (i.e. L2 misses/L3 hits) will improve the latency; reduce contention with sibling physical cores and increase performance. Note the value of this node may overlap with its siblings. Sample with: MEM_LOAD_RETIRED.L3_HIT_PS. Related metrics: tma_info_memory_latency, tma_mem_latency", + "PublicDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited). Avoiding private cache misses (i.e. L2 misses/L3 hits) will improve the latency; reduce contention with sibling physical cores and increase performance. Note the value of this node may overlap with its siblings. Sample with: MEM_LOAD_RETIRED.L3_HIT_PS. Related metrics: tma_info_bottleneck_memory_latency, tma_mem_latency", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)", - "MetricExpr": "DECODE.LCP / tma_info_clks", + "MetricExpr": "DECODE.LCP / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_lcp", "MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb", + "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation)", + "DefaultMetricgroupName": "TopdownL2", "MetricExpr": "max(0, tma_retiring - tma_heavy_operations)", - "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", + "MetricGroup": "Default;Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_light_operations", "MetricThreshold": "tma_light_operations > 0.6", + "MetricgroupNoGroup": "TopdownL2;Default", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation). This correlates with total number of instructions used by the program. A uops-per-instruction (see UopPI metric) ratio of 1 or less should be expected for decently optimized software running on Intel Core/Xeon products. While this often indicates efficient X86 instructions were executed; high value does not necessarily mean better performance cannot be achieved. Sample with: INST_RETIRED.PREC_DIST", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Load operations", - "MetricExpr": "UOPS_DISPATCHED.PORT_2_3_10 / (3 * tma_info_core_clks)", + "MetricExpr": "UOPS_DISPATCHED.PORT_2_3_10 / (3 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_load_op_utilization", "MetricThreshold": "tma_load_op_utilization > 0.6", @@ -1238,7 +1522,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles where the Second-level TLB (STLB) was missed by load accesses, performing a hardware page walk", - "MetricExpr": "DTLB_LOAD_MISSES.WALK_ACTIVE / tma_info_clks", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_ACTIVE / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL5;tma_L5_group;tma_dtlb_load_group", "MetricName": "tma_load_stlb_miss", "MetricThreshold": "tma_load_stlb_miss > 0.05 & (tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1246,7 +1530,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from local memory", - "MetricExpr": "71 * tma_info_average_frequency * MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "71 * tma_info_system_average_frequency * MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Server;TopdownL5;tma_L5_group;tma_mem_latency_group", "MetricName": "tma_local_dram", "MetricThreshold": "tma_local_dram > 0.1 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1256,7 +1540,7 @@ { "BriefDescription": "This metric represents fraction of cycles the CPU spent handling cache misses due to lock operations", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(16 * max(0, MEM_INST_RETIRED.LOCK_LOADS - L2_RQSTS.ALL_RFO) + MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES * (10 * L2_RQSTS.RFO_HIT + min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO))) / tma_info_clks", + "MetricExpr": "(16 * max(0, MEM_INST_RETIRED.LOCK_LOADS - L2_RQSTS.ALL_RFO) + MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES * (10 * L2_RQSTS.RFO_HIT + min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO))) / tma_info_thread_clks", "MetricGroup": "Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_l1_bound_group", "MetricName": "tma_lock_latency", "MetricThreshold": "tma_lock_latency > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1265,16 +1549,18 @@ }, { "BriefDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears", + "DefaultMetricgroupName": "TopdownL2", "MetricExpr": "max(0, tma_bad_speculation - tma_branch_mispredicts)", - "MetricGroup": "BadSpec;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", + "MetricGroup": "BadSpec;Default;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.1 & tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL2;Default", "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears. These slots are either wasted by uops fetched prior to the clear; or stalls the out-of-order portion of the machine needs to recover its state after the clear. For example; this can happen due to memory ordering Nukes (e.g. Memory Disambiguation) or Self-Modifying-Code (SMC) nukes. Sample with: MACHINE_CLEARS.COUNT. Related metrics: tma_clears_resteers, tma_contested_accesses, tma_data_sharing, tma_false_sharing, tma_l1_bound, tma_microcode_sequencer, tma_ms_switches, tma_remote_cache", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to memory bandwidth Allocation feature (RDT's memory bandwidth throttling).", - "MetricExpr": "INT_MISC.MBA_STALLS / tma_info_clks", + "MetricExpr": "INT_MISC.MBA_STALLS / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;Server;TopdownL5;tma_L5_group;tma_mem_bandwidth_group", "MetricName": "tma_mba_stalls", "MetricThreshold": "tma_mba_stalls > 0.1 & (tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1282,34 +1568,36 @@ }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_clks", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW", "MetricName": "tma_mem_bandwidth", "MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_sq_full", + "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_sq_full", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat", "MetricName": "tma_mem_latency", "MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM). This metric does not aggregate requests from other Logical Processors/Physical Cores/sockets (see Uncore counters for that). Related metrics: tma_info_memory_latency, tma_l3_hit_latency", + "PublicDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM). This metric does not aggregate requests from other Logical Processors/Physical Cores/sockets (see Uncore counters for that). Related metrics: tma_info_bottleneck_memory_latency, tma_l3_hit_latency", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck", - "MetricExpr": "topdown\\-mem\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_slots", - "MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", + "DefaultMetricgroupName": "TopdownL2", + "MetricExpr": "topdown\\-mem\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_thread_slots", + "MetricGroup": "Backend;Default;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2;Default", "PublicDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck. Memory Bound estimates fraction of slots where pipeline is likely stalled due to demand load or store instructions. This accounts mainly for (1) non-completed in-flight memory demand loads which coincides with execution units starvation; in addition to (2) cases where stores could impose backpressure on the pipeline when many of them get buffered at the same time (less common out of the two).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to LFENCE Instructions.", - "MetricExpr": "13 * MISC2_RETIRED.LFENCE / tma_info_clks", + "MetricExpr": "13 * MISC2_RETIRED.LFENCE / tma_info_thread_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_serializing_operation_group", "MetricName": "tma_memory_fence", "MetricThreshold": "tma_memory_fence > 0.05 & (tma_serializing_operation > 0.1 & (tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))))", @@ -1317,7 +1605,8 @@ }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring memory operations -- uops for memory load or store accesses.", - "MetricExpr": "tma_light_operations * MEM_UOP_RETIRED.ANY / (tma_retiring * tma_info_slots)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "tma_light_operations * MEM_UOP_RETIRED.ANY / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_memory_operations", "MetricThreshold": "tma_memory_operations > 0.1 & tma_light_operations > 0.6", @@ -1325,7 +1614,7 @@ }, { "BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit", - "MetricExpr": "UOPS_RETIRED.MS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.MS / tma_info_thread_slots", "MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS", "MetricName": "tma_microcode_sequencer", "MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1", @@ -1334,25 +1623,25 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage", - "MetricExpr": "tma_branch_mispredicts / tma_bad_speculation * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks", + "MetricExpr": "tma_branch_mispredicts / tma_bad_speculation * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks", "MetricGroup": "BadSpec;BrMispredicts;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueBM", "MetricName": "tma_mispredicts_resteers", "MetricThreshold": "tma_mispredicts_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost, tma_info_mispredictions", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost, tma_info_bottleneck_mispredictions", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline)", - "MetricExpr": "(IDQ.MITE_CYCLES_ANY - IDQ.MITE_CYCLES_OK) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.MITE_CYCLES_ANY - IDQ.MITE_CYCLES_OK) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_mite", - "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 6 > 0.35)", + "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 6 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline). This pipeline is used for code that was not pre-cached in the DSB or LSD. For example; inefficiencies due to asymmetric decoders; use of long immediate or LCP can manifest as MITE fetch bandwidth bottleneck. Sample with: FRONTEND_RETIRED.ANY_DSB_MISS", "ScaleUnit": "100%" }, { "BriefDescription": "The Mixing_Vectors metric gives the percentage of injected blend uops out of all uops issued", - "MetricExpr": "160 * ASSISTS.SSE_AVX_MIX / tma_info_clks", + "MetricExpr": "160 * ASSISTS.SSE_AVX_MIX / tma_info_thread_clks", "MetricGroup": "TopdownL5;tma_L5_group;tma_issueMV;tma_ports_utilized_0_group", "MetricName": "tma_mixing_vectors", "MetricThreshold": "tma_mixing_vectors > 0.05", @@ -1361,7 +1650,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)", - "MetricExpr": "3 * cpu@UOPS_RETIRED.MS\\,cmask\\=1\\,edge@ / (tma_retiring * tma_info_slots / UOPS_ISSUED.ANY) / tma_info_clks", + "MetricExpr": "3 * cpu@UOPS_RETIRED.MS\\,cmask\\=1\\,edge@ / (tma_retiring * tma_info_thread_slots / UOPS_ISSUED.ANY) / tma_info_thread_clks", "MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO", "MetricName": "tma_ms_switches", "MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -1370,7 +1659,7 @@ }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring branch instructions that were not fused", - "MetricExpr": "tma_light_operations * (BR_INST_RETIRED.ALL_BRANCHES - INST_RETIRED.MACRO_FUSED) / (tma_retiring * tma_info_slots)", + "MetricExpr": "tma_light_operations * (BR_INST_RETIRED.ALL_BRANCHES - INST_RETIRED.MACRO_FUSED) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_non_fused_branches", "MetricThreshold": "tma_non_fused_branches > 0.1 & tma_light_operations > 0.6", @@ -1379,7 +1668,7 @@ }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring NOP (no op) instructions", - "MetricExpr": "tma_light_operations * INST_RETIRED.NOP / (tma_retiring * tma_info_slots)", + "MetricExpr": "tma_light_operations * INST_RETIRED.NOP / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_nop_instructions", "MetricThreshold": "tma_nop_instructions > 0.1 & tma_light_operations > 0.6", @@ -1388,6 +1677,7 @@ }, { "BriefDescription": "This metric represents the remaining light uops fraction the CPU has executed - remaining means not covered by other sibling nodes", + "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "max(0, tma_light_operations - (tma_fp_arith + tma_int_operations + tma_memory_operations + tma_fused_instructions + tma_non_fused_branches + tma_nop_instructions))", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_other_light_ops", @@ -1397,7 +1687,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of slots the CPU retired uops as a result of handing Page Faults", - "MetricExpr": "99 * ASSISTS.PAGE_FAULT / tma_info_slots", + "MetricExpr": "99 * ASSISTS.PAGE_FAULT / tma_info_thread_slots", "MetricGroup": "TopdownL5;tma_L5_group;tma_assists_group", "MetricName": "tma_page_faults", "MetricThreshold": "tma_page_faults > 0.05", @@ -1406,7 +1696,7 @@ }, { "BriefDescription": "This metric roughly estimates (based on idle latencies) how often the CPU was stalled on accesses to external 3D-Xpoint (Crystal Ridge, a.k.a", - "MetricExpr": "(((1 - ((19 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + 10 * (MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS))) / (19 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + 10 * (MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + (25 * (MEM_LOAD_RETIRED.LOCAL_PMM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) if #has_pmem > 0 else 0) + 33 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_PMM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) if #has_pmem > 0 else 0))) if #has_pmem > 0 else 0)) * (MEMORY_ACTIVITY.STALLS_L3_MISS / tma_info_clks) if 1e6 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_PMM + MEM_LOAD_RETIRED.LOCAL_PMM) > MEM_LOAD_RETIRED.L1_MISS else 0) if #has_pmem > 0 else 0)", + "MetricExpr": "(((1 - ((19 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + 10 * (MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS))) / (19 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + 10 * (MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS)) + (25 * (MEM_LOAD_RETIRED.LOCAL_PMM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) if #has_pmem > 0 else 0) + 33 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_PMM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) if #has_pmem > 0 else 0))) if #has_pmem > 0 else 0)) * (MEMORY_ACTIVITY.STALLS_L3_MISS / tma_info_thread_clks) if 1e6 * (MEM_LOAD_L3_MISS_RETIRED.REMOTE_PMM + MEM_LOAD_RETIRED.LOCAL_PMM) > MEM_LOAD_RETIRED.L1_MISS else 0) if #has_pmem > 0 else 0)", "MetricGroup": "MemoryBound;Server;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_pmm_bound", "MetricThreshold": "tma_pmm_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1415,7 +1705,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 0 ([SNB+] ALU; [HSW+] ALU and 2nd branch)", - "MetricExpr": "UOPS_DISPATCHED.PORT_0 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_0 / tma_info_core_core_clks", "MetricGroup": "Compute;TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_0", "MetricThreshold": "tma_port_0 > 0.6", @@ -1424,7 +1714,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 1 (ALU)", - "MetricExpr": "UOPS_DISPATCHED.PORT_1 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_1 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_1", "MetricThreshold": "tma_port_1 > 0.6", @@ -1433,7 +1723,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 6 ([HSW+]Primary Branch and simple ALU)", - "MetricExpr": "UOPS_DISPATCHED.PORT_6 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_6 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_6", "MetricThreshold": "tma_port_6 > 0.6", @@ -1442,7 +1732,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)", - "MetricExpr": "((cpu@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - EXE_ACTIVITY.BOUND_ON_LOADS) + (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * cpu@EXE_ACTIVITY.2_PORTS_UTIL\\,umask\\=0xc@)) / tma_info_clks if ARITH.DIV_ACTIVE < CYCLE_ACTIVITY.STALLS_TOTAL - EXE_ACTIVITY.BOUND_ON_LOADS else (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * cpu@EXE_ACTIVITY.2_PORTS_UTIL\\,umask\\=0xc@) / tma_info_clks)", + "MetricExpr": "((cpu@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - EXE_ACTIVITY.BOUND_ON_LOADS) + (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * cpu@EXE_ACTIVITY.2_PORTS_UTIL\\,umask\\=0xc@)) / tma_info_thread_clks if ARITH.DIV_ACTIVE < CYCLE_ACTIVITY.STALLS_TOTAL - EXE_ACTIVITY.BOUND_ON_LOADS else (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * cpu@EXE_ACTIVITY.2_PORTS_UTIL\\,umask\\=0xc@) / tma_info_thread_clks)", "MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_ports_utilization", "MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -1451,7 +1741,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed no uops on any execution port (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "cpu@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ / tma_info_clks + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - EXE_ACTIVITY.BOUND_ON_LOADS) / tma_info_clks", + "MetricExpr": "cpu@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ / tma_info_thread_clks + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - EXE_ACTIVITY.BOUND_ON_LOADS) / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_0", "MetricThreshold": "tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1460,7 +1750,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the CPU executed total of 1 uop per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "EXE_ACTIVITY.1_PORTS_UTIL / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.1_PORTS_UTIL / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issueL1;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_1", "MetricThreshold": "tma_ports_utilized_1 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1469,7 +1759,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 2 uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "EXE_ACTIVITY.2_PORTS_UTIL / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.2_PORTS_UTIL / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issue2P;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_2", "MetricThreshold": "tma_ports_utilized_2 > 0.15 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1478,7 +1768,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 3 or more uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "UOPS_EXECUTED.CYCLES_GE_3 / tma_info_clks", + "MetricExpr": "UOPS_EXECUTED.CYCLES_GE_3 / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_3m", "MetricThreshold": "tma_ports_utilized_3m > 0.7 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1487,7 +1777,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from remote cache in other sockets including synchronizations issues", - "MetricExpr": "(135.5 * tma_info_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM + 135.5 * tma_info_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "(135.5 * tma_info_system_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM + 135.5 * tma_info_system_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Offcore;Server;Snoop;TopdownL5;tma_L5_group;tma_issueSyncxn;tma_mem_latency_group", "MetricName": "tma_remote_cache", "MetricThreshold": "tma_remote_cache > 0.05 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1496,7 +1786,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from remote memory", - "MetricExpr": "149 * tma_info_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "149 * tma_info_system_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Server;Snoop;TopdownL5;tma_L5_group;tma_mem_latency_group", "MetricName": "tma_remote_dram", "MetricThreshold": "tma_remote_dram > 0.1 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1505,16 +1795,18 @@ }, { "BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired", - "MetricExpr": "topdown\\-retiring / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_slots", - "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "topdown\\-retiring / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_thread_slots", + "MetricGroup": "Default;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired. Ideally; all pipeline slots would be attributed to the Retiring category. Retiring of 100% would indicate the maximum Pipeline_Width throughput was achieved. Maximizing Retiring typically increases the Instructions-per-cycle (see IPC metric). Note that a high Retiring value does not necessary mean there is no room for more performance. For example; Heavy-operations or Microcode Assists are categorized under Retiring. They often indicate suboptimal performance and can often be optimized or avoided. Sample with: UOPS_RETIRED.SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU issue-pipeline was stalled due to serializing operations", - "MetricExpr": "RESOURCE_STALLS.SCOREBOARD / tma_info_clks", + "MetricExpr": "RESOURCE_STALLS.SCOREBOARD / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL5;tma_L5_group;tma_issueSO;tma_ports_utilized_0_group", "MetricName": "tma_serializing_operation", "MetricThreshold": "tma_serializing_operation > 0.1 & (tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)))", @@ -1523,7 +1815,7 @@ }, { "BriefDescription": "This metric represents Shuffle (cross \"vector lane\" data transfers) uops fraction the CPU has retired.", - "MetricExpr": "INT_VEC_RETIRED.SHUFFLES / (tma_retiring * tma_info_slots)", + "MetricExpr": "INT_VEC_RETIRED.SHUFFLES / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "HPC;Pipeline;TopdownL4;tma_L4_group;tma_int_operations_group", "MetricName": "tma_shuffles", "MetricThreshold": "tma_shuffles > 0.1 & (tma_int_operations > 0.1 & tma_light_operations > 0.6)", @@ -1531,7 +1823,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to PAUSE Instructions", - "MetricExpr": "CPU_CLK_UNHALTED.PAUSE / tma_info_clks", + "MetricExpr": "CPU_CLK_UNHALTED.PAUSE / tma_info_thread_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_serializing_operation_group", "MetricName": "tma_slow_pause", "MetricThreshold": "tma_slow_pause > 0.05 & (tma_serializing_operation > 0.1 & (tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))))", @@ -1540,7 +1832,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles handling memory load split accesses - load that cross 64-byte cache line boundary", - "MetricExpr": "tma_info_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_split_loads", "MetricThreshold": "tma_split_loads > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1549,7 +1841,7 @@ }, { "BriefDescription": "This metric represents rate of split store accesses", - "MetricExpr": "MEM_INST_RETIRED.SPLIT_STORES / tma_info_core_clks", + "MetricExpr": "MEM_INST_RETIRED.SPLIT_STORES / tma_info_core_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_issueSpSt;tma_store_bound_group", "MetricName": "tma_split_stores", "MetricThreshold": "tma_split_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1558,16 +1850,16 @@ }, { "BriefDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors)", - "MetricExpr": "(XQ.FULL_CYCLES + L1D_PEND_MISS.L2_STALLS) / tma_info_clks", + "MetricExpr": "(XQ.FULL_CYCLES + L1D_PEND_MISS.L2_STALLS) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueBW;tma_l3_bound_group", "MetricName": "tma_sq_full", "MetricThreshold": "tma_sq_full > 0.3 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_mem_bandwidth", + "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_mem_bandwidth", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write", - "MetricExpr": "EXE_ACTIVITY.BOUND_ON_STORES / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.BOUND_ON_STORES / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1576,7 +1868,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of cycles when the memory subsystem had loads blocked since they could not forward data from earlier (in program order) overlapping stores", - "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_clks", + "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_store_fwd_blk", "MetricThreshold": "tma_store_fwd_blk > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1585,7 +1877,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles the CPU spent handling L1D store misses", - "MetricExpr": "(MEM_STORE_RETIRED.L2_HIT * 10 * (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) + (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_clks", + "MetricExpr": "(MEM_STORE_RETIRED.L2_HIT * 10 * (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) + (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_thread_clks", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_issueSL;tma_store_bound_group", "MetricName": "tma_store_latency", "MetricThreshold": "tma_store_latency > 0.1 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1594,7 +1886,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Store operations", - "MetricExpr": "(UOPS_DISPATCHED.PORT_4_9 + UOPS_DISPATCHED.PORT_7_8) / (4 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED.PORT_4_9 + UOPS_DISPATCHED.PORT_7_8) / (4 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_store_op_utilization", "MetricThreshold": "tma_store_op_utilization > 0.6", @@ -1611,7 +1903,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles where the STLB was missed by store accesses, performing a hardware page walk", - "MetricExpr": "DTLB_STORE_MISSES.WALK_ACTIVE / tma_info_core_clks", + "MetricExpr": "DTLB_STORE_MISSES.WALK_ACTIVE / tma_info_core_core_clks", "MetricGroup": "MemoryTLB;TopdownL5;tma_L5_group;tma_dtlb_store_group", "MetricName": "tma_store_stlb_miss", "MetricThreshold": "tma_store_stlb_miss > 0.05 & (tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1619,7 +1911,7 @@ }, { "BriefDescription": "This metric estimates how often CPU was stalled due to Streaming store memory accesses; Streaming store optimize out a read request required by RFO stores", - "MetricExpr": "9 * OCR.STREAMING_WR.ANY_RESPONSE / tma_info_clks", + "MetricExpr": "9 * OCR.STREAMING_WR.ANY_RESPONSE / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueSmSt;tma_store_bound_group", "MetricName": "tma_streaming_stores", "MetricThreshold": "tma_streaming_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1628,7 +1920,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to new branch address clears", - "MetricExpr": "INT_MISC.UNKNOWN_BRANCH_CYCLES / tma_info_clks", + "MetricExpr": "INT_MISC.UNKNOWN_BRANCH_CYCLES / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;TopdownL4;tma_L4_group;tma_branch_resteers_group", "MetricName": "tma_unknown_branches", "MetricThreshold": "tma_unknown_branches > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", @@ -1671,5 +1963,17 @@ "MetricGroup": "transaction", "MetricName": "tsx_transactional_cycles", "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uncore operating frequency in GHz", + "MetricExpr": "UNC_CHA_CLOCKTICKS / (source_count(UNC_CHA_CLOCKTICKS) * #num_packages) / 1e9 / duration_time", + "MetricName": "uncore_frequency", + "ScaleUnit": "1GHz" + }, + { + "BriefDescription": "Intel(R) Ultra Path Interconnect (UPI) data transmit bandwidth (MB/sec)", + "MetricExpr": "UNC_UPI_TxL_FLITS.ALL_DATA * 7.111111111111111 / 1e6 / duration_time", + "MetricName": "upi_data_transmit_bw", + "ScaleUnit": "1MB/s" } ] diff --git a/tools/perf/pmu-events/arch/x86/sapphirerapids/uncore-interconnect.json b/tools/perf/pmu-events/arch/x86/sapphirerapids/uncore-interconnect.json index 08faf38115d9..6800de05c836 100644 --- a/tools/perf/pmu-events/arch/x86/sapphirerapids/uncore-interconnect.json +++ b/tools/perf/pmu-events/arch/x86/sapphirerapids/uncore-interconnect.json @@ -464,7 +464,7 @@ "Unit": "M2M" }, { - "BriefDescription": "Counts the time when FM didn? do d2c for fill reads (cross tile case)", + "BriefDescription": "Counts the time when FM didn't do d2c for fill reads (cross tile case)", "EventCode": "0x4a", "EventName": "UNC_M2M_DIRECT2CORE_NOT_TAKEN_NOTFORKED", "PerPkg": "1", diff --git a/tools/perf/pmu-events/arch/x86/sapphirerapids/uncore-memory.json b/tools/perf/pmu-events/arch/x86/sapphirerapids/uncore-memory.json index 225333561295..3ff9e9b722c8 100644 --- a/tools/perf/pmu-events/arch/x86/sapphirerapids/uncore-memory.json +++ b/tools/perf/pmu-events/arch/x86/sapphirerapids/uncore-memory.json @@ -2480,11 +2480,11 @@ "Unit": "iMC" }, { - "BriefDescription": "DRAM Precharge commands. : Precharge due to (?)", + "BriefDescription": "DRAM Precharge commands", "EventCode": "0x03", "EventName": "UNC_M_PRE_COUNT.PGT", "PerPkg": "1", - "PublicDescription": "DRAM Precharge commands. : Precharge due to (?) : Counts the number of DRAM Precharge commands sent on this channel.", + "PublicDescription": "DRAM Precharge commands. Counts the number of DRAM Precharge commands sent on this channel.", "UMask": "0x88", "Unit": "iMC" }, @@ -3236,7 +3236,7 @@ "Unit": "iMC" }, { - "BriefDescription": "2LM Tag check hit due to memory read (bug?)", + "BriefDescription": "2LM Tag check hit due to memory read", "EventCode": "0xd3", "EventName": "UNC_M_TAGCHK.NM_RD_HIT", "PerPkg": "1", @@ -3244,7 +3244,7 @@ "Unit": "iMC" }, { - "BriefDescription": "2LM Tag check hit due to memory write (bug?)", + "BriefDescription": "2LM Tag check hit due to memory write", "EventCode": "0xd3", "EventName": "UNC_M_TAGCHK.NM_WR_HIT", "PerPkg": "1", diff --git a/tools/perf/pmu-events/arch/x86/skylake/floating-point.json b/tools/perf/pmu-events/arch/x86/skylake/floating-point.json index 4d494a5cabbf..5891bd74af60 100644 --- a/tools/perf/pmu-events/arch/x86/skylake/floating-point.json +++ b/tools/perf/pmu-events/arch/x86/skylake/floating-point.json @@ -32,6 +32,14 @@ "UMask": "0x20" }, { + "BriefDescription": "Number of SSE/AVX computational 128-bit packed single and 256-bit packed double precision FP instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, 1 for each element. Applies to SSE* and AVX* packed single precision and packed double precision FP instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB count twice as they perform 2 calculations per element.", + "EventCode": "0xC7", + "EventName": "FP_ARITH_INST_RETIRED.4_FLOPS", + "PublicDescription": "Number of SSE/AVX computational 128-bit packed single precision and 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point and packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", + "SampleAfterValue": "1000003", + "UMask": "0x18" + }, + { "BriefDescription": "Counts once for most SIMD scalar computational floating-point instructions retired. Counts twice for DPP and FM(N)ADD/SUB instructions retired.", "EventCode": "0xC7", "EventName": "FP_ARITH_INST_RETIRED.SCALAR", diff --git a/tools/perf/pmu-events/arch/x86/skylake/metricgroups.json b/tools/perf/pmu-events/arch/x86/skylake/metricgroups.json new file mode 100644 index 000000000000..a151ba9cccb0 --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/skylake/metricgroups.json @@ -0,0 +1,113 @@ +{ + "Backend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Bad": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BadSpec": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BigFoot": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BrMispredicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Branches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CacheMisses": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CodeGen": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Compute": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Cor": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSBmiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DataSharing": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Fed": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Flops": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpScalar": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Frontend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "HPC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IcMiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "InsType": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "L2Evicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "LSD": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MachineClears": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBound": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryTLB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_BW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_Lat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MicroSeq": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "OS": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Offcore": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PGO": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Pipeline": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PortsUtil": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Prefetches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Ret": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Retire": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SMT": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Server": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Snoop": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SoC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL1": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL2": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL3mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "TopdownL5": "Metrics for top-down breakdown at level 5", + "TopdownL6": "Metrics for top-down breakdown at level 6", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_L5_group": "Metrics for top-down breakdown at level 5", + "tma_L6_group": "Metrics for top-down breakdown at level 6", + "tma_alu_op_utilization_group": "Metrics contributing to tma_alu_op_utilization category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_branch_resteers_group": "Metrics contributing to tma_branch_resteers category", + "tma_core_bound_group": "Metrics contributing to tma_core_bound category", + "tma_dram_bound_group": "Metrics contributing to tma_dram_bound category", + "tma_dtlb_load_group": "Metrics contributing to tma_dtlb_load category", + "tma_dtlb_store_group": "Metrics contributing to tma_dtlb_store category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_fp_arith_group": "Metrics contributing to tma_fp_arith category", + "tma_fp_vector_group": "Metrics contributing to tma_fp_vector category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_heavy_operations_group": "Metrics contributing to tma_heavy_operations category", + "tma_issue2P": "Metrics related by the issue $issue2P", + "tma_issueBC": "Metrics related by the issue $issueBC", + "tma_issueBM": "Metrics related by the issue $issueBM", + "tma_issueBW": "Metrics related by the issue $issueBW", + "tma_issueD0": "Metrics related by the issue $issueD0", + "tma_issueFB": "Metrics related by the issue $issueFB", + "tma_issueFL": "Metrics related by the issue $issueFL", + "tma_issueL1": "Metrics related by the issue $issueL1", + "tma_issueLat": "Metrics related by the issue $issueLat", + "tma_issueMC": "Metrics related by the issue $issueMC", + "tma_issueMS": "Metrics related by the issue $issueMS", + "tma_issueMV": "Metrics related by the issue $issueMV", + "tma_issueRFO": "Metrics related by the issue $issueRFO", + "tma_issueSL": "Metrics related by the issue $issueSL", + "tma_issueSO": "Metrics related by the issue $issueSO", + "tma_issueSmSt": "Metrics related by the issue $issueSmSt", + "tma_issueSpSt": "Metrics related by the issue $issueSpSt", + "tma_issueSyncxn": "Metrics related by the issue $issueSyncxn", + "tma_issueTLB": "Metrics related by the issue $issueTLB", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_l3_bound_group": "Metrics contributing to tma_l3_bound category", + "tma_light_operations_group": "Metrics contributing to tma_light_operations category", + "tma_load_op_utilization_group": "Metrics contributing to tma_load_op_utilization category", + "tma_mem_latency_group": "Metrics contributing to tma_mem_latency category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_microcode_sequencer_group": "Metrics contributing to tma_microcode_sequencer category", + "tma_mite_group": "Metrics contributing to tma_mite category", + "tma_ports_utilization_group": "Metrics contributing to tma_ports_utilization category", + "tma_ports_utilized_0_group": "Metrics contributing to tma_ports_utilized_0 category", + "tma_ports_utilized_3m_group": "Metrics contributing to tma_ports_utilized_3m category", + "tma_retiring_group": "Metrics contributing to tma_retiring category", + "tma_serializing_operation_group": "Metrics contributing to tma_serializing_operation category", + "tma_store_bound_group": "Metrics contributing to tma_store_bound category", + "tma_store_op_utilization_group": "Metrics contributing to tma_store_op_utilization category" +} diff --git a/tools/perf/pmu-events/arch/x86/skylake/pipeline.json b/tools/perf/pmu-events/arch/x86/skylake/pipeline.json index 2dfc3af08eff..cc800fb8180a 100644 --- a/tools/perf/pmu-events/arch/x86/skylake/pipeline.json +++ b/tools/perf/pmu-events/arch/x86/skylake/pipeline.json @@ -26,12 +26,21 @@ "UMask": "0x4" }, { - "BriefDescription": "Conditional branch instructions retired.", + "BriefDescription": "Conditional branch instructions retired. [This event is alias to BR_INST_RETIRED.CONDITIONAL]", + "Errata": "SKL091", + "EventCode": "0xC4", + "EventName": "BR_INST_RETIRED.COND", + "PublicDescription": "This event counts conditional branch instructions retired. [This event is alias to BR_INST_RETIRED.CONDITIONAL]", + "SampleAfterValue": "400009", + "UMask": "0x1" + }, + { + "BriefDescription": "Conditional branch instructions retired. [This event is alias to BR_INST_RETIRED.COND]", "Errata": "SKL091", "EventCode": "0xC4", "EventName": "BR_INST_RETIRED.CONDITIONAL", "PEBS": "1", - "PublicDescription": "This event counts conditional branch instructions retired.", + "PublicDescription": "This event counts conditional branch instructions retired. [This event is alias to BR_INST_RETIRED.COND]", "SampleAfterValue": "400009", "UMask": "0x1" }, @@ -405,9 +414,9 @@ "UMask": "0x1" }, { - "AnyThread": "1", "BriefDescription": "Clears speculative count", "CounterMask": "1", + "EdgeDetect": "1", "EventCode": "0x0D", "EventName": "INT_MISC.CLEARS_COUNT", "PublicDescription": "Counts the number of speculative clears due to any type of branch misprediction or machine clears", diff --git a/tools/perf/pmu-events/arch/x86/skylake/skl-metrics.json b/tools/perf/pmu-events/arch/x86/skylake/skl-metrics.json index a6d212b349f5..2ed88842b880 100644 --- a/tools/perf/pmu-events/arch/x86/skylake/skl-metrics.json +++ b/tools/perf/pmu-events/arch/x86/skylake/skl-metrics.json @@ -50,7 +50,7 @@ }, { "BriefDescription": "Uncore frequency per die [GHZ]", - "MetricExpr": "tma_info_socket_clks / #num_dies / duration_time / 1e9", + "MetricExpr": "tma_info_system_socket_clks / #num_dies / duration_time / 1e9", "MetricGroup": "SoC", "MetricName": "UNCORE_FREQ" }, @@ -71,7 +71,7 @@ }, { "BriefDescription": "This metric estimates how often memory load accesses were aliased by preceding stores (in program order) with a 4K address offset", - "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_clks", + "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_4k_aliasing", "MetricThreshold": "tma_4k_aliasing > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -80,7 +80,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution ports for ALU operations.", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5 + UOPS_DISPATCHED_PORT.PORT_6) / tma_info_slots", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5 + UOPS_DISPATCHED_PORT.PORT_6) / tma_info_thread_slots", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_alu_op_utilization", "MetricThreshold": "tma_alu_op_utilization > 0.6", @@ -88,7 +88,7 @@ }, { "BriefDescription": "This metric estimates fraction of slots the CPU retired uops delivered by the Microcode_Sequencer as a result of Assists", - "MetricExpr": "100 * (FP_ASSIST.ANY + OTHER_ASSISTS.ANY) / tma_info_slots", + "MetricExpr": "100 * (FP_ASSIST.ANY + OTHER_ASSISTS.ANY) / tma_info_thread_slots", "MetricGroup": "TopdownL4;tma_L4_group;tma_microcode_sequencer_group", "MetricName": "tma_assists", "MetricThreshold": "tma_assists > 0.1 & (tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1)", @@ -97,19 +97,21 @@ }, { "BriefDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend", - "MetricExpr": "1 - tma_frontend_bound - (UOPS_ISSUED.ANY + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_slots", + "MetricExpr": "1 - tma_frontend_bound - (UOPS_ISSUED.ANY + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend. Backend is the portion of the processor core where the out-of-order scheduler dispatches ready uops into their respective execution units; and once completed these uops get retired according to program order. For example; stalls due to data-cache misses or stalls due to the divider unit being overloaded are both categorized under Backend Bound. Backend Bound is further divided into two main categories: Memory Bound and Core Bound.", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations", - "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_slots", + "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots wasted due to incorrect speculations. This include slots used to issue uops that do not eventually get retired and slots for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work due to miss-predicted branches are categorized under Bad Speculation category. Incorrect data speculation followed by Memory Ordering Nukes is another example.", "ScaleUnit": "100%" }, @@ -120,12 +122,13 @@ "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15", - "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_branch_misprediction_cost, tma_info_mispredictions, tma_mispredicts_resteers", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_info_bottleneck_mispredictions, tma_mispredicts_resteers", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers", - "MetricExpr": "INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks + tma_unknown_branches", + "MetricExpr": "INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks + tma_unknown_branches", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteers", "MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -143,7 +146,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Machine Clears", - "MetricExpr": "(1 - BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT)) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks", + "MetricExpr": "(1 - BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT)) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks", "MetricGroup": "BadSpec;MachineClears;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueMC", "MetricName": "tma_clears_resteers", "MetricThreshold": "tma_clears_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", @@ -153,7 +156,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to contested accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(18.5 * tma_info_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_HITM + 16.5 * tma_info_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "(18.5 * tma_info_system_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_HITM + 16.5 * tma_info_system_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_contested_accesses", "MetricThreshold": "tma_contested_accesses > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -167,13 +170,14 @@ "MetricGroup": "Backend;Compute;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck. Shortage in hardware compute resources; or dependencies in software's instructions are both categorized under Core Bound. Hence it may indicate the machine ran out of an out-of-order resource; certain execution units are overloaded or dependencies in program's data- or instruction-flow are limiting the performance (e.g. FP-chained long-latency arithmetic operations).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to data-sharing accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "16.5 * tma_info_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "16.5 * tma_info_system_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_data_sharing", "MetricThreshold": "tma_data_sharing > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -182,16 +186,16 @@ }, { "BriefDescription": "This metric represents fraction of cycles where decoder-0 was the only active decoder", - "MetricExpr": "(cpu@INST_DECODED.DECODERS\\,cmask\\=1@ - cpu@INST_DECODED.DECODERS\\,cmask\\=2@) / tma_info_core_clks / 2", + "MetricExpr": "(cpu@INST_DECODED.DECODERS\\,cmask\\=1@ - cpu@INST_DECODED.DECODERS\\,cmask\\=2@) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL4;tma_L4_group;tma_issueD0;tma_mite_group", "MetricName": "tma_decoder0_alone", - "MetricThreshold": "tma_decoder0_alone > 0.1 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35))", + "MetricThreshold": "tma_decoder0_alone > 0.1 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35))", "PublicDescription": "This metric represents fraction of cycles where decoder-0 was the only active decoder. Related metrics: tma_few_uops_instructions", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles where the Divider unit was active", - "MetricExpr": "ARITH.DIVIDER_ACTIVE / tma_info_clks", + "MetricExpr": "ARITH.DIVIDER_ACTIVE / tma_info_thread_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_divider", "MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -201,7 +205,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks - tma_l2_bound", + "MetricExpr": "CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_thread_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks - tma_l2_bound", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -210,45 +214,45 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline", - "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSB;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_dsb", - "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline. For example; inefficient utilization of the DSB cache structure or bank conflict when reading from it; are categorized here.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_dsb_switches", "MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb, tma_lcp", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "min(9 * cpu@DTLB_LOAD_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_LOAD_MISSES.WALK_ACTIVE, max(CYCLE_ACTIVITY.CYCLES_MEM_ANY - CYCLE_ACTIVITY.CYCLES_L1D_MISS, 0)) / tma_info_clks", + "MetricExpr": "min(9 * cpu@DTLB_LOAD_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_LOAD_MISSES.WALK_ACTIVE, max(CYCLE_ACTIVITY.CYCLES_MEM_ANY - CYCLE_ACTIVITY.CYCLES_L1D_MISS, 0)) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group", "MetricName": "tma_dtlb_load", "MetricThreshold": "tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses. TLBs (Translation Look-aside Buffers) are processor caches for recently used entries out of the Page Tables that are used to map virtual- to physical-addresses by the operating system. This metric approximates the potential delay of demand loads missing the first-level data TLB (assuming worst case scenario with back to back misses to different pages). This includes hitting in the second-level TLB (STLB) as well as performing a hardware page walk on an STLB miss. Sample with: MEM_INST_RETIRED.STLB_MISS_LOADS_PS. Related metrics: tma_dtlb_store, tma_info_memory_data_tlbs", + "PublicDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses. TLBs (Translation Look-aside Buffers) are processor caches for recently used entries out of the Page Tables that are used to map virtual- to physical-addresses by the operating system. This metric approximates the potential delay of demand loads missing the first-level data TLB (assuming worst case scenario with back to back misses to different pages). This includes hitting in the second-level TLB (STLB) as well as performing a hardware page walk on an STLB miss. Sample with: MEM_INST_RETIRED.STLB_MISS_LOADS_PS. Related metrics: tma_dtlb_store, tma_info_bottleneck_memory_data_tlbs", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses", - "MetricExpr": "(9 * cpu@DTLB_STORE_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_STORE_MISSES.WALK_ACTIVE) / tma_info_core_clks", + "MetricExpr": "(9 * cpu@DTLB_STORE_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_STORE_MISSES.WALK_ACTIVE) / tma_info_core_core_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_store_bound_group", "MetricName": "tma_dtlb_store", "MetricThreshold": "tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses. As with ordinary data caching; focus on improving data locality and reducing working-set size to reduce DTLB overhead. Additionally; consider using profile-guided optimization (PGO) to collocate frequently-used data on the same page. Try using larger page sizes for large amounts of frequently-used data. Sample with: MEM_INST_RETIRED.STLB_MISS_STORES_PS. Related metrics: tma_dtlb_load, tma_info_memory_data_tlbs", + "PublicDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses. As with ordinary data caching; focus on improving data locality and reducing working-set size to reduce DTLB overhead. Additionally; consider using profile-guided optimization (PGO) to collocate frequently-used data on the same page. Try using larger page sizes for large amounts of frequently-used data. Sample with: MEM_INST_RETIRED.STLB_MISS_STORES_PS. Related metrics: tma_dtlb_load, tma_info_bottleneck_memory_data_tlbs", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates how often CPU was handling synchronizations due to False Sharing", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "22 * tma_info_average_frequency * OFFCORE_RESPONSE.DEMAND_RFO.L3_HIT.SNOOP_HITM / tma_info_clks", + "MetricExpr": "22 * tma_info_system_average_frequency * OFFCORE_RESPONSE.DEMAND_RFO.L3_HIT.SNOOP_HITM / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_store_bound_group", "MetricName": "tma_false_sharing", "MetricThreshold": "tma_false_sharing > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -258,11 +262,11 @@ { "BriefDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "tma_info_load_miss_real_latency * cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@ / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@ / tma_info_thread_clks", "MetricGroup": "MemoryBW;TopdownL4;tma_L4_group;tma_issueBW;tma_issueSL;tma_issueSmSt;tma_l1_bound_group", "MetricName": "tma_fb_full", "MetricThreshold": "tma_fb_full > 0.3", - "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", + "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", "ScaleUnit": "100%" }, { @@ -270,16 +274,18 @@ "MetricExpr": "tma_frontend_bound - tma_fetch_latency", "MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", "MetricName": "tma_fetch_bandwidth", - "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb, tma_lcp", + "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues", - "MetricExpr": "4 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE / tma_info_slots", + "MetricExpr": "4 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE / tma_info_thread_slots", "MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: FRONTEND_RETIRED.LATENCY_GE_16_PS;FRONTEND_RETIRED.LATENCY_GE_8_PS", "ScaleUnit": "100%" }, @@ -341,10 +347,11 @@ }, { "BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend", - "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_slots", + "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_thread_slots", "MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound. Sample with: FRONTEND_RETIRED.LATENCY_GE_4_PS", "ScaleUnit": "100%" }, @@ -359,16 +366,17 @@ }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences", - "MetricExpr": "(UOPS_RETIRED.RETIRE_SLOTS + UOPS_RETIRED.MACRO_FUSED - INST_RETIRED.ANY) / tma_info_slots", + "MetricExpr": "(UOPS_RETIRED.RETIRE_SLOTS + UOPS_RETIRED.MACRO_FUSED - INST_RETIRED.ANY) / tma_info_thread_slots", "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_heavy_operations", "MetricThreshold": "tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to instruction cache misses", - "MetricExpr": "(ICACHE_16B.IFDATA_STALL + 2 * cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@) / tma_info_clks", + "MetricExpr": "(ICACHE_16B.IFDATA_STALL + 2 * cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@) / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;IcMiss;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_icache_misses", "MetricThreshold": "tma_icache_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -376,220 +384,231 @@ "ScaleUnit": "100%" }, { - "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", - "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time", - "MetricGroup": "Power;Summary", - "MetricName": "tma_info_average_frequency" + "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", + "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_thread_slots / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BrMispredicts;tma_issueBM", + "MetricName": "tma_info_bad_spec_branch_misprediction_cost", + "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_info_bottleneck_mispredictions, tma_mispredicts_resteers" + }, + { + "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", + "MetricExpr": "tma_info_inst_mix_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_indirect", + "MetricThreshold": "tma_info_bad_spec_ipmisp_indirect < 1e3" + }, + { + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BadSpec;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmispredict", + "MetricThreshold": "tma_info_bad_spec_ipmispredict < 200" + }, + { + "BriefDescription": "Probability of Core Bound bottleneck hidden by SMT-profiling artifacts", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(100 * (1 - tma_core_bound / tma_ports_utilization if tma_core_bound < tma_ports_utilization else 1) if tma_info_system_smt_2t_utilization > 0.5 else 0)", + "MetricGroup": "Cor;SMT", + "MetricName": "tma_info_botlnk_l0_core_bound_likely", + "MetricThreshold": "tma_info_botlnk_l0_core_bound_likely > 0.5" + }, + { + "BriefDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_fetch_latency * tma_dsb_switches / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches) + tma_fetch_bandwidth * tma_mite / (tma_dsb + tma_mite))", + "MetricGroup": "DSBmiss;Fed;tma_issueFB", + "MetricName": "tma_info_botlnk_l2_dsb_misses", + "MetricThreshold": "tma_info_botlnk_l2_dsb_misses > 10", + "PublicDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp" + }, + { + "BriefDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck", + "MetricExpr": "100 * (tma_fetch_latency * tma_icache_misses / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", + "MetricGroup": "Fed;FetchLat;IcMiss;tma_issueFL", + "MetricName": "tma_info_botlnk_l2_ic_misses", + "MetricThreshold": "tma_info_botlnk_l2_ic_misses > 5", + "PublicDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck. Related metrics: " }, { "BriefDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses)", "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "100 * tma_fetch_latency * (tma_itlb_misses + tma_icache_misses + tma_unknown_branches) / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)", "MetricGroup": "BigFoot;Fed;Frontend;IcMiss;MemoryTLB;tma_issueBC", - "MetricName": "tma_info_big_code", - "MetricThreshold": "tma_info_big_code > 20", - "PublicDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses). Related metrics: tma_info_branching_overhead" + "MetricName": "tma_info_bottleneck_big_code", + "MetricThreshold": "tma_info_bottleneck_big_code > 20", + "PublicDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses). Related metrics: tma_info_bottleneck_branching_overhead" }, { - "BriefDescription": "Branch instructions per taken branch.", - "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", - "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_bptkbranch" + "BriefDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls)", + "MetricExpr": "100 * ((BR_INST_RETIRED.CONDITIONAL + 3 * BR_INST_RETIRED.NEAR_CALL + (BR_INST_RETIRED.NEAR_TAKEN - (BR_INST_RETIRED.CONDITIONAL - BR_INST_RETIRED.NOT_TAKEN) - 2 * BR_INST_RETIRED.NEAR_CALL)) / tma_info_thread_slots)", + "MetricGroup": "Ret;tma_issueBC", + "MetricName": "tma_info_bottleneck_branching_overhead", + "MetricThreshold": "tma_info_bottleneck_branching_overhead > 10", + "PublicDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls). Related metrics: tma_info_bottleneck_big_code" }, { - "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", - "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_slots / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_branch_misprediction_cost", - "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_info_mispredictions, tma_mispredicts_resteers" + "BriefDescription": "Total pipeline cost of instruction fetch bandwidth related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_frontend_bound - tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) - tma_info_bottleneck_big_code", + "MetricGroup": "Fed;FetchBW;Frontend", + "MetricName": "tma_info_bottleneck_instruction_fetch_bw", + "MetricThreshold": "tma_info_bottleneck_instruction_fetch_bw > 20" }, { - "BriefDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls)", - "MetricExpr": "100 * ((BR_INST_RETIRED.CONDITIONAL + 3 * BR_INST_RETIRED.NEAR_CALL + (BR_INST_RETIRED.NEAR_TAKEN - (BR_INST_RETIRED.CONDITIONAL - BR_INST_RETIRED.NOT_TAKEN) - 2 * BR_INST_RETIRED.NEAR_CALL)) / tma_info_slots)", - "MetricGroup": "Ret;tma_issueBC", - "MetricName": "tma_info_branching_overhead", - "MetricThreshold": "tma_info_branching_overhead > 10", - "PublicDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls). Related metrics: tma_info_big_code" + "BriefDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_bandwidth / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_sq_full / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full))) + tma_l1_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_fb_full / (tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk))", + "MetricGroup": "Mem;MemoryBW;Offcore;tma_issueBW", + "MetricName": "tma_info_bottleneck_memory_bandwidth", + "MetricThreshold": "tma_info_bottleneck_memory_bandwidth > 20", + "PublicDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks. Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full" }, { - "BriefDescription": "Fraction of branches that are CALL or RET", - "MetricExpr": "(BR_INST_RETIRED.NEAR_CALL + BR_INST_RETIRED.NEAR_RETURN) / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_callret" + "BriefDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_l1_bound / max(tma_memory_bound, tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_load / max(tma_l1_bound, tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk)) + tma_store_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_store / (tma_dtlb_store + tma_false_sharing + tma_split_stores + tma_store_latency)))", + "MetricGroup": "Mem;MemoryTLB;Offcore;tma_issueTLB", + "MetricName": "tma_info_bottleneck_memory_data_tlbs", + "MetricThreshold": "tma_info_bottleneck_memory_data_tlbs > 20", + "PublicDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs). Related metrics: tma_dtlb_load, tma_dtlb_store" }, { - "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "Pipeline", - "MetricName": "tma_info_clks" + "BriefDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_latency / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_l3_hit_latency / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full)) + tma_l2_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound))", + "MetricGroup": "Mem;MemoryLat;Offcore;tma_issueLat", + "MetricName": "tma_info_bottleneck_memory_latency", + "MetricThreshold": "tma_info_bottleneck_memory_latency > 20", + "PublicDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches). Related metrics: tma_l3_hit_latency, tma_mem_latency" }, { - "BriefDescription": "STLB (2nd level TLB) code speculative misses per kilo instruction (misses of any page-size that complete the page walk)", - "MetricExpr": "1e3 * ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", - "MetricGroup": "Fed;MemoryTLB", - "MetricName": "tma_info_code_stlb_mpki" + "BriefDescription": "Total pipeline cost of Branch Misprediction related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", + "MetricGroup": "Bad;BadSpec;BrMispredicts;tma_issueBM", + "MetricName": "tma_info_bottleneck_mispredictions", + "MetricThreshold": "tma_info_bottleneck_mispredictions > 20", + "PublicDescription": "Total pipeline cost of Branch Misprediction related bottlenecks. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers" + }, + { + "BriefDescription": "Fraction of branches that are CALL or RET", + "MetricExpr": "(BR_INST_RETIRED.NEAR_CALL + BR_INST_RETIRED.NEAR_RETURN) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_callret" }, { "BriefDescription": "Fraction of branches that are non-taken conditionals", "MetricExpr": "BR_INST_RETIRED.NOT_TAKEN / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Bad;Branches;CodeGen;PGO", - "MetricName": "tma_info_cond_nt" + "MetricName": "tma_info_branches_cond_nt" }, { "BriefDescription": "Fraction of branches that are taken conditionals", "MetricExpr": "(BR_INST_RETIRED.CONDITIONAL - BR_INST_RETIRED.NOT_TAKEN) / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Bad;Branches;CodeGen;PGO", - "MetricName": "tma_info_cond_tk" + "MetricName": "tma_info_branches_cond_tk" }, { - "BriefDescription": "Probability of Core Bound bottleneck hidden by SMT-profiling artifacts", + "BriefDescription": "Fraction of branches that are unconditional (direct or indirect) jumps", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(100 * (1 - tma_core_bound / tma_ports_utilization if tma_core_bound < tma_ports_utilization else 1) if tma_info_smt_2t_utilization > 0.5 else 0)", - "MetricGroup": "Cor;SMT", - "MetricName": "tma_info_core_bound_likely", - "MetricThreshold": "tma_info_core_bound_likely > 0.5" + "MetricExpr": "(BR_INST_RETIRED.NEAR_TAKEN - (BR_INST_RETIRED.CONDITIONAL - BR_INST_RETIRED.NOT_TAKEN) - 2 * BR_INST_RETIRED.NEAR_CALL) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_jump" }, { "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", - "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_clks))", + "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_thread_clks))", "MetricGroup": "SMT", - "MetricName": "tma_info_core_clks" + "MetricName": "tma_info_core_core_clks" }, { "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks", "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_coreipc" + "MetricName": "tma_info_core_coreipc" }, { - "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricExpr": "1 / tma_info_ipc", - "MetricGroup": "Mem;Pipeline", - "MetricName": "tma_info_cpi" - }, - { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": "HPC;Summary", - "MetricName": "tma_info_cpu_utilization" + "BriefDescription": "Floating Point Operations Per Cycle", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / tma_info_core_core_clks", + "MetricGroup": "Flops;Ret", + "MetricName": "tma_info_core_flopc" }, { - "BriefDescription": "Average Parallel L2 cache miss data reads", - "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", - "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_data_l2_mlp" + "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", + "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@) / (2 * tma_info_core_core_clks)", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_core_fp_arith_utilization", + "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." }, { - "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricExpr": "64 * (UNC_ARB_TRK_REQUESTS.ALL + UNC_ARB_COH_TRK_REQUESTS.ALL) / 1e6 / duration_time / 1e3", - "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", - "MetricName": "tma_info_dram_bw_use", - "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_info_memory_bandwidth, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", + "MetricExpr": "UOPS_EXECUTED.THREAD / (UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1)", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", + "MetricName": "tma_info_core_ilp" }, { "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", "MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)", "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", - "MetricName": "tma_info_dsb_coverage", - "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_misses, tma_info_iptb, tma_lcp" - }, - { - "BriefDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * (tma_fetch_latency * tma_dsb_switches / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches) + tma_fetch_bandwidth * tma_mite / (tma_dsb + tma_mite))", - "MetricGroup": "DSBmiss;Fed;tma_issueFB", - "MetricName": "tma_info_dsb_misses", - "MetricThreshold": "tma_info_dsb_misses > 10", - "PublicDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb, tma_lcp" + "MetricName": "tma_info_frontend_dsb_coverage", + "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 4 > 0.35", + "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_inst_mix_iptb, tma_lcp" }, { "BriefDescription": "Average number of cycles of a switch from the DSB fetch-unit to MITE fetch unit - see DSB_Switches tree node for details.", "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / DSB2MITE_SWITCHES.COUNT", "MetricGroup": "DSBmiss", - "MetricName": "tma_info_dsb_switch_cost" - }, - { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", - "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", - "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", - "MetricName": "tma_info_execute" - }, - { - "BriefDescription": "The ratio of Executed- by Issued-Uops", - "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", - "MetricGroup": "Cor;Pipeline", - "MetricName": "tma_info_execute_per_issue", - "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." - }, - { - "BriefDescription": "Fill Buffer (FB) hits per kilo instructions for retired demand loads (L1D misses that merge into ongoing miss-handling entries)", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.FB_HIT / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_fb_hpki" + "MetricName": "tma_info_frontend_dsb_switch_cost" }, { "BriefDescription": "Average number of Uops issued by front-end when it issued something", "MetricExpr": "UOPS_ISSUED.ANY / cpu@UOPS_ISSUED.ANY\\,cmask\\=1@", "MetricGroup": "Fed;FetchBW", - "MetricName": "tma_info_fetch_upc" + "MetricName": "tma_info_frontend_fetch_upc" }, { - "BriefDescription": "Floating Point Operations Per Cycle", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / tma_info_core_clks", - "MetricGroup": "Flops;Ret", - "MetricName": "tma_info_flopc" - }, - { - "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@) / (2 * tma_info_core_clks)", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_fp_arith_utilization", - "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." + "BriefDescription": "Average Latency for L1 instruction cache misses", + "MetricExpr": "ICACHE_16B.IFDATA_STALL / cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@ + 2", + "MetricGroup": "Fed;FetchLat;IcMiss", + "MetricName": "tma_info_frontend_icache_miss_latency" }, { - "BriefDescription": "Giga Floating Point Operations Per Second", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / 1e9 / duration_time", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_gflops", - "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + "BriefDescription": "Instructions per non-speculative DSB miss (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / FRONTEND_RETIRED.ANY_DSB_MISS", + "MetricGroup": "DSBmiss;Fed", + "MetricName": "tma_info_frontend_ipdsb_miss_ret", + "MetricThreshold": "tma_info_frontend_ipdsb_miss_ret < 50" }, { - "BriefDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck", - "MetricExpr": "100 * (tma_fetch_latency * tma_icache_misses / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", - "MetricGroup": "Fed;FetchLat;IcMiss;tma_issueFL", - "MetricName": "tma_info_ic_misses", - "MetricThreshold": "tma_info_ic_misses > 5", - "PublicDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck. Related metrics: " + "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_inst_mix_instructions / BACLEARS.ANY", + "MetricGroup": "Fed", + "MetricName": "tma_info_frontend_ipunknown_branch" }, { - "BriefDescription": "Average Latency for L1 instruction cache misses", - "MetricExpr": "ICACHE_16B.IFDATA_STALL / cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@ + 2", - "MetricGroup": "Fed;FetchLat;IcMiss", - "MetricName": "tma_info_icache_miss_latency" + "BriefDescription": "L2 cache true code cacheline misses per kilo instruction", + "MetricExpr": "1e3 * FRONTEND_RETIRED.L2_MISS / INST_RETIRED.ANY", + "MetricGroup": "IcMiss", + "MetricName": "tma_info_frontend_l2mpki_code" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", - "MetricExpr": "UOPS_EXECUTED.THREAD / (UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1)", - "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", - "MetricName": "tma_info_ilp" + "BriefDescription": "L2 cache speculative code cacheline misses per kilo instruction", + "MetricExpr": "1e3 * L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", + "MetricGroup": "IcMiss", + "MetricName": "tma_info_frontend_l2mpki_code_all" }, { - "BriefDescription": "Total pipeline cost of instruction fetch bandwidth related bottlenecks", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * (tma_frontend_bound - tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) - tma_info_big_code", - "MetricGroup": "Fed;FetchBW;Frontend", - "MetricName": "tma_info_instruction_fetch_bw", - "MetricThreshold": "tma_info_instruction_fetch_bw > 20" + "BriefDescription": "Branch instructions per taken branch.", + "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;PGO", + "MetricName": "tma_info_inst_mix_bptkbranch" }, { "BriefDescription": "Total number of retired Instructions", "MetricExpr": "INST_RETIRED.ANY", "MetricGroup": "Summary;TmaL1;tma_L1_group", - "MetricName": "tma_info_instructions", + "MetricName": "tma_info_inst_mix_instructions", "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" }, { @@ -597,416 +616,404 @@ "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "INST_RETIRED.ANY / (cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0x3c@)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_iparith", - "MetricThreshold": "tma_info_iparith < 10", + "MetricName": "tma_info_inst_mix_iparith", + "MetricThreshold": "tma_info_inst_mix_iparith < 10", "PublicDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate). May undercount due to FMA double counting. Approximated prior to BDW." }, { "BriefDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx128", - "MetricThreshold": "tma_info_iparith_avx128 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx128", + "MetricThreshold": "tma_info_inst_mix_iparith_avx128 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx256", - "MetricThreshold": "tma_info_iparith_avx256 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx256", + "MetricThreshold": "tma_info_inst_mix_iparith_avx256 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_dp", - "MetricThreshold": "tma_info_iparith_scalar_dp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_dp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_dp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_SINGLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_sp", - "MetricThreshold": "tma_info_iparith_scalar_sp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_sp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_sp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per Branch (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Branches;Fed;InsType", - "MetricName": "tma_info_ipbranch", - "MetricThreshold": "tma_info_ipbranch < 8" - }, - { - "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": "Ret;Summary", - "MetricName": "tma_info_ipc" + "MetricName": "tma_info_inst_mix_ipbranch", + "MetricThreshold": "tma_info_inst_mix_ipbranch < 8" }, { "BriefDescription": "Instructions per (near) call (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_CALL", "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_ipcall", - "MetricThreshold": "tma_info_ipcall < 200" - }, - { - "BriefDescription": "Instructions per non-speculative DSB miss (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / FRONTEND_RETIRED.ANY_DSB_MISS", - "MetricGroup": "DSBmiss;Fed", - "MetricName": "tma_info_ipdsb_miss_ret", - "MetricThreshold": "tma_info_ipdsb_miss_ret < 50" - }, - { - "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", - "MetricGroup": "Branches;OS", - "MetricName": "tma_info_ipfarbranch", - "MetricThreshold": "tma_info_ipfarbranch < 1e6" + "MetricName": "tma_info_inst_mix_ipcall", + "MetricThreshold": "tma_info_inst_mix_ipcall < 200" }, { "BriefDescription": "Instructions per Floating Point (FP) Operation (lower number means higher occurrence rate)", "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_ipflop", - "MetricThreshold": "tma_info_ipflop < 10" + "MetricName": "tma_info_inst_mix_ipflop", + "MetricThreshold": "tma_info_inst_mix_ipflop < 10" }, { "BriefDescription": "Instructions per Load (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_LOADS", "MetricGroup": "InsType", - "MetricName": "tma_info_ipload", - "MetricThreshold": "tma_info_ipload < 3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", - "MetricExpr": "tma_info_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_indirect", - "MetricThreshold": "tma_info_ipmisp_indirect < 1e3" - }, - { - "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BadSpec;BrMispredicts", - "MetricName": "tma_info_ipmispredict", - "MetricThreshold": "tma_info_ipmispredict < 200" + "MetricName": "tma_info_inst_mix_ipload", + "MetricThreshold": "tma_info_inst_mix_ipload < 3" }, { "BriefDescription": "Instructions per Store (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_STORES", "MetricGroup": "InsType", - "MetricName": "tma_info_ipstore", - "MetricThreshold": "tma_info_ipstore < 8" + "MetricName": "tma_info_inst_mix_ipstore", + "MetricThreshold": "tma_info_inst_mix_ipstore < 8" }, { "BriefDescription": "Instructions per Software prefetch instruction (of any type: NTA/T0/T1/T2/Prefetch) (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / cpu@SW_PREFETCH_ACCESS.T0\\,umask\\=0xF@", "MetricGroup": "Prefetches", - "MetricName": "tma_info_ipswpf", - "MetricThreshold": "tma_info_ipswpf < 100" + "MetricName": "tma_info_inst_mix_ipswpf", + "MetricThreshold": "tma_info_inst_mix_ipswpf < 100" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW;Frontend;PGO;tma_issueFB", - "MetricName": "tma_info_iptb", - "MetricThreshold": "tma_info_iptb < 9", - "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_lcp" - }, - { - "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", - "MetricExpr": "tma_info_instructions / BACLEARS.ANY", - "MetricGroup": "Fed", - "MetricName": "tma_info_ipunknown_branch" + "MetricName": "tma_info_inst_mix_iptb", + "MetricThreshold": "tma_info_inst_mix_iptb < 9", + "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_lcp" }, { - "BriefDescription": "Fraction of branches that are unconditional (direct or indirect) jumps", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(BR_INST_RETIRED.NEAR_TAKEN - (BR_INST_RETIRED.CONDITIONAL - BR_INST_RETIRED.NOT_TAKEN) - 2 * BR_INST_RETIRED.NEAR_CALL) / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_jump" + "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_core_l1d_cache_fill_bw" }, { - "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_cpi" + "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_core_l2_cache_fill_bw" }, { - "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_utilization", - "MetricThreshold": "tma_info_kernel_utilization > 0.05" + "BriefDescription": "Average per-core data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * OFFCORE_REQUESTS.ALL_REQUESTS / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_core_l3_cache_access_bw" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", + "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw" + "MetricName": "tma_info_memory_core_l3_cache_fill_bw" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "tma_info_l1d_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw_1t" + "BriefDescription": "Fill Buffer (FB) hits per kilo instructions for retired demand loads (L1D misses that merge into ongoing miss-handling entries)", + "MetricExpr": "1e3 * MEM_LOAD_RETIRED.FB_HIT / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_fb_hpki" }, { "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L1_MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki" + "MetricName": "tma_info_memory_l1mpki" }, { "BriefDescription": "L1 cache true misses per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.ALL_DEMAND_DATA_RD / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki_load" - }, - { - "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw" - }, - { - "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "tma_info_l2_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw_1t" + "MetricName": "tma_info_memory_l1mpki_load" }, { "BriefDescription": "L2 cache hits per kilo instruction for all request types (including speculative)", "MetricExpr": "1e3 * (L2_RQSTS.REFERENCES - L2_RQSTS.MISS) / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_all" + "MetricName": "tma_info_memory_l2hpki_all" }, { "BriefDescription": "L2 cache hits per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_HIT / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_load" + "MetricName": "tma_info_memory_l2hpki_load" }, { "BriefDescription": "L2 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L2_MISS / INST_RETIRED.ANY", "MetricGroup": "Backend;CacheMisses;Mem", - "MetricName": "tma_info_l2mpki" + "MetricName": "tma_info_memory_l2mpki" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all request types (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem;Offcore", - "MetricName": "tma_info_l2mpki_all" - }, - { - "BriefDescription": "L2 cache true code cacheline misses per kilo instruction", - "MetricExpr": "1e3 * FRONTEND_RETIRED.L2_MISS / INST_RETIRED.ANY", - "MetricGroup": "IcMiss", - "MetricName": "tma_info_l2mpki_code" - }, - { - "BriefDescription": "L2 cache speculative code cacheline misses per kilo instruction", - "MetricExpr": "1e3 * L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", - "MetricGroup": "IcMiss", - "MetricName": "tma_info_l2mpki_code_all" + "MetricName": "tma_info_memory_l2mpki_all" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2mpki_load" - }, - { - "BriefDescription": "Average per-core data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * OFFCORE_REQUESTS.ALL_REQUESTS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw" + "MetricName": "tma_info_memory_l2mpki_load" }, { - "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_access_bw", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw_1t" + "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L3_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l3mpki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw" + "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", + "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_RETIRED.L1_MISS + MEM_LOAD_RETIRED.FB_HIT)", + "MetricGroup": "Mem;MemoryBound;MemoryLat", + "MetricName": "tma_info_memory_load_miss_real_latency" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw_1t" + "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", + "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", + "MetricGroup": "Mem;MemoryBW;MemoryBound", + "MetricName": "tma_info_memory_mlp", + "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" }, { - "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L3_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l3mpki" + "BriefDescription": "Average Parallel L2 cache miss data reads", + "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", + "MetricGroup": "Memory_BW;Offcore", + "MetricName": "tma_info_memory_oro_data_l2_mlp" }, { "BriefDescription": "Average Latency for L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS.DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l2_miss_latency" + "MetricName": "tma_info_memory_oro_load_l2_miss_latency" }, { "BriefDescription": "Average Parallel L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_DATA_RD", "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_load_l2_mlp" + "MetricName": "tma_info_memory_oro_load_l2_mlp" }, { - "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", - "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_RETIRED.L1_MISS + MEM_LOAD_RETIRED.FB_HIT)", - "MetricGroup": "Mem;MemoryBound;MemoryLat", - "MetricName": "tma_info_load_miss_real_latency" + "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l1d_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l1d_cache_fill_bw_1t" + }, + { + "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l2_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l2_cache_fill_bw_1t" + }, + { + "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_access_bw", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_thread_l3_cache_access_bw_1t" + }, + { + "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l3_cache_fill_bw_1t" + }, + { + "BriefDescription": "STLB (2nd level TLB) code speculative misses per kilo instruction (misses of any page-size that complete the page walk)", + "MetricExpr": "1e3 * ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricGroup": "Fed;MemoryTLB", + "MetricName": "tma_info_memory_tlb_code_stlb_mpki" }, { "BriefDescription": "STLB (2nd level TLB) data load speculative misses per kilo instruction (misses of any page-size that complete the page walk)", "MetricExpr": "1e3 * DTLB_LOAD_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_load_stlb_mpki" + "MetricName": "tma_info_memory_tlb_load_stlb_mpki" + }, + { + "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", + "MetricConstraint": "NO_GROUP_EVENTS_NMI", + "MetricExpr": "(ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING + EPT.WALK_PENDING) / (2 * tma_info_core_core_clks)", + "MetricGroup": "Mem;MemoryTLB", + "MetricName": "tma_info_memory_tlb_page_walks_utilization", + "MetricThreshold": "tma_info_memory_tlb_page_walks_utilization > 0.5" + }, + { + "BriefDescription": "STLB (2nd level TLB) data store speculative misses per kilo instruction (misses of any page-size that complete the page walk)", + "MetricExpr": "1e3 * DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricGroup": "Mem;MemoryTLB", + "MetricName": "tma_info_memory_tlb_store_stlb_mpki" + }, + { + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", + "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", + "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", + "MetricName": "tma_info_pipeline_execute" + }, + { + "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@", + "MetricGroup": "Pipeline;Ret", + "MetricName": "tma_info_pipeline_retire" + }, + { + "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time", + "MetricGroup": "Power;Summary", + "MetricName": "tma_info_system_average_frequency" + }, + { + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricGroup": "HPC;Summary", + "MetricName": "tma_info_system_cpu_utilization" + }, + { + "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", + "MetricExpr": "64 * (UNC_ARB_TRK_REQUESTS.ALL + UNC_ARB_COH_TRK_REQUESTS.ALL) / 1e6 / duration_time / 1e3", + "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", + "MetricName": "tma_info_system_dram_bw_use", + "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_mem_bandwidth, tma_sq_full" + }, + { + "BriefDescription": "Giga Floating Point Operations Per Second", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / 1e9 / duration_time", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_system_gflops", + "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + }, + { + "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", + "MetricGroup": "Branches;OS", + "MetricName": "tma_info_system_ipfarbranch", + "MetricThreshold": "tma_info_system_ipfarbranch < 1e6" + }, + { + "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_cpi" + }, + { + "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_utilization", + "MetricThreshold": "tma_info_system_kernel_utilization > 0.05" }, { "BriefDescription": "Average number of parallel data read requests to external memory", "MetricExpr": "UNC_ARB_TRK_OCCUPANCY.DATA_READ / UNC_ARB_TRK_OCCUPANCY.DATA_READ@thresh\\=1@", "MetricGroup": "Mem;MemoryBW;SoC", - "MetricName": "tma_info_mem_parallel_reads", + "MetricName": "tma_info_system_mem_parallel_reads", "PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches" }, { "BriefDescription": "Average number of parallel requests to external memory", "MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_OCCUPANCY.CYCLES_WITH_ANY_REQUEST", "MetricGroup": "Mem;SoC", - "MetricName": "tma_info_mem_parallel_requests", + "MetricName": "tma_info_system_mem_parallel_requests", "PublicDescription": "Average number of parallel requests to external memory. Accounts for all requests" }, { "BriefDescription": "Average latency of data read request to external memory (in nanoseconds)", - "MetricExpr": "1e9 * (UNC_ARB_TRK_OCCUPANCY.DATA_READ / UNC_ARB_TRK_REQUESTS.DATA_READ) / (tma_info_socket_clks / duration_time)", + "MetricExpr": "1e9 * (UNC_ARB_TRK_OCCUPANCY.DATA_READ / UNC_ARB_TRK_REQUESTS.DATA_READ) / (tma_info_system_socket_clks / duration_time)", "MetricGroup": "Mem;MemoryLat;SoC", - "MetricName": "tma_info_mem_read_latency", + "MetricName": "tma_info_system_mem_read_latency", "PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)" }, { "BriefDescription": "Average latency of all requests to external memory (in Uncore cycles)", "MetricExpr": "UNC_ARB_TRK_OCCUPANCY.ALL / UNC_ARB_TRK_REQUESTS.ALL", "MetricGroup": "Mem;SoC", - "MetricName": "tma_info_mem_request_latency" + "MetricName": "tma_info_system_mem_request_latency" }, { - "BriefDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_bandwidth / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_sq_full / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full))) + tma_l1_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_fb_full / (tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk))", - "MetricGroup": "Mem;MemoryBW;Offcore;tma_issueBW", - "MetricName": "tma_info_memory_bandwidth", - "MetricThreshold": "tma_info_memory_bandwidth > 20", - "PublicDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks. Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", + "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / (CPU_CLK_UNHALTED.REF_XCLK_ANY / 2) if #SMT_on else 0)", + "MetricGroup": "SMT", + "MetricName": "tma_info_system_smt_2t_utilization" }, { - "BriefDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * tma_memory_bound * (tma_l1_bound / max(tma_memory_bound, tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_load / max(tma_l1_bound, tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk)) + tma_store_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_store / (tma_dtlb_store + tma_false_sharing + tma_split_stores + tma_store_latency)))", - "MetricGroup": "Mem;MemoryTLB;Offcore;tma_issueTLB", - "MetricName": "tma_info_memory_data_tlbs", - "MetricThreshold": "tma_info_memory_data_tlbs > 20", - "PublicDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs). Related metrics: tma_dtlb_load, tma_dtlb_store" + "BriefDescription": "Socket actual clocks when any core is active on that socket", + "MetricExpr": "UNC_CLOCK.SOCKET", + "MetricGroup": "SoC", + "MetricName": "tma_info_system_socket_clks" }, { - "BriefDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_latency / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_l3_hit_latency / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full)) + tma_l2_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound))", - "MetricGroup": "Mem;MemoryLat;Offcore;tma_issueLat", - "MetricName": "tma_info_memory_latency", - "MetricThreshold": "tma_info_memory_latency > 20", - "PublicDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches). Related metrics: tma_l3_hit_latency, tma_mem_latency" + "BriefDescription": "Average Frequency Utilization relative nominal frequency", + "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricGroup": "Power", + "MetricName": "tma_info_system_turbo_utilization" }, { - "BriefDescription": "Total pipeline cost of Branch Misprediction related bottlenecks", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * (tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", - "MetricGroup": "Bad;BadSpec;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_mispredictions", - "MetricThreshold": "tma_info_mispredictions > 20", - "PublicDescription": "Total pipeline cost of Branch Misprediction related bottlenecks. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost, tma_mispredicts_resteers" + "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "Pipeline", + "MetricName": "tma_info_thread_clks" }, { - "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", - "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", - "MetricGroup": "Mem;MemoryBW;MemoryBound", - "MetricName": "tma_info_mlp", - "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" + "BriefDescription": "Cycles Per Instruction (per Logical Processor)", + "MetricExpr": "1 / tma_info_thread_ipc", + "MetricGroup": "Mem;Pipeline", + "MetricName": "tma_info_thread_cpi" }, { - "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", - "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING + EPT.WALK_PENDING) / (2 * tma_info_core_clks)", - "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_page_walks_utilization", - "MetricThreshold": "tma_info_page_walks_utilization > 0.5" + "BriefDescription": "The ratio of Executed- by Issued-Uops", + "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", + "MetricGroup": "Cor;Pipeline", + "MetricName": "tma_info_thread_execute_per_issue", + "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." }, { - "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@", - "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_retire" + "BriefDescription": "Instructions Per Cycle (per Logical Processor)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks", + "MetricGroup": "Ret;Summary", + "MetricName": "tma_info_thread_ipc" }, { "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", - "MetricExpr": "4 * tma_info_core_clks", + "MetricExpr": "4 * tma_info_core_core_clks", "MetricGroup": "TmaL1;tma_L1_group", - "MetricName": "tma_info_slots" - }, - { - "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", - "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / (CPU_CLK_UNHALTED.REF_XCLK_ANY / 2) if #SMT_on else 0)", - "MetricGroup": "SMT", - "MetricName": "tma_info_smt_2t_utilization" - }, - { - "BriefDescription": "Socket actual clocks when any core is active on that socket", - "MetricExpr": "UNC_CLOCK.SOCKET", - "MetricGroup": "SoC", - "MetricName": "tma_info_socket_clks" - }, - { - "BriefDescription": "STLB (2nd level TLB) data store speculative misses per kilo instruction (misses of any page-size that complete the page walk)", - "MetricExpr": "1e3 * DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", - "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_store_stlb_mpki" - }, - { - "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", - "MetricGroup": "Power", - "MetricName": "tma_info_turbo_utilization" + "MetricName": "tma_info_thread_slots" }, { "BriefDescription": "Uops Per Instruction", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / INST_RETIRED.ANY", "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_uoppi", - "MetricThreshold": "tma_info_uoppi > 1.05" + "MetricName": "tma_info_thread_uoppi", + "MetricThreshold": "tma_info_thread_uoppi > 1.05" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW", - "MetricName": "tma_info_uptb", - "MetricThreshold": "tma_info_uptb < 6" + "MetricName": "tma_info_thread_uptb", + "MetricThreshold": "tma_info_thread_uptb < 6" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses", - "MetricExpr": "ICACHE_64B.IFTAG_STALL / tma_info_clks", + "MetricExpr": "ICACHE_64B.IFTAG_STALL / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -1015,7 +1022,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled without loads missing the L1 data cache", - "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_clks, 0)", + "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_thread_clks, 0)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_issueL1;tma_issueMC;tma_memory_bound_group", "MetricName": "tma_l1_bound", "MetricThreshold": "tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1025,7 +1032,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled due to L2 cache accesses by loads", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) / (MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@) * ((CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks)", + "MetricExpr": "MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) / (MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@) * ((CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l2_bound", "MetricThreshold": "tma_l2_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1034,7 +1041,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L2_MISS - CYCLE_ACTIVITY.STALLS_L3_MISS) / tma_info_clks", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L2_MISS - CYCLE_ACTIVITY.STALLS_L3_MISS) / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1043,20 +1050,20 @@ }, { "BriefDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited)", - "MetricExpr": "6.5 * tma_info_average_frequency * MEM_LOAD_RETIRED.L3_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "6.5 * tma_info_system_average_frequency * MEM_LOAD_RETIRED.L3_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "MemoryLat;TopdownL4;tma_L4_group;tma_issueLat;tma_l3_bound_group", "MetricName": "tma_l3_hit_latency", "MetricThreshold": "tma_l3_hit_latency > 0.1 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited). Avoiding private cache misses (i.e. L2 misses/L3 hits) will improve the latency; reduce contention with sibling physical cores and increase performance. Note the value of this node may overlap with its siblings. Sample with: MEM_LOAD_RETIRED.L3_HIT_PS. Related metrics: tma_info_memory_latency, tma_mem_latency", + "PublicDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited). Avoiding private cache misses (i.e. L2 misses/L3 hits) will improve the latency; reduce contention with sibling physical cores and increase performance. Note the value of this node may overlap with its siblings. Sample with: MEM_LOAD_RETIRED.L3_HIT_PS. Related metrics: tma_info_bottleneck_memory_latency, tma_mem_latency", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)", - "MetricExpr": "ILD_STALL.LCP / tma_info_clks", + "MetricExpr": "ILD_STALL.LCP / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_lcp", "MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb", + "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb", "ScaleUnit": "100%" }, { @@ -1065,12 +1072,13 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_light_operations", "MetricThreshold": "tma_light_operations > 0.6", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation). This correlates with total number of instructions used by the program. A uops-per-instruction (see UopPI metric) ratio of 1 or less should be expected for decently optimized software running on Intel Core/Xeon products. While this often indicates efficient X86 instructions were executed; high value does not necessarily mean better performance cannot be achieved. Sample with: INST_RETIRED.PREC_DIST", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Load operations", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 + UOPS_DISPATCHED_PORT.PORT_7 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 + UOPS_DISPATCHED_PORT.PORT_7 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_load_op_utilization", "MetricThreshold": "tma_load_op_utilization > 0.6", @@ -1088,7 +1096,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles where the Second-level TLB (STLB) was missed by load accesses, performing a hardware page walk", - "MetricExpr": "DTLB_LOAD_MISSES.WALK_ACTIVE / tma_info_clks", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_ACTIVE / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL5;tma_L5_group;tma_dtlb_load_group", "MetricName": "tma_load_stlb_miss", "MetricThreshold": "tma_load_stlb_miss > 0.05 & (tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1096,7 +1104,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU spent handling cache misses due to lock operations", - "MetricExpr": "(12 * max(0, MEM_INST_RETIRED.LOCK_LOADS - L2_RQSTS.ALL_RFO) + MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES * (9 * L2_RQSTS.RFO_HIT + min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO))) / tma_info_clks", + "MetricExpr": "(12 * max(0, MEM_INST_RETIRED.LOCK_LOADS - L2_RQSTS.ALL_RFO) + MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES * (9 * L2_RQSTS.RFO_HIT + min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO))) / tma_info_thread_clks", "MetricGroup": "Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_l1_bound_group", "MetricName": "tma_lock_latency", "MetricThreshold": "tma_lock_latency > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1110,25 +1118,26 @@ "MetricGroup": "BadSpec;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.1 & tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears. These slots are either wasted by uops fetched prior to the clear; or stalls the out-of-order portion of the machine needs to recover its state after the clear. For example; this can happen due to memory ordering Nukes (e.g. Memory Disambiguation) or Self-Modifying-Code (SMC) nukes. Sample with: MACHINE_CLEARS.COUNT. Related metrics: tma_clears_resteers, tma_contested_accesses, tma_data_sharing, tma_false_sharing, tma_l1_bound, tma_microcode_sequencer, tma_ms_switches, tma_remote_cache", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_clks", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW", "MetricName": "tma_mem_bandwidth", "MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_sq_full", + "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_sq_full", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat", "MetricName": "tma_mem_latency", "MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM). This metric does not aggregate requests from other Logical Processors/Physical Cores/sockets (see Uncore counters for that). Related metrics: tma_info_memory_latency, tma_l3_hit_latency", + "PublicDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM). This metric does not aggregate requests from other Logical Processors/Physical Cores/sockets (see Uncore counters for that). Related metrics: tma_info_bottleneck_memory_latency, tma_l3_hit_latency", "ScaleUnit": "100%" }, { @@ -1138,6 +1147,7 @@ "MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck. Memory Bound estimates fraction of slots where pipeline is likely stalled due to demand load or store instructions. This accounts mainly for (1) non-completed in-flight memory demand loads which coincides with execution units starvation; in addition to (2) cases where stores could impose backpressure on the pipeline when many of them get buffered at the same time (less common out of the two).", "ScaleUnit": "100%" }, @@ -1151,7 +1161,7 @@ }, { "BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_thread_slots", "MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS", "MetricName": "tma_microcode_sequencer", "MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1", @@ -1160,19 +1170,19 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage", - "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks", + "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks", "MetricGroup": "BadSpec;BrMispredicts;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueBM", "MetricName": "tma_mispredicts_resteers", "MetricThreshold": "tma_mispredicts_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost, tma_info_mispredictions", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost, tma_info_bottleneck_mispredictions", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline)", - "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_mite", - "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline). This pipeline is used for code that was not pre-cached in the DSB or LSD. For example; inefficiencies due to asymmetric decoders; use of long immediate or LCP can manifest as MITE fetch bandwidth bottleneck. Sample with: FRONTEND_RETIRED.ANY_DSB_MISS", "ScaleUnit": "100%" }, @@ -1187,7 +1197,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)", - "MetricExpr": "2 * IDQ.MS_SWITCHES / tma_info_clks", + "MetricExpr": "2 * IDQ.MS_SWITCHES / tma_info_thread_clks", "MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO", "MetricName": "tma_ms_switches", "MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -1223,7 +1233,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 0 ([SNB+] ALU; [HSW+] ALU and 2nd branch)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_core_clks", "MetricGroup": "Compute;TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_0", "MetricThreshold": "tma_port_0 > 0.6", @@ -1232,7 +1242,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 1 (ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_1", "MetricThreshold": "tma_port_1 > 0.6", @@ -1241,7 +1251,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 2 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_2", "MetricThreshold": "tma_port_2 > 0.6", @@ -1250,7 +1260,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 3 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_3", "MetricThreshold": "tma_port_3 > 0.6", @@ -1268,7 +1278,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 5 ([SNB+] Branches and ALU; [HSW+] ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_5", "MetricThreshold": "tma_port_5 > 0.6", @@ -1277,7 +1287,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 6 ([HSW+]Primary Branch and simple ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_6 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_6 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_6", "MetricThreshold": "tma_port_6 > 0.6", @@ -1286,7 +1296,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 7 ([HSW+]simple Store-address)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_7 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_7 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_store_op_utilization_group", "MetricName": "tma_port_7", "MetricThreshold": "tma_port_7 > 0.6", @@ -1295,7 +1305,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)", - "MetricExpr": "((EXE_ACTIVITY.EXE_BOUND_0_PORTS + (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL)) / tma_info_clks if ARITH.DIVIDER_ACTIVE < CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY else (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_clks)", + "MetricExpr": "((EXE_ACTIVITY.EXE_BOUND_0_PORTS + (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL)) / tma_info_thread_clks if ARITH.DIVIDER_ACTIVE < CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY else (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_thread_clks)", "MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_ports_utilization", "MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -1304,7 +1314,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed no uops on any execution port (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "(UOPS_EXECUTED.CORE_CYCLES_NONE / 2 if #SMT_on else CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_core_clks", + "MetricExpr": "(UOPS_EXECUTED.CORE_CYCLES_NONE / 2 if #SMT_on else CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_0", "MetricThreshold": "tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1313,7 +1323,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the CPU executed total of 1 uop per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((UOPS_EXECUTED.CORE_CYCLES_GE_1 - UOPS_EXECUTED.CORE_CYCLES_GE_2) / 2 if #SMT_on else EXE_ACTIVITY.1_PORTS_UTIL) / tma_info_core_clks", + "MetricExpr": "((UOPS_EXECUTED.CORE_CYCLES_GE_1 - UOPS_EXECUTED.CORE_CYCLES_GE_2) / 2 if #SMT_on else EXE_ACTIVITY.1_PORTS_UTIL) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issueL1;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_1", "MetricThreshold": "tma_ports_utilized_1 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1322,7 +1332,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 2 uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((UOPS_EXECUTED.CORE_CYCLES_GE_2 - UOPS_EXECUTED.CORE_CYCLES_GE_3) / 2 if #SMT_on else EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_core_clks", + "MetricExpr": "((UOPS_EXECUTED.CORE_CYCLES_GE_2 - UOPS_EXECUTED.CORE_CYCLES_GE_3) / 2 if #SMT_on else EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issue2P;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_2", "MetricThreshold": "tma_ports_utilized_2 > 0.15 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1331,7 +1341,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 3 or more uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise).", - "MetricExpr": "(UOPS_EXECUTED.CORE_CYCLES_GE_3 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_3) / tma_info_core_clks", + "MetricExpr": "(UOPS_EXECUTED.CORE_CYCLES_GE_3 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_3) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_3m", "MetricThreshold": "tma_ports_utilized_3m > 0.7 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1339,16 +1349,17 @@ }, { "BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired. Ideally; all pipeline slots would be attributed to the Retiring category. Retiring of 100% would indicate the maximum Pipeline_Width throughput was achieved. Maximizing Retiring typically increases the Instructions-per-cycle (see IPC metric). Note that a high Retiring value does not necessary mean there is no room for more performance. For example; Heavy-operations or Microcode Assists are categorized under Retiring. They often indicate suboptimal performance and can often be optimized or avoided. Sample with: UOPS_RETIRED.RETIRE_SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU issue-pipeline was stalled due to serializing operations", - "MetricExpr": "PARTIAL_RAT_STALLS.SCOREBOARD / tma_info_clks", + "MetricExpr": "PARTIAL_RAT_STALLS.SCOREBOARD / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL5;tma_L5_group;tma_issueSO;tma_ports_utilized_0_group", "MetricName": "tma_serializing_operation", "MetricThreshold": "tma_serializing_operation > 0.1 & (tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)))", @@ -1358,7 +1369,7 @@ { "BriefDescription": "This metric estimates fraction of cycles handling memory load split accesses - load that cross 64-byte cache line boundary", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "tma_info_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_split_loads", "MetricThreshold": "tma_split_loads > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1367,7 +1378,7 @@ }, { "BriefDescription": "This metric represents rate of split store accesses", - "MetricExpr": "MEM_INST_RETIRED.SPLIT_STORES / tma_info_core_clks", + "MetricExpr": "MEM_INST_RETIRED.SPLIT_STORES / tma_info_core_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_issueSpSt;tma_store_bound_group", "MetricName": "tma_split_stores", "MetricThreshold": "tma_split_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1376,16 +1387,16 @@ }, { "BriefDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors)", - "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_clks", + "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_core_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueBW;tma_l3_bound_group", "MetricName": "tma_sq_full", "MetricThreshold": "tma_sq_full > 0.3 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_mem_bandwidth", + "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_mem_bandwidth", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write", - "MetricExpr": "EXE_ACTIVITY.BOUND_ON_STORES / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.BOUND_ON_STORES / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1394,7 +1405,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of cycles when the memory subsystem had loads blocked since they could not forward data from earlier (in program order) overlapping stores", - "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_clks", + "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_store_fwd_blk", "MetricThreshold": "tma_store_fwd_blk > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1404,7 +1415,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU spent handling L1D store misses", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(L2_RQSTS.RFO_HIT * 9 * (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) + (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_clks", + "MetricExpr": "(L2_RQSTS.RFO_HIT * 9 * (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) + (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_thread_clks", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_issueSL;tma_store_bound_group", "MetricName": "tma_store_latency", "MetricThreshold": "tma_store_latency > 0.1 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1413,7 +1424,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Store operations", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_core_clks", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_store_op_utilization", "MetricThreshold": "tma_store_op_utilization > 0.6", @@ -1429,7 +1440,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles where the STLB was missed by store accesses, performing a hardware page walk", - "MetricExpr": "DTLB_STORE_MISSES.WALK_ACTIVE / tma_info_core_clks", + "MetricExpr": "DTLB_STORE_MISSES.WALK_ACTIVE / tma_info_core_core_clks", "MetricGroup": "MemoryTLB;TopdownL5;tma_L5_group;tma_dtlb_store_group", "MetricName": "tma_store_stlb_miss", "MetricThreshold": "tma_store_stlb_miss > 0.05 & (tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1437,7 +1448,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to new branch address clears", - "MetricExpr": "9 * BACLEARS.ANY / tma_info_clks", + "MetricExpr": "9 * BACLEARS.ANY / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;TopdownL4;tma_L4_group;tma_branch_resteers_group", "MetricName": "tma_unknown_branches", "MetricThreshold": "tma_unknown_branches > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", diff --git a/tools/perf/pmu-events/arch/x86/skylakex/floating-point.json b/tools/perf/pmu-events/arch/x86/skylakex/floating-point.json index 64dd36387209..384b3c551a1f 100644 --- a/tools/perf/pmu-events/arch/x86/skylakex/floating-point.json +++ b/tools/perf/pmu-events/arch/x86/skylakex/floating-point.json @@ -32,6 +32,14 @@ "UMask": "0x20" }, { + "BriefDescription": "Number of SSE/AVX computational 128-bit packed single and 256-bit packed double precision FP instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, 1 for each element. Applies to SSE* and AVX* packed single precision and packed double precision FP instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB count twice as they perform 2 calculations per element.", + "EventCode": "0xC7", + "EventName": "FP_ARITH_INST_RETIRED.4_FLOPS", + "PublicDescription": "Number of SSE/AVX computational 128-bit packed single precision and 256-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 2 or/and 4 computation operations, one for each element. Applies to SSE* and AVX* packed single precision floating-point and packed double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX RCP14 RSQRT14 SQRT DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", + "SampleAfterValue": "1000003", + "UMask": "0x18" + }, + { "BriefDescription": "Counts number of SSE/AVX computational 512-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed double precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT14 RCP14 FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element.", "EventCode": "0xC7", "EventName": "FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE", @@ -48,6 +56,22 @@ "UMask": "0x80" }, { + "BriefDescription": "Number of SSE/AVX computational 256-bit packed single precision and 512-bit packed double precision FP instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, 1 for each element. Applies to SSE* and AVX* packed single precision and double precision FP instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RSQRT14 RCP RCP14 DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB count twice as they perform 2 calculations per element.", + "EventCode": "0xC7", + "EventName": "FP_ARITH_INST_RETIRED.8_FLOPS", + "PublicDescription": "Number of SSE/AVX computational 256-bit packed single precision and 512-bit packed double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 8 computation operations, one for each element. Applies to SSE* and AVX* packed single precision and double precision floating-point instructions: ADD SUB HADD HSUB SUBADD MUL DIV MIN MAX SQRT RSQRT RSQRT14 RCP RCP14 DPP FM(N)ADD/SUB. DPP and FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", + "SampleAfterValue": "1000003", + "UMask": "0x18" + }, + { + "BriefDescription": "Counts once for most SIMD scalar computational floating-point instructions retired. Counts twice for DPP and FM(N)ADD/SUB instructions retired.", + "EventCode": "0xC7", + "EventName": "FP_ARITH_INST_RETIRED.SCALAR", + "PublicDescription": "Counts once for most SIMD scalar computational single precision and double precision floating-point instructions retired; some instructions will count twice as noted below. Each count represents 1 computational operation. Applies to SIMD scalar single precision floating-point instructions: ADD SUB MUL DIV MIN MAX SQRT RSQRT RCP FM(N)ADD/SUB. FM(N)ADD/SUB instructions count twice as they perform 2 calculations per element. The DAZ and FTZ flags in the MXCSR register need to be set when using these events.", + "SampleAfterValue": "2000003", + "UMask": "0x3" + }, + { "BriefDescription": "Counts once for most SIMD scalar computational double precision floating-point instructions retired. Counts twice for DPP and FM(N)ADD/SUB instructions retired.", "EventCode": "0xC7", "EventName": "FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", @@ -64,6 +88,13 @@ "UMask": "0x2" }, { + "BriefDescription": "Number of any Vector retired FP arithmetic instructions", + "EventCode": "0xC7", + "EventName": "FP_ARITH_INST_RETIRED.VECTOR", + "SampleAfterValue": "2000003", + "UMask": "0xfc" + }, + { "BriefDescription": "Cycles with any input/output SSE or FP assist", "CounterMask": "1", "EventCode": "0xCA", diff --git a/tools/perf/pmu-events/arch/x86/skylakex/metricgroups.json b/tools/perf/pmu-events/arch/x86/skylakex/metricgroups.json new file mode 100644 index 000000000000..bc6a9a4d27a9 --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/skylakex/metricgroups.json @@ -0,0 +1,114 @@ +{ + "Backend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Bad": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BadSpec": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BigFoot": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BrMispredicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Branches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CacheMisses": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CodeGen": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Compute": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Cor": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSBmiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DataSharing": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Fed": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Flops": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpScalar": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Frontend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "HPC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IcMiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "InsType": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IoBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "L2Evicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "LSD": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MachineClears": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBound": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryTLB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_BW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_Lat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MicroSeq": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "OS": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Offcore": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PGO": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Pipeline": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PortsUtil": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Prefetches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Ret": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Retire": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SMT": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Server": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Snoop": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SoC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL1": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL2": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL3mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "TopdownL5": "Metrics for top-down breakdown at level 5", + "TopdownL6": "Metrics for top-down breakdown at level 6", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_L5_group": "Metrics for top-down breakdown at level 5", + "tma_L6_group": "Metrics for top-down breakdown at level 6", + "tma_alu_op_utilization_group": "Metrics contributing to tma_alu_op_utilization category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_branch_resteers_group": "Metrics contributing to tma_branch_resteers category", + "tma_core_bound_group": "Metrics contributing to tma_core_bound category", + "tma_dram_bound_group": "Metrics contributing to tma_dram_bound category", + "tma_dtlb_load_group": "Metrics contributing to tma_dtlb_load category", + "tma_dtlb_store_group": "Metrics contributing to tma_dtlb_store category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_fp_arith_group": "Metrics contributing to tma_fp_arith category", + "tma_fp_vector_group": "Metrics contributing to tma_fp_vector category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_heavy_operations_group": "Metrics contributing to tma_heavy_operations category", + "tma_issue2P": "Metrics related by the issue $issue2P", + "tma_issueBC": "Metrics related by the issue $issueBC", + "tma_issueBM": "Metrics related by the issue $issueBM", + "tma_issueBW": "Metrics related by the issue $issueBW", + "tma_issueD0": "Metrics related by the issue $issueD0", + "tma_issueFB": "Metrics related by the issue $issueFB", + "tma_issueFL": "Metrics related by the issue $issueFL", + "tma_issueL1": "Metrics related by the issue $issueL1", + "tma_issueLat": "Metrics related by the issue $issueLat", + "tma_issueMC": "Metrics related by the issue $issueMC", + "tma_issueMS": "Metrics related by the issue $issueMS", + "tma_issueMV": "Metrics related by the issue $issueMV", + "tma_issueRFO": "Metrics related by the issue $issueRFO", + "tma_issueSL": "Metrics related by the issue $issueSL", + "tma_issueSO": "Metrics related by the issue $issueSO", + "tma_issueSmSt": "Metrics related by the issue $issueSmSt", + "tma_issueSpSt": "Metrics related by the issue $issueSpSt", + "tma_issueSyncxn": "Metrics related by the issue $issueSyncxn", + "tma_issueTLB": "Metrics related by the issue $issueTLB", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_l3_bound_group": "Metrics contributing to tma_l3_bound category", + "tma_light_operations_group": "Metrics contributing to tma_light_operations category", + "tma_load_op_utilization_group": "Metrics contributing to tma_load_op_utilization category", + "tma_mem_latency_group": "Metrics contributing to tma_mem_latency category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_microcode_sequencer_group": "Metrics contributing to tma_microcode_sequencer category", + "tma_mite_group": "Metrics contributing to tma_mite category", + "tma_ports_utilization_group": "Metrics contributing to tma_ports_utilization category", + "tma_ports_utilized_0_group": "Metrics contributing to tma_ports_utilized_0 category", + "tma_ports_utilized_3m_group": "Metrics contributing to tma_ports_utilized_3m category", + "tma_retiring_group": "Metrics contributing to tma_retiring category", + "tma_serializing_operation_group": "Metrics contributing to tma_serializing_operation category", + "tma_store_bound_group": "Metrics contributing to tma_store_bound category", + "tma_store_op_utilization_group": "Metrics contributing to tma_store_op_utilization category" +} diff --git a/tools/perf/pmu-events/arch/x86/skylakex/pipeline.json b/tools/perf/pmu-events/arch/x86/skylakex/pipeline.json index 0f06e314fe36..31a1663d57f8 100644 --- a/tools/perf/pmu-events/arch/x86/skylakex/pipeline.json +++ b/tools/perf/pmu-events/arch/x86/skylakex/pipeline.json @@ -26,12 +26,21 @@ "UMask": "0x4" }, { - "BriefDescription": "Conditional branch instructions retired.", + "BriefDescription": "Conditional branch instructions retired. [This event is alias to BR_INST_RETIRED.CONDITIONAL]", + "Errata": "SKL091", + "EventCode": "0xC4", + "EventName": "BR_INST_RETIRED.COND", + "PublicDescription": "This event counts conditional branch instructions retired. [This event is alias to BR_INST_RETIRED.CONDITIONAL]", + "SampleAfterValue": "400009", + "UMask": "0x1" + }, + { + "BriefDescription": "Conditional branch instructions retired. [This event is alias to BR_INST_RETIRED.COND]", "Errata": "SKL091", "EventCode": "0xC4", "EventName": "BR_INST_RETIRED.CONDITIONAL", "PEBS": "1", - "PublicDescription": "This event counts conditional branch instructions retired.", + "PublicDescription": "This event counts conditional branch instructions retired. [This event is alias to BR_INST_RETIRED.COND]", "SampleAfterValue": "400009", "UMask": "0x1" }, @@ -414,6 +423,16 @@ "UMask": "0x1" }, { + "BriefDescription": "Clears speculative count", + "CounterMask": "1", + "EdgeDetect": "1", + "EventCode": "0x0D", + "EventName": "INT_MISC.CLEARS_COUNT", + "PublicDescription": "Counts the number of speculative clears due to any type of branch misprediction or machine clears", + "SampleAfterValue": "2000003", + "UMask": "0x1" + }, + { "BriefDescription": "Cycles the issue-stage is waiting for front-end to fetch from resteered path following branch misprediction or machine clear events.", "EventCode": "0x0D", "EventName": "INT_MISC.CLEAR_RESTEER_CYCLES", diff --git a/tools/perf/pmu-events/arch/x86/skylakex/skx-metrics.json b/tools/perf/pmu-events/arch/x86/skylakex/skx-metrics.json index fa2f7f126a30..507d39efacc8 100644 --- a/tools/perf/pmu-events/arch/x86/skylakex/skx-metrics.json +++ b/tools/perf/pmu-events/arch/x86/skylakex/skx-metrics.json @@ -50,11 +50,220 @@ }, { "BriefDescription": "Uncore frequency per die [GHZ]", - "MetricExpr": "tma_info_socket_clks / #num_dies / duration_time / 1e9", + "MetricExpr": "tma_info_system_socket_clks / #num_dies / duration_time / 1e9", "MetricGroup": "SoC", "MetricName": "UNCORE_FREQ" }, { + "BriefDescription": "Cycles per instruction retired; indicating how much time each executed instruction took; in units of cycles.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD / INST_RETIRED.ANY", + "MetricName": "cpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "CPU operating frequency (in GHz)", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD / CPU_CLK_UNHALTED.REF_TSC * #SYSTEM_TSC_FREQ / 1e9", + "MetricName": "cpu_operating_frequency", + "ScaleUnit": "1GHz" + }, + { + "BriefDescription": "Percentage of time spent in the active CPU power state C0", + "MetricExpr": "tma_info_system_cpu_utilization", + "MetricName": "cpu_utilization", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for 2 megabyte page sizes) caused by demand data loads to the total number of completed instructions", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_COMPLETED_2M_4M / INST_RETIRED.ANY", + "MetricName": "dtlb_2mb_large_page_load_mpi", + "PublicDescription": "Ratio of number of completed page walks (for 2 megabyte page sizes) caused by demand data loads to the total number of completed instructions. This implies it missed in the Data Translation Lookaside Buffer (DTLB) and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data loads to the total number of completed instructions", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "dtlb_load_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data loads to the total number of completed instructions. This implies it missed in the DTLB and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data stores to the total number of completed instructions", + "MetricExpr": "DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "dtlb_store_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by demand data stores to the total number of completed instructions. This implies it missed in the DTLB and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Bandwidth of IO reads that are initiated by end device controllers that are requesting memory from the CPU.", + "MetricExpr": "(UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART0 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART1 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART2 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART3) * 4 / 1e6 / duration_time", + "MetricName": "io_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Bandwidth of IO writes that are initiated by end device controllers that are writing memory to the CPU.", + "MetricExpr": "(UNC_IIO_PAYLOAD_BYTES_IN.MEM_WRITE.PART0 + UNC_IIO_PAYLOAD_BYTES_IN.MEM_WRITE.PART1 + UNC_IIO_PAYLOAD_BYTES_IN.MEM_WRITE.PART2 + UNC_IIO_PAYLOAD_BYTES_IN.MEM_WRITE.PART3) * 4 / 1e6 / duration_time", + "MetricName": "io_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for 2 megabyte and 4 megabyte page sizes) caused by a code fetch to the total number of completed instructions", + "MetricExpr": "ITLB_MISSES.WALK_COMPLETED_2M_4M / INST_RETIRED.ANY", + "MetricName": "itlb_large_page_mpi", + "PublicDescription": "Ratio of number of completed page walks (for 2 megabyte and 4 megabyte page sizes) caused by a code fetch to the total number of completed instructions. This implies it missed in the Instruction Translation Lookaside Buffer (ITLB) and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed page walks (for all page sizes) caused by a code fetch to the total number of completed instructions", + "MetricExpr": "ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricName": "itlb_mpi", + "PublicDescription": "Ratio of number of completed page walks (for all page sizes) caused by a code fetch to the total number of completed instructions. This implies it missed in the ITLB (Instruction TLB) and further levels of TLB.", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read requests missing in L1 instruction cache (includes prefetches) to the total number of completed instructions", + "MetricExpr": "L2_RQSTS.ALL_CODE_RD / INST_RETIRED.ANY", + "MetricName": "l1_i_code_read_misses_with_prefetches_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of demand load requests hitting in L1 data cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_RETIRED.L1_HIT / INST_RETIRED.ANY", + "MetricName": "l1d_demand_data_read_hits_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of requests missing L1 data cache (includes data+rfo w/ prefetches) to the total number of completed instructions", + "MetricExpr": "L1D.REPLACEMENT / INST_RETIRED.ANY", + "MetricName": "l1d_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read request missing L2 cache to the total number of completed instructions", + "MetricExpr": "L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", + "MetricName": "l2_demand_code_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed demand load requests hitting in L2 cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_RETIRED.L2_HIT / INST_RETIRED.ANY", + "MetricName": "l2_demand_data_read_hits_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of completed data read request missing L2 cache to the total number of completed instructions", + "MetricExpr": "MEM_LOAD_RETIRED.L2_MISS / INST_RETIRED.ANY", + "MetricName": "l2_demand_data_read_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of requests missing L2 cache (includes code+data+rfo w/ prefetches) to the total number of completed instructions", + "MetricExpr": "L2_LINES_IN.ALL / INST_RETIRED.ANY", + "MetricName": "l2_mpi", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Ratio of number of code read requests missing last level core cache (includes demand w/ prefetches) to the total number of completed instructions", + "MetricExpr": "cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x12CC0233@ / INST_RETIRED.ANY", + "MetricName": "llc_code_read_mpi_demand_plus_prefetch", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand and prefetch data read miss (read memory access) in nano seconds", + "MetricExpr": "1e9 * (cha@UNC_CHA_TOR_OCCUPANCY.IA_MISS\\,config1\\=0x40433@ / cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40433@) / (UNC_CHA_CLOCKTICKS / (#num_cores / #num_packages * #num_packages)) * duration_time", + "MetricName": "llc_data_read_demand_plus_prefetch_miss_latency", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand and prefetch data read miss (read memory access) addressed to local memory in nano seconds", + "MetricExpr": "1e9 * (cha@UNC_CHA_TOR_OCCUPANCY.IA_MISS\\,config1\\=0x40432@ / cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40432@) / (UNC_CHA_CLOCKTICKS / (#num_cores / #num_packages * #num_packages)) * duration_time", + "MetricName": "llc_data_read_demand_plus_prefetch_miss_latency_for_local_requests", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Average latency of a last level cache (LLC) demand and prefetch data read miss (read memory access) addressed to remote memory in nano seconds", + "MetricExpr": "1e9 * (cha@UNC_CHA_TOR_OCCUPANCY.IA_MISS\\,config1\\=0x40431@ / cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40431@) / (UNC_CHA_CLOCKTICKS / (#num_cores / #num_packages * #num_packages)) * duration_time", + "MetricName": "llc_data_read_demand_plus_prefetch_miss_latency_for_remote_requests", + "ScaleUnit": "1ns" + }, + { + "BriefDescription": "Ratio of number of data read requests missing last level core cache (includes demand w/ prefetches) to the total number of completed instructions", + "MetricExpr": "cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x12D40433@ / INST_RETIRED.ANY", + "MetricName": "llc_data_read_mpi_demand_plus_prefetch", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "Bandwidth (MB/sec) of read requests that miss the last level cache (LLC) and go to local memory.", + "MetricExpr": "UNC_CHA_REQUESTS.READS_LOCAL * 64 / 1e6 / duration_time", + "MetricName": "llc_miss_local_memory_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Bandwidth (MB/sec) of write requests that miss the last level cache (LLC) and go to local memory.", + "MetricExpr": "UNC_CHA_REQUESTS.WRITES_LOCAL * 64 / 1e6 / duration_time", + "MetricName": "llc_miss_local_memory_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Bandwidth (MB/sec) of read requests that miss the last level cache (LLC) and go to remote memory.", + "MetricExpr": "UNC_CHA_REQUESTS.READS_REMOTE * 64 / 1e6 / duration_time", + "MetricName": "llc_miss_remote_memory_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "The ratio of number of completed memory load instructions to the total number completed instructions", + "MetricExpr": "MEM_INST_RETIRED.ALL_LOADS / INST_RETIRED.ANY", + "MetricName": "loads_per_instr", + "ScaleUnit": "1per_instr" + }, + { + "BriefDescription": "DDR memory read bandwidth (MB/sec)", + "MetricExpr": "UNC_M_CAS_COUNT.RD * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_read", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "DDR memory bandwidth (MB/sec)", + "MetricExpr": "(UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_total", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "DDR memory write bandwidth (MB/sec)", + "MetricExpr": "UNC_M_CAS_COUNT.WR * 64 / 1e6 / duration_time", + "MetricName": "memory_bandwidth_write", + "ScaleUnit": "1MB/s" + }, + { + "BriefDescription": "Memory read that miss the last level cache (LLC) addressed to local DRAM as a percentage of total memory read accesses, does not include LLC prefetches.", + "MetricExpr": "cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40432@ / (cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40432@ + cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40431@)", + "MetricName": "numa_reads_addressed_to_local_dram", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Memory reads that miss the last level cache (LLC) addressed to remote DRAM as a percentage of total memory read accesses, does not include LLC prefetches.", + "MetricExpr": "cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40431@ / (cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40432@ + cha@UNC_CHA_TOR_INSERTS.IA_MISS\\,config1\\=0x40431@)", + "MetricName": "numa_reads_addressed_to_remote_dram", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from decoded instruction cache (decoded stream buffer or DSB) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.DSB_UOPS / UOPS_ISSUED.ANY", + "MetricName": "percent_uops_delivered_from_decoded_icache", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from legacy decode pipeline (Micro-instruction Translation Engine or MITE) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.MITE_UOPS / UOPS_ISSUED.ANY", + "MetricName": "percent_uops_delivered_from_legacy_decode_pipeline", + "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uops delivered from microcode sequencer (MS) as a percent of total uops delivered to Instruction Decode Queue", + "MetricExpr": "IDQ.MS_UOPS / UOPS_ISSUED.ANY", + "MetricName": "percent_uops_delivered_from_microcode_sequencer", + "ScaleUnit": "100%" + }, + { "BriefDescription": "Percentage of cycles spent in System Management Interrupts.", "MetricExpr": "((msr@aperf@ - cycles) / msr@aperf@ if msr@smi@ > 0 else 0)", "MetricGroup": "smi", @@ -70,8 +279,14 @@ "ScaleUnit": "1SMI#" }, { + "BriefDescription": "The ratio of number of completed memory store instructions to the total number completed instructions", + "MetricExpr": "MEM_INST_RETIRED.ALL_STORES / INST_RETIRED.ANY", + "MetricName": "stores_per_instr", + "ScaleUnit": "1per_instr" + }, + { "BriefDescription": "This metric estimates how often memory load accesses were aliased by preceding stores (in program order) with a 4K address offset", - "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_clks", + "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_4k_aliasing", "MetricThreshold": "tma_4k_aliasing > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -80,7 +295,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution ports for ALU operations.", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5 + UOPS_DISPATCHED_PORT.PORT_6) / tma_info_slots", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_0 + UOPS_DISPATCHED_PORT.PORT_1 + UOPS_DISPATCHED_PORT.PORT_5 + UOPS_DISPATCHED_PORT.PORT_6) / tma_info_thread_slots", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_alu_op_utilization", "MetricThreshold": "tma_alu_op_utilization > 0.6", @@ -88,7 +303,7 @@ }, { "BriefDescription": "This metric estimates fraction of slots the CPU retired uops delivered by the Microcode_Sequencer as a result of Assists", - "MetricExpr": "100 * (FP_ASSIST.ANY + OTHER_ASSISTS.ANY) / tma_info_slots", + "MetricExpr": "100 * (FP_ASSIST.ANY + OTHER_ASSISTS.ANY) / tma_info_thread_slots", "MetricGroup": "TopdownL4;tma_L4_group;tma_microcode_sequencer_group", "MetricName": "tma_assists", "MetricThreshold": "tma_assists > 0.1 & (tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1)", @@ -97,19 +312,21 @@ }, { "BriefDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend", - "MetricExpr": "1 - tma_frontend_bound - (UOPS_ISSUED.ANY + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_slots", + "MetricExpr": "1 - tma_frontend_bound - (UOPS_ISSUED.ANY + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend. Backend is the portion of the processor core where the out-of-order scheduler dispatches ready uops into their respective execution units; and once completed these uops get retired according to program order. For example; stalls due to data-cache misses or stalls due to the divider unit being overloaded are both categorized under Backend Bound. Backend Bound is further divided into two main categories: Memory Bound and Core Bound.", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations", - "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_slots", + "MetricExpr": "(UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (INT_MISC.RECOVERY_CYCLES_ANY / 2 if #SMT_on else INT_MISC.RECOVERY_CYCLES)) / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots wasted due to incorrect speculations. This include slots used to issue uops that do not eventually get retired and slots for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work due to miss-predicted branches are categorized under Bad Speculation category. Incorrect data speculation followed by Memory Ordering Nukes is another example.", "ScaleUnit": "100%" }, @@ -120,12 +337,13 @@ "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15", - "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_branch_misprediction_cost, tma_info_mispredictions, tma_mispredicts_resteers", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_info_bottleneck_mispredictions, tma_mispredicts_resteers", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers", - "MetricExpr": "INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks + tma_unknown_branches", + "MetricExpr": "INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks + tma_unknown_branches", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteers", "MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -143,7 +361,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Machine Clears", - "MetricExpr": "(1 - BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT)) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks", + "MetricExpr": "(1 - BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT)) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks", "MetricGroup": "BadSpec;MachineClears;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueMC", "MetricName": "tma_clears_resteers", "MetricThreshold": "tma_clears_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", @@ -153,7 +371,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to contested accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(44 * tma_info_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_HITM * (OFFCORE_RESPONSE.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE / (OFFCORE_RESPONSE.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE + OFFCORE_RESPONSE.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) + 44 * tma_info_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "(44 * tma_info_system_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_HITM * (OFFCORE_RESPONSE.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE / (OFFCORE_RESPONSE.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE + OFFCORE_RESPONSE.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) + 44 * tma_info_system_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_contested_accesses", "MetricThreshold": "tma_contested_accesses > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -167,13 +385,14 @@ "MetricGroup": "Backend;Compute;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck. Shortage in hardware compute resources; or dependencies in software's instructions are both categorized under Core Bound. Hence it may indicate the machine ran out of an out-of-order resource; certain execution units are overloaded or dependencies in program's data- or instruction-flow are limiting the performance (e.g. FP-chained long-latency arithmetic operations).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to data-sharing accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "44 * tma_info_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_L3_HIT_RETIRED.XSNP_HITM * (1 - OFFCORE_RESPONSE.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE / (OFFCORE_RESPONSE.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE + OFFCORE_RESPONSE.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "44 * tma_info_system_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_HIT + MEM_LOAD_L3_HIT_RETIRED.XSNP_HITM * (1 - OFFCORE_RESPONSE.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE / (OFFCORE_RESPONSE.DEMAND_DATA_RD.L3_HIT.HITM_OTHER_CORE + OFFCORE_RESPONSE.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_data_sharing", "MetricThreshold": "tma_data_sharing > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -182,16 +401,16 @@ }, { "BriefDescription": "This metric represents fraction of cycles where decoder-0 was the only active decoder", - "MetricExpr": "(cpu@INST_DECODED.DECODERS\\,cmask\\=1@ - cpu@INST_DECODED.DECODERS\\,cmask\\=2@) / tma_info_core_clks / 2", + "MetricExpr": "(cpu@INST_DECODED.DECODERS\\,cmask\\=1@ - cpu@INST_DECODED.DECODERS\\,cmask\\=2@) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL4;tma_L4_group;tma_issueD0;tma_mite_group", "MetricName": "tma_decoder0_alone", - "MetricThreshold": "tma_decoder0_alone > 0.1 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35))", + "MetricThreshold": "tma_decoder0_alone > 0.1 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35))", "PublicDescription": "This metric represents fraction of cycles where decoder-0 was the only active decoder. Related metrics: tma_few_uops_instructions", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles where the Divider unit was active", - "MetricExpr": "ARITH.DIVIDER_ACTIVE / tma_info_clks", + "MetricExpr": "ARITH.DIVIDER_ACTIVE / tma_info_thread_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_divider", "MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -201,7 +420,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks - tma_l2_bound", + "MetricExpr": "CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_thread_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks - tma_l2_bound", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -210,45 +429,45 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline", - "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_DSB_CYCLES_ANY_UOPS - IDQ.ALL_DSB_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSB;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_dsb", - "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline. For example; inefficient utilization of the DSB cache structure or bank conflict when reading from it; are categorized here.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_dsb_switches", "MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb, tma_lcp", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "min(9 * cpu@DTLB_LOAD_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_LOAD_MISSES.WALK_ACTIVE, max(CYCLE_ACTIVITY.CYCLES_MEM_ANY - CYCLE_ACTIVITY.CYCLES_L1D_MISS, 0)) / tma_info_clks", + "MetricExpr": "min(9 * cpu@DTLB_LOAD_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_LOAD_MISSES.WALK_ACTIVE, max(CYCLE_ACTIVITY.CYCLES_MEM_ANY - CYCLE_ACTIVITY.CYCLES_L1D_MISS, 0)) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group", "MetricName": "tma_dtlb_load", "MetricThreshold": "tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses. TLBs (Translation Look-aside Buffers) are processor caches for recently used entries out of the Page Tables that are used to map virtual- to physical-addresses by the operating system. This metric approximates the potential delay of demand loads missing the first-level data TLB (assuming worst case scenario with back to back misses to different pages). This includes hitting in the second-level TLB (STLB) as well as performing a hardware page walk on an STLB miss. Sample with: MEM_INST_RETIRED.STLB_MISS_LOADS_PS. Related metrics: tma_dtlb_store, tma_info_memory_data_tlbs", + "PublicDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses. TLBs (Translation Look-aside Buffers) are processor caches for recently used entries out of the Page Tables that are used to map virtual- to physical-addresses by the operating system. This metric approximates the potential delay of demand loads missing the first-level data TLB (assuming worst case scenario with back to back misses to different pages). This includes hitting in the second-level TLB (STLB) as well as performing a hardware page walk on an STLB miss. Sample with: MEM_INST_RETIRED.STLB_MISS_LOADS_PS. Related metrics: tma_dtlb_store, tma_info_bottleneck_memory_data_tlbs", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses", - "MetricExpr": "(9 * cpu@DTLB_STORE_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_STORE_MISSES.WALK_ACTIVE) / tma_info_core_clks", + "MetricExpr": "(9 * cpu@DTLB_STORE_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_STORE_MISSES.WALK_ACTIVE) / tma_info_core_core_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_store_bound_group", "MetricName": "tma_dtlb_store", "MetricThreshold": "tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses. As with ordinary data caching; focus on improving data locality and reducing working-set size to reduce DTLB overhead. Additionally; consider using profile-guided optimization (PGO) to collocate frequently-used data on the same page. Try using larger page sizes for large amounts of frequently-used data. Sample with: MEM_INST_RETIRED.STLB_MISS_STORES_PS. Related metrics: tma_dtlb_load, tma_info_memory_data_tlbs", + "PublicDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses. As with ordinary data caching; focus on improving data locality and reducing working-set size to reduce DTLB overhead. Additionally; consider using profile-guided optimization (PGO) to collocate frequently-used data on the same page. Try using larger page sizes for large amounts of frequently-used data. Sample with: MEM_INST_RETIRED.STLB_MISS_STORES_PS. Related metrics: tma_dtlb_load, tma_info_bottleneck_memory_data_tlbs", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates how often CPU was handling synchronizations due to False Sharing", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(110 * tma_info_average_frequency * (OFFCORE_RESPONSE.DEMAND_RFO.L3_MISS.REMOTE_HITM + OFFCORE_RESPONSE.PF_L2_RFO.L3_MISS.REMOTE_HITM) + 47.5 * tma_info_average_frequency * (OFFCORE_RESPONSE.DEMAND_RFO.L3_HIT.HITM_OTHER_CORE + OFFCORE_RESPONSE.PF_L2_RFO.L3_HIT.HITM_OTHER_CORE)) / tma_info_clks", + "MetricExpr": "(110 * tma_info_system_average_frequency * (OFFCORE_RESPONSE.DEMAND_RFO.L3_MISS.REMOTE_HITM + OFFCORE_RESPONSE.PF_L2_RFO.L3_MISS.REMOTE_HITM) + 47.5 * tma_info_system_average_frequency * (OFFCORE_RESPONSE.DEMAND_RFO.L3_HIT.HITM_OTHER_CORE + OFFCORE_RESPONSE.PF_L2_RFO.L3_HIT.HITM_OTHER_CORE)) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_store_bound_group", "MetricName": "tma_false_sharing", "MetricThreshold": "tma_false_sharing > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -258,11 +477,11 @@ { "BriefDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "tma_info_load_miss_real_latency * cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@ / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@ / tma_info_thread_clks", "MetricGroup": "MemoryBW;TopdownL4;tma_L4_group;tma_issueBW;tma_issueSL;tma_issueSmSt;tma_l1_bound_group", "MetricName": "tma_fb_full", "MetricThreshold": "tma_fb_full > 0.3", - "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", + "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", "ScaleUnit": "100%" }, { @@ -270,16 +489,18 @@ "MetricExpr": "tma_frontend_bound - tma_fetch_latency", "MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", "MetricName": "tma_fetch_bandwidth", - "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb, tma_lcp", + "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues", - "MetricExpr": "4 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE / tma_info_slots", + "MetricExpr": "4 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE / tma_info_thread_slots", "MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: FRONTEND_RETIRED.LATENCY_GE_16_PS;FRONTEND_RETIRED.LATENCY_GE_8_PS", "ScaleUnit": "100%" }, @@ -350,10 +571,11 @@ }, { "BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend", - "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_slots", + "MetricExpr": "IDQ_UOPS_NOT_DELIVERED.CORE / tma_info_thread_slots", "MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound. Sample with: FRONTEND_RETIRED.LATENCY_GE_4_PS", "ScaleUnit": "100%" }, @@ -368,16 +590,17 @@ }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences", - "MetricExpr": "(UOPS_RETIRED.RETIRE_SLOTS + UOPS_RETIRED.MACRO_FUSED - INST_RETIRED.ANY) / tma_info_slots", + "MetricExpr": "(UOPS_RETIRED.RETIRE_SLOTS + UOPS_RETIRED.MACRO_FUSED - INST_RETIRED.ANY) / tma_info_thread_slots", "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_heavy_operations", "MetricThreshold": "tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to instruction cache misses", - "MetricExpr": "(ICACHE_16B.IFDATA_STALL + 2 * cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@) / tma_info_clks", + "MetricExpr": "(ICACHE_16B.IFDATA_STALL + 2 * cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@) / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;IcMiss;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_icache_misses", "MetricThreshold": "tma_icache_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -385,686 +608,692 @@ "ScaleUnit": "100%" }, { - "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", - "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time", - "MetricGroup": "Power;Summary", - "MetricName": "tma_info_average_frequency" + "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", + "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_thread_slots / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BrMispredicts;tma_issueBM", + "MetricName": "tma_info_bad_spec_branch_misprediction_cost", + "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_info_bottleneck_mispredictions, tma_mispredicts_resteers" + }, + { + "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", + "MetricExpr": "tma_info_inst_mix_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_indirect", + "MetricThreshold": "tma_info_bad_spec_ipmisp_indirect < 1e3" + }, + { + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_core_ipmispredict", + "MetricGroup": "Bad;BadSpec;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmispredict", + "MetricThreshold": "tma_info_bad_spec_ipmispredict < 200" + }, + { + "BriefDescription": "Probability of Core Bound bottleneck hidden by SMT-profiling artifacts", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(100 * (1 - tma_core_bound / tma_ports_utilization if tma_core_bound < tma_ports_utilization else 1) if tma_info_system_smt_2t_utilization > 0.5 else 0)", + "MetricGroup": "Cor;SMT", + "MetricName": "tma_info_botlnk_l0_core_bound_likely", + "MetricThreshold": "tma_info_botlnk_l0_core_bound_likely > 0.5" + }, + { + "BriefDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_fetch_latency * tma_dsb_switches / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches) + tma_fetch_bandwidth * tma_mite / (tma_dsb + tma_mite))", + "MetricGroup": "DSBmiss;Fed;tma_issueFB", + "MetricName": "tma_info_botlnk_l2_dsb_misses", + "MetricThreshold": "tma_info_botlnk_l2_dsb_misses > 10", + "PublicDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp" + }, + { + "BriefDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck", + "MetricExpr": "100 * (tma_fetch_latency * tma_icache_misses / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", + "MetricGroup": "Fed;FetchLat;IcMiss;tma_issueFL", + "MetricName": "tma_info_botlnk_l2_ic_misses", + "MetricThreshold": "tma_info_botlnk_l2_ic_misses > 5", + "PublicDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck. Related metrics: " }, { "BriefDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses)", "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "100 * tma_fetch_latency * (tma_itlb_misses + tma_icache_misses + tma_unknown_branches) / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)", "MetricGroup": "BigFoot;Fed;Frontend;IcMiss;MemoryTLB;tma_issueBC", - "MetricName": "tma_info_big_code", - "MetricThreshold": "tma_info_big_code > 20", - "PublicDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses). Related metrics: tma_info_branching_overhead" + "MetricName": "tma_info_bottleneck_big_code", + "MetricThreshold": "tma_info_bottleneck_big_code > 20", + "PublicDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses). Related metrics: tma_info_bottleneck_branching_overhead" }, { - "BriefDescription": "Branch instructions per taken branch.", - "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", - "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_bptkbranch" + "BriefDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls)", + "MetricExpr": "100 * ((BR_INST_RETIRED.CONDITIONAL + 3 * BR_INST_RETIRED.NEAR_CALL + (BR_INST_RETIRED.NEAR_TAKEN - (BR_INST_RETIRED.CONDITIONAL - BR_INST_RETIRED.NOT_TAKEN) - 2 * BR_INST_RETIRED.NEAR_CALL)) / tma_info_thread_slots)", + "MetricGroup": "Ret;tma_issueBC", + "MetricName": "tma_info_bottleneck_branching_overhead", + "MetricThreshold": "tma_info_bottleneck_branching_overhead > 10", + "PublicDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls). Related metrics: tma_info_bottleneck_big_code" }, { - "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", - "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_slots / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_branch_misprediction_cost", - "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_info_mispredictions, tma_mispredicts_resteers" + "BriefDescription": "Total pipeline cost of instruction fetch bandwidth related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_frontend_bound - tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) - tma_info_bottleneck_big_code", + "MetricGroup": "Fed;FetchBW;Frontend", + "MetricName": "tma_info_bottleneck_instruction_fetch_bw", + "MetricThreshold": "tma_info_bottleneck_instruction_fetch_bw > 20" }, { - "BriefDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls)", - "MetricExpr": "100 * ((BR_INST_RETIRED.CONDITIONAL + 3 * BR_INST_RETIRED.NEAR_CALL + (BR_INST_RETIRED.NEAR_TAKEN - (BR_INST_RETIRED.CONDITIONAL - BR_INST_RETIRED.NOT_TAKEN) - 2 * BR_INST_RETIRED.NEAR_CALL)) / tma_info_slots)", - "MetricGroup": "Ret;tma_issueBC", - "MetricName": "tma_info_branching_overhead", - "MetricThreshold": "tma_info_branching_overhead > 10", - "PublicDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls). Related metrics: tma_info_big_code" + "BriefDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_bandwidth / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_sq_full / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full))) + tma_l1_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_fb_full / (tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk))", + "MetricGroup": "Mem;MemoryBW;Offcore;tma_issueBW", + "MetricName": "tma_info_bottleneck_memory_bandwidth", + "MetricThreshold": "tma_info_bottleneck_memory_bandwidth > 20", + "PublicDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks. Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full" }, { - "BriefDescription": "Fraction of branches that are CALL or RET", - "MetricExpr": "(BR_INST_RETIRED.NEAR_CALL + BR_INST_RETIRED.NEAR_RETURN) / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_callret" + "BriefDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_l1_bound / max(tma_memory_bound, tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_load / max(tma_l1_bound, tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk)) + tma_store_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_store / (tma_dtlb_store + tma_false_sharing + tma_split_stores + tma_store_latency)))", + "MetricGroup": "Mem;MemoryTLB;Offcore;tma_issueTLB", + "MetricName": "tma_info_bottleneck_memory_data_tlbs", + "MetricThreshold": "tma_info_bottleneck_memory_data_tlbs > 20", + "PublicDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs). Related metrics: tma_dtlb_load, tma_dtlb_store" }, { - "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "Pipeline", - "MetricName": "tma_info_clks" + "BriefDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_latency / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_l3_hit_latency / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full)) + tma_l2_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound))", + "MetricGroup": "Mem;MemoryLat;Offcore;tma_issueLat", + "MetricName": "tma_info_bottleneck_memory_latency", + "MetricThreshold": "tma_info_bottleneck_memory_latency > 20", + "PublicDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches). Related metrics: tma_l3_hit_latency, tma_mem_latency" }, { - "BriefDescription": "STLB (2nd level TLB) code speculative misses per kilo instruction (misses of any page-size that complete the page walk)", - "MetricExpr": "1e3 * ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", - "MetricGroup": "Fed;MemoryTLB", - "MetricName": "tma_info_code_stlb_mpki" + "BriefDescription": "Total pipeline cost of Branch Misprediction related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", + "MetricGroup": "Bad;BadSpec;BrMispredicts;tma_issueBM", + "MetricName": "tma_info_bottleneck_mispredictions", + "MetricThreshold": "tma_info_bottleneck_mispredictions > 20", + "PublicDescription": "Total pipeline cost of Branch Misprediction related bottlenecks. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers" + }, + { + "BriefDescription": "Fraction of branches that are CALL or RET", + "MetricExpr": "(BR_INST_RETIRED.NEAR_CALL + BR_INST_RETIRED.NEAR_RETURN) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_callret" }, { "BriefDescription": "Fraction of branches that are non-taken conditionals", "MetricExpr": "BR_INST_RETIRED.NOT_TAKEN / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Bad;Branches;CodeGen;PGO", - "MetricName": "tma_info_cond_nt" + "MetricName": "tma_info_branches_cond_nt" }, { "BriefDescription": "Fraction of branches that are taken conditionals", "MetricExpr": "(BR_INST_RETIRED.CONDITIONAL - BR_INST_RETIRED.NOT_TAKEN) / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Bad;Branches;CodeGen;PGO", - "MetricName": "tma_info_cond_tk" + "MetricName": "tma_info_branches_cond_tk" }, { - "BriefDescription": "Probability of Core Bound bottleneck hidden by SMT-profiling artifacts", + "BriefDescription": "Fraction of branches that are unconditional (direct or indirect) jumps", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(100 * (1 - tma_core_bound / tma_ports_utilization if tma_core_bound < tma_ports_utilization else 1) if tma_info_smt_2t_utilization > 0.5 else 0)", - "MetricGroup": "Cor;SMT", - "MetricName": "tma_info_core_bound_likely", - "MetricThreshold": "tma_info_core_bound_likely > 0.5" + "MetricExpr": "(BR_INST_RETIRED.NEAR_TAKEN - (BR_INST_RETIRED.CONDITIONAL - BR_INST_RETIRED.NOT_TAKEN) - 2 * BR_INST_RETIRED.NEAR_CALL) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_jump" }, { "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", - "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_clks))", + "MetricExpr": "(CPU_CLK_UNHALTED.THREAD / 2 * (1 + CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_XCLK) if #core_wide < 1 else (CPU_CLK_UNHALTED.THREAD_ANY / 2 if #SMT_on else tma_info_thread_clks))", "MetricGroup": "SMT", - "MetricName": "tma_info_core_clks" + "MetricName": "tma_info_core_core_clks" }, { "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks", "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_coreipc" + "MetricName": "tma_info_core_coreipc" }, { - "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricExpr": "1 / tma_info_ipc", - "MetricGroup": "Mem;Pipeline", - "MetricName": "tma_info_cpi" + "BriefDescription": "Floating Point Operations Per Cycle", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * (FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE) + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / tma_info_core_core_clks", + "MetricGroup": "Flops;Ret", + "MetricName": "tma_info_core_flopc" }, { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": "HPC;Summary", - "MetricName": "tma_info_cpu_utilization" + "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", + "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@) / (2 * tma_info_core_core_clks)", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_core_fp_arith_utilization", + "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." }, { - "BriefDescription": "Average Parallel L2 cache miss data reads", - "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", - "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_data_l2_mlp" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", + "MetricExpr": "UOPS_EXECUTED.THREAD / (UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1)", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", + "MetricName": "tma_info_core_ilp" }, { - "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time", - "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", - "MetricName": "tma_info_dram_bw_use", - "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_info_memory_bandwidth, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear)", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BadSpec;BrMispredicts;TopdownL1;tma_L1_group", + "MetricName": "tma_info_core_ipmispredict", + "MetricgroupNoGroup": "TopdownL1" }, { "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", "MetricExpr": "IDQ.DSB_UOPS / (IDQ.DSB_UOPS + IDQ.MITE_UOPS + IDQ.MS_UOPS)", "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", - "MetricName": "tma_info_dsb_coverage", - "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 4 > 0.35", - "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_misses, tma_info_iptb, tma_lcp" - }, - { - "BriefDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * (tma_fetch_latency * tma_dsb_switches / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches) + tma_fetch_bandwidth * tma_mite / (tma_dsb + tma_mite))", - "MetricGroup": "DSBmiss;Fed;tma_issueFB", - "MetricName": "tma_info_dsb_misses", - "MetricThreshold": "tma_info_dsb_misses > 10", - "PublicDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb, tma_lcp" + "MetricName": "tma_info_frontend_dsb_coverage", + "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 4 > 0.35", + "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_inst_mix_iptb, tma_lcp" }, { "BriefDescription": "Average number of cycles of a switch from the DSB fetch-unit to MITE fetch unit - see DSB_Switches tree node for details.", "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / DSB2MITE_SWITCHES.COUNT", "MetricGroup": "DSBmiss", - "MetricName": "tma_info_dsb_switch_cost" - }, - { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", - "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", - "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", - "MetricName": "tma_info_execute" - }, - { - "BriefDescription": "The ratio of Executed- by Issued-Uops", - "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", - "MetricGroup": "Cor;Pipeline", - "MetricName": "tma_info_execute_per_issue", - "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." - }, - { - "BriefDescription": "Fill Buffer (FB) hits per kilo instructions for retired demand loads (L1D misses that merge into ongoing miss-handling entries)", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.FB_HIT / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_fb_hpki" + "MetricName": "tma_info_frontend_dsb_switch_cost" }, { "BriefDescription": "Average number of Uops issued by front-end when it issued something", "MetricExpr": "UOPS_ISSUED.ANY / cpu@UOPS_ISSUED.ANY\\,cmask\\=1@", "MetricGroup": "Fed;FetchBW", - "MetricName": "tma_info_fetch_upc" - }, - { - "BriefDescription": "Floating Point Operations Per Cycle", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * (FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE) + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / tma_info_core_clks", - "MetricGroup": "Flops;Ret", - "MetricName": "tma_info_flopc" + "MetricName": "tma_info_frontend_fetch_upc" }, { - "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@) / (2 * tma_info_core_clks)", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_fp_arith_utilization", - "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." + "BriefDescription": "Average Latency for L1 instruction cache misses", + "MetricExpr": "ICACHE_16B.IFDATA_STALL / cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@ + 2", + "MetricGroup": "Fed;FetchLat;IcMiss", + "MetricName": "tma_info_frontend_icache_miss_latency" }, { - "BriefDescription": "Giga Floating Point Operations Per Second", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * (FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE) + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / 1e9 / duration_time", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_gflops", - "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + "BriefDescription": "Instructions per non-speculative DSB miss (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / FRONTEND_RETIRED.ANY_DSB_MISS", + "MetricGroup": "DSBmiss;Fed", + "MetricName": "tma_info_frontend_ipdsb_miss_ret", + "MetricThreshold": "tma_info_frontend_ipdsb_miss_ret < 50" }, { - "BriefDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck", - "MetricExpr": "100 * (tma_fetch_latency * tma_icache_misses / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", - "MetricGroup": "Fed;FetchLat;IcMiss;tma_issueFL", - "MetricName": "tma_info_ic_misses", - "MetricThreshold": "tma_info_ic_misses > 5", - "PublicDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck. Related metrics: " + "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_inst_mix_instructions / BACLEARS.ANY", + "MetricGroup": "Fed", + "MetricName": "tma_info_frontend_ipunknown_branch" }, { - "BriefDescription": "Average Latency for L1 instruction cache misses", - "MetricExpr": "ICACHE_16B.IFDATA_STALL / cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@ + 2", - "MetricGroup": "Fed;FetchLat;IcMiss", - "MetricName": "tma_info_icache_miss_latency" + "BriefDescription": "L2 cache true code cacheline misses per kilo instruction", + "MetricExpr": "1e3 * FRONTEND_RETIRED.L2_MISS / INST_RETIRED.ANY", + "MetricGroup": "IcMiss", + "MetricName": "tma_info_frontend_l2mpki_code" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", - "MetricExpr": "UOPS_EXECUTED.THREAD / (UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1)", - "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", - "MetricName": "tma_info_ilp" + "BriefDescription": "L2 cache speculative code cacheline misses per kilo instruction", + "MetricExpr": "1e3 * L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", + "MetricGroup": "IcMiss", + "MetricName": "tma_info_frontend_l2mpki_code_all" }, { - "BriefDescription": "Total pipeline cost of instruction fetch bandwidth related bottlenecks", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * (tma_frontend_bound - tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) - tma_info_big_code", - "MetricGroup": "Fed;FetchBW;Frontend", - "MetricName": "tma_info_instruction_fetch_bw", - "MetricThreshold": "tma_info_instruction_fetch_bw > 20" + "BriefDescription": "Branch instructions per taken branch.", + "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;PGO", + "MetricName": "tma_info_inst_mix_bptkbranch" }, { "BriefDescription": "Total number of retired Instructions", "MetricExpr": "INST_RETIRED.ANY", "MetricGroup": "Summary;TmaL1;tma_L1_group", - "MetricName": "tma_info_instructions", + "MetricName": "tma_info_inst_mix_instructions", "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" }, { - "BriefDescription": "Average IO (network or disk) Bandwidth Use for Reads [GB / sec]", - "MetricExpr": "(UNC_IIO_DATA_REQ_OF_CPU.MEM_WRITE.PART0 + UNC_IIO_DATA_REQ_OF_CPU.MEM_WRITE.PART1 + UNC_IIO_DATA_REQ_OF_CPU.MEM_WRITE.PART2 + UNC_IIO_DATA_REQ_OF_CPU.MEM_WRITE.PART3) * 4 / 1e9 / duration_time", - "MetricGroup": "IoBW;Mem;Server;SoC", - "MetricName": "tma_info_io_read_bw" - }, - { - "BriefDescription": "Average IO (network or disk) Bandwidth Use for Writes [GB / sec]", - "MetricExpr": "(UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART0 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART1 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART2 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART3) * 4 / 1e9 / duration_time", - "MetricGroup": "IoBW;Mem;Server;SoC", - "MetricName": "tma_info_io_write_bw" - }, - { "BriefDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate)", "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "INST_RETIRED.ANY / (cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_iparith", - "MetricThreshold": "tma_info_iparith < 10", + "MetricName": "tma_info_inst_mix_iparith", + "MetricThreshold": "tma_info_inst_mix_iparith < 10", "PublicDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate). May undercount due to FMA double counting. Approximated prior to BDW." }, { "BriefDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx128", - "MetricThreshold": "tma_info_iparith_avx128 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx128", + "MetricThreshold": "tma_info_inst_mix_iparith_avx128 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx256", - "MetricThreshold": "tma_info_iparith_avx256 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx256", + "MetricThreshold": "tma_info_inst_mix_iparith_avx256 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic AVX 512-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx512", - "MetricThreshold": "tma_info_iparith_avx512 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx512", + "MetricThreshold": "tma_info_inst_mix_iparith_avx512 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX 512-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_dp", - "MetricThreshold": "tma_info_iparith_scalar_dp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_dp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_dp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_SINGLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_sp", - "MetricThreshold": "tma_info_iparith_scalar_sp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_sp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_sp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per Branch (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Branches;Fed;InsType", - "MetricName": "tma_info_ipbranch", - "MetricThreshold": "tma_info_ipbranch < 8" - }, - { - "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": "Ret;Summary", - "MetricName": "tma_info_ipc" + "MetricName": "tma_info_inst_mix_ipbranch", + "MetricThreshold": "tma_info_inst_mix_ipbranch < 8" }, { "BriefDescription": "Instructions per (near) call (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_CALL", "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_ipcall", - "MetricThreshold": "tma_info_ipcall < 200" - }, - { - "BriefDescription": "Instructions per non-speculative DSB miss (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / FRONTEND_RETIRED.ANY_DSB_MISS", - "MetricGroup": "DSBmiss;Fed", - "MetricName": "tma_info_ipdsb_miss_ret", - "MetricThreshold": "tma_info_ipdsb_miss_ret < 50" - }, - { - "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", - "MetricGroup": "Branches;OS", - "MetricName": "tma_info_ipfarbranch", - "MetricThreshold": "tma_info_ipfarbranch < 1e6" + "MetricName": "tma_info_inst_mix_ipcall", + "MetricThreshold": "tma_info_inst_mix_ipcall < 200" }, { "BriefDescription": "Instructions per Floating Point (FP) Operation (lower number means higher occurrence rate)", "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * (FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE) + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_ipflop", - "MetricThreshold": "tma_info_ipflop < 10" + "MetricName": "tma_info_inst_mix_ipflop", + "MetricThreshold": "tma_info_inst_mix_ipflop < 10" }, { "BriefDescription": "Instructions per Load (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_LOADS", "MetricGroup": "InsType", - "MetricName": "tma_info_ipload", - "MetricThreshold": "tma_info_ipload < 3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", - "MetricExpr": "tma_info_instructions / (UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * cpu@BR_MISP_EXEC.ALL_BRANCHES\\,umask\\=0xE4@)", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_indirect", - "MetricThreshold": "tma_info_ipmisp_indirect < 1e3" - }, - { - "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BadSpec;BrMispredicts", - "MetricName": "tma_info_ipmispredict", - "MetricThreshold": "tma_info_ipmispredict < 200" + "MetricName": "tma_info_inst_mix_ipload", + "MetricThreshold": "tma_info_inst_mix_ipload < 3" }, { "BriefDescription": "Instructions per Store (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_STORES", "MetricGroup": "InsType", - "MetricName": "tma_info_ipstore", - "MetricThreshold": "tma_info_ipstore < 8" + "MetricName": "tma_info_inst_mix_ipstore", + "MetricThreshold": "tma_info_inst_mix_ipstore < 8" }, { "BriefDescription": "Instructions per Software prefetch instruction (of any type: NTA/T0/T1/T2/Prefetch) (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / cpu@SW_PREFETCH_ACCESS.T0\\,umask\\=0xF@", "MetricGroup": "Prefetches", - "MetricName": "tma_info_ipswpf", - "MetricThreshold": "tma_info_ipswpf < 100" + "MetricName": "tma_info_inst_mix_ipswpf", + "MetricThreshold": "tma_info_inst_mix_ipswpf < 100" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW;Frontend;PGO;tma_issueFB", - "MetricName": "tma_info_iptb", - "MetricThreshold": "tma_info_iptb < 9", - "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_lcp" + "MetricName": "tma_info_inst_mix_iptb", + "MetricThreshold": "tma_info_inst_mix_iptb < 9", + "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_lcp" }, { - "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", - "MetricExpr": "tma_info_instructions / BACLEARS.ANY", - "MetricGroup": "Fed", - "MetricName": "tma_info_ipunknown_branch" + "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_core_l1d_cache_fill_bw" }, { - "BriefDescription": "Fraction of branches that are unconditional (direct or indirect) jumps", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(BR_INST_RETIRED.NEAR_TAKEN - (BR_INST_RETIRED.CONDITIONAL - BR_INST_RETIRED.NOT_TAKEN) - 2 * BR_INST_RETIRED.NEAR_CALL) / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_jump" + "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_core_l2_cache_fill_bw" }, { - "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_cpi" + "BriefDescription": "Rate of non silent evictions from the L2 cache per Kilo instruction", + "MetricExpr": "1e3 * L2_LINES_OUT.NON_SILENT / tma_info_inst_mix_instructions", + "MetricGroup": "L2Evicts;Mem;Server", + "MetricName": "tma_info_memory_core_l2_evictions_nonsilent_pki" }, { - "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_utilization", - "MetricThreshold": "tma_info_kernel_utilization > 0.05" + "BriefDescription": "Rate of silent evictions from the L2 cache per Kilo instruction where the evicted lines are dropped (no writeback to L3 or memory)", + "MetricExpr": "1e3 * L2_LINES_OUT.SILENT / tma_info_inst_mix_instructions", + "MetricGroup": "L2Evicts;Mem;Server", + "MetricName": "tma_info_memory_core_l2_evictions_silent_pki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw" + "BriefDescription": "Average per-core data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * OFFCORE_REQUESTS.ALL_REQUESTS / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_core_l3_cache_access_bw" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "tma_info_l1d_cache_fill_bw", + "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw_1t" + "MetricName": "tma_info_memory_core_l3_cache_fill_bw" + }, + { + "BriefDescription": "Fill Buffer (FB) hits per kilo instructions for retired demand loads (L1D misses that merge into ongoing miss-handling entries)", + "MetricExpr": "1e3 * MEM_LOAD_RETIRED.FB_HIT / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_fb_hpki" }, { "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L1_MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki" + "MetricName": "tma_info_memory_l1mpki" }, { "BriefDescription": "L1 cache true misses per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.ALL_DEMAND_DATA_RD / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki_load" - }, - { - "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw" - }, - { - "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "tma_info_l2_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw_1t" - }, - { - "BriefDescription": "Rate of non silent evictions from the L2 cache per Kilo instruction", - "MetricExpr": "1e3 * L2_LINES_OUT.NON_SILENT / tma_info_instructions", - "MetricGroup": "L2Evicts;Mem;Server", - "MetricName": "tma_info_l2_evictions_nonsilent_pki" - }, - { - "BriefDescription": "Rate of silent evictions from the L2 cache per Kilo instruction where the evicted lines are dropped (no writeback to L3 or memory)", - "MetricExpr": "1e3 * L2_LINES_OUT.SILENT / tma_info_instructions", - "MetricGroup": "L2Evicts;Mem;Server", - "MetricName": "tma_info_l2_evictions_silent_pki" + "MetricName": "tma_info_memory_l1mpki_load" }, { "BriefDescription": "L2 cache hits per kilo instruction for all request types (including speculative)", "MetricExpr": "1e3 * (L2_RQSTS.REFERENCES - L2_RQSTS.MISS) / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_all" + "MetricName": "tma_info_memory_l2hpki_all" }, { "BriefDescription": "L2 cache hits per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_HIT / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_load" + "MetricName": "tma_info_memory_l2hpki_load" }, { "BriefDescription": "L2 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L2_MISS / INST_RETIRED.ANY", "MetricGroup": "Backend;CacheMisses;Mem", - "MetricName": "tma_info_l2mpki" + "MetricName": "tma_info_memory_l2mpki" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all request types (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem;Offcore", - "MetricName": "tma_info_l2mpki_all" - }, - { - "BriefDescription": "L2 cache true code cacheline misses per kilo instruction", - "MetricExpr": "1e3 * FRONTEND_RETIRED.L2_MISS / INST_RETIRED.ANY", - "MetricGroup": "IcMiss", - "MetricName": "tma_info_l2mpki_code" - }, - { - "BriefDescription": "L2 cache speculative code cacheline misses per kilo instruction", - "MetricExpr": "1e3 * L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", - "MetricGroup": "IcMiss", - "MetricName": "tma_info_l2mpki_code_all" + "MetricName": "tma_info_memory_l2mpki_all" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2mpki_load" + "MetricName": "tma_info_memory_l2mpki_load" }, { - "BriefDescription": "Average per-core data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * OFFCORE_REQUESTS.ALL_REQUESTS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw" - }, - { - "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_access_bw", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw_1t" + "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L3_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l3mpki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw" + "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", + "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_RETIRED.L1_MISS + MEM_LOAD_RETIRED.FB_HIT)", + "MetricGroup": "Mem;MemoryBound;MemoryLat", + "MetricName": "tma_info_memory_load_miss_real_latency" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw_1t" + "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", + "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", + "MetricGroup": "Mem;MemoryBW;MemoryBound", + "MetricName": "tma_info_memory_mlp", + "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" }, { - "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L3_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l3mpki" + "BriefDescription": "Average Parallel L2 cache miss data reads", + "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", + "MetricGroup": "Memory_BW;Offcore", + "MetricName": "tma_info_memory_oro_data_l2_mlp" }, { "BriefDescription": "Average Latency for L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS.DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l2_miss_latency" + "MetricName": "tma_info_memory_oro_load_l2_miss_latency" }, { "BriefDescription": "Average Parallel L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_DATA_RD", "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_load_l2_mlp" + "MetricName": "tma_info_memory_oro_load_l2_mlp" }, { - "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", - "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_RETIRED.L1_MISS + MEM_LOAD_RETIRED.FB_HIT)", - "MetricGroup": "Mem;MemoryBound;MemoryLat", - "MetricName": "tma_info_load_miss_real_latency" + "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l1d_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l1d_cache_fill_bw_1t" + }, + { + "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l2_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l2_cache_fill_bw_1t" + }, + { + "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_access_bw", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_thread_l3_cache_access_bw_1t" + }, + { + "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l3_cache_fill_bw_1t" + }, + { + "BriefDescription": "STLB (2nd level TLB) code speculative misses per kilo instruction (misses of any page-size that complete the page walk)", + "MetricExpr": "1e3 * ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricGroup": "Fed;MemoryTLB", + "MetricName": "tma_info_memory_tlb_code_stlb_mpki" }, { "BriefDescription": "STLB (2nd level TLB) data load speculative misses per kilo instruction (misses of any page-size that complete the page walk)", "MetricExpr": "1e3 * DTLB_LOAD_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_load_stlb_mpki" + "MetricName": "tma_info_memory_tlb_load_stlb_mpki" }, { - "BriefDescription": "Average latency of data read request to external DRAM memory [in nanoseconds]", - "MetricExpr": "1e9 * (UNC_M_RPQ_OCCUPANCY / UNC_M_RPQ_INSERTS) / imc_0@event\\=0x0@", - "MetricGroup": "Mem;MemoryLat;Server;SoC", - "MetricName": "tma_info_mem_dram_read_latency", - "PublicDescription": "Average latency of data read request to external DRAM memory [in nanoseconds]. Accounts for demand loads and L1/L2 data-read prefetches" + "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", + "MetricConstraint": "NO_GROUP_EVENTS_NMI", + "MetricExpr": "(ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING + EPT.WALK_PENDING) / (2 * tma_info_core_core_clks)", + "MetricGroup": "Mem;MemoryTLB", + "MetricName": "tma_info_memory_tlb_page_walks_utilization", + "MetricThreshold": "tma_info_memory_tlb_page_walks_utilization > 0.5" }, { - "BriefDescription": "Average number of parallel data read requests to external memory", - "MetricExpr": "UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD / UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD@thresh\\=1@", - "MetricGroup": "Mem;MemoryBW;SoC", - "MetricName": "tma_info_mem_parallel_reads", - "PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches" + "BriefDescription": "STLB (2nd level TLB) data store speculative misses per kilo instruction (misses of any page-size that complete the page walk)", + "MetricExpr": "1e3 * DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricGroup": "Mem;MemoryTLB", + "MetricName": "tma_info_memory_tlb_store_stlb_mpki" }, { - "BriefDescription": "Average latency of data read request to external memory (in nanoseconds)", - "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD / UNC_CHA_TOR_INSERTS.IA_MISS_DRD) / (tma_info_socket_clks / duration_time)", - "MetricGroup": "Mem;MemoryLat;SoC", - "MetricName": "tma_info_mem_read_latency", - "PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", + "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", + "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", + "MetricName": "tma_info_pipeline_execute" }, { - "BriefDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_bandwidth / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_sq_full / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full))) + tma_l1_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_fb_full / (tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk))", - "MetricGroup": "Mem;MemoryBW;Offcore;tma_issueBW", - "MetricName": "tma_info_memory_bandwidth", - "MetricThreshold": "tma_info_memory_bandwidth > 20", - "PublicDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks. Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@", + "MetricGroup": "Pipeline;Ret", + "MetricName": "tma_info_pipeline_retire" }, { - "BriefDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * tma_memory_bound * (tma_l1_bound / max(tma_memory_bound, tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_load / max(tma_l1_bound, tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk)) + tma_store_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_store / (tma_dtlb_store + tma_false_sharing + tma_split_stores + tma_store_latency)))", - "MetricGroup": "Mem;MemoryTLB;Offcore;tma_issueTLB", - "MetricName": "tma_info_memory_data_tlbs", - "MetricThreshold": "tma_info_memory_data_tlbs > 20", - "PublicDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs). Related metrics: tma_dtlb_load, tma_dtlb_store" + "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time", + "MetricGroup": "Power;Summary", + "MetricName": "tma_info_system_average_frequency" }, { - "BriefDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_latency / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_l3_hit_latency / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full)) + tma_l2_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound))", - "MetricGroup": "Mem;MemoryLat;Offcore;tma_issueLat", - "MetricName": "tma_info_memory_latency", - "MetricThreshold": "tma_info_memory_latency > 20", - "PublicDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches). Related metrics: tma_l3_hit_latency, tma_mem_latency" + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricGroup": "HPC;Summary", + "MetricName": "tma_info_system_cpu_utilization" }, { - "BriefDescription": "Total pipeline cost of Branch Misprediction related bottlenecks", + "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", + "MetricExpr": "64 * (UNC_M_CAS_COUNT.RD + UNC_M_CAS_COUNT.WR) / 1e9 / duration_time", + "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", + "MetricName": "tma_info_system_dram_bw_use", + "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_mem_bandwidth, tma_sq_full" + }, + { + "BriefDescription": "Giga Floating Point Operations Per Second", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "100 * (tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", - "MetricGroup": "Bad;BadSpec;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_mispredictions", - "MetricThreshold": "tma_info_mispredictions > 20", - "PublicDescription": "Total pipeline cost of Branch Misprediction related bottlenecks. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost, tma_mispredicts_resteers" + "MetricExpr": "(FP_ARITH_INST_RETIRED.SCALAR_SINGLE + FP_ARITH_INST_RETIRED.SCALAR_DOUBLE + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * (FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE) + 8 * (FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE + FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE) + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / 1e9 / duration_time", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_system_gflops", + "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." }, { - "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", - "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", - "MetricGroup": "Mem;MemoryBW;MemoryBound", - "MetricName": "tma_info_mlp", - "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" + "BriefDescription": "Average IO (network or disk) Bandwidth Use for Reads [GB / sec]", + "MetricExpr": "(UNC_IIO_DATA_REQ_OF_CPU.MEM_WRITE.PART0 + UNC_IIO_DATA_REQ_OF_CPU.MEM_WRITE.PART1 + UNC_IIO_DATA_REQ_OF_CPU.MEM_WRITE.PART2 + UNC_IIO_DATA_REQ_OF_CPU.MEM_WRITE.PART3) * 4 / 1e9 / duration_time", + "MetricGroup": "IoBW;Mem;Server;SoC", + "MetricName": "tma_info_system_io_read_bw" }, { - "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", - "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING + EPT.WALK_PENDING) / (2 * tma_info_core_clks)", - "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_page_walks_utilization", - "MetricThreshold": "tma_info_page_walks_utilization > 0.5" + "BriefDescription": "Average IO (network or disk) Bandwidth Use for Writes [GB / sec]", + "MetricExpr": "(UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART0 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART1 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART2 + UNC_IIO_DATA_REQ_OF_CPU.MEM_READ.PART3) * 4 / 1e9 / duration_time", + "MetricGroup": "IoBW;Mem;Server;SoC", + "MetricName": "tma_info_system_io_write_bw" + }, + { + "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", + "MetricGroup": "Branches;OS", + "MetricName": "tma_info_system_ipfarbranch", + "MetricThreshold": "tma_info_system_ipfarbranch < 1e6" + }, + { + "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_cpi" + }, + { + "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_utilization", + "MetricThreshold": "tma_info_system_kernel_utilization > 0.05" + }, + { + "BriefDescription": "Average latency of data read request to external DRAM memory [in nanoseconds]", + "MetricExpr": "1e9 * (UNC_M_RPQ_OCCUPANCY / UNC_M_RPQ_INSERTS) / imc_0@event\\=0x0@", + "MetricGroup": "Mem;MemoryLat;Server;SoC", + "MetricName": "tma_info_system_mem_dram_read_latency", + "PublicDescription": "Average latency of data read request to external DRAM memory [in nanoseconds]. Accounts for demand loads and L1/L2 data-read prefetches" + }, + { + "BriefDescription": "Average number of parallel data read requests to external memory", + "MetricExpr": "UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD / UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD@thresh\\=1@", + "MetricGroup": "Mem;MemoryBW;SoC", + "MetricName": "tma_info_system_mem_parallel_reads", + "PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches" + }, + { + "BriefDescription": "Average latency of data read request to external memory (in nanoseconds)", + "MetricExpr": "1e9 * (UNC_CHA_TOR_OCCUPANCY.IA_MISS_DRD / UNC_CHA_TOR_INSERTS.IA_MISS_DRD) / (tma_info_system_socket_clks / duration_time)", + "MetricGroup": "Mem;MemoryLat;SoC", + "MetricName": "tma_info_system_mem_read_latency", + "PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)" }, { "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for baseline license level 0", - "MetricExpr": "(CORE_POWER.LVL0_TURBO_LICENSE / 2 / tma_info_core_clks if #SMT_on else CORE_POWER.LVL0_TURBO_LICENSE / tma_info_core_clks)", + "MetricExpr": "(CORE_POWER.LVL0_TURBO_LICENSE / 2 / tma_info_core_core_clks if #SMT_on else CORE_POWER.LVL0_TURBO_LICENSE / tma_info_core_core_clks)", "MetricGroup": "Power", - "MetricName": "tma_info_power_license0_utilization", + "MetricName": "tma_info_system_power_license0_utilization", "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for baseline license level 0. This includes non-AVX codes, SSE, AVX 128-bit, and low-current AVX 256-bit codes." }, { "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 1", - "MetricExpr": "(CORE_POWER.LVL1_TURBO_LICENSE / 2 / tma_info_core_clks if #SMT_on else CORE_POWER.LVL1_TURBO_LICENSE / tma_info_core_clks)", + "MetricExpr": "(CORE_POWER.LVL1_TURBO_LICENSE / 2 / tma_info_core_core_clks if #SMT_on else CORE_POWER.LVL1_TURBO_LICENSE / tma_info_core_core_clks)", "MetricGroup": "Power", - "MetricName": "tma_info_power_license1_utilization", - "MetricThreshold": "tma_info_power_license1_utilization > 0.5", + "MetricName": "tma_info_system_power_license1_utilization", + "MetricThreshold": "tma_info_system_power_license1_utilization > 0.5", "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 1. This includes high current AVX 256-bit instructions as well as low current AVX 512-bit instructions." }, { "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 2 (introduced in SKX)", - "MetricExpr": "(CORE_POWER.LVL2_TURBO_LICENSE / 2 / tma_info_core_clks if #SMT_on else CORE_POWER.LVL2_TURBO_LICENSE / tma_info_core_clks)", + "MetricExpr": "(CORE_POWER.LVL2_TURBO_LICENSE / 2 / tma_info_core_core_clks if #SMT_on else CORE_POWER.LVL2_TURBO_LICENSE / tma_info_core_core_clks)", "MetricGroup": "Power", - "MetricName": "tma_info_power_license2_utilization", - "MetricThreshold": "tma_info_power_license2_utilization > 0.5", + "MetricName": "tma_info_system_power_license2_utilization", + "MetricThreshold": "tma_info_system_power_license2_utilization > 0.5", "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 2 (introduced in SKX). This includes high current AVX 512-bit instructions." }, { - "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / cpu@UOPS_RETIRED.RETIRE_SLOTS\\,cmask\\=1@", - "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_retire" - }, - { - "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", - "MetricExpr": "4 * tma_info_core_clks", - "MetricGroup": "TmaL1;tma_L1_group", - "MetricName": "tma_info_slots" - }, - { "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / (CPU_CLK_UNHALTED.REF_XCLK_ANY / 2) if #SMT_on else 0)", "MetricGroup": "SMT", - "MetricName": "tma_info_smt_2t_utilization" + "MetricName": "tma_info_system_smt_2t_utilization" }, { "BriefDescription": "Socket actual clocks when any core is active on that socket", "MetricExpr": "cha_0@event\\=0x0@", "MetricGroup": "SoC", - "MetricName": "tma_info_socket_clks" - }, - { - "BriefDescription": "STLB (2nd level TLB) data store speculative misses per kilo instruction (misses of any page-size that complete the page walk)", - "MetricExpr": "1e3 * DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", - "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_store_stlb_mpki" + "MetricName": "tma_info_system_socket_clks" }, { "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC", "MetricGroup": "Power", - "MetricName": "tma_info_turbo_utilization" + "MetricName": "tma_info_system_turbo_utilization" + }, + { + "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "Pipeline", + "MetricName": "tma_info_thread_clks" + }, + { + "BriefDescription": "Cycles Per Instruction (per Logical Processor)", + "MetricExpr": "1 / tma_info_thread_ipc", + "MetricGroup": "Mem;Pipeline", + "MetricName": "tma_info_thread_cpi" + }, + { + "BriefDescription": "The ratio of Executed- by Issued-Uops", + "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", + "MetricGroup": "Cor;Pipeline", + "MetricName": "tma_info_thread_execute_per_issue", + "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." + }, + { + "BriefDescription": "Instructions Per Cycle (per Logical Processor)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks", + "MetricGroup": "Ret;Summary", + "MetricName": "tma_info_thread_ipc" + }, + { + "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", + "MetricExpr": "4 * tma_info_core_core_clks", + "MetricGroup": "TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots" }, { "BriefDescription": "Uops Per Instruction", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / INST_RETIRED.ANY", "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_uoppi", - "MetricThreshold": "tma_info_uoppi > 1.05" + "MetricName": "tma_info_thread_uoppi", + "MetricThreshold": "tma_info_thread_uoppi > 1.05" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW", - "MetricName": "tma_info_uptb", - "MetricThreshold": "tma_info_uptb < 6" + "MetricName": "tma_info_thread_uptb", + "MetricThreshold": "tma_info_thread_uptb < 6" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses", - "MetricExpr": "ICACHE_64B.IFTAG_STALL / tma_info_clks", + "MetricExpr": "ICACHE_64B.IFTAG_STALL / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -1073,7 +1302,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled without loads missing the L1 data cache", - "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_clks, 0)", + "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_thread_clks, 0)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_issueL1;tma_issueMC;tma_memory_bound_group", "MetricName": "tma_l1_bound", "MetricThreshold": "tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1083,7 +1312,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled due to L2 cache accesses by loads", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) / (MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@) * ((CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks)", + "MetricExpr": "MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) / (MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + cpu@L1D_PEND_MISS.FB_FULL\\,cmask\\=1@) * ((CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l2_bound", "MetricThreshold": "tma_l2_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1092,7 +1321,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L2_MISS - CYCLE_ACTIVITY.STALLS_L3_MISS) / tma_info_clks", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L2_MISS - CYCLE_ACTIVITY.STALLS_L3_MISS) / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1101,20 +1330,20 @@ }, { "BriefDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited)", - "MetricExpr": "17 * tma_info_average_frequency * MEM_LOAD_RETIRED.L3_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "17 * tma_info_system_average_frequency * MEM_LOAD_RETIRED.L3_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "MemoryLat;TopdownL4;tma_L4_group;tma_issueLat;tma_l3_bound_group", "MetricName": "tma_l3_hit_latency", "MetricThreshold": "tma_l3_hit_latency > 0.1 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited). Avoiding private cache misses (i.e. L2 misses/L3 hits) will improve the latency; reduce contention with sibling physical cores and increase performance. Note the value of this node may overlap with its siblings. Sample with: MEM_LOAD_RETIRED.L3_HIT_PS. Related metrics: tma_info_memory_latency, tma_mem_latency", + "PublicDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited). Avoiding private cache misses (i.e. L2 misses/L3 hits) will improve the latency; reduce contention with sibling physical cores and increase performance. Note the value of this node may overlap with its siblings. Sample with: MEM_LOAD_RETIRED.L3_HIT_PS. Related metrics: tma_info_bottleneck_memory_latency, tma_mem_latency", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)", - "MetricExpr": "ILD_STALL.LCP / tma_info_clks", + "MetricExpr": "ILD_STALL.LCP / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_lcp", "MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb", + "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb", "ScaleUnit": "100%" }, { @@ -1123,12 +1352,13 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_light_operations", "MetricThreshold": "tma_light_operations > 0.6", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation). This correlates with total number of instructions used by the program. A uops-per-instruction (see UopPI metric) ratio of 1 or less should be expected for decently optimized software running on Intel Core/Xeon products. While this often indicates efficient X86 instructions were executed; high value does not necessarily mean better performance cannot be achieved. Sample with: INST_RETIRED.PREC_DIST", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Load operations", - "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 + UOPS_DISPATCHED_PORT.PORT_7 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED_PORT.PORT_2 + UOPS_DISPATCHED_PORT.PORT_3 + UOPS_DISPATCHED_PORT.PORT_7 - UOPS_DISPATCHED_PORT.PORT_4) / (2 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_load_op_utilization", "MetricThreshold": "tma_load_op_utilization > 0.6", @@ -1146,7 +1376,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles where the Second-level TLB (STLB) was missed by load accesses, performing a hardware page walk", - "MetricExpr": "DTLB_LOAD_MISSES.WALK_ACTIVE / tma_info_clks", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_ACTIVE / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL5;tma_L5_group;tma_dtlb_load_group", "MetricName": "tma_load_stlb_miss", "MetricThreshold": "tma_load_stlb_miss > 0.05 & (tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1154,7 +1384,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from local memory", - "MetricExpr": "59.5 * tma_info_average_frequency * MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "59.5 * tma_info_system_average_frequency * MEM_LOAD_L3_MISS_RETIRED.LOCAL_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Server;TopdownL5;tma_L5_group;tma_mem_latency_group", "MetricName": "tma_local_dram", "MetricThreshold": "tma_local_dram > 0.1 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1163,7 +1393,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU spent handling cache misses due to lock operations", - "MetricExpr": "(12 * max(0, MEM_INST_RETIRED.LOCK_LOADS - L2_RQSTS.ALL_RFO) + MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES * (11 * L2_RQSTS.RFO_HIT + min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO))) / tma_info_clks", + "MetricExpr": "(12 * max(0, MEM_INST_RETIRED.LOCK_LOADS - L2_RQSTS.ALL_RFO) + MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES * (11 * L2_RQSTS.RFO_HIT + min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO))) / tma_info_thread_clks", "MetricGroup": "Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_l1_bound_group", "MetricName": "tma_lock_latency", "MetricThreshold": "tma_lock_latency > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1177,25 +1407,26 @@ "MetricGroup": "BadSpec;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.1 & tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears. These slots are either wasted by uops fetched prior to the clear; or stalls the out-of-order portion of the machine needs to recover its state after the clear. For example; this can happen due to memory ordering Nukes (e.g. Memory Disambiguation) or Self-Modifying-Code (SMC) nukes. Sample with: MACHINE_CLEARS.COUNT. Related metrics: tma_clears_resteers, tma_contested_accesses, tma_data_sharing, tma_false_sharing, tma_l1_bound, tma_microcode_sequencer, tma_ms_switches, tma_remote_cache", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_clks", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW", "MetricName": "tma_mem_bandwidth", "MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_sq_full", + "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_sq_full", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat", "MetricName": "tma_mem_latency", "MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM). This metric does not aggregate requests from other Logical Processors/Physical Cores/sockets (see Uncore counters for that). Related metrics: tma_info_memory_latency, tma_l3_hit_latency", + "PublicDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM). This metric does not aggregate requests from other Logical Processors/Physical Cores/sockets (see Uncore counters for that). Related metrics: tma_info_bottleneck_memory_latency, tma_l3_hit_latency", "ScaleUnit": "100%" }, { @@ -1205,6 +1436,7 @@ "MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck. Memory Bound estimates fraction of slots where pipeline is likely stalled due to demand load or store instructions. This accounts mainly for (1) non-completed in-flight memory demand loads which coincides with execution units starvation; in addition to (2) cases where stores could impose backpressure on the pipeline when many of them get buffered at the same time (less common out of the two).", "ScaleUnit": "100%" }, @@ -1218,7 +1450,7 @@ }, { "BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_thread_slots", "MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS", "MetricName": "tma_microcode_sequencer", "MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1", @@ -1227,19 +1459,19 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage", - "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks", + "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks", "MetricGroup": "BadSpec;BrMispredicts;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueBM", "MetricName": "tma_mispredicts_resteers", "MetricThreshold": "tma_mispredicts_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost, tma_info_mispredictions", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost, tma_info_bottleneck_mispredictions", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline)", - "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.ALL_MITE_CYCLES_ANY_UOPS - IDQ.ALL_MITE_CYCLES_4_UOPS) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_mite", - "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 4 > 0.35)", + "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 4 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline). This pipeline is used for code that was not pre-cached in the DSB or LSD. For example; inefficiencies due to asymmetric decoders; use of long immediate or LCP can manifest as MITE fetch bandwidth bottleneck. Sample with: FRONTEND_RETIRED.ANY_DSB_MISS", "ScaleUnit": "100%" }, @@ -1254,7 +1486,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)", - "MetricExpr": "2 * IDQ.MS_SWITCHES / tma_info_clks", + "MetricExpr": "2 * IDQ.MS_SWITCHES / tma_info_thread_clks", "MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO", "MetricName": "tma_ms_switches", "MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -1290,7 +1522,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 0 ([SNB+] ALU; [HSW+] ALU and 2nd branch)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_0 / tma_info_core_core_clks", "MetricGroup": "Compute;TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_0", "MetricThreshold": "tma_port_0 > 0.6", @@ -1299,7 +1531,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 1 (ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_1 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_1", "MetricThreshold": "tma_port_1 > 0.6", @@ -1308,7 +1540,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 2 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_2 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_2", "MetricThreshold": "tma_port_2 > 0.6", @@ -1317,7 +1549,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 3 ([SNB+]Loads and Store-address; [ICL+] Loads)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_3 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_load_op_utilization_group", "MetricName": "tma_port_3", "MetricThreshold": "tma_port_3 > 0.6", @@ -1335,7 +1567,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 5 ([SNB+] Branches and ALU; [HSW+] ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_5 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_5", "MetricThreshold": "tma_port_5 > 0.6", @@ -1344,7 +1576,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 6 ([HSW+]Primary Branch and simple ALU)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_6 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_6 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_6", "MetricThreshold": "tma_port_6 > 0.6", @@ -1353,7 +1585,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 7 ([HSW+]simple Store-address)", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_7 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_7 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_store_op_utilization_group", "MetricName": "tma_port_7", "MetricThreshold": "tma_port_7 > 0.6", @@ -1362,7 +1594,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)", - "MetricExpr": "((EXE_ACTIVITY.EXE_BOUND_0_PORTS + (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL)) / tma_info_clks if ARITH.DIVIDER_ACTIVE < CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY else (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_clks)", + "MetricExpr": "((EXE_ACTIVITY.EXE_BOUND_0_PORTS + (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL)) / tma_info_thread_clks if ARITH.DIVIDER_ACTIVE < CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY else (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_thread_clks)", "MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_ports_utilization", "MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -1371,7 +1603,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed no uops on any execution port (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "(UOPS_EXECUTED.CORE_CYCLES_NONE / 2 if #SMT_on else CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_core_clks", + "MetricExpr": "(UOPS_EXECUTED.CORE_CYCLES_NONE / 2 if #SMT_on else CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_0", "MetricThreshold": "tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1380,7 +1612,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the CPU executed total of 1 uop per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((UOPS_EXECUTED.CORE_CYCLES_GE_1 - UOPS_EXECUTED.CORE_CYCLES_GE_2) / 2 if #SMT_on else EXE_ACTIVITY.1_PORTS_UTIL) / tma_info_core_clks", + "MetricExpr": "((UOPS_EXECUTED.CORE_CYCLES_GE_1 - UOPS_EXECUTED.CORE_CYCLES_GE_2) / 2 if #SMT_on else EXE_ACTIVITY.1_PORTS_UTIL) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issueL1;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_1", "MetricThreshold": "tma_ports_utilized_1 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1389,7 +1621,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 2 uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "((UOPS_EXECUTED.CORE_CYCLES_GE_2 - UOPS_EXECUTED.CORE_CYCLES_GE_3) / 2 if #SMT_on else EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_core_clks", + "MetricExpr": "((UOPS_EXECUTED.CORE_CYCLES_GE_2 - UOPS_EXECUTED.CORE_CYCLES_GE_3) / 2 if #SMT_on else EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issue2P;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_2", "MetricThreshold": "tma_ports_utilized_2 > 0.15 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1398,7 +1630,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 3 or more uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise).", - "MetricExpr": "(UOPS_EXECUTED.CORE_CYCLES_GE_3 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_3) / tma_info_core_clks", + "MetricExpr": "(UOPS_EXECUTED.CORE_CYCLES_GE_3 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_3) / tma_info_core_core_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_3m", "MetricThreshold": "tma_ports_utilized_3m > 0.7 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1407,7 +1639,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from remote cache in other sockets including synchronizations issues", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(89.5 * tma_info_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM + 89.5 * tma_info_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "(89.5 * tma_info_system_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_HITM + 89.5 * tma_info_system_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_FWD) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Offcore;Server;Snoop;TopdownL5;tma_L5_group;tma_issueSyncxn;tma_mem_latency_group", "MetricName": "tma_remote_cache", "MetricThreshold": "tma_remote_cache > 0.05 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1416,7 +1648,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling loads from remote memory", - "MetricExpr": "127 * tma_info_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "127 * tma_info_system_average_frequency * MEM_LOAD_L3_MISS_RETIRED.REMOTE_DRAM * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Server;Snoop;TopdownL5;tma_L5_group;tma_mem_latency_group", "MetricName": "tma_remote_dram", "MetricThreshold": "tma_remote_dram > 0.1 & (tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1425,16 +1657,17 @@ }, { "BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired", - "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_slots", + "MetricExpr": "UOPS_RETIRED.RETIRE_SLOTS / tma_info_thread_slots", "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL1", "PublicDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired. Ideally; all pipeline slots would be attributed to the Retiring category. Retiring of 100% would indicate the maximum Pipeline_Width throughput was achieved. Maximizing Retiring typically increases the Instructions-per-cycle (see IPC metric). Note that a high Retiring value does not necessary mean there is no room for more performance. For example; Heavy-operations or Microcode Assists are categorized under Retiring. They often indicate suboptimal performance and can often be optimized or avoided. Sample with: UOPS_RETIRED.RETIRE_SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU issue-pipeline was stalled due to serializing operations", - "MetricExpr": "PARTIAL_RAT_STALLS.SCOREBOARD / tma_info_clks", + "MetricExpr": "PARTIAL_RAT_STALLS.SCOREBOARD / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL5;tma_L5_group;tma_issueSO;tma_ports_utilized_0_group", "MetricName": "tma_serializing_operation", "MetricThreshold": "tma_serializing_operation > 0.1 & (tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)))", @@ -1444,7 +1677,7 @@ { "BriefDescription": "This metric estimates fraction of cycles handling memory load split accesses - load that cross 64-byte cache line boundary", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "tma_info_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_split_loads", "MetricThreshold": "tma_split_loads > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1453,7 +1686,7 @@ }, { "BriefDescription": "This metric represents rate of split store accesses", - "MetricExpr": "MEM_INST_RETIRED.SPLIT_STORES / tma_info_core_clks", + "MetricExpr": "MEM_INST_RETIRED.SPLIT_STORES / tma_info_core_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_issueSpSt;tma_store_bound_group", "MetricName": "tma_split_stores", "MetricThreshold": "tma_split_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1462,16 +1695,16 @@ }, { "BriefDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors)", - "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_clks", + "MetricExpr": "(OFFCORE_REQUESTS_BUFFER.SQ_FULL / 2 if #SMT_on else OFFCORE_REQUESTS_BUFFER.SQ_FULL) / tma_info_core_core_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueBW;tma_l3_bound_group", "MetricName": "tma_sq_full", "MetricThreshold": "tma_sq_full > 0.3 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_mem_bandwidth", + "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_mem_bandwidth", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write", - "MetricExpr": "EXE_ACTIVITY.BOUND_ON_STORES / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.BOUND_ON_STORES / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1480,7 +1713,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of cycles when the memory subsystem had loads blocked since they could not forward data from earlier (in program order) overlapping stores", - "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_clks", + "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_store_fwd_blk", "MetricThreshold": "tma_store_fwd_blk > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1490,7 +1723,7 @@ { "BriefDescription": "This metric estimates fraction of cycles the CPU spent handling L1D store misses", "MetricConstraint": "NO_GROUP_EVENTS_NMI", - "MetricExpr": "(L2_RQSTS.RFO_HIT * 11 * (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) + (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_clks", + "MetricExpr": "(L2_RQSTS.RFO_HIT * 11 * (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) + (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_thread_clks", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_issueSL;tma_store_bound_group", "MetricName": "tma_store_latency", "MetricThreshold": "tma_store_latency > 0.1 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1499,7 +1732,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Store operations", - "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED_PORT.PORT_4 / tma_info_core_core_clks", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_store_op_utilization", "MetricThreshold": "tma_store_op_utilization > 0.6", @@ -1515,7 +1748,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles where the STLB was missed by store accesses, performing a hardware page walk", - "MetricExpr": "DTLB_STORE_MISSES.WALK_ACTIVE / tma_info_core_clks", + "MetricExpr": "DTLB_STORE_MISSES.WALK_ACTIVE / tma_info_core_core_clks", "MetricGroup": "MemoryTLB;TopdownL5;tma_L5_group;tma_dtlb_store_group", "MetricName": "tma_store_stlb_miss", "MetricThreshold": "tma_store_stlb_miss > 0.05 & (tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1523,7 +1756,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to new branch address clears", - "MetricExpr": "9 * BACLEARS.ANY / tma_info_clks", + "MetricExpr": "9 * BACLEARS.ANY / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;TopdownL4;tma_L4_group;tma_branch_resteers_group", "MetricName": "tma_unknown_branches", "MetricThreshold": "tma_unknown_branches > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", @@ -1566,5 +1799,17 @@ "MetricGroup": "transaction", "MetricName": "tsx_transactional_cycles", "ScaleUnit": "100%" + }, + { + "BriefDescription": "Uncore operating frequency in GHz", + "MetricExpr": "UNC_CHA_CLOCKTICKS / (#num_cores / #num_packages * #num_packages) / 1e9 / duration_time", + "MetricName": "uncore_frequency", + "ScaleUnit": "1GHz" + }, + { + "BriefDescription": "Intel(R) Ultra Path Interconnect (UPI) data transmit bandwidth (MB/sec)", + "MetricExpr": "UNC_UPI_TxL_FLITS.ALL_DATA * 7.111111111111111 / 1e6 / duration_time", + "MetricName": "upi_data_transmit_bw", + "ScaleUnit": "1MB/s" } ] diff --git a/tools/perf/pmu-events/arch/x86/snowridgex/cache.json b/tools/perf/pmu-events/arch/x86/snowridgex/cache.json index 0ab90e3bf76b..c6be60584522 100644 --- a/tools/perf/pmu-events/arch/x86/snowridgex/cache.json +++ b/tools/perf/pmu-events/arch/x86/snowridgex/cache.json @@ -72,6 +72,7 @@ "BriefDescription": "Counts the number of cycles the core is stalled due to an instruction cache or TLB miss which hit in the L2, LLC, DRAM or MMIO (Non-DRAM).", "EventCode": "0x34", "EventName": "MEM_BOUND_STALLS.IFETCH", + "PublicDescription": "Counts the number of cycles the core is stalled due to an instruction cache or translation lookaside buffer (TLB) miss which hit in the L2, LLC, DRAM or MMIO (Non-DRAM).", "SampleAfterValue": "200003", "UMask": "0x38" }, @@ -437,6 +438,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT", "MSRIndex": "0x1a6,0x1a7", @@ -446,6 +448,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_HITM", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM", "MSRIndex": "0x1a6,0x1a7", @@ -455,6 +458,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_HIT_NO_FWD", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_NO_FWD", "MSRIndex": "0x1a6,0x1a7", @@ -464,6 +468,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_HIT_WITH_FWD", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD", "MSRIndex": "0x1a6,0x1a7", @@ -473,6 +478,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_MISS", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_MISS", "MSRIndex": "0x1a6,0x1a7", @@ -482,6 +488,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_HIT.SNOOP_NOT_NEEDED", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_NOT_NEEDED", "MSRIndex": "0x1a6,0x1a7", diff --git a/tools/perf/pmu-events/arch/x86/snowridgex/memory.json b/tools/perf/pmu-events/arch/x86/snowridgex/memory.json index 18621909d1a9..c02eb0e836ad 100644 --- a/tools/perf/pmu-events/arch/x86/snowridgex/memory.json +++ b/tools/perf/pmu-events/arch/x86/snowridgex/memory.json @@ -96,6 +96,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_MISS", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.L3_MISS", "MSRIndex": "0x1a6,0x1a7", @@ -105,6 +106,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.L3_MISS_LOCAL", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.L3_MISS_LOCAL", "MSRIndex": "0x1a6,0x1a7", diff --git a/tools/perf/pmu-events/arch/x86/snowridgex/other.json b/tools/perf/pmu-events/arch/x86/snowridgex/other.json index 00ae180ded25..fefbc383b840 100644 --- a/tools/perf/pmu-events/arch/x86/snowridgex/other.json +++ b/tools/perf/pmu-events/arch/x86/snowridgex/other.json @@ -1,6 +1,7 @@ [ { "BriefDescription": "This event is deprecated. Refer to new event BUS_LOCK.SELF_LOCKS", + "Deprecated": "1", "EdgeDetect": "1", "EventCode": "0x63", "EventName": "BUS_LOCK.ALL", @@ -16,6 +17,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event BUS_LOCK.BLOCK_CYCLES", + "Deprecated": "1", "EventCode": "0x63", "EventName": "BUS_LOCK.CYCLES_OTHER_BLOCK", "SampleAfterValue": "200003", @@ -23,6 +25,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event BUS_LOCK.LOCK_CYCLES", + "Deprecated": "1", "EventCode": "0x63", "EventName": "BUS_LOCK.CYCLES_SELF_BLOCK", "SampleAfterValue": "200003", @@ -46,6 +49,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event MEM_BOUND_STALLS.LOAD_DRAM_HIT", + "Deprecated": "1", "EventCode": "0x34", "EventName": "C0_STALLS.LOAD_DRAM_HIT", "SampleAfterValue": "200003", @@ -53,6 +57,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event MEM_BOUND_STALLS.LOAD_L2_HIT", + "Deprecated": "1", "EventCode": "0x34", "EventName": "C0_STALLS.LOAD_L2_HIT", "SampleAfterValue": "200003", @@ -60,6 +65,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event MEM_BOUND_STALLS.LOAD_LLC_HIT", + "Deprecated": "1", "EventCode": "0x34", "EventName": "C0_STALLS.LOAD_LLC_HIT", "SampleAfterValue": "200003", @@ -207,6 +213,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.ANY_RESPONSE", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.ANY_RESPONSE", "MSRIndex": "0x1a6,0x1a7", @@ -216,6 +223,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.DRAM", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.DRAM", "MSRIndex": "0x1a6,0x1a7", @@ -225,6 +233,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.LOCAL_DRAM", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.LOCAL_DRAM", "MSRIndex": "0x1a6,0x1a7", @@ -234,6 +243,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event OCR.DEMAND_DATA_AND_L1PF_RD.OUTSTANDING", + "Deprecated": "1", "EventCode": "0XB7", "EventName": "OCR.DEMAND_DATA_RD.OUTSTANDING", "MSRIndex": "0x1a6", diff --git a/tools/perf/pmu-events/arch/x86/snowridgex/pipeline.json b/tools/perf/pmu-events/arch/x86/snowridgex/pipeline.json index 9dd8c909facc..c483c0838e08 100644 --- a/tools/perf/pmu-events/arch/x86/snowridgex/pipeline.json +++ b/tools/perf/pmu-events/arch/x86/snowridgex/pipeline.json @@ -165,6 +165,7 @@ }, { "BriefDescription": "This event is deprecated.", + "Deprecated": "1", "EventCode": "0xcd", "EventName": "CYCLES_DIV_BUSY.ANY", "SampleAfterValue": "2000003" @@ -283,6 +284,7 @@ }, { "BriefDescription": "This event is deprecated. Refer to new event TOPDOWN_BAD_SPECULATION.FASTNUKE", + "Deprecated": "1", "EventCode": "0x73", "EventName": "TOPDOWN_BAD_SPECULATION.MONUKE", "SampleAfterValue": "1000003", @@ -338,6 +340,7 @@ }, { "BriefDescription": "This event is deprecated.", + "Deprecated": "1", "EventCode": "0x74", "EventName": "TOPDOWN_BE_BOUND.STORE_BUFFER", "SampleAfterValue": "1000003", diff --git a/tools/perf/pmu-events/arch/x86/snowridgex/uncore-interconnect.json b/tools/perf/pmu-events/arch/x86/snowridgex/uncore-interconnect.json index de3840078e21..7e2895f7fe3d 100644 --- a/tools/perf/pmu-events/arch/x86/snowridgex/uncore-interconnect.json +++ b/tools/perf/pmu-events/arch/x86/snowridgex/uncore-interconnect.json @@ -590,7 +590,7 @@ "EventCode": "0x0C", "EventName": "UNC_I_TxS_REQUEST_OCCUPANCY", "PerPkg": "1", - "PublicDescription": "Outbound Request Queue Occupancy : Accumultes the number of outstanding outbound requests from the IRP to the switch (towards the devices). This can be used in conjuection with the allocations event in order to calculate average latency of outbound requests.", + "PublicDescription": "Outbound Request Queue Occupancy : Accumulates the number of outstanding outbound requests from the IRP to the switch (towards the devices). This can be used in conjunction with the allocations event in order to calculate average latency of outbound requests.", "Unit": "IRP" }, { @@ -5570,7 +5570,7 @@ "Unit": "M2M" }, { - "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Regular : Channel 0", + "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Regular : Channel 0", "EventCode": "0x4D", "EventName": "UNC_M2M_WPQ_NO_REG_CRD.CHN0", "PerPkg": "1", @@ -5578,7 +5578,7 @@ "Unit": "M2M" }, { - "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Regular : Channel 1", + "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Regular : Channel 1", "EventCode": "0x4D", "EventName": "UNC_M2M_WPQ_NO_REG_CRD.CHN1", "PerPkg": "1", @@ -5586,7 +5586,7 @@ "Unit": "M2M" }, { - "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Regular : Channel 2", + "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Regular : Channel 2", "EventCode": "0x4D", "EventName": "UNC_M2M_WPQ_NO_REG_CRD.CHN2", "PerPkg": "1", @@ -5594,7 +5594,7 @@ "Unit": "M2M" }, { - "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Special : Channel 0", + "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Special : Channel 0", "EventCode": "0x4E", "EventName": "UNC_M2M_WPQ_NO_SPEC_CRD.CHN0", "PerPkg": "1", @@ -5602,7 +5602,7 @@ "Unit": "M2M" }, { - "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Special : Channel 1", + "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Special : Channel 1", "EventCode": "0x4E", "EventName": "UNC_M2M_WPQ_NO_SPEC_CRD.CHN1", "PerPkg": "1", @@ -5610,7 +5610,7 @@ "Unit": "M2M" }, { - "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Special : Channel 2", + "BriefDescription": "M2M->iMC WPQ Cycles w/Credits - Special : Channel 2", "EventCode": "0x4E", "EventName": "UNC_M2M_WPQ_NO_SPEC_CRD.CHN2", "PerPkg": "1", diff --git a/tools/perf/pmu-events/arch/x86/snowridgex/uncore-io.json b/tools/perf/pmu-events/arch/x86/snowridgex/uncore-io.json index 996028071ee4..ecdd6f0f8e8f 100644 --- a/tools/perf/pmu-events/arch/x86/snowridgex/uncore-io.json +++ b/tools/perf/pmu-events/arch/x86/snowridgex/uncore-io.json @@ -34,7 +34,6 @@ "EventCode": "0xff", "EventName": "UNC_IIO_BANDWIDTH_IN.PART0_FREERUN", "PerPkg": "1", - "PublicDescription": "UNC_IIO_BANDWIDTH_IN.PART0_FREERUN", "UMask": "0x20", "Unit": "iio_free_running" }, @@ -43,7 +42,6 @@ "EventCode": "0xff", "EventName": "UNC_IIO_BANDWIDTH_IN.PART1_FREERUN", "PerPkg": "1", - "PublicDescription": "UNC_IIO_BANDWIDTH_IN.PART1_FREERUN", "UMask": "0x21", "Unit": "iio_free_running" }, @@ -52,7 +50,6 @@ "EventCode": "0xff", "EventName": "UNC_IIO_BANDWIDTH_IN.PART2_FREERUN", "PerPkg": "1", - "PublicDescription": "UNC_IIO_BANDWIDTH_IN.PART2_FREERUN", "UMask": "0x22", "Unit": "iio_free_running" }, @@ -61,7 +58,6 @@ "EventCode": "0xff", "EventName": "UNC_IIO_BANDWIDTH_IN.PART3_FREERUN", "PerPkg": "1", - "PublicDescription": "UNC_IIO_BANDWIDTH_IN.PART3_FREERUN", "UMask": "0x23", "Unit": "iio_free_running" }, @@ -70,7 +66,6 @@ "EventCode": "0xff", "EventName": "UNC_IIO_BANDWIDTH_IN.PART4_FREERUN", "PerPkg": "1", - "PublicDescription": "UNC_IIO_BANDWIDTH_IN.PART4_FREERUN", "UMask": "0x24", "Unit": "iio_free_running" }, @@ -79,7 +74,6 @@ "EventCode": "0xff", "EventName": "UNC_IIO_BANDWIDTH_IN.PART5_FREERUN", "PerPkg": "1", - "PublicDescription": "UNC_IIO_BANDWIDTH_IN.PART5_FREERUN", "UMask": "0x25", "Unit": "iio_free_running" }, @@ -88,7 +82,6 @@ "EventCode": "0xff", "EventName": "UNC_IIO_BANDWIDTH_IN.PART6_FREERUN", "PerPkg": "1", - "PublicDescription": "UNC_IIO_BANDWIDTH_IN.PART6_FREERUN", "UMask": "0x26", "Unit": "iio_free_running" }, @@ -97,7 +90,6 @@ "EventCode": "0xff", "EventName": "UNC_IIO_BANDWIDTH_IN.PART7_FREERUN", "PerPkg": "1", - "PublicDescription": "UNC_IIO_BANDWIDTH_IN.PART7_FREERUN", "UMask": "0x27", "Unit": "iio_free_running" }, diff --git a/tools/perf/pmu-events/arch/x86/snowridgex/uncore-memory.json b/tools/perf/pmu-events/arch/x86/snowridgex/uncore-memory.json index 530e9b71b92a..b80911d498dd 100644 --- a/tools/perf/pmu-events/arch/x86/snowridgex/uncore-memory.json +++ b/tools/perf/pmu-events/arch/x86/snowridgex/uncore-memory.json @@ -130,7 +130,6 @@ "EventCode": "0xff", "EventName": "UNC_M_CLOCKTICKS_FREERUN", "PerPkg": "1", - "PublicDescription": "UNC_M_CLOCKTICKS_FREERUN", "UMask": "0x10", "Unit": "imc_free_running" }, @@ -322,7 +321,7 @@ "EventCode": "0x02", "EventName": "UNC_M_PRE_COUNT.PGT", "PerPkg": "1", - "PublicDescription": "DRAM Precharge commands. : Precharge due to page table : Counts the number of DRAM Precharge commands sent on this channel. : Prechages from Page Table", + "PublicDescription": "DRAM Precharge commands. : Precharge due to page table : Counts the number of DRAM Precharge commands sent on this channel. : Precharges from Page Table", "UMask": "0x10", "Unit": "iMC" }, @@ -497,7 +496,7 @@ "EventCode": "0x82", "EventName": "UNC_M_WPQ_OCCUPANCY_PCH0", "PerPkg": "1", - "PublicDescription": "Write Pending Queue Occupancy : Accumulates the occupancies of the Write Pending Queue each cycle. This can then be used to calculate both the average queue occupancy (in conjunction with the number of cycles not empty) and the average latency (in conjunction with the number of allocations). The WPQ is used to schedule write out to the memory controller and to track the writes. Requests allocate into the WPQ soon after they enter the memory controller, and need credits for an entry in this buffer before being sent from the HA to the iMC. They deallocate after being issued to DRAM. Write requests themselves are able to complete (from the perspective of the rest of the system) as soon they have posted to the iMC. This is not to be confused with actually performing the write to DRAM. Therefore, the average latency for this queue is actually not useful for deconstruction intermediate write latencies. So, we provide filtering based on if the request has posted or not. By using the not posted filter, we can track how long writes spent in the iMC before completions were sent to the HA. The posted filter, on the other hand, provides information about how much queueing is actually happenning in the iMC for writes before they are actually issued to memory. High average occupancies will generally coincide with high write major mode counts.", + "PublicDescription": "Write Pending Queue Occupancy : Accumulates the occupancies of the Write Pending Queue each cycle. This can then be used to calculate both the average queue occupancy (in conjunction with the number of cycles not empty) and the average latency (in conjunction with the number of allocations). The WPQ is used to schedule write out to the memory controller and to track the writes. Requests allocate into the WPQ soon after they enter the memory controller, and need credits for an entry in this buffer before being sent from the HA to the iMC. They deallocate after being issued to DRAM. Write requests themselves are able to complete (from the perspective of the rest of the system) as soon they have posted to the iMC. This is not to be confused with actually performing the write to DRAM. Therefore, the average latency for this queue is actually not useful for deconstruction intermediate write latencies. So, we provide filtering based on if the request has posted or not. By using the not posted filter, we can track how long writes spent in the iMC before completions were sent to the HA. The posted filter, on the other hand, provides information about how much queueing is actually happening in the iMC for writes before they are actually issued to memory. High average occupancies will generally coincide with high write major mode counts.", "Unit": "iMC" }, { @@ -505,7 +504,7 @@ "EventCode": "0x83", "EventName": "UNC_M_WPQ_OCCUPANCY_PCH1", "PerPkg": "1", - "PublicDescription": "Write Pending Queue Occupancy : Accumulates the occupancies of the Write Pending Queue each cycle. This can then be used to calculate both the average queue occupancy (in conjunction with the number of cycles not empty) and the average latency (in conjunction with the number of allocations). The WPQ is used to schedule write out to the memory controller and to track the writes. Requests allocate into the WPQ soon after they enter the memory controller, and need credits for an entry in this buffer before being sent from the HA to the iMC. They deallocate after being issued to DRAM. Write requests themselves are able to complete (from the perspective of the rest of the system) as soon they have posted to the iMC. This is not to be confused with actually performing the write to DRAM. Therefore, the average latency for this queue is actually not useful for deconstruction intermediate write latencies. So, we provide filtering based on if the request has posted or not. By using the not posted filter, we can track how long writes spent in the iMC before completions were sent to the HA. The posted filter, on the other hand, provides information about how much queueing is actually happenning in the iMC for writes before they are actually issued to memory. High average occupancies will generally coincide with high write major mode counts.", + "PublicDescription": "Write Pending Queue Occupancy : Accumulates the occupancies of the Write Pending Queue each cycle. This can then be used to calculate both the average queue occupancy (in conjunction with the number of cycles not empty) and the average latency (in conjunction with the number of allocations). The WPQ is used to schedule write out to the memory controller and to track the writes. Requests allocate into the WPQ soon after they enter the memory controller, and need credits for an entry in this buffer before being sent from the HA to the iMC. They deallocate after being issued to DRAM. Write requests themselves are able to complete (from the perspective of the rest of the system) as soon they have posted to the iMC. This is not to be confused with actually performing the write to DRAM. Therefore, the average latency for this queue is actually not useful for deconstruction intermediate write latencies. So, we provide filtering based on if the request has posted or not. By using the not posted filter, we can track how long writes spent in the iMC before completions were sent to the HA. The posted filter, on the other hand, provides information about how much queueing is actually happening in the iMC for writes before they are actually issued to memory. High average occupancies will generally coincide with high write major mode counts.", "Unit": "iMC" }, { diff --git a/tools/perf/pmu-events/arch/x86/snowridgex/uncore-power.json b/tools/perf/pmu-events/arch/x86/snowridgex/uncore-power.json index 27fc155f1223..a61ffca2dfea 100644 --- a/tools/perf/pmu-events/arch/x86/snowridgex/uncore-power.json +++ b/tools/perf/pmu-events/arch/x86/snowridgex/uncore-power.json @@ -149,7 +149,7 @@ "EventCode": "0x80", "EventName": "UNC_P_POWER_STATE_OCCUPANCY.CORES_C0", "PerPkg": "1", - "PublicDescription": "Number of cores in C-State : C0 and C1 : This is an occupancy event that tracks the number of cores that are in the chosen C-State. It can be used by itself to get the average number of cores in that C-state with threshholding to generate histograms, or with other PCU events and occupancy triggering to capture other details.", + "PublicDescription": "Number of cores in C-State : C0 and C1 : This is an occupancy event that tracks the number of cores that are in the chosen C-State. It can be used by itself to get the average number of cores in that C-state with thresholding to generate histograms, or with other PCU events and occupancy triggering to capture other details.", "Unit": "PCU" }, { @@ -157,7 +157,7 @@ "EventCode": "0x80", "EventName": "UNC_P_POWER_STATE_OCCUPANCY.CORES_C3", "PerPkg": "1", - "PublicDescription": "Number of cores in C-State : C3 : This is an occupancy event that tracks the number of cores that are in the chosen C-State. It can be used by itself to get the average number of cores in that C-state with threshholding to generate histograms, or with other PCU events and occupancy triggering to capture other details.", + "PublicDescription": "Number of cores in C-State : C3 : This is an occupancy event that tracks the number of cores that are in the chosen C-State. It can be used by itself to get the average number of cores in that C-state with thresholding to generate histograms, or with other PCU events and occupancy triggering to capture other details.", "Unit": "PCU" }, { @@ -165,7 +165,7 @@ "EventCode": "0x80", "EventName": "UNC_P_POWER_STATE_OCCUPANCY.CORES_C6", "PerPkg": "1", - "PublicDescription": "Number of cores in C-State : C6 and C7 : This is an occupancy event that tracks the number of cores that are in the chosen C-State. It can be used by itself to get the average number of cores in that C-state with threshholding to generate histograms, or with other PCU events and occupancy triggering to capture other details.", + "PublicDescription": "Number of cores in C-State : C6 and C7 : This is an occupancy event that tracks the number of cores that are in the chosen C-State. It can be used by itself to get the average number of cores in that C-state with thresholding to generate histograms, or with other PCU events and occupancy triggering to capture other details.", "Unit": "PCU" }, { diff --git a/tools/perf/pmu-events/arch/x86/tigerlake/cache.json b/tools/perf/pmu-events/arch/x86/tigerlake/cache.json index 738249a6f488..c54fb65d3259 100644 --- a/tools/perf/pmu-events/arch/x86/tigerlake/cache.json +++ b/tools/perf/pmu-events/arch/x86/tigerlake/cache.json @@ -323,6 +323,16 @@ "UMask": "0x2" }, { + "BriefDescription": "Retired instructions with at least 1 uncacheable load or lock.", + "Data_LA": "1", + "EventCode": "0xd4", + "EventName": "MEM_LOAD_MISC_RETIRED.UC", + "PEBS": "1", + "PublicDescription": "Retired instructions with at least one load to uncacheable memory-type, or at least one cache-line split locked access", + "SampleAfterValue": "100007", + "UMask": "0x4" + }, + { "BriefDescription": "Number of completed demand load requests that missed the L1, but hit the FB(fill buffer), because a preceding miss to the same cacheline initiated the line to be brought into L1, but data is not yet ready in L1.", "Data_LA": "1", "EventCode": "0xd1", @@ -511,6 +521,14 @@ "UMask": "0x4" }, { + "BriefDescription": "Counts bus locks, accounts for cache line split locks and UC locks.", + "EventCode": "0xf4", + "EventName": "SQ_MISC.BUS_LOCK", + "PublicDescription": "Counts the more expensive bus lock needed to enforce cache coherency for certain memory accesses that need to be done atomically. Can be created by issuing an atomic instruction (via the LOCK prefix) which causes a cache line split or accesses uncacheable memory.", + "SampleAfterValue": "100003", + "UMask": "0x10" + }, + { "BriefDescription": "Cycles the superQ cannot take any more entries.", "EventCode": "0xf4", "EventName": "SQ_MISC.SQ_FULL", diff --git a/tools/perf/pmu-events/arch/x86/tigerlake/metricgroups.json b/tools/perf/pmu-events/arch/x86/tigerlake/metricgroups.json new file mode 100644 index 000000000000..a151ba9cccb0 --- /dev/null +++ b/tools/perf/pmu-events/arch/x86/tigerlake/metricgroups.json @@ -0,0 +1,113 @@ +{ + "Backend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Bad": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BadSpec": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BigFoot": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "BrMispredicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Branches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CacheMisses": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "CodeGen": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Compute": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Cor": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DSBmiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "DataSharing": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Fed": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FetchLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Flops": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpScalar": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "FpVector": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Frontend": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "HPC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "IcMiss": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "InsType": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "L2Evicts": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "LSD": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MachineClears": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryBound": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryLat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MemoryTLB": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_BW": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Memory_Lat": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "MicroSeq": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "OS": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Offcore": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PGO": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Pipeline": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "PortsUtil": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Power": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Prefetches": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Ret": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Retire": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SMT": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Server": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Snoop": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "SoC": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "Summary": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL1": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL2": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TmaL3mem": "Grouping from Top-down Microarchitecture Analysis Metrics spreadsheet", + "TopdownL1": "Metrics for top-down breakdown at level 1", + "TopdownL2": "Metrics for top-down breakdown at level 2", + "TopdownL3": "Metrics for top-down breakdown at level 3", + "TopdownL4": "Metrics for top-down breakdown at level 4", + "TopdownL5": "Metrics for top-down breakdown at level 5", + "TopdownL6": "Metrics for top-down breakdown at level 6", + "tma_L1_group": "Metrics for top-down breakdown at level 1", + "tma_L2_group": "Metrics for top-down breakdown at level 2", + "tma_L3_group": "Metrics for top-down breakdown at level 3", + "tma_L4_group": "Metrics for top-down breakdown at level 4", + "tma_L5_group": "Metrics for top-down breakdown at level 5", + "tma_L6_group": "Metrics for top-down breakdown at level 6", + "tma_alu_op_utilization_group": "Metrics contributing to tma_alu_op_utilization category", + "tma_backend_bound_group": "Metrics contributing to tma_backend_bound category", + "tma_bad_speculation_group": "Metrics contributing to tma_bad_speculation category", + "tma_branch_resteers_group": "Metrics contributing to tma_branch_resteers category", + "tma_core_bound_group": "Metrics contributing to tma_core_bound category", + "tma_dram_bound_group": "Metrics contributing to tma_dram_bound category", + "tma_dtlb_load_group": "Metrics contributing to tma_dtlb_load category", + "tma_dtlb_store_group": "Metrics contributing to tma_dtlb_store category", + "tma_fetch_bandwidth_group": "Metrics contributing to tma_fetch_bandwidth category", + "tma_fetch_latency_group": "Metrics contributing to tma_fetch_latency category", + "tma_fp_arith_group": "Metrics contributing to tma_fp_arith category", + "tma_fp_vector_group": "Metrics contributing to tma_fp_vector category", + "tma_frontend_bound_group": "Metrics contributing to tma_frontend_bound category", + "tma_heavy_operations_group": "Metrics contributing to tma_heavy_operations category", + "tma_issue2P": "Metrics related by the issue $issue2P", + "tma_issueBC": "Metrics related by the issue $issueBC", + "tma_issueBM": "Metrics related by the issue $issueBM", + "tma_issueBW": "Metrics related by the issue $issueBW", + "tma_issueD0": "Metrics related by the issue $issueD0", + "tma_issueFB": "Metrics related by the issue $issueFB", + "tma_issueFL": "Metrics related by the issue $issueFL", + "tma_issueL1": "Metrics related by the issue $issueL1", + "tma_issueLat": "Metrics related by the issue $issueLat", + "tma_issueMC": "Metrics related by the issue $issueMC", + "tma_issueMS": "Metrics related by the issue $issueMS", + "tma_issueMV": "Metrics related by the issue $issueMV", + "tma_issueRFO": "Metrics related by the issue $issueRFO", + "tma_issueSL": "Metrics related by the issue $issueSL", + "tma_issueSO": "Metrics related by the issue $issueSO", + "tma_issueSmSt": "Metrics related by the issue $issueSmSt", + "tma_issueSpSt": "Metrics related by the issue $issueSpSt", + "tma_issueSyncxn": "Metrics related by the issue $issueSyncxn", + "tma_issueTLB": "Metrics related by the issue $issueTLB", + "tma_l1_bound_group": "Metrics contributing to tma_l1_bound category", + "tma_l3_bound_group": "Metrics contributing to tma_l3_bound category", + "tma_light_operations_group": "Metrics contributing to tma_light_operations category", + "tma_load_op_utilization_group": "Metrics contributing to tma_load_op_utilization category", + "tma_mem_latency_group": "Metrics contributing to tma_mem_latency category", + "tma_memory_bound_group": "Metrics contributing to tma_memory_bound category", + "tma_microcode_sequencer_group": "Metrics contributing to tma_microcode_sequencer category", + "tma_mite_group": "Metrics contributing to tma_mite category", + "tma_ports_utilization_group": "Metrics contributing to tma_ports_utilization category", + "tma_ports_utilized_0_group": "Metrics contributing to tma_ports_utilized_0 category", + "tma_ports_utilized_3m_group": "Metrics contributing to tma_ports_utilized_3m category", + "tma_retiring_group": "Metrics contributing to tma_retiring category", + "tma_serializing_operation_group": "Metrics contributing to tma_serializing_operation category", + "tma_store_bound_group": "Metrics contributing to tma_store_bound category", + "tma_store_op_utilization_group": "Metrics contributing to tma_store_op_utilization category" +} diff --git a/tools/perf/pmu-events/arch/x86/tigerlake/pipeline.json b/tools/perf/pmu-events/arch/x86/tigerlake/pipeline.json index a0aeeb801fd7..020801cbd7e3 100644 --- a/tools/perf/pmu-events/arch/x86/tigerlake/pipeline.json +++ b/tools/perf/pmu-events/arch/x86/tigerlake/pipeline.json @@ -395,6 +395,7 @@ { "BriefDescription": "Clears speculative count", "CounterMask": "1", + "EdgeDetect": "1", "EventCode": "0x0d", "EventName": "INT_MISC.CLEARS_COUNT", "PublicDescription": "Counts the number of speculative clears due to any type of branch misprediction or machine clears", diff --git a/tools/perf/pmu-events/arch/x86/tigerlake/tgl-metrics.json b/tools/perf/pmu-events/arch/x86/tigerlake/tgl-metrics.json index 4c80d6be6cf1..83346911aa63 100644 --- a/tools/perf/pmu-events/arch/x86/tigerlake/tgl-metrics.json +++ b/tools/perf/pmu-events/arch/x86/tigerlake/tgl-metrics.json @@ -79,7 +79,7 @@ }, { "BriefDescription": "This metric estimates how often memory load accesses were aliased by preceding stores (in program order) with a 4K address offset", - "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_clks", + "MetricExpr": "LD_BLOCKS_PARTIAL.ADDRESS_ALIAS / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_4k_aliasing", "MetricThreshold": "tma_4k_aliasing > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -88,7 +88,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution ports for ALU operations.", - "MetricExpr": "(UOPS_DISPATCHED.PORT_0 + UOPS_DISPATCHED.PORT_1 + UOPS_DISPATCHED.PORT_5 + UOPS_DISPATCHED.PORT_6) / (4 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED.PORT_0 + UOPS_DISPATCHED.PORT_1 + UOPS_DISPATCHED.PORT_5 + UOPS_DISPATCHED.PORT_6) / (4 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_alu_op_utilization", "MetricThreshold": "tma_alu_op_utilization > 0.6", @@ -96,7 +96,7 @@ }, { "BriefDescription": "This metric estimates fraction of slots the CPU retired uops delivered by the Microcode_Sequencer as a result of Assists", - "MetricExpr": "100 * ASSISTS.ANY / tma_info_slots", + "MetricExpr": "100 * ASSISTS.ANY / tma_info_thread_slots", "MetricGroup": "TopdownL4;tma_L4_group;tma_microcode_sequencer_group", "MetricName": "tma_assists", "MetricThreshold": "tma_assists > 0.1 & (tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1)", @@ -105,25 +105,29 @@ }, { "BriefDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend", - "MetricExpr": "topdown\\-be\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 5 * cpu@INT_MISC.RECOVERY_CYCLES\\,cmask\\=1\\,edge@ / tma_info_slots", - "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "topdown\\-be\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 5 * cpu@INT_MISC.RECOVERY_CYCLES\\,cmask\\=1\\,edge@ / tma_info_thread_slots", + "MetricGroup": "Default;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_backend_bound", "MetricThreshold": "tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots where no uops are being delivered due to a lack of required resources for accepting new uops in the Backend. Backend is the portion of the processor core where the out-of-order scheduler dispatches ready uops into their respective execution units; and once completed these uops get retired according to program order. For example; stalls due to data-cache misses or stalls due to the divider unit being overloaded are both categorized under Backend Bound. Backend Bound is further divided into two main categories: Memory Bound and Core Bound. Sample with: TOPDOWN.BACKEND_BOUND_SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This category represents fraction of slots wasted due to incorrect speculations", + "DefaultMetricgroupName": "TopdownL1", "MetricExpr": "max(1 - (tma_frontend_bound + tma_backend_bound + tma_retiring), 0)", - "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", + "MetricGroup": "Default;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_bad_speculation", "MetricThreshold": "tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots wasted due to incorrect speculations. This include slots used to issue uops that do not eventually get retired and slots for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work due to miss-predicted branches are categorized under Bad Speculation category. Incorrect data speculation followed by Memory Ordering Nukes is another example.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring branch instructions.", - "MetricExpr": "tma_light_operations * BR_INST_RETIRED.ALL_BRANCHES / (tma_retiring * tma_info_slots)", + "MetricExpr": "tma_light_operations * BR_INST_RETIRED.ALL_BRANCHES / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_branch_instructions", "MetricThreshold": "tma_branch_instructions > 0.1 & tma_light_operations > 0.6", @@ -135,12 +139,13 @@ "MetricGroup": "BadSpec;BrMispredicts;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueBM", "MetricName": "tma_branch_mispredicts", "MetricThreshold": "tma_branch_mispredicts > 0.1 & tma_bad_speculation > 0.15", - "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_branch_misprediction_cost, tma_info_mispredictions, tma_mispredicts_resteers", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Branch Misprediction. These slots are either wasted by uops fetched from an incorrectly speculated program path; or stalls when the out-of-order part of the machine needs to recover its state from a speculative path. Sample with: BR_MISP_RETIRED.ALL_BRANCHES. Related metrics: tma_info_bad_spec_branch_misprediction_cost, tma_info_bottleneck_mispredictions, tma_mispredicts_resteers", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers", - "MetricExpr": "INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks + tma_unknown_branches", + "MetricExpr": "INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks + tma_unknown_branches", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_branch_resteers", "MetricThreshold": "tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -158,7 +163,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Machine Clears", - "MetricExpr": "(1 - BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT)) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks", + "MetricExpr": "(1 - BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT)) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks", "MetricGroup": "BadSpec;MachineClears;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueMC", "MetricName": "tma_clears_resteers", "MetricThreshold": "tma_clears_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", @@ -168,7 +173,7 @@ { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to contested accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(49 * tma_info_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_FWD * (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM / (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM + OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) + 48 * tma_info_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "(49 * tma_info_system_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_FWD * (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM / (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM + OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) + 48 * tma_info_system_average_frequency * MEM_LOAD_L3_HIT_RETIRED.XSNP_MISS) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_contested_accesses", "MetricThreshold": "tma_contested_accesses > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -181,13 +186,14 @@ "MetricGroup": "Backend;Compute;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_core_bound", "MetricThreshold": "tma_core_bound > 0.1 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where Core non-memory issues were of a bottleneck. Shortage in hardware compute resources; or dependencies in software's instructions are both categorized under Core Bound. Hence it may indicate the machine ran out of an out-of-order resource; certain execution units are overloaded or dependencies in program's data- or instruction-flow are limiting the performance (e.g. FP-chained long-latency arithmetic operations).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles while the memory subsystem was handling synchronizations due to data-sharing accesses", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "48 * tma_info_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_NO_FWD + MEM_LOAD_L3_HIT_RETIRED.XSNP_FWD * (1 - OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM / (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM + OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "48 * tma_info_system_average_frequency * (MEM_LOAD_L3_HIT_RETIRED.XSNP_NO_FWD + MEM_LOAD_L3_HIT_RETIRED.XSNP_FWD * (1 - OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM / (OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HITM + OCR.DEMAND_DATA_RD.L3_HIT.SNOOP_HIT_WITH_FWD))) * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_l3_bound_group", "MetricName": "tma_data_sharing", "MetricThreshold": "tma_data_sharing > 0.05 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -196,16 +202,16 @@ }, { "BriefDescription": "This metric represents fraction of cycles where decoder-0 was the only active decoder", - "MetricExpr": "(cpu@INST_DECODED.DECODERS\\,cmask\\=1@ - cpu@INST_DECODED.DECODERS\\,cmask\\=2@) / tma_info_core_clks / 2", + "MetricExpr": "(cpu@INST_DECODED.DECODERS\\,cmask\\=1@ - cpu@INST_DECODED.DECODERS\\,cmask\\=2@) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL4;tma_L4_group;tma_issueD0;tma_mite_group", "MetricName": "tma_decoder0_alone", - "MetricThreshold": "tma_decoder0_alone > 0.1 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 5 > 0.35))", + "MetricThreshold": "tma_decoder0_alone > 0.1 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 5 > 0.35))", "PublicDescription": "This metric represents fraction of cycles where decoder-0 was the only active decoder. Related metrics: tma_few_uops_instructions", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles where the Divider unit was active", - "MetricExpr": "ARITH.DIVIDER_ACTIVE / tma_info_clks", + "MetricExpr": "ARITH.DIVIDER_ACTIVE / tma_info_thread_clks", "MetricGroup": "TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_divider", "MetricThreshold": "tma_divider > 0.2 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -215,7 +221,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled on accesses to external memory (DRAM) by loads", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks - tma_l2_bound", + "MetricExpr": "CYCLE_ACTIVITY.STALLS_L3_MISS / tma_info_thread_clks + (CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks - tma_l2_bound", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_dram_bound", "MetricThreshold": "tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -224,43 +230,43 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline", - "MetricExpr": "(IDQ.DSB_CYCLES_ANY - IDQ.DSB_CYCLES_OK) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.DSB_CYCLES_ANY - IDQ.DSB_CYCLES_OK) / tma_info_core_core_clks / 2", "MetricGroup": "DSB;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_dsb", - "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 5 > 0.35)", + "MetricThreshold": "tma_dsb > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 5 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to DSB (decoded uop cache) fetch pipeline. For example; inefficient utilization of the DSB cache structure or bank conflict when reading from it; are categorized here.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines", - "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_clks", + "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_dsb_switches", "MetricThreshold": "tma_dsb_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb, tma_lcp", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to switches from DSB to MITE pipelines. The DSB (decoded i-cache) is a Uop Cache where the front-end directly delivers Uops (micro operations) avoiding heavy x86 decoding. The DSB pipeline has shorter latency and delivered higher bandwidth than the MITE (legacy instruction decode pipeline). Switching between the two pipelines can cause penalties hence this metric measures the exposed penalty. Sample with: FRONTEND_RETIRED.DSB_MISS_PS. Related metrics: tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses", - "MetricExpr": "min(7 * cpu@DTLB_LOAD_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_LOAD_MISSES.WALK_ACTIVE, max(CYCLE_ACTIVITY.CYCLES_MEM_ANY - CYCLE_ACTIVITY.CYCLES_L1D_MISS, 0)) / tma_info_clks", + "MetricExpr": "min(7 * cpu@DTLB_LOAD_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_LOAD_MISSES.WALK_ACTIVE, max(CYCLE_ACTIVITY.CYCLES_MEM_ANY - CYCLE_ACTIVITY.CYCLES_L1D_MISS, 0)) / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_l1_bound_group", "MetricName": "tma_dtlb_load", "MetricThreshold": "tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses. TLBs (Translation Look-aside Buffers) are processor caches for recently used entries out of the Page Tables that are used to map virtual- to physical-addresses by the operating system. This metric approximates the potential delay of demand loads missing the first-level data TLB (assuming worst case scenario with back to back misses to different pages). This includes hitting in the second-level TLB (STLB) as well as performing a hardware page walk on an STLB miss. Sample with: MEM_INST_RETIRED.STLB_MISS_LOADS_PS. Related metrics: tma_dtlb_store, tma_info_memory_data_tlbs", + "PublicDescription": "This metric roughly estimates the fraction of cycles where the Data TLB (DTLB) was missed by load accesses. TLBs (Translation Look-aside Buffers) are processor caches for recently used entries out of the Page Tables that are used to map virtual- to physical-addresses by the operating system. This metric approximates the potential delay of demand loads missing the first-level data TLB (assuming worst case scenario with back to back misses to different pages). This includes hitting in the second-level TLB (STLB) as well as performing a hardware page walk on an STLB miss. Sample with: MEM_INST_RETIRED.STLB_MISS_LOADS_PS. Related metrics: tma_dtlb_store, tma_info_bottleneck_memory_data_tlbs", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses", - "MetricExpr": "(7 * cpu@DTLB_STORE_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_STORE_MISSES.WALK_ACTIVE) / tma_info_core_clks", + "MetricExpr": "(7 * cpu@DTLB_STORE_MISSES.STLB_HIT\\,cmask\\=1@ + DTLB_STORE_MISSES.WALK_ACTIVE) / tma_info_core_core_clks", "MetricGroup": "MemoryTLB;TopdownL4;tma_L4_group;tma_issueTLB;tma_store_bound_group", "MetricName": "tma_dtlb_store", "MetricThreshold": "tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses. As with ordinary data caching; focus on improving data locality and reducing working-set size to reduce DTLB overhead. Additionally; consider using profile-guided optimization (PGO) to collocate frequently-used data on the same page. Try using larger page sizes for large amounts of frequently-used data. Sample with: MEM_INST_RETIRED.STLB_MISS_STORES_PS. Related metrics: tma_dtlb_load, tma_info_memory_data_tlbs", + "PublicDescription": "This metric roughly estimates the fraction of cycles spent handling first-level data TLB store misses. As with ordinary data caching; focus on improving data locality and reducing working-set size to reduce DTLB overhead. Additionally; consider using profile-guided optimization (PGO) to collocate frequently-used data on the same page. Try using larger page sizes for large amounts of frequently-used data. Sample with: MEM_INST_RETIRED.STLB_MISS_STORES_PS. Related metrics: tma_dtlb_load, tma_info_bottleneck_memory_data_tlbs", "ScaleUnit": "100%" }, { "BriefDescription": "This metric roughly estimates how often CPU was handling synchronizations due to False Sharing", - "MetricExpr": "54 * tma_info_average_frequency * OCR.DEMAND_RFO.L3_HIT.SNOOP_HITM / tma_info_clks", + "MetricExpr": "54 * tma_info_system_average_frequency * OCR.DEMAND_RFO.L3_HIT.SNOOP_HITM / tma_info_thread_clks", "MetricGroup": "DataSharing;Offcore;Snoop;TopdownL4;tma_L4_group;tma_issueSyncxn;tma_store_bound_group", "MetricName": "tma_false_sharing", "MetricThreshold": "tma_false_sharing > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -269,11 +275,11 @@ }, { "BriefDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed", - "MetricExpr": "L1D_PEND_MISS.FB_FULL / tma_info_clks", + "MetricExpr": "L1D_PEND_MISS.FB_FULL / tma_info_thread_clks", "MetricGroup": "MemoryBW;TopdownL4;tma_L4_group;tma_issueBW;tma_issueSL;tma_issueSmSt;tma_l1_bound_group", "MetricName": "tma_fb_full", "MetricThreshold": "tma_fb_full > 0.3", - "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", + "PublicDescription": "This metric does a *rough estimation* of how often L1D Fill Buffer unavailability limited additional L1D miss memory access requests to proceed. The higher the metric value; the deeper the memory hierarchy level the misses are satisfied from (metric values >1 are valid). Often it hints on approaching bandwidth limits (to L2 cache; L3 cache or external memory). Related metrics: tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full, tma_store_latency, tma_streaming_stores", "ScaleUnit": "100%" }, { @@ -281,16 +287,18 @@ "MetricExpr": "max(0, tma_frontend_bound - tma_fetch_latency)", "MetricGroup": "FetchBW;Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group;tma_issueFB", "MetricName": "tma_fetch_bandwidth", - "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 5 > 0.35", - "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb, tma_lcp", + "MetricThreshold": "tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 5 > 0.35", + "MetricgroupNoGroup": "TopdownL2", + "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend bandwidth issues. For example; inefficiencies at the instruction decoders; or restrictions for caching in the DSB (decoded uops cache) are categorized under Fetch Bandwidth. In such cases; the Frontend typically delivers suboptimal amount of uops to the Backend. Sample with: FRONTEND_RETIRED.LATENCY_GE_2_BUBBLES_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_1_PS;FRONTEND_RETIRED.LATENCY_GE_2_PS. Related metrics: tma_dsb_switches, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues", - "MetricExpr": "(5 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE - INT_MISC.UOP_DROPPING) / tma_info_slots", + "MetricExpr": "(5 * IDQ_UOPS_NOT_DELIVERED.CYCLES_0_UOPS_DELIV.CORE - INT_MISC.UOP_DROPPING) / tma_info_thread_slots", "MetricGroup": "Frontend;TmaL2;TopdownL2;tma_L2_group;tma_frontend_bound_group", "MetricName": "tma_fetch_latency", "MetricThreshold": "tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU was stalled due to Frontend latency issues. For example; instruction-cache misses; iTLB misses or fetch stalls after a branch misprediction are categorized under Frontend Latency. In such cases; the Frontend eventually delivers no uops for some period. Sample with: FRONTEND_RETIRED.LATENCY_GE_16_PS;FRONTEND_RETIRED.LATENCY_GE_8_PS", "ScaleUnit": "100%" }, @@ -305,6 +313,7 @@ }, { "BriefDescription": "This metric represents overall arithmetic floating-point (FP) operations fraction the CPU has executed (retired)", + "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "tma_x87_use + tma_fp_scalar + tma_fp_vector", "MetricGroup": "HPC;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_fp_arith", @@ -314,7 +323,7 @@ }, { "BriefDescription": "This metric approximates arithmetic floating-point (FP) scalar uops fraction the CPU has retired", - "MetricExpr": "cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ / (tma_retiring * tma_info_slots)", + "MetricExpr": "cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL4;tma_L4_group;tma_fp_arith_group;tma_issue2P", "MetricName": "tma_fp_scalar", "MetricThreshold": "tma_fp_scalar > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6)", @@ -323,7 +332,7 @@ }, { "BriefDescription": "This metric approximates arithmetic floating-point (FP) vector uops fraction the CPU has retired aggregated across all vector widths", - "MetricExpr": "cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@ / (tma_retiring * tma_info_slots)", + "MetricExpr": "cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@ / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL4;tma_L4_group;tma_fp_arith_group;tma_issue2P", "MetricName": "tma_fp_vector", "MetricThreshold": "tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6)", @@ -332,7 +341,7 @@ }, { "BriefDescription": "This metric approximates arithmetic FP vector uops fraction the CPU has retired for 128-bit wide vectors", - "MetricExpr": "(FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL5;tma_L5_group;tma_fp_vector_group;tma_issue2P", "MetricName": "tma_fp_vector_128b", "MetricThreshold": "tma_fp_vector_128b > 0.1 & (tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6))", @@ -341,7 +350,7 @@ }, { "BriefDescription": "This metric approximates arithmetic FP vector uops fraction the CPU has retired for 256-bit wide vectors", - "MetricExpr": "(FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL5;tma_L5_group;tma_fp_vector_group;tma_issue2P", "MetricName": "tma_fp_vector_256b", "MetricThreshold": "tma_fp_vector_256b > 0.1 & (tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6))", @@ -350,7 +359,7 @@ }, { "BriefDescription": "This metric approximates arithmetic FP vector uops fraction the CPU has retired for 512-bit wide vectors", - "MetricExpr": "(FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / (tma_retiring * tma_info_slots)", + "MetricExpr": "(FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Compute;Flops;TopdownL5;tma_L5_group;tma_fp_vector_group;tma_issue2P", "MetricName": "tma_fp_vector_512b", "MetricThreshold": "tma_fp_vector_512b > 0.1 & (tma_fp_vector > 0.1 & (tma_fp_arith > 0.2 & tma_light_operations > 0.6))", @@ -359,10 +368,12 @@ }, { "BriefDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend", - "MetricExpr": "topdown\\-fe\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) - INT_MISC.UOP_DROPPING / tma_info_slots", - "MetricGroup": "PGO;TmaL1;TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "topdown\\-fe\\-bound / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) - INT_MISC.UOP_DROPPING / tma_info_thread_slots", + "MetricGroup": "Default;PGO;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_frontend_bound", "MetricThreshold": "tma_frontend_bound > 0.15", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots where the processor's Frontend undersupplies its Backend. Frontend denotes the first part of the processor core responsible to fetch operations that are executed later on by the Backend part. Within the Frontend; a branch predictor predicts the next address to fetch; cache-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into micro-operations (uops). Ideally the Frontend can issue Pipeline_Width uops every cycle to the Backend. Frontend Bound denotes unutilized issue-slots when there is no Backend stall; i.e. bubbles where Frontend delivered no uops while Backend could have accepted them. For example; stalls due to instruction-cache misses would be categorized under Frontend Bound. Sample with: FRONTEND_RETIRED.LATENCY_GE_4_PS", "ScaleUnit": "100%" }, @@ -372,12 +383,13 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_heavy_operations", "MetricThreshold": "tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring heavy-weight operations -- instructions that require two or more uops or micro-coded sequences. This highly-correlates with the uop length of these instructions/sequences.", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to instruction cache misses", - "MetricExpr": "ICACHE_16B.IFDATA_STALL / tma_info_clks", + "MetricExpr": "ICACHE_16B.IFDATA_STALL / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;IcMiss;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_icache_misses", "MetricThreshold": "tma_icache_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -385,688 +397,696 @@ "ScaleUnit": "100%" }, { - "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", - "MetricExpr": "tma_info_turbo_utilization * TSC / 1e9 / duration_time", - "MetricGroup": "Power;Summary", - "MetricName": "tma_info_average_frequency" + "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_thread_slots / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BrMispredicts;tma_issueBM", + "MetricName": "tma_info_bad_spec_branch_misprediction_cost", + "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_info_bottleneck_mispredictions, tma_mispredicts_resteers" + }, + { + "BriefDescription": "Instructions per retired mispredicts for conditional non-taken branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_NTAKEN", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_cond_ntaken", + "MetricThreshold": "tma_info_bad_spec_ipmisp_cond_ntaken < 200" + }, + { + "BriefDescription": "Instructions per retired mispredicts for conditional taken branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_TAKEN", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_cond_taken", + "MetricThreshold": "tma_info_bad_spec_ipmisp_cond_taken < 200" + }, + { + "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.INDIRECT", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_indirect", + "MetricThreshold": "tma_info_bad_spec_ipmisp_indirect < 1e3" + }, + { + "BriefDescription": "Instructions per retired mispredicts for return branches (lower number means higher occurrence rate).", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.RET", + "MetricGroup": "Bad;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmisp_ret", + "MetricThreshold": "tma_info_bad_spec_ipmisp_ret < 500" + }, + { + "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;BadSpec;BrMispredicts", + "MetricName": "tma_info_bad_spec_ipmispredict", + "MetricThreshold": "tma_info_bad_spec_ipmispredict < 200" + }, + { + "BriefDescription": "Probability of Core Bound bottleneck hidden by SMT-profiling artifacts", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(100 * (1 - tma_core_bound / tma_ports_utilization if tma_core_bound < tma_ports_utilization else 1) if tma_info_system_smt_2t_utilization > 0.5 else 0)", + "MetricGroup": "Cor;SMT", + "MetricName": "tma_info_botlnk_l0_core_bound_likely", + "MetricThreshold": "tma_info_botlnk_l0_core_bound_likely > 0.5" + }, + { + "BriefDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_fetch_latency * tma_dsb_switches / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches) + tma_fetch_bandwidth * tma_mite / (tma_dsb + tma_lsd + tma_mite))", + "MetricGroup": "DSBmiss;Fed;tma_issueFB", + "MetricName": "tma_info_botlnk_l2_dsb_misses", + "MetricThreshold": "tma_info_botlnk_l2_dsb_misses > 10", + "PublicDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb, tma_lcp" + }, + { + "BriefDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck", + "MetricExpr": "100 * (tma_fetch_latency * tma_icache_misses / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", + "MetricGroup": "Fed;FetchLat;IcMiss;tma_issueFL", + "MetricName": "tma_info_botlnk_l2_ic_misses", + "MetricThreshold": "tma_info_botlnk_l2_ic_misses > 5", + "PublicDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck. Related metrics: " }, { "BriefDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses)", "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "100 * tma_fetch_latency * (tma_itlb_misses + tma_icache_misses + tma_unknown_branches) / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)", "MetricGroup": "BigFoot;Fed;Frontend;IcMiss;MemoryTLB;tma_issueBC", - "MetricName": "tma_info_big_code", - "MetricThreshold": "tma_info_big_code > 20", - "PublicDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses). Related metrics: tma_info_branching_overhead" + "MetricName": "tma_info_bottleneck_big_code", + "MetricThreshold": "tma_info_bottleneck_big_code > 20", + "PublicDescription": "Total pipeline cost of instruction fetch related bottlenecks by large code footprint programs (i-side cache; TLB and BTB misses). Related metrics: tma_info_bottleneck_branching_overhead" }, { - "BriefDescription": "Branch instructions per taken branch.", - "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", - "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_bptkbranch" + "BriefDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls)", + "MetricExpr": "100 * ((BR_INST_RETIRED.COND + 3 * BR_INST_RETIRED.NEAR_CALL + (BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL)) / tma_info_thread_slots)", + "MetricGroup": "Ret;tma_issueBC", + "MetricName": "tma_info_bottleneck_branching_overhead", + "MetricThreshold": "tma_info_bottleneck_branching_overhead > 10", + "PublicDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls). Related metrics: tma_info_bottleneck_big_code" }, { - "BriefDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear)", - "MetricExpr": "(tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) * tma_info_slots / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_branch_misprediction_cost", - "PublicDescription": "Branch Misprediction Cost: Fraction of TMA slots wasted per non-speculative branch misprediction (retired JEClear). Related metrics: tma_branch_mispredicts, tma_info_mispredictions, tma_mispredicts_resteers" + "BriefDescription": "Total pipeline cost of instruction fetch bandwidth related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_frontend_bound - tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) - tma_info_bottleneck_big_code", + "MetricGroup": "Fed;FetchBW;Frontend", + "MetricName": "tma_info_bottleneck_instruction_fetch_bw", + "MetricThreshold": "tma_info_bottleneck_instruction_fetch_bw > 20" }, { - "BriefDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls)", - "MetricExpr": "100 * ((BR_INST_RETIRED.COND + 3 * BR_INST_RETIRED.NEAR_CALL + (BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL)) / tma_info_slots)", - "MetricGroup": "Ret;tma_issueBC", - "MetricName": "tma_info_branching_overhead", - "MetricThreshold": "tma_info_branching_overhead > 10", - "PublicDescription": "Total pipeline cost of branch related instructions (used for program control-flow including function calls). Related metrics: tma_info_big_code" + "BriefDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks", + "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_bandwidth / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_sq_full / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full))) + tma_l1_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_fb_full / (tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk))", + "MetricGroup": "Mem;MemoryBW;Offcore;tma_issueBW", + "MetricName": "tma_info_bottleneck_memory_bandwidth", + "MetricThreshold": "tma_info_bottleneck_memory_bandwidth > 20", + "PublicDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks. Related metrics: tma_fb_full, tma_info_system_dram_bw_use, tma_mem_bandwidth, tma_sq_full" }, { - "BriefDescription": "Fraction of branches that are CALL or RET", - "MetricExpr": "(BR_INST_RETIRED.NEAR_CALL + BR_INST_RETIRED.NEAR_RETURN) / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_callret" + "BriefDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_l1_bound / max(tma_memory_bound, tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_load / max(tma_l1_bound, tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk)) + tma_store_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_store / (tma_dtlb_store + tma_false_sharing + tma_split_stores + tma_store_latency + tma_streaming_stores)))", + "MetricGroup": "Mem;MemoryTLB;Offcore;tma_issueTLB", + "MetricName": "tma_info_bottleneck_memory_data_tlbs", + "MetricThreshold": "tma_info_bottleneck_memory_data_tlbs > 20", + "PublicDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs). Related metrics: tma_dtlb_load, tma_dtlb_store" }, { - "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "Pipeline", - "MetricName": "tma_info_clks" + "BriefDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_latency / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_l3_hit_latency / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full)) + tma_l2_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound))", + "MetricGroup": "Mem;MemoryLat;Offcore;tma_issueLat", + "MetricName": "tma_info_bottleneck_memory_latency", + "MetricThreshold": "tma_info_bottleneck_memory_latency > 20", + "PublicDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches). Related metrics: tma_l3_hit_latency, tma_mem_latency" }, { - "BriefDescription": "STLB (2nd level TLB) code speculative misses per kilo instruction (misses of any page-size that complete the page walk)", - "MetricExpr": "1e3 * ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", - "MetricGroup": "Fed;MemoryTLB", - "MetricName": "tma_info_code_stlb_mpki" + "BriefDescription": "Total pipeline cost of Branch Misprediction related bottlenecks", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "100 * (tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", + "MetricGroup": "Bad;BadSpec;BrMispredicts;tma_issueBM", + "MetricName": "tma_info_bottleneck_mispredictions", + "MetricThreshold": "tma_info_bottleneck_mispredictions > 20", + "PublicDescription": "Total pipeline cost of Branch Misprediction related bottlenecks. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost, tma_mispredicts_resteers" + }, + { + "BriefDescription": "Fraction of branches that are CALL or RET", + "MetricExpr": "(BR_INST_RETIRED.NEAR_CALL + BR_INST_RETIRED.NEAR_RETURN) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_callret" }, { "BriefDescription": "Fraction of branches that are non-taken conditionals", "MetricExpr": "BR_INST_RETIRED.COND_NTAKEN / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Bad;Branches;CodeGen;PGO", - "MetricName": "tma_info_cond_nt" + "MetricName": "tma_info_branches_cond_nt" }, { "BriefDescription": "Fraction of branches that are taken conditionals", "MetricExpr": "BR_INST_RETIRED.COND_TAKEN / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Bad;Branches;CodeGen;PGO", - "MetricName": "tma_info_cond_tk" + "MetricName": "tma_info_branches_cond_tk" }, { - "BriefDescription": "Probability of Core Bound bottleneck hidden by SMT-profiling artifacts", - "MetricExpr": "(100 * (1 - tma_core_bound / tma_ports_utilization if tma_core_bound < tma_ports_utilization else 1) if tma_info_smt_2t_utilization > 0.5 else 0)", - "MetricGroup": "Cor;SMT", - "MetricName": "tma_info_core_bound_likely", - "MetricThreshold": "tma_info_core_bound_likely > 0.5" + "BriefDescription": "Fraction of branches that are unconditional (direct or indirect) jumps", + "MetricExpr": "(BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL) / BR_INST_RETIRED.ALL_BRANCHES", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_jump" + }, + { + "BriefDescription": "Fraction of branches of other types (not individually covered by other metrics in Info.Branches group)", + "MetricExpr": "1 - (tma_info_branches_cond_nt + tma_info_branches_cond_tk + tma_info_branches_callret + tma_info_branches_jump)", + "MetricGroup": "Bad;Branches", + "MetricName": "tma_info_branches_other_branches" }, { "BriefDescription": "Core actual clocks when any Logical Processor is active on the Physical Core", "MetricExpr": "CPU_CLK_UNHALTED.DISTRIBUTED", "MetricGroup": "SMT", - "MetricName": "tma_info_core_clks" + "MetricName": "tma_info_core_core_clks" }, { "BriefDescription": "Instructions Per Cycle across hyper-threads (per physical core)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_core_clks", + "MetricExpr": "INST_RETIRED.ANY / tma_info_core_core_clks", "MetricGroup": "Ret;SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_coreipc" + "MetricName": "tma_info_core_coreipc" }, { - "BriefDescription": "Cycles Per Instruction (per Logical Processor)", - "MetricExpr": "1 / tma_info_ipc", - "MetricGroup": "Mem;Pipeline", - "MetricName": "tma_info_cpi" - }, - { - "BriefDescription": "Average CPU Utilization", - "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", - "MetricGroup": "HPC;Summary", - "MetricName": "tma_info_cpu_utilization" + "BriefDescription": "Floating Point Operations Per Cycle", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * cpu@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE\\,umask\\=0x18@ + 8 * cpu@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE\\,umask\\=0x60@ + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / tma_info_core_core_clks", + "MetricGroup": "Flops;Ret", + "MetricName": "tma_info_core_flopc" }, { - "BriefDescription": "Average Parallel L2 cache miss data reads", - "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", - "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_data_l2_mlp" + "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@) / (2 * tma_info_core_core_clks)", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_core_fp_arith_utilization", + "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." }, { - "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", - "MetricExpr": "64 * (arb@event\\=0x81\\,umask\\=0x1@ + arb@event\\=0x84\\,umask\\=0x1@) / 1e6 / duration_time / 1e3", - "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", - "MetricName": "tma_info_dram_bw_use", - "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_info_memory_bandwidth, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", + "MetricExpr": "UOPS_EXECUTED.THREAD / (UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1)", + "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", + "MetricName": "tma_info_core_ilp" }, { "BriefDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache)", "MetricExpr": "IDQ.DSB_UOPS / UOPS_ISSUED.ANY", "MetricGroup": "DSB;Fed;FetchBW;tma_issueFB", - "MetricName": "tma_info_dsb_coverage", - "MetricThreshold": "tma_info_dsb_coverage < 0.7 & tma_info_ipc / 5 > 0.35", - "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_misses, tma_info_iptb, tma_lcp" - }, - { - "BriefDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck", - "MetricExpr": "100 * (tma_fetch_latency * tma_dsb_switches / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches) + tma_fetch_bandwidth * tma_mite / (tma_dsb + tma_lsd + tma_mite))", - "MetricGroup": "DSBmiss;Fed;tma_issueFB", - "MetricName": "tma_info_dsb_misses", - "MetricThreshold": "tma_info_dsb_misses > 10", - "PublicDescription": "Total pipeline cost of DSB (uop cache) misses - subset of the Instruction_Fetch_BW Bottleneck. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_iptb, tma_lcp" + "MetricName": "tma_info_frontend_dsb_coverage", + "MetricThreshold": "tma_info_frontend_dsb_coverage < 0.7 & tma_info_thread_ipc / 5 > 0.35", + "PublicDescription": "Fraction of Uops delivered by the DSB (aka Decoded ICache; or Uop Cache). Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_inst_mix_iptb, tma_lcp" }, { "BriefDescription": "Average number of cycles of a switch from the DSB fetch-unit to MITE fetch unit - see DSB_Switches tree node for details.", "MetricExpr": "DSB2MITE_SWITCHES.PENALTY_CYCLES / cpu@DSB2MITE_SWITCHES.PENALTY_CYCLES\\,cmask\\=1\\,edge@", "MetricGroup": "DSBmiss", - "MetricName": "tma_info_dsb_switch_cost" - }, - { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", - "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", - "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", - "MetricName": "tma_info_execute" - }, - { - "BriefDescription": "The ratio of Executed- by Issued-Uops", - "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", - "MetricGroup": "Cor;Pipeline", - "MetricName": "tma_info_execute_per_issue", - "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." - }, - { - "BriefDescription": "Fill Buffer (FB) hits per kilo instructions for retired demand loads (L1D misses that merge into ongoing miss-handling entries)", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.FB_HIT / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_fb_hpki" + "MetricName": "tma_info_frontend_dsb_switch_cost" }, { "BriefDescription": "Average number of Uops issued by front-end when it issued something", "MetricExpr": "UOPS_ISSUED.ANY / cpu@UOPS_ISSUED.ANY\\,cmask\\=1@", "MetricGroup": "Fed;FetchBW", - "MetricName": "tma_info_fetch_upc" + "MetricName": "tma_info_frontend_fetch_upc" }, { - "BriefDescription": "Floating Point Operations Per Cycle", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * cpu@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE\\,umask\\=0x18@ + 8 * cpu@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE\\,umask\\=0x60@ + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / tma_info_core_clks", - "MetricGroup": "Flops;Ret", - "MetricName": "tma_info_flopc" + "BriefDescription": "Average Latency for L1 instruction cache misses", + "MetricExpr": "ICACHE_16B.IFDATA_STALL / cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@", + "MetricGroup": "Fed;FetchLat;IcMiss", + "MetricName": "tma_info_frontend_icache_miss_latency" }, { - "BriefDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width)", - "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@) / (2 * tma_info_core_clks)", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_fp_arith_utilization", - "PublicDescription": "Actual per-core usage of the Floating Point non-X87 execution units (regardless of precision or vector-width). Values > 1 are possible due to ([BDW+] Fused-Multiply Add (FMA) counting - common; [ADL+] use all of ADD/MUL/FMA in Scalar or 128/256-bit vectors - less common)." + "BriefDescription": "Instructions per non-speculative DSB miss (lower number means higher occurrence rate)", + "MetricExpr": "INST_RETIRED.ANY / FRONTEND_RETIRED.ANY_DSB_MISS", + "MetricGroup": "DSBmiss;Fed", + "MetricName": "tma_info_frontend_ipdsb_miss_ret", + "MetricThreshold": "tma_info_frontend_ipdsb_miss_ret < 50" }, { - "BriefDescription": "Giga Floating Point Operations Per Second", - "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * cpu@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE\\,umask\\=0x18@ + 8 * cpu@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE\\,umask\\=0x60@ + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / 1e9 / duration_time", - "MetricGroup": "Cor;Flops;HPC", - "MetricName": "tma_info_gflops", - "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." + "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", + "MetricExpr": "tma_info_inst_mix_instructions / BACLEARS.ANY", + "MetricGroup": "Fed", + "MetricName": "tma_info_frontend_ipunknown_branch" }, { - "BriefDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck", - "MetricExpr": "100 * (tma_fetch_latency * tma_icache_misses / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", - "MetricGroup": "Fed;FetchLat;IcMiss;tma_issueFL", - "MetricName": "tma_info_ic_misses", - "MetricThreshold": "tma_info_ic_misses > 5", - "PublicDescription": "Total pipeline cost of Instruction Cache misses - subset of the Big_Code Bottleneck. Related metrics: " + "BriefDescription": "L2 cache true code cacheline misses per kilo instruction", + "MetricExpr": "1e3 * FRONTEND_RETIRED.L2_MISS / INST_RETIRED.ANY", + "MetricGroup": "IcMiss", + "MetricName": "tma_info_frontend_l2mpki_code" }, { - "BriefDescription": "Average Latency for L1 instruction cache misses", - "MetricExpr": "ICACHE_16B.IFDATA_STALL / cpu@ICACHE_16B.IFDATA_STALL\\,cmask\\=1\\,edge@", - "MetricGroup": "Fed;FetchLat;IcMiss", - "MetricName": "tma_info_icache_miss_latency" + "BriefDescription": "L2 cache speculative code cacheline misses per kilo instruction", + "MetricExpr": "1e3 * L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", + "MetricGroup": "IcMiss", + "MetricName": "tma_info_frontend_l2mpki_code_all" }, { - "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-core", - "MetricExpr": "UOPS_EXECUTED.THREAD / (UOPS_EXECUTED.CORE_CYCLES_GE_1 / 2 if #SMT_on else UOPS_EXECUTED.CORE_CYCLES_GE_1)", - "MetricGroup": "Backend;Cor;Pipeline;PortsUtil", - "MetricName": "tma_info_ilp" + "BriefDescription": "Fraction of Uops delivered by the LSD (Loop Stream Detector; aka Loop Cache)", + "MetricExpr": "LSD.UOPS / UOPS_ISSUED.ANY", + "MetricGroup": "Fed;LSD", + "MetricName": "tma_info_frontend_lsd_coverage" }, { - "BriefDescription": "Total pipeline cost of instruction fetch bandwidth related bottlenecks", - "MetricExpr": "100 * (tma_frontend_bound - tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches)) - tma_info_big_code", - "MetricGroup": "Fed;FetchBW;Frontend", - "MetricName": "tma_info_instruction_fetch_bw", - "MetricThreshold": "tma_info_instruction_fetch_bw > 20" + "BriefDescription": "Branch instructions per taken branch.", + "MetricExpr": "BR_INST_RETIRED.ALL_BRANCHES / BR_INST_RETIRED.NEAR_TAKEN", + "MetricGroup": "Branches;Fed;PGO", + "MetricName": "tma_info_inst_mix_bptkbranch" }, { "BriefDescription": "Total number of retired Instructions", "MetricExpr": "INST_RETIRED.ANY", "MetricGroup": "Summary;TmaL1;tma_L1_group", - "MetricName": "tma_info_instructions", + "MetricName": "tma_info_inst_mix_instructions", "PublicDescription": "Total number of retired Instructions. Sample with: INST_RETIRED.PREC_DIST" }, { "BriefDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + cpu@FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE\\,umask\\=0xfc@)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_iparith", - "MetricThreshold": "tma_info_iparith < 10", + "MetricName": "tma_info_inst_mix_iparith", + "MetricThreshold": "tma_info_inst_mix_iparith < 10", "PublicDescription": "Instructions per FP Arithmetic instruction (lower number means higher occurrence rate). May undercount due to FMA double counting. Approximated prior to BDW." }, { "BriefDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx128", - "MetricThreshold": "tma_info_iparith_avx128 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx128", + "MetricThreshold": "tma_info_inst_mix_iparith_avx128 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX/SSE 128-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.256B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx256", - "MetricThreshold": "tma_info_iparith_avx256 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx256", + "MetricThreshold": "tma_info_inst_mix_iparith_avx256 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX* 256-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic AVX 512-bit instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (FP_ARITH_INST_RETIRED.512B_PACKED_DOUBLE + FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE)", "MetricGroup": "Flops;FpVector;InsType", - "MetricName": "tma_info_iparith_avx512", - "MetricThreshold": "tma_info_iparith_avx512 < 10", + "MetricName": "tma_info_inst_mix_iparith_avx512", + "MetricThreshold": "tma_info_inst_mix_iparith_avx512 < 10", "PublicDescription": "Instructions per FP Arithmetic AVX 512-bit instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_DOUBLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_dp", - "MetricThreshold": "tma_info_iparith_scalar_dp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_dp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_dp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Double-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / FP_ARITH_INST_RETIRED.SCALAR_SINGLE", "MetricGroup": "Flops;FpScalar;InsType", - "MetricName": "tma_info_iparith_scalar_sp", - "MetricThreshold": "tma_info_iparith_scalar_sp < 10", + "MetricName": "tma_info_inst_mix_iparith_scalar_sp", + "MetricThreshold": "tma_info_inst_mix_iparith_scalar_sp < 10", "PublicDescription": "Instructions per FP Arithmetic Scalar Single-Precision instruction (lower number means higher occurrence rate). May undercount due to FMA double counting." }, { "BriefDescription": "Instructions per Branch (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.ALL_BRANCHES", "MetricGroup": "Branches;Fed;InsType", - "MetricName": "tma_info_ipbranch", - "MetricThreshold": "tma_info_ipbranch < 8" - }, - { - "BriefDescription": "Instructions Per Cycle (per Logical Processor)", - "MetricExpr": "INST_RETIRED.ANY / tma_info_clks", - "MetricGroup": "Ret;Summary", - "MetricName": "tma_info_ipc" + "MetricName": "tma_info_inst_mix_ipbranch", + "MetricThreshold": "tma_info_inst_mix_ipbranch < 8" }, { "BriefDescription": "Instructions per (near) call (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_CALL", "MetricGroup": "Branches;Fed;PGO", - "MetricName": "tma_info_ipcall", - "MetricThreshold": "tma_info_ipcall < 200" - }, - { - "BriefDescription": "Instructions per non-speculative DSB miss (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / FRONTEND_RETIRED.ANY_DSB_MISS", - "MetricGroup": "DSBmiss;Fed", - "MetricName": "tma_info_ipdsb_miss_ret", - "MetricThreshold": "tma_info_ipdsb_miss_ret < 50" - }, - { - "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", - "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", - "MetricGroup": "Branches;OS", - "MetricName": "tma_info_ipfarbranch", - "MetricThreshold": "tma_info_ipfarbranch < 1e6" + "MetricName": "tma_info_inst_mix_ipcall", + "MetricThreshold": "tma_info_inst_mix_ipcall < 200" }, { "BriefDescription": "Instructions per Floating Point (FP) Operation (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / (cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * cpu@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE\\,umask\\=0x18@ + 8 * cpu@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE\\,umask\\=0x60@ + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE)", "MetricGroup": "Flops;InsType", - "MetricName": "tma_info_ipflop", - "MetricThreshold": "tma_info_ipflop < 10" + "MetricName": "tma_info_inst_mix_ipflop", + "MetricThreshold": "tma_info_inst_mix_ipflop < 10" }, { "BriefDescription": "Instructions per Load (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_LOADS", "MetricGroup": "InsType", - "MetricName": "tma_info_ipload", - "MetricThreshold": "tma_info_ipload < 3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for conditional non-taken branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_NTAKEN", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_cond_ntaken", - "MetricThreshold": "tma_info_ipmisp_cond_ntaken < 200" - }, - { - "BriefDescription": "Instructions per retired mispredicts for conditional taken branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.COND_TAKEN", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_cond_taken", - "MetricThreshold": "tma_info_ipmisp_cond_taken < 200" - }, - { - "BriefDescription": "Instructions per retired mispredicts for indirect CALL or JMP branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.INDIRECT", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_indirect", - "MetricThreshold": "tma_info_ipmisp_indirect < 1e3" - }, - { - "BriefDescription": "Instructions per retired mispredicts for return branches (lower number means higher occurrence rate).", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.RET", - "MetricGroup": "Bad;BrMispredicts", - "MetricName": "tma_info_ipmisp_ret", - "MetricThreshold": "tma_info_ipmisp_ret < 500" - }, - { - "BriefDescription": "Number of Instructions per non-speculative Branch Misprediction (JEClear) (lower number means higher occurrence rate)", - "MetricExpr": "INST_RETIRED.ANY / BR_MISP_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;BadSpec;BrMispredicts", - "MetricName": "tma_info_ipmispredict", - "MetricThreshold": "tma_info_ipmispredict < 200" + "MetricName": "tma_info_inst_mix_ipload", + "MetricThreshold": "tma_info_inst_mix_ipload < 3" }, { "BriefDescription": "Instructions per Store (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / MEM_INST_RETIRED.ALL_STORES", "MetricGroup": "InsType", - "MetricName": "tma_info_ipstore", - "MetricThreshold": "tma_info_ipstore < 8" + "MetricName": "tma_info_inst_mix_ipstore", + "MetricThreshold": "tma_info_inst_mix_ipstore < 8" }, { "BriefDescription": "Instructions per Software prefetch instruction (of any type: NTA/T0/T1/T2/Prefetch) (lower number means higher occurrence rate)", "MetricExpr": "INST_RETIRED.ANY / cpu@SW_PREFETCH_ACCESS.T0\\,umask\\=0xF@", "MetricGroup": "Prefetches", - "MetricName": "tma_info_ipswpf", - "MetricThreshold": "tma_info_ipswpf < 100" + "MetricName": "tma_info_inst_mix_ipswpf", + "MetricThreshold": "tma_info_inst_mix_ipswpf < 100" }, { "BriefDescription": "Instruction per taken branch", "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW;Frontend;PGO;tma_issueFB", - "MetricName": "tma_info_iptb", - "MetricThreshold": "tma_info_iptb < 11", - "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_lcp" + "MetricName": "tma_info_inst_mix_iptb", + "MetricThreshold": "tma_info_inst_mix_iptb < 11", + "PublicDescription": "Instruction per taken branch. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_lcp" }, { - "BriefDescription": "Instructions per speculative Unknown Branch Misprediction (BAClear) (lower number means higher occurrence rate)", - "MetricExpr": "tma_info_instructions / BACLEARS.ANY", - "MetricGroup": "Fed", - "MetricName": "tma_info_ipunknown_branch" - }, - { - "BriefDescription": "Fraction of branches that are unconditional (direct or indirect) jumps", - "MetricExpr": "(BR_INST_RETIRED.NEAR_TAKEN - BR_INST_RETIRED.COND_TAKEN - 2 * BR_INST_RETIRED.NEAR_CALL) / BR_INST_RETIRED.ALL_BRANCHES", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_jump" + "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_core_l1d_cache_fill_bw" }, { - "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_cpi" + "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_core_l2_cache_fill_bw" }, { - "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", - "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", - "MetricGroup": "OS", - "MetricName": "tma_info_kernel_utilization", - "MetricThreshold": "tma_info_kernel_utilization > 0.05" + "BriefDescription": "Average per-core data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * OFFCORE_REQUESTS.ALL_REQUESTS / 1e9 / duration_time", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_core_l3_cache_access_bw" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "64 * L1D.REPLACEMENT / 1e9 / duration_time", + "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw" + "MetricName": "tma_info_memory_core_l3_cache_fill_bw" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", - "MetricExpr": "tma_info_l1d_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l1d_cache_fill_bw_1t" + "BriefDescription": "Fill Buffer (FB) hits per kilo instructions for retired demand loads (L1D misses that merge into ongoing miss-handling entries)", + "MetricExpr": "1e3 * MEM_LOAD_RETIRED.FB_HIT / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_fb_hpki" }, { "BriefDescription": "L1 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L1_MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki" + "MetricName": "tma_info_memory_l1mpki" }, { "BriefDescription": "L1 cache true misses per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.ALL_DEMAND_DATA_RD / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l1mpki_load" - }, - { - "BriefDescription": "Average per-core data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "64 * L2_LINES_IN.ALL / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw" - }, - { - "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", - "MetricExpr": "tma_info_l2_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l2_cache_fill_bw_1t" + "MetricName": "tma_info_memory_l1mpki_load" }, { "BriefDescription": "L2 cache hits per kilo instruction for all request types (including speculative)", "MetricExpr": "1e3 * (L2_RQSTS.REFERENCES - L2_RQSTS.MISS) / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_all" + "MetricName": "tma_info_memory_l2hpki_all" }, { "BriefDescription": "L2 cache hits per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_HIT / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2hpki_load" + "MetricName": "tma_info_memory_l2hpki_load" }, { "BriefDescription": "L2 cache true misses per kilo instruction for retired demand loads", "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L2_MISS / INST_RETIRED.ANY", "MetricGroup": "Backend;CacheMisses;Mem", - "MetricName": "tma_info_l2mpki" + "MetricName": "tma_info_memory_l2mpki" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all request types (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem;Offcore", - "MetricName": "tma_info_l2mpki_all" - }, - { - "BriefDescription": "L2 cache true code cacheline misses per kilo instruction", - "MetricExpr": "1e3 * FRONTEND_RETIRED.L2_MISS / INST_RETIRED.ANY", - "MetricGroup": "IcMiss", - "MetricName": "tma_info_l2mpki_code" - }, - { - "BriefDescription": "L2 cache speculative code cacheline misses per kilo instruction", - "MetricExpr": "1e3 * L2_RQSTS.CODE_RD_MISS / INST_RETIRED.ANY", - "MetricGroup": "IcMiss", - "MetricName": "tma_info_l2mpki_code_all" + "MetricName": "tma_info_memory_l2mpki_all" }, { "BriefDescription": "L2 cache ([RKL+] true) misses per kilo instruction for all demand loads (including speculative)", "MetricExpr": "1e3 * L2_RQSTS.DEMAND_DATA_RD_MISS / INST_RETIRED.ANY", "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l2mpki_load" - }, - { - "BriefDescription": "Average per-core data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * OFFCORE_REQUESTS.ALL_REQUESTS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw" + "MetricName": "tma_info_memory_l2mpki_load" }, { - "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_access_bw", - "MetricGroup": "Mem;MemoryBW;Offcore", - "MetricName": "tma_info_l3_cache_access_bw_1t" + "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", + "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L3_MISS / INST_RETIRED.ANY", + "MetricGroup": "CacheMisses;Mem", + "MetricName": "tma_info_memory_l3mpki" }, { - "BriefDescription": "Average per-core data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "64 * LONGEST_LAT_CACHE.MISS / 1e9 / duration_time", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw" + "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", + "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_RETIRED.L1_MISS + MEM_LOAD_RETIRED.FB_HIT)", + "MetricGroup": "Mem;MemoryBound;MemoryLat", + "MetricName": "tma_info_memory_load_miss_real_latency" }, { - "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", - "MetricExpr": "tma_info_l3_cache_fill_bw", - "MetricGroup": "Mem;MemoryBW", - "MetricName": "tma_info_l3_cache_fill_bw_1t" + "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", + "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", + "MetricGroup": "Mem;MemoryBW;MemoryBound", + "MetricName": "tma_info_memory_mlp", + "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" }, { - "BriefDescription": "L3 cache true misses per kilo instruction for retired demand loads", - "MetricExpr": "1e3 * MEM_LOAD_RETIRED.L3_MISS / INST_RETIRED.ANY", - "MetricGroup": "CacheMisses;Mem", - "MetricName": "tma_info_l3mpki" + "BriefDescription": "Average Parallel L2 cache miss data reads", + "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD / OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD", + "MetricGroup": "Memory_BW;Offcore", + "MetricName": "tma_info_memory_oro_data_l2_mlp" }, { "BriefDescription": "Average Latency for L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / OFFCORE_REQUESTS.DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l2_miss_latency" + "MetricName": "tma_info_memory_oro_load_l2_miss_latency" }, { "BriefDescription": "Average Parallel L2 cache miss demand Loads", "MetricExpr": "OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD / cpu@OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD\\,cmask\\=1@", "MetricGroup": "Memory_BW;Offcore", - "MetricName": "tma_info_load_l2_mlp" + "MetricName": "tma_info_memory_oro_load_l2_mlp" }, { "BriefDescription": "Average Latency for L3 cache miss demand Loads", "MetricExpr": "cpu@OFFCORE_REQUESTS_OUTSTANDING.DEMAND_DATA_RD\\,umask\\=0x10@ / OFFCORE_REQUESTS.L3_MISS_DEMAND_DATA_RD", "MetricGroup": "Memory_Lat;Offcore", - "MetricName": "tma_info_load_l3_miss_latency" + "MetricName": "tma_info_memory_oro_load_l3_miss_latency" }, { - "BriefDescription": "Actual Average Latency for L1 data-cache miss demand load operations (in core cycles)", - "MetricExpr": "L1D_PEND_MISS.PENDING / (MEM_LOAD_RETIRED.L1_MISS + MEM_LOAD_RETIRED.FB_HIT)", - "MetricGroup": "Mem;MemoryBound;MemoryLat", - "MetricName": "tma_info_load_miss_real_latency" + "BriefDescription": "Average per-thread data fill bandwidth to the L1 data cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l1d_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l1d_cache_fill_bw_1t" + }, + { + "BriefDescription": "Average per-thread data fill bandwidth to the L2 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l2_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l2_cache_fill_bw_1t" + }, + { + "BriefDescription": "Average per-thread data access bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_access_bw", + "MetricGroup": "Mem;MemoryBW;Offcore", + "MetricName": "tma_info_memory_thread_l3_cache_access_bw_1t" + }, + { + "BriefDescription": "Average per-thread data fill bandwidth to the L3 cache [GB / sec]", + "MetricExpr": "tma_info_memory_core_l3_cache_fill_bw", + "MetricGroup": "Mem;MemoryBW", + "MetricName": "tma_info_memory_thread_l3_cache_fill_bw_1t" + }, + { + "BriefDescription": "STLB (2nd level TLB) code speculative misses per kilo instruction (misses of any page-size that complete the page walk)", + "MetricExpr": "1e3 * ITLB_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricGroup": "Fed;MemoryTLB", + "MetricName": "tma_info_memory_tlb_code_stlb_mpki" }, { "BriefDescription": "STLB (2nd level TLB) data load speculative misses per kilo instruction (misses of any page-size that complete the page walk)", "MetricExpr": "1e3 * DTLB_LOAD_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_load_stlb_mpki" + "MetricName": "tma_info_memory_tlb_load_stlb_mpki" }, { - "BriefDescription": "Fraction of Uops delivered by the LSD (Loop Stream Detector; aka Loop Cache)", - "MetricExpr": "LSD.UOPS / UOPS_ISSUED.ANY", - "MetricGroup": "Fed;LSD", - "MetricName": "tma_info_lsd_coverage" + "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", + "MetricExpr": "(ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING) / (2 * tma_info_core_core_clks)", + "MetricGroup": "Mem;MemoryTLB", + "MetricName": "tma_info_memory_tlb_page_walks_utilization", + "MetricThreshold": "tma_info_memory_tlb_page_walks_utilization > 0.5" }, { - "BriefDescription": "Average number of parallel data read requests to external memory", - "MetricExpr": "UNC_ARB_DAT_OCCUPANCY.RD / UNC_ARB_DAT_OCCUPANCY.RD@cmask\\=1@", - "MetricGroup": "Mem;MemoryBW;SoC", - "MetricName": "tma_info_mem_parallel_reads", - "PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches" + "BriefDescription": "STLB (2nd level TLB) data store speculative misses per kilo instruction (misses of any page-size that complete the page walk)", + "MetricExpr": "1e3 * DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", + "MetricGroup": "Mem;MemoryTLB", + "MetricName": "tma_info_memory_tlb_store_stlb_mpki" }, { - "BriefDescription": "Average latency of data read request to external memory (in nanoseconds)", - "MetricExpr": "(UNC_ARB_TRK_OCCUPANCY.RD + UNC_ARB_DAT_OCCUPANCY.RD) / UNC_ARB_TRK_REQUESTS.RD", - "MetricGroup": "Mem;MemoryLat;SoC", - "MetricName": "tma_info_mem_read_latency", - "PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)" + "BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is execution) per-thread", + "MetricExpr": "UOPS_EXECUTED.THREAD / cpu@UOPS_EXECUTED.THREAD\\,cmask\\=1@", + "MetricGroup": "Cor;Pipeline;PortsUtil;SMT", + "MetricName": "tma_info_pipeline_execute" }, { - "BriefDescription": "Average latency of all requests to external memory (in Uncore cycles)", - "MetricExpr": "(UNC_ARB_TRK_OCCUPANCY.ALL + UNC_ARB_DAT_OCCUPANCY.RD) / arb@event\\=0x81\\,umask\\=0x1@", - "MetricGroup": "Mem;SoC", - "MetricName": "tma_info_mem_request_latency" + "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", + "MetricConstraint": "NO_GROUP_EVENTS", + "MetricExpr": "tma_retiring * tma_info_thread_slots / cpu@UOPS_RETIRED.SLOTS\\,cmask\\=1@", + "MetricGroup": "Pipeline;Ret", + "MetricName": "tma_info_pipeline_retire" }, { - "BriefDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks", - "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_bandwidth / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_sq_full / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full))) + tma_l1_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_fb_full / (tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk))", - "MetricGroup": "Mem;MemoryBW;Offcore;tma_issueBW", - "MetricName": "tma_info_memory_bandwidth", - "MetricThreshold": "tma_info_memory_bandwidth > 20", - "PublicDescription": "Total pipeline cost of (external) Memory Bandwidth related bottlenecks. Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_mem_bandwidth, tma_sq_full" + "BriefDescription": "Measured Average Frequency for unhalted processors [GHz]", + "MetricExpr": "tma_info_system_turbo_utilization * TSC / 1e9 / duration_time", + "MetricGroup": "Power;Summary", + "MetricName": "tma_info_system_average_frequency" }, { - "BriefDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs)", - "MetricExpr": "100 * tma_memory_bound * (tma_l1_bound / max(tma_memory_bound, tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_load / max(tma_l1_bound, tma_4k_aliasing + tma_dtlb_load + tma_fb_full + tma_lock_latency + tma_split_loads + tma_store_fwd_blk)) + tma_store_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_dtlb_store / (tma_dtlb_store + tma_false_sharing + tma_split_stores + tma_store_latency + tma_streaming_stores)))", - "MetricGroup": "Mem;MemoryTLB;Offcore;tma_issueTLB", - "MetricName": "tma_info_memory_data_tlbs", - "MetricThreshold": "tma_info_memory_data_tlbs > 20", - "PublicDescription": "Total pipeline cost of Memory Address Translation related bottlenecks (data-side TLBs). Related metrics: tma_dtlb_load, tma_dtlb_store" + "BriefDescription": "Average CPU Utilization", + "MetricExpr": "CPU_CLK_UNHALTED.REF_TSC / TSC", + "MetricGroup": "HPC;Summary", + "MetricName": "tma_info_system_cpu_utilization" }, { - "BriefDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches)", - "MetricExpr": "100 * tma_memory_bound * (tma_dram_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_mem_latency / (tma_mem_bandwidth + tma_mem_latency)) + tma_l3_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound) * (tma_l3_hit_latency / (tma_contested_accesses + tma_data_sharing + tma_l3_hit_latency + tma_sq_full)) + tma_l2_bound / (tma_dram_bound + tma_l1_bound + tma_l2_bound + tma_l3_bound + tma_store_bound))", - "MetricGroup": "Mem;MemoryLat;Offcore;tma_issueLat", - "MetricName": "tma_info_memory_latency", - "MetricThreshold": "tma_info_memory_latency > 20", - "PublicDescription": "Total pipeline cost of Memory Latency related bottlenecks (external memory and off-core caches). Related metrics: tma_l3_hit_latency, tma_mem_latency" + "BriefDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]", + "MetricExpr": "64 * (arb@event\\=0x81\\,umask\\=0x1@ + arb@event\\=0x84\\,umask\\=0x1@) / 1e6 / duration_time / 1e3", + "MetricGroup": "HPC;Mem;MemoryBW;SoC;tma_issueBW", + "MetricName": "tma_info_system_dram_bw_use", + "PublicDescription": "Average external Memory Bandwidth Use for reads and writes [GB / sec]. Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_mem_bandwidth, tma_sq_full" }, { - "BriefDescription": "Total pipeline cost of Branch Misprediction related bottlenecks", - "MetricExpr": "100 * (tma_branch_mispredicts + tma_fetch_latency * tma_mispredicts_resteers / (tma_branch_resteers + tma_dsb_switches + tma_icache_misses + tma_itlb_misses + tma_lcp + tma_ms_switches))", - "MetricGroup": "Bad;BadSpec;BrMispredicts;tma_issueBM", - "MetricName": "tma_info_mispredictions", - "MetricThreshold": "tma_info_mispredictions > 20", - "PublicDescription": "Total pipeline cost of Branch Misprediction related bottlenecks. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost, tma_mispredicts_resteers" + "BriefDescription": "Giga Floating Point Operations Per Second", + "MetricExpr": "(cpu@FP_ARITH_INST_RETIRED.SCALAR_SINGLE\\,umask\\=0x03@ + 2 * FP_ARITH_INST_RETIRED.128B_PACKED_DOUBLE + 4 * cpu@FP_ARITH_INST_RETIRED.128B_PACKED_SINGLE\\,umask\\=0x18@ + 8 * cpu@FP_ARITH_INST_RETIRED.256B_PACKED_SINGLE\\,umask\\=0x60@ + 16 * FP_ARITH_INST_RETIRED.512B_PACKED_SINGLE) / 1e9 / duration_time", + "MetricGroup": "Cor;Flops;HPC", + "MetricName": "tma_info_system_gflops", + "PublicDescription": "Giga Floating Point Operations Per Second. Aggregate across all supported options of: FP precisions, scalar and vector instructions, vector-width and AMX engine." }, { - "BriefDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss", - "MetricExpr": "L1D_PEND_MISS.PENDING / L1D_PEND_MISS.PENDING_CYCLES", - "MetricGroup": "Mem;MemoryBW;MemoryBound", - "MetricName": "tma_info_mlp", - "PublicDescription": "Memory-Level-Parallelism (average number of L1 miss demand load when there is at least one such miss. Per-Logical Processor)" + "BriefDescription": "Instructions per Far Branch ( Far Branches apply upon transition from application to operating system, handling interrupts, exceptions) [lower number means higher occurrence rate]", + "MetricExpr": "INST_RETIRED.ANY / BR_INST_RETIRED.FAR_BRANCH:u", + "MetricGroup": "Branches;OS", + "MetricName": "tma_info_system_ipfarbranch", + "MetricThreshold": "tma_info_system_ipfarbranch < 1e6" }, { - "BriefDescription": "Fraction of branches of other types (not individually covered by other metrics in Info.Branches group)", - "MetricExpr": "1 - (tma_info_cond_nt + tma_info_cond_tk + tma_info_callret + tma_info_jump)", - "MetricGroup": "Bad;Branches", - "MetricName": "tma_info_other_branches" + "BriefDescription": "Cycles Per Instruction for the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / INST_RETIRED.ANY_P:k", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_cpi" }, { - "BriefDescription": "Utilization of the core's Page Walker(s) serving STLB misses triggered by instruction/Load/Store accesses", - "MetricExpr": "(ITLB_MISSES.WALK_PENDING + DTLB_LOAD_MISSES.WALK_PENDING + DTLB_STORE_MISSES.WALK_PENDING) / (2 * tma_info_core_clks)", - "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_page_walks_utilization", - "MetricThreshold": "tma_info_page_walks_utilization > 0.5" + "BriefDescription": "Fraction of cycles spent in the Operating System (OS) Kernel mode", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD_P:k / CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "OS", + "MetricName": "tma_info_system_kernel_utilization", + "MetricThreshold": "tma_info_system_kernel_utilization > 0.05" + }, + { + "BriefDescription": "Average number of parallel data read requests to external memory", + "MetricExpr": "UNC_ARB_DAT_OCCUPANCY.RD / UNC_ARB_DAT_OCCUPANCY.RD@cmask\\=1@", + "MetricGroup": "Mem;MemoryBW;SoC", + "MetricName": "tma_info_system_mem_parallel_reads", + "PublicDescription": "Average number of parallel data read requests to external memory. Accounts for demand loads and L1/L2 prefetches" + }, + { + "BriefDescription": "Average latency of data read request to external memory (in nanoseconds)", + "MetricExpr": "(UNC_ARB_TRK_OCCUPANCY.RD + UNC_ARB_DAT_OCCUPANCY.RD) / UNC_ARB_TRK_REQUESTS.RD", + "MetricGroup": "Mem;MemoryLat;SoC", + "MetricName": "tma_info_system_mem_read_latency", + "PublicDescription": "Average latency of data read request to external memory (in nanoseconds). Accounts for demand loads and L1/L2 prefetches. ([RKL+]memory-controller only)" + }, + { + "BriefDescription": "Average latency of all requests to external memory (in Uncore cycles)", + "MetricExpr": "(UNC_ARB_TRK_OCCUPANCY.ALL + UNC_ARB_DAT_OCCUPANCY.RD) / arb@event\\=0x81\\,umask\\=0x1@", + "MetricGroup": "Mem;SoC", + "MetricName": "tma_info_system_mem_request_latency" }, { "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for baseline license level 0", - "MetricExpr": "CORE_POWER.LVL0_TURBO_LICENSE / tma_info_core_clks", + "MetricExpr": "CORE_POWER.LVL0_TURBO_LICENSE / tma_info_core_core_clks", "MetricGroup": "Power", - "MetricName": "tma_info_power_license0_utilization", + "MetricName": "tma_info_system_power_license0_utilization", "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for baseline license level 0. This includes non-AVX codes, SSE, AVX 128-bit, and low-current AVX 256-bit codes." }, { "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 1", - "MetricExpr": "CORE_POWER.LVL1_TURBO_LICENSE / tma_info_core_clks", + "MetricExpr": "CORE_POWER.LVL1_TURBO_LICENSE / tma_info_core_core_clks", "MetricGroup": "Power", - "MetricName": "tma_info_power_license1_utilization", - "MetricThreshold": "tma_info_power_license1_utilization > 0.5", + "MetricName": "tma_info_system_power_license1_utilization", + "MetricThreshold": "tma_info_system_power_license1_utilization > 0.5", "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 1. This includes high current AVX 256-bit instructions as well as low current AVX 512-bit instructions." }, { "BriefDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 2 (introduced in SKX)", - "MetricExpr": "CORE_POWER.LVL2_TURBO_LICENSE / tma_info_core_clks", + "MetricExpr": "CORE_POWER.LVL2_TURBO_LICENSE / tma_info_core_core_clks", "MetricGroup": "Power", - "MetricName": "tma_info_power_license2_utilization", - "MetricThreshold": "tma_info_power_license2_utilization > 0.5", + "MetricName": "tma_info_system_power_license2_utilization", + "MetricThreshold": "tma_info_system_power_license2_utilization > 0.5", "PublicDescription": "Fraction of Core cycles where the core was running with power-delivery for license level 2 (introduced in SKX). This includes high current AVX 512-bit instructions." }, { - "BriefDescription": "Average number of Uops retired in cycles where at least one uop has retired.", - "MetricExpr": "tma_retiring * tma_info_slots / cpu@UOPS_RETIRED.SLOTS\\,cmask\\=1@", - "MetricGroup": "Pipeline;Ret", - "MetricName": "tma_info_retire" + "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", + "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_DISTRIBUTED if #SMT_on else 0)", + "MetricGroup": "SMT", + "MetricName": "tma_info_system_smt_2t_utilization" }, { - "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", - "MetricExpr": "TOPDOWN.SLOTS", - "MetricGroup": "TmaL1;tma_L1_group", - "MetricName": "tma_info_slots" + "BriefDescription": "Average Frequency Utilization relative nominal frequency", + "MetricExpr": "tma_info_thread_clks / CPU_CLK_UNHALTED.REF_TSC", + "MetricGroup": "Power", + "MetricName": "tma_info_system_turbo_utilization" }, { - "BriefDescription": "Fraction of Physical Core issue-slots utilized by this Logical Processor", - "MetricExpr": "(tma_info_slots / (TOPDOWN.SLOTS / 2) if #SMT_on else 1)", - "MetricGroup": "SMT;TmaL1;tma_L1_group", - "MetricName": "tma_info_slots_utilization" + "BriefDescription": "Per-Logical Processor actual clocks when the Logical Processor is active.", + "MetricExpr": "CPU_CLK_UNHALTED.THREAD", + "MetricGroup": "Pipeline", + "MetricName": "tma_info_thread_clks" }, { - "BriefDescription": "Fraction of cycles where both hardware Logical Processors were active", - "MetricExpr": "(1 - CPU_CLK_UNHALTED.ONE_THREAD_ACTIVE / CPU_CLK_UNHALTED.REF_DISTRIBUTED if #SMT_on else 0)", - "MetricGroup": "SMT", - "MetricName": "tma_info_smt_2t_utilization" + "BriefDescription": "Cycles Per Instruction (per Logical Processor)", + "MetricExpr": "1 / tma_info_thread_ipc", + "MetricGroup": "Mem;Pipeline", + "MetricName": "tma_info_thread_cpi" }, { - "BriefDescription": "STLB (2nd level TLB) data store speculative misses per kilo instruction (misses of any page-size that complete the page walk)", - "MetricExpr": "1e3 * DTLB_STORE_MISSES.WALK_COMPLETED / INST_RETIRED.ANY", - "MetricGroup": "Mem;MemoryTLB", - "MetricName": "tma_info_store_stlb_mpki" + "BriefDescription": "The ratio of Executed- by Issued-Uops", + "MetricExpr": "UOPS_EXECUTED.THREAD / UOPS_ISSUED.ANY", + "MetricGroup": "Cor;Pipeline", + "MetricName": "tma_info_thread_execute_per_issue", + "PublicDescription": "The ratio of Executed- by Issued-Uops. Ratio > 1 suggests high rate of uop micro-fusions. Ratio < 1 suggest high rate of \"execute\" at rename stage." }, { - "BriefDescription": "Average Frequency Utilization relative nominal frequency", - "MetricExpr": "tma_info_clks / CPU_CLK_UNHALTED.REF_TSC", - "MetricGroup": "Power", - "MetricName": "tma_info_turbo_utilization" + "BriefDescription": "Instructions Per Cycle (per Logical Processor)", + "MetricExpr": "INST_RETIRED.ANY / tma_info_thread_clks", + "MetricGroup": "Ret;Summary", + "MetricName": "tma_info_thread_ipc" + }, + { + "BriefDescription": "Total issue-pipeline slots (per-Physical Core till ICL; per-Logical Processor ICL onward)", + "MetricExpr": "TOPDOWN.SLOTS", + "MetricGroup": "TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots" + }, + { + "BriefDescription": "Fraction of Physical Core issue-slots utilized by this Logical Processor", + "MetricExpr": "(tma_info_thread_slots / (TOPDOWN.SLOTS / 2) if #SMT_on else 1)", + "MetricGroup": "SMT;TmaL1;tma_L1_group", + "MetricName": "tma_info_thread_slots_utilization" }, { "BriefDescription": "Uops Per Instruction", - "MetricExpr": "tma_retiring * tma_info_slots / INST_RETIRED.ANY", + "MetricExpr": "tma_retiring * tma_info_thread_slots / INST_RETIRED.ANY", "MetricGroup": "Pipeline;Ret;Retire", - "MetricName": "tma_info_uoppi", - "MetricThreshold": "tma_info_uoppi > 1.05" + "MetricName": "tma_info_thread_uoppi", + "MetricThreshold": "tma_info_thread_uoppi > 1.05" }, { "BriefDescription": "Instruction per taken branch", - "MetricExpr": "tma_retiring * tma_info_slots / BR_INST_RETIRED.NEAR_TAKEN", + "MetricExpr": "tma_retiring * tma_info_thread_slots / BR_INST_RETIRED.NEAR_TAKEN", "MetricGroup": "Branches;Fed;FetchBW", - "MetricName": "tma_info_uptb", - "MetricThreshold": "tma_info_uptb < 7.5" + "MetricName": "tma_info_thread_uptb", + "MetricThreshold": "tma_info_thread_uptb < 7.5" }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Instruction TLB (ITLB) misses", - "MetricExpr": "ICACHE_64B.IFTAG_STALL / tma_info_clks", + "MetricExpr": "ICACHE_64B.IFTAG_STALL / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;MemoryTLB;TopdownL3;tma_L3_group;tma_fetch_latency_group", "MetricName": "tma_itlb_misses", "MetricThreshold": "tma_itlb_misses > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -1075,7 +1095,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled without loads missing the L1 data cache", - "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_clks, 0)", + "MetricExpr": "max((CYCLE_ACTIVITY.STALLS_MEM_ANY - CYCLE_ACTIVITY.STALLS_L1D_MISS) / tma_info_thread_clks, 0)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_issueL1;tma_issueMC;tma_memory_bound_group", "MetricName": "tma_l1_bound", "MetricThreshold": "tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1085,7 +1105,7 @@ { "BriefDescription": "This metric estimates how often the CPU was stalled due to L2 cache accesses by loads", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) / (MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + L1D_PEND_MISS.FB_FULL_PERIODS) * ((CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_clks)", + "MetricExpr": "MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) / (MEM_LOAD_RETIRED.L2_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS) + L1D_PEND_MISS.FB_FULL_PERIODS) * ((CYCLE_ACTIVITY.STALLS_L1D_MISS - CYCLE_ACTIVITY.STALLS_L2_MISS) / tma_info_thread_clks)", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l2_bound", "MetricThreshold": "tma_l2_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1094,7 +1114,7 @@ }, { "BriefDescription": "This metric estimates how often the CPU was stalled due to loads accesses to L3 cache or contended with a sibling Core", - "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L2_MISS - CYCLE_ACTIVITY.STALLS_L3_MISS) / tma_info_clks", + "MetricExpr": "(CYCLE_ACTIVITY.STALLS_L2_MISS - CYCLE_ACTIVITY.STALLS_L3_MISS) / tma_info_thread_clks", "MetricGroup": "CacheMisses;MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_l3_bound", "MetricThreshold": "tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1103,20 +1123,20 @@ }, { "BriefDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited)", - "MetricExpr": "17.5 * tma_info_average_frequency * MEM_LOAD_RETIRED.L3_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_clks", + "MetricExpr": "17.5 * tma_info_system_average_frequency * MEM_LOAD_RETIRED.L3_HIT * (1 + MEM_LOAD_RETIRED.FB_HIT / MEM_LOAD_RETIRED.L1_MISS / 2) / tma_info_thread_clks", "MetricGroup": "MemoryLat;TopdownL4;tma_L4_group;tma_issueLat;tma_l3_bound_group", "MetricName": "tma_l3_hit_latency", "MetricThreshold": "tma_l3_hit_latency > 0.1 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited). Avoiding private cache misses (i.e. L2 misses/L3 hits) will improve the latency; reduce contention with sibling physical cores and increase performance. Note the value of this node may overlap with its siblings. Sample with: MEM_LOAD_RETIRED.L3_HIT_PS. Related metrics: tma_info_memory_latency, tma_mem_latency", + "PublicDescription": "This metric represents fraction of cycles with demand load accesses that hit the L3 cache under unloaded scenarios (possibly L3 latency limited). Avoiding private cache misses (i.e. L2 misses/L3 hits) will improve the latency; reduce contention with sibling physical cores and increase performance. Note the value of this node may overlap with its siblings. Sample with: MEM_LOAD_RETIRED.L3_HIT_PS. Related metrics: tma_info_bottleneck_memory_latency, tma_mem_latency", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs)", - "MetricExpr": "ILD_STALL.LCP / tma_info_clks", + "MetricExpr": "ILD_STALL.LCP / tma_info_thread_clks", "MetricGroup": "FetchLat;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueFB", "MetricName": "tma_lcp", "MetricThreshold": "tma_lcp > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", - "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_dsb_coverage, tma_info_dsb_misses, tma_info_iptb", + "PublicDescription": "This metric represents fraction of cycles CPU was stalled due to Length Changing Prefixes (LCPs). Using proper compiler flags or Intel Compiler by default will certainly avoid this. #Link: Optimization Guide about LCP BKMs. Related metrics: tma_dsb_switches, tma_fetch_bandwidth, tma_info_botlnk_l2_dsb_misses, tma_info_frontend_dsb_coverage, tma_info_inst_mix_iptb", "ScaleUnit": "100%" }, { @@ -1125,12 +1145,13 @@ "MetricGroup": "Retire;TmaL2;TopdownL2;tma_L2_group;tma_retiring_group", "MetricName": "tma_light_operations", "MetricThreshold": "tma_light_operations > 0.6", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots where the CPU was retiring light-weight operations -- instructions that require no more than one uop (micro-operation). This correlates with total number of instructions used by the program. A uops-per-instruction (see UopPI metric) ratio of 1 or less should be expected for decently optimized software running on Intel Core/Xeon products. While this often indicates efficient X86 instructions were executed; high value does not necessarily mean better performance cannot be achieved. Sample with: INST_RETIRED.PREC_DIST", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Load operations", - "MetricExpr": "UOPS_DISPATCHED.PORT_2_3 / (2 * tma_info_core_clks)", + "MetricExpr": "UOPS_DISPATCHED.PORT_2_3 / (2 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_load_op_utilization", "MetricThreshold": "tma_load_op_utilization > 0.6", @@ -1147,7 +1168,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles where the Second-level TLB (STLB) was missed by load accesses, performing a hardware page walk", - "MetricExpr": "DTLB_LOAD_MISSES.WALK_ACTIVE / tma_info_clks", + "MetricExpr": "DTLB_LOAD_MISSES.WALK_ACTIVE / tma_info_thread_clks", "MetricGroup": "MemoryTLB;TopdownL5;tma_L5_group;tma_dtlb_load_group", "MetricName": "tma_load_stlb_miss", "MetricThreshold": "tma_load_stlb_miss > 0.05 & (tma_dtlb_load > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1156,7 +1177,7 @@ { "BriefDescription": "This metric represents fraction of cycles the CPU spent handling cache misses due to lock operations", "MetricConstraint": "NO_GROUP_EVENTS", - "MetricExpr": "(16 * max(0, MEM_INST_RETIRED.LOCK_LOADS - L2_RQSTS.ALL_RFO) + MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES * (10 * L2_RQSTS.RFO_HIT + min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO))) / tma_info_clks", + "MetricExpr": "(16 * max(0, MEM_INST_RETIRED.LOCK_LOADS - L2_RQSTS.ALL_RFO) + MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES * (10 * L2_RQSTS.RFO_HIT + min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO))) / tma_info_thread_clks", "MetricGroup": "Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_l1_bound_group", "MetricName": "tma_lock_latency", "MetricThreshold": "tma_lock_latency > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1165,10 +1186,10 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to LSD (Loop Stream Detector) unit", - "MetricExpr": "(LSD.CYCLES_ACTIVE - LSD.CYCLES_OK) / tma_info_core_clks / 2", + "MetricExpr": "(LSD.CYCLES_ACTIVE - LSD.CYCLES_OK) / tma_info_core_core_clks / 2", "MetricGroup": "FetchBW;LSD;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_lsd", - "MetricThreshold": "tma_lsd > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 5 > 0.35)", + "MetricThreshold": "tma_lsd > 0.15 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 5 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to LSD (Loop Stream Detector) unit. LSD typically does well sustaining Uop supply. However; in some rare cases; optimal uop-delivery could not be reached for small loops whose size (in terms of number of uops) does not suit well the LSD structure.", "ScaleUnit": "100%" }, @@ -1178,25 +1199,26 @@ "MetricGroup": "BadSpec;MachineClears;TmaL2;TopdownL2;tma_L2_group;tma_bad_speculation_group;tma_issueMC;tma_issueSyncxn", "MetricName": "tma_machine_clears", "MetricThreshold": "tma_machine_clears > 0.1 & tma_bad_speculation > 0.15", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the CPU has wasted due to Machine Clears. These slots are either wasted by uops fetched prior to the clear; or stalls the out-of-order portion of the machine needs to recover its state after the clear. For example; this can happen due to memory ordering Nukes (e.g. Memory Disambiguation) or Self-Modifying-Code (SMC) nukes. Sample with: MACHINE_CLEARS.COUNT. Related metrics: tma_clears_resteers, tma_contested_accesses, tma_data_sharing, tma_false_sharing, tma_l1_bound, tma_microcode_sequencer, tma_ms_switches, tma_remote_cache", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_clks", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, cpu@OFFCORE_REQUESTS_OUTSTANDING.ALL_DATA_RD\\,cmask\\=4@) / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueBW", "MetricName": "tma_mem_bandwidth", "MetricThreshold": "tma_mem_bandwidth > 0.2 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_sq_full", + "PublicDescription": "This metric estimates fraction of cycles where the core's performance was likely hurt due to approaching bandwidth limits of external memory (DRAM). The underlying heuristic assumes that a similar off-core traffic is generated by all IA cores. This metric does not aggregate non-data-read requests by this logical processor; requests from other IA Logical Processors/Physical Cores/sockets; or other non-IA devices like GPU; hence the maximum external memory bandwidth limits may or may not be approached when this metric is flagged (see Uncore counters for that). Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_sq_full", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM)", - "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_clks - tma_mem_bandwidth", + "MetricExpr": "min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DATA_RD) / tma_info_thread_clks - tma_mem_bandwidth", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_dram_bound_group;tma_issueLat", "MetricName": "tma_mem_latency", "MetricThreshold": "tma_mem_latency > 0.1 & (tma_dram_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM). This metric does not aggregate requests from other Logical Processors/Physical Cores/sockets (see Uncore counters for that). Related metrics: tma_info_memory_latency, tma_l3_hit_latency", + "PublicDescription": "This metric estimates fraction of cycles where the performance was likely hurt due to latency from external memory (DRAM). This metric does not aggregate requests from other Logical Processors/Physical Cores/sockets (see Uncore counters for that). Related metrics: tma_info_bottleneck_memory_latency, tma_l3_hit_latency", "ScaleUnit": "100%" }, { @@ -1205,11 +1227,13 @@ "MetricGroup": "Backend;TmaL2;TopdownL2;tma_L2_group;tma_backend_bound_group", "MetricName": "tma_memory_bound", "MetricThreshold": "tma_memory_bound > 0.2 & tma_backend_bound > 0.2", + "MetricgroupNoGroup": "TopdownL2", "PublicDescription": "This metric represents fraction of slots the Memory subsystem within the Backend was a bottleneck. Memory Bound estimates fraction of slots where pipeline is likely stalled due to demand load or store instructions. This accounts mainly for (1) non-completed in-flight memory demand loads which coincides with execution units starvation; in addition to (2) cases where stores could impose backpressure on the pipeline when many of them get buffered at the same time (less common out of the two).", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring memory operations -- uops for memory load or store accesses.", + "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "tma_light_operations * MEM_INST_RETIRED.ANY / INST_RETIRED.ANY", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_memory_operations", @@ -1218,7 +1242,7 @@ }, { "BriefDescription": "This metric represents fraction of slots the CPU was retiring uops fetched by the Microcode Sequencer (MS) unit", - "MetricExpr": "tma_retiring * tma_info_slots / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_slots", + "MetricExpr": "tma_retiring * tma_info_thread_slots / UOPS_ISSUED.ANY * IDQ.MS_UOPS / tma_info_thread_slots", "MetricGroup": "MicroSeq;TopdownL3;tma_L3_group;tma_heavy_operations_group;tma_issueMC;tma_issueMS", "MetricName": "tma_microcode_sequencer", "MetricThreshold": "tma_microcode_sequencer > 0.05 & tma_heavy_operations > 0.1", @@ -1227,28 +1251,28 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage", - "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_clks", + "MetricExpr": "BR_MISP_RETIRED.ALL_BRANCHES / (BR_MISP_RETIRED.ALL_BRANCHES + MACHINE_CLEARS.COUNT) * INT_MISC.CLEAR_RESTEER_CYCLES / tma_info_thread_clks", "MetricGroup": "BadSpec;BrMispredicts;TopdownL4;tma_L4_group;tma_branch_resteers_group;tma_issueBM", "MetricName": "tma_mispredicts_resteers", "MetricThreshold": "tma_mispredicts_resteers > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", - "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_branch_misprediction_cost, tma_info_mispredictions", + "PublicDescription": "This metric represents fraction of cycles the CPU was stalled due to Branch Resteers as a result of Branch Misprediction at execution stage. Sample with: INT_MISC.CLEAR_RESTEER_CYCLES. Related metrics: tma_branch_mispredicts, tma_info_bad_spec_branch_misprediction_cost, tma_info_bottleneck_mispredictions", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline)", - "MetricExpr": "(IDQ.MITE_CYCLES_ANY - IDQ.MITE_CYCLES_OK) / tma_info_core_clks / 2", + "MetricExpr": "(IDQ.MITE_CYCLES_ANY - IDQ.MITE_CYCLES_OK) / tma_info_core_core_clks / 2", "MetricGroup": "DSBmiss;FetchBW;TopdownL3;tma_L3_group;tma_fetch_bandwidth_group", "MetricName": "tma_mite", - "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 5 > 0.35)", + "MetricThreshold": "tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 5 > 0.35)", "PublicDescription": "This metric represents Core fraction of cycles in which CPU was likely limited due to the MITE pipeline (the legacy decode pipeline). This pipeline is used for code that was not pre-cached in the DSB or LSD. For example; inefficiencies due to asymmetric decoders; use of long immediate or LCP can manifest as MITE fetch bandwidth bottleneck. Sample with: FRONTEND_RETIRED.ANY_DSB_MISS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles where (only) 4 uops were delivered by the MITE pipeline", - "MetricExpr": "(cpu@IDQ.MITE_UOPS\\,cmask\\=4@ - cpu@IDQ.MITE_UOPS\\,cmask\\=5@) / tma_info_clks", + "MetricExpr": "(cpu@IDQ.MITE_UOPS\\,cmask\\=4@ - cpu@IDQ.MITE_UOPS\\,cmask\\=5@) / tma_info_thread_clks", "MetricGroup": "DSBmiss;FetchBW;TopdownL4;tma_L4_group;tma_mite_group", "MetricName": "tma_mite_4wide", - "MetricThreshold": "tma_mite_4wide > 0.05 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_ipc / 5 > 0.35))", + "MetricThreshold": "tma_mite_4wide > 0.05 & (tma_mite > 0.1 & (tma_fetch_bandwidth > 0.1 & tma_frontend_bound > 0.15 & tma_info_thread_ipc / 5 > 0.35))", "ScaleUnit": "100%" }, { @@ -1262,7 +1286,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles when the CPU was stalled due to switches of uop delivery to the Microcode Sequencer (MS)", - "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_clks", + "MetricExpr": "3 * IDQ.MS_SWITCHES / tma_info_thread_clks", "MetricGroup": "FetchLat;MicroSeq;TopdownL3;tma_L3_group;tma_fetch_latency_group;tma_issueMC;tma_issueMS;tma_issueMV;tma_issueSO", "MetricName": "tma_ms_switches", "MetricThreshold": "tma_ms_switches > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15)", @@ -1271,7 +1295,7 @@ }, { "BriefDescription": "This metric represents fraction of slots where the CPU was retiring NOP (no op) instructions", - "MetricExpr": "tma_light_operations * INST_RETIRED.NOP / (tma_retiring * tma_info_slots)", + "MetricExpr": "tma_light_operations * INST_RETIRED.NOP / (tma_retiring * tma_info_thread_slots)", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_nop_instructions", "MetricThreshold": "tma_nop_instructions > 0.1 & tma_light_operations > 0.6", @@ -1280,6 +1304,7 @@ }, { "BriefDescription": "This metric represents the remaining light uops fraction the CPU has executed - remaining means not covered by other sibling nodes", + "MetricConstraint": "NO_GROUP_EVENTS", "MetricExpr": "max(0, tma_light_operations - (tma_fp_arith + tma_memory_operations + tma_branch_instructions + tma_nop_instructions))", "MetricGroup": "Pipeline;TopdownL3;tma_L3_group;tma_light_operations_group", "MetricName": "tma_other_light_ops", @@ -1289,7 +1314,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 0 ([SNB+] ALU; [HSW+] ALU and 2nd branch)", - "MetricExpr": "UOPS_DISPATCHED.PORT_0 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_0 / tma_info_core_core_clks", "MetricGroup": "Compute;TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_0", "MetricThreshold": "tma_port_0 > 0.6", @@ -1298,7 +1323,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 1 (ALU)", - "MetricExpr": "UOPS_DISPATCHED.PORT_1 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_1 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_1", "MetricThreshold": "tma_port_1 > 0.6", @@ -1307,7 +1332,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 5 ([SNB+] Branches and ALU; [HSW+] ALU)", - "MetricExpr": "UOPS_DISPATCHED.PORT_5 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_5 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_5", "MetricThreshold": "tma_port_5 > 0.6", @@ -1316,7 +1341,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port 6 ([HSW+]Primary Branch and simple ALU)", - "MetricExpr": "UOPS_DISPATCHED.PORT_6 / tma_info_core_clks", + "MetricExpr": "UOPS_DISPATCHED.PORT_6 / tma_info_core_core_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_alu_op_utilization_group;tma_issue2P", "MetricName": "tma_port_6", "MetricThreshold": "tma_port_6 > 0.6", @@ -1325,7 +1350,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles the CPU performance was potentially limited due to Core computation issues (non divider-related)", - "MetricExpr": "((cpu@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) + (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL)) / tma_info_clks if ARITH.DIVIDER_ACTIVE < CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY else (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_clks)", + "MetricExpr": "((cpu@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) + (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL)) / tma_info_thread_clks if ARITH.DIVIDER_ACTIVE < CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY else (EXE_ACTIVITY.1_PORTS_UTIL + tma_retiring * EXE_ACTIVITY.2_PORTS_UTIL) / tma_info_thread_clks)", "MetricGroup": "PortsUtil;TopdownL3;tma_L3_group;tma_core_bound_group", "MetricName": "tma_ports_utilization", "MetricThreshold": "tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)", @@ -1334,7 +1359,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed no uops on any execution port (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "cpu@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ / tma_info_clks + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_clks", + "MetricExpr": "cpu@EXE_ACTIVITY.3_PORTS_UTIL\\,umask\\=0x80@ / tma_info_thread_clks + tma_serializing_operation * (CYCLE_ACTIVITY.STALLS_TOTAL - CYCLE_ACTIVITY.STALLS_MEM_ANY) / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_0", "MetricThreshold": "tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1343,7 +1368,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles where the CPU executed total of 1 uop per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "EXE_ACTIVITY.1_PORTS_UTIL / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.1_PORTS_UTIL / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issueL1;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_1", "MetricThreshold": "tma_ports_utilized_1 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1352,7 +1377,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 2 uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "EXE_ACTIVITY.2_PORTS_UTIL / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.2_PORTS_UTIL / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_issue2P;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_2", "MetricThreshold": "tma_ports_utilized_2 > 0.15 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1361,7 +1386,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles CPU executed total of 3 or more uops per cycle on all execution ports (Logical Processor cycles since ICL, Physical Core cycles otherwise)", - "MetricExpr": "UOPS_EXECUTED.CYCLES_GE_3 / tma_info_clks", + "MetricExpr": "UOPS_EXECUTED.CYCLES_GE_3 / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL4;tma_L4_group;tma_ports_utilization_group", "MetricName": "tma_ports_utilized_3m", "MetricThreshold": "tma_ports_utilized_3m > 0.7 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))", @@ -1370,16 +1395,18 @@ }, { "BriefDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired", - "MetricExpr": "topdown\\-retiring / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_slots", - "MetricGroup": "TmaL1;TopdownL1;tma_L1_group", + "DefaultMetricgroupName": "TopdownL1", + "MetricExpr": "topdown\\-retiring / (topdown\\-fe\\-bound + topdown\\-bad\\-spec + topdown\\-retiring + topdown\\-be\\-bound) + 0 * tma_info_thread_slots", + "MetricGroup": "Default;TmaL1;TopdownL1;tma_L1_group", "MetricName": "tma_retiring", "MetricThreshold": "tma_retiring > 0.7 | tma_heavy_operations > 0.1", + "MetricgroupNoGroup": "TopdownL1;Default", "PublicDescription": "This category represents fraction of slots utilized by useful work i.e. issued uops that eventually get retired. Ideally; all pipeline slots would be attributed to the Retiring category. Retiring of 100% would indicate the maximum Pipeline_Width throughput was achieved. Maximizing Retiring typically increases the Instructions-per-cycle (see IPC metric). Note that a high Retiring value does not necessary mean there is no room for more performance. For example; Heavy-operations or Microcode Assists are categorized under Retiring. They often indicate suboptimal performance and can often be optimized or avoided. Sample with: UOPS_RETIRED.SLOTS", "ScaleUnit": "100%" }, { "BriefDescription": "This metric represents fraction of cycles the CPU issue-pipeline was stalled due to serializing operations", - "MetricExpr": "RESOURCE_STALLS.SCOREBOARD / tma_info_clks", + "MetricExpr": "RESOURCE_STALLS.SCOREBOARD / tma_info_thread_clks", "MetricGroup": "PortsUtil;TopdownL5;tma_L5_group;tma_issueSO;tma_ports_utilized_0_group", "MetricName": "tma_serializing_operation", "MetricThreshold": "tma_serializing_operation > 0.1 & (tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2)))", @@ -1388,7 +1415,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to PAUSE Instructions", - "MetricExpr": "140 * MISC_RETIRED.PAUSE_INST / tma_info_clks", + "MetricExpr": "140 * MISC_RETIRED.PAUSE_INST / tma_info_thread_clks", "MetricGroup": "TopdownL6;tma_L6_group;tma_serializing_operation_group", "MetricName": "tma_slow_pause", "MetricThreshold": "tma_slow_pause > 0.05 & (tma_serializing_operation > 0.1 & (tma_ports_utilized_0 > 0.2 & (tma_ports_utilization > 0.15 & (tma_core_bound > 0.1 & tma_backend_bound > 0.2))))", @@ -1397,7 +1424,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles handling memory load split accesses - load that cross 64-byte cache line boundary", - "MetricExpr": "tma_info_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_clks", + "MetricExpr": "tma_info_memory_load_miss_real_latency * LD_BLOCKS.NO_SR / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_split_loads", "MetricThreshold": "tma_split_loads > 0.2 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1406,7 +1433,7 @@ }, { "BriefDescription": "This metric represents rate of split store accesses", - "MetricExpr": "MEM_INST_RETIRED.SPLIT_STORES / tma_info_core_clks", + "MetricExpr": "MEM_INST_RETIRED.SPLIT_STORES / tma_info_core_core_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_issueSpSt;tma_store_bound_group", "MetricName": "tma_split_stores", "MetricThreshold": "tma_split_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1415,16 +1442,16 @@ }, { "BriefDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors)", - "MetricExpr": "L1D_PEND_MISS.L2_STALL / tma_info_clks", + "MetricExpr": "L1D_PEND_MISS.L2_STALL / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueBW;tma_l3_bound_group", "MetricName": "tma_sq_full", "MetricThreshold": "tma_sq_full > 0.3 & (tma_l3_bound > 0.05 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", - "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_dram_bw_use, tma_info_memory_bandwidth, tma_mem_bandwidth", + "PublicDescription": "This metric measures fraction of cycles where the Super Queue (SQ) was full taking into account all request-types and both hardware SMT threads (Logical Processors). Related metrics: tma_fb_full, tma_info_bottleneck_memory_bandwidth, tma_info_system_dram_bw_use, tma_mem_bandwidth", "ScaleUnit": "100%" }, { "BriefDescription": "This metric estimates how often CPU was stalled due to RFO store memory accesses; RFO store issue a read-for-ownership request before the write", - "MetricExpr": "EXE_ACTIVITY.BOUND_ON_STORES / tma_info_clks", + "MetricExpr": "EXE_ACTIVITY.BOUND_ON_STORES / tma_info_thread_clks", "MetricGroup": "MemoryBound;TmaL3mem;TopdownL3;tma_L3_group;tma_memory_bound_group", "MetricName": "tma_store_bound", "MetricThreshold": "tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)", @@ -1433,7 +1460,7 @@ }, { "BriefDescription": "This metric roughly estimates fraction of cycles when the memory subsystem had loads blocked since they could not forward data from earlier (in program order) overlapping stores", - "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_clks", + "MetricExpr": "13 * LD_BLOCKS.STORE_FORWARD / tma_info_thread_clks", "MetricGroup": "TopdownL4;tma_L4_group;tma_l1_bound_group", "MetricName": "tma_store_fwd_blk", "MetricThreshold": "tma_store_fwd_blk > 0.1 & (tma_l1_bound > 0.1 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1442,7 +1469,7 @@ }, { "BriefDescription": "This metric estimates fraction of cycles the CPU spent handling L1D store misses", - "MetricExpr": "(L2_RQSTS.RFO_HIT * 10 * (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) + (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_clks", + "MetricExpr": "(L2_RQSTS.RFO_HIT * 10 * (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) + (1 - MEM_INST_RETIRED.LOCK_LOADS / MEM_INST_RETIRED.ALL_STORES) * min(CPU_CLK_UNHALTED.THREAD, OFFCORE_REQUESTS_OUTSTANDING.CYCLES_WITH_DEMAND_RFO)) / tma_info_thread_clks", "MetricGroup": "MemoryLat;Offcore;TopdownL4;tma_L4_group;tma_issueRFO;tma_issueSL;tma_store_bound_group", "MetricName": "tma_store_latency", "MetricThreshold": "tma_store_latency > 0.1 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1451,7 +1478,7 @@ }, { "BriefDescription": "This metric represents Core fraction of cycles CPU dispatched uops on execution port for Store operations", - "MetricExpr": "(UOPS_DISPATCHED.PORT_4_9 + UOPS_DISPATCHED.PORT_7_8) / (4 * tma_info_core_clks)", + "MetricExpr": "(UOPS_DISPATCHED.PORT_4_9 + UOPS_DISPATCHED.PORT_7_8) / (4 * tma_info_core_core_clks)", "MetricGroup": "TopdownL5;tma_L5_group;tma_ports_utilized_3m_group", "MetricName": "tma_store_op_utilization", "MetricThreshold": "tma_store_op_utilization > 0.6", @@ -1468,7 +1495,7 @@ }, { "BriefDescription": "This metric estimates the fraction of cycles where the STLB was missed by store accesses, performing a hardware page walk", - "MetricExpr": "DTLB_STORE_MISSES.WALK_ACTIVE / tma_info_core_clks", + "MetricExpr": "DTLB_STORE_MISSES.WALK_ACTIVE / tma_info_core_core_clks", "MetricGroup": "MemoryTLB;TopdownL5;tma_L5_group;tma_dtlb_store_group", "MetricName": "tma_store_stlb_miss", "MetricThreshold": "tma_store_stlb_miss > 0.05 & (tma_dtlb_store > 0.05 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2)))", @@ -1476,7 +1503,7 @@ }, { "BriefDescription": "This metric estimates how often CPU was stalled due to Streaming store memory accesses; Streaming store optimize out a read request required by RFO stores", - "MetricExpr": "9 * OCR.STREAMING_WR.ANY_RESPONSE / tma_info_clks", + "MetricExpr": "9 * OCR.STREAMING_WR.ANY_RESPONSE / tma_info_thread_clks", "MetricGroup": "MemoryBW;Offcore;TopdownL4;tma_L4_group;tma_issueSmSt;tma_store_bound_group", "MetricName": "tma_streaming_stores", "MetricThreshold": "tma_streaming_stores > 0.2 & (tma_store_bound > 0.2 & (tma_memory_bound > 0.2 & tma_backend_bound > 0.2))", @@ -1485,7 +1512,7 @@ }, { "BriefDescription": "This metric represents fraction of cycles the CPU was stalled due to new branch address clears", - "MetricExpr": "10 * BACLEARS.ANY / tma_info_clks", + "MetricExpr": "10 * BACLEARS.ANY / tma_info_thread_clks", "MetricGroup": "BigFoot;FetchLat;TopdownL4;tma_L4_group;tma_branch_resteers_group", "MetricName": "tma_unknown_branches", "MetricThreshold": "tma_unknown_branches > 0.05 & (tma_branch_resteers > 0.05 & (tma_fetch_latency > 0.1 & tma_frontend_bound > 0.15))", diff --git a/tools/perf/pmu-events/empty-pmu-events.c b/tools/perf/pmu-events/empty-pmu-events.c index e74defb5284f..a630c617e879 100644 --- a/tools/perf/pmu-events/empty-pmu-events.c +++ b/tools/perf/pmu-events/empty-pmu-events.c @@ -420,3 +420,8 @@ int pmu_for_each_sys_metric(pmu_metric_iter_fn fn __maybe_unused, void *data __m { return 0; } + +const char *describe_metricgroup(const char *group __maybe_unused) +{ + return NULL; +} diff --git a/tools/perf/pmu-events/jevents.py b/tools/perf/pmu-events/jevents.py index ca99b9cfe4ad..12e80bb7939b 100755 --- a/tools/perf/pmu-events/jevents.py +++ b/tools/perf/pmu-events/jevents.py @@ -37,6 +37,8 @@ _pending_metrics = [] _pending_metrics_tblname = None # Global BigCString shared by all structures. _bcs = None +# Map from the name of a metric group to a description of the group. +_metricgroups = {} # Order specific JsonEvent attributes will be visited. _json_event_attributes = [ # cmp_sevent related attributes. @@ -51,8 +53,9 @@ _json_event_attributes = [ # Attributes that are in pmu_metric rather than pmu_event. _json_metric_attributes = [ - 'metric_name', 'metric_group', 'metric_expr', 'metric_threshold', 'desc', - 'long_desc', 'unit', 'compat', 'aggr_mode', 'event_grouping' + 'pmu', 'metric_name', 'metric_group', 'metric_expr', 'metric_threshold', + 'desc', 'long_desc', 'unit', 'compat', 'metricgroup_no_group', + 'default_metricgroup_name', 'aggr_mode', 'event_grouping' ] # Attributes that are bools or enum int values, encoded as '0', '1',... _json_enum_attributes = ['aggr_mode', 'deprecated', 'event_grouping', 'perpkg'] @@ -303,6 +306,8 @@ class JsonEvent: self.deprecated = jd.get('Deprecated') self.metric_name = jd.get('MetricName') self.metric_group = jd.get('MetricGroup') + self.metricgroup_no_group = jd.get('MetricgroupNoGroup') + self.default_metricgroup_name = jd.get('DefaultMetricgroupName') self.event_grouping = convert_metric_constraint(jd.get('MetricConstraint')) self.metric_expr = None if 'MetricExpr' in jd: @@ -389,11 +394,11 @@ def read_json_events(path: str, topic: str) -> Sequence[JsonEvent]: except BaseException as err: print(f"Exception processing {path}") raise - metrics: list[Tuple[str, metric.Expression]] = [] + metrics: list[Tuple[str, str, metric.Expression]] = [] for event in events: event.topic = topic if event.metric_name and '-' not in event.metric_name: - metrics.append((event.metric_name, event.metric_expr)) + metrics.append((event.pmu, event.metric_name, event.metric_expr)) updates = metric.RewriteMetricsInTermsOfOthers(metrics) if updates: for event in events: @@ -510,6 +515,17 @@ def preprocess_one_file(parents: Sequence[str], item: os.DirEntry) -> None: if not item.is_file() or not item.name.endswith('.json'): return + if item.name == 'metricgroups.json': + metricgroup_descriptions = json.load(open(item.path)) + for mgroup in metricgroup_descriptions: + assert len(mgroup) > 1, parents + description = f"{metricgroup_descriptions[mgroup]}\\000" + mgroup = f"{mgroup}\\000" + _bcs.add(mgroup) + _bcs.add(description) + _metricgroups[mgroup] = description + return + topic = get_topic(item.name) for event in read_json_events(item.path, topic): if event.name: @@ -546,7 +562,7 @@ def process_one_file(parents: Sequence[str], item: os.DirEntry) -> None: # Ignore other directories. If the file name does not have a .json # extension, ignore it. It could be a readme.txt for instance. - if not item.is_file() or not item.name.endswith('.json'): + if not item.is_file() or not item.name.endswith('.json') or item.name == 'metricgroups.json': return add_events_table_entries(item, get_topic(item.name)) @@ -909,6 +925,38 @@ int pmu_for_each_sys_metric(pmu_metric_iter_fn fn, void *data) } """) +def print_metricgroups() -> None: + _args.output_file.write(""" +static const int metricgroups[][2] = { +""") + for mgroup in sorted(_metricgroups): + description = _metricgroups[mgroup] + _args.output_file.write( + f'\t{{ {_bcs.offsets[mgroup]}, {_bcs.offsets[description]} }}, /* {mgroup} => {description} */\n' + ) + _args.output_file.write(""" +}; + +const char *describe_metricgroup(const char *group) +{ + int low = 0, high = (int)ARRAY_SIZE(metricgroups) - 1; + + while (low <= high) { + int mid = (low + high) / 2; + const char *mgroup = &big_c_string[metricgroups[mid][0]]; + int cmp = strcmp(mgroup, group); + + if (cmp == 0) { + return &big_c_string[metricgroups[mid][1]]; + } else if (cmp < 0) { + low = mid + 1; + } else { + high = mid - 1; + } + } + return NULL; +} +""") def main() -> None: global _args @@ -991,7 +1039,7 @@ struct compact_pmu_event { print_mapping_table(archs) print_system_mapping_table() - + print_metricgroups() if __name__ == '__main__': main() diff --git a/tools/perf/pmu-events/metric.py b/tools/perf/pmu-events/metric.py index 8ec0ba884673..af58b74d1644 100644 --- a/tools/perf/pmu-events/metric.py +++ b/tools/perf/pmu-events/metric.py @@ -552,28 +552,34 @@ def ParsePerfJson(orig: str) -> Expression: return _Constify(eval(compile(parsed, orig, 'eval'))) -def RewriteMetricsInTermsOfOthers(metrics: List[Tuple[str, Expression]] - )-> Dict[str, Expression]: +def RewriteMetricsInTermsOfOthers(metrics: List[Tuple[str, str, Expression]] + )-> Dict[Tuple[str, str], Expression]: """Shorten metrics by rewriting in terms of others. Args: - metrics (list): pairs of metric names and their expressions. + metrics (list): pmus, metric names and their expressions. Returns: - Dict: mapping from a metric name to a shortened expression. + Dict: mapping from a pmu, metric name pair to a shortened expression. """ - updates: Dict[str, Expression] = dict() - for outer_name, outer_expression in metrics: + updates: Dict[Tuple[str, str], Expression] = dict() + for outer_pmu, outer_name, outer_expression in metrics: + if outer_pmu is None: + outer_pmu = 'cpu' updated = outer_expression while True: - for inner_name, inner_expression in metrics: + for inner_pmu, inner_name, inner_expression in metrics: + if inner_pmu is None: + inner_pmu = 'cpu' + if inner_pmu.lower() != outer_pmu.lower(): + continue if inner_name.lower() == outer_name.lower(): continue - if inner_name in updates: - inner_expression = updates[inner_name] + if (inner_pmu, inner_name) in updates: + inner_expression = updates[(inner_pmu, inner_name)] updated = updated.Substitute(inner_name, inner_expression) if updated.Equals(outer_expression): break - if outer_name in updates and updated.Equals(updates[outer_name]): + if (outer_pmu, outer_name) in updates and updated.Equals(updates[(outer_pmu, outer_name)]): break - updates[outer_name] = updated + updates[(outer_pmu, outer_name)] = updated return updates diff --git a/tools/perf/pmu-events/metric_test.py b/tools/perf/pmu-events/metric_test.py index 40a3c7d8b2bc..ee22ff43ddd7 100755 --- a/tools/perf/pmu-events/metric_test.py +++ b/tools/perf/pmu-events/metric_test.py @@ -158,9 +158,9 @@ class TestMetricExpressions(unittest.TestCase): def test_RewriteMetricsInTermsOfOthers(self): Expression.__eq__ = lambda e1, e2: e1.Equals(e2) - before = [('m1', ParsePerfJson('a + b + c + d')), - ('m2', ParsePerfJson('a + b + c'))] - after = {'m1': ParsePerfJson('m2 + d')} + before = [('cpu', 'm1', ParsePerfJson('a + b + c + d')), + ('cpu', 'm2', ParsePerfJson('a + b + c'))] + after = {('cpu', 'm1'): ParsePerfJson('m2 + d')} self.assertEqual(RewriteMetricsInTermsOfOthers(before), after) Expression.__eq__ = None diff --git a/tools/perf/pmu-events/pmu-events.h b/tools/perf/pmu-events/pmu-events.h index b7dff8f1021f..caf59f23cd64 100644 --- a/tools/perf/pmu-events/pmu-events.h +++ b/tools/perf/pmu-events/pmu-events.h @@ -51,6 +51,7 @@ struct pmu_event { }; struct pmu_metric { + const char *pmu; const char *metric_name; const char *metric_group; const char *metric_expr; @@ -59,6 +60,8 @@ struct pmu_metric { const char *compat; const char *desc; const char *long_desc; + const char *metricgroup_no_group; + const char *default_metricgroup_name; enum aggr_mode_class aggr_mode; enum metric_event_groups event_grouping; }; @@ -91,4 +94,6 @@ const struct pmu_metrics_table *find_sys_metrics_table(const char *name); int pmu_for_each_sys_event(pmu_event_iter_fn fn, void *data); int pmu_for_each_sys_metric(pmu_metric_iter_fn fn, void *data); +const char *describe_metricgroup(const char *group); + #endif |