aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
AgeCommit message (Collapse)AuthorFilesLines
2024-04-09drm/amdgpu: Refine IB schedule error loggingGravatar Lijo Lazar 1-2/+5
Downgrade to debug information when IBs are skipped. Also, use dev_* to identify the device. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Asad Kamal <asad.kamal@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-03-04drm/amdgpu: change vm->task_info handlingGravatar Shashank Sharma 1-7/+11
This patch changes the handling and lifecycle of vm->task_info object. The major changes are: - vm->task_info is a dynamically allocated ptr now, and its uasge is reference counted. - introducing two new helper funcs for task_info lifecycle management - amdgpu_vm_get_task_info: reference counts up task_info before returning this info - amdgpu_vm_put_task_info: reference counts down task_info - last put to task_info() frees task_info from the vm. This patch also does logistical changes required for existing usage of vm->task_info. V2: Do not block all the prints when task_info not found (Felix) V3: Fixed review comments from Felix - Fix wrong indentation - No debug message for -ENOMEM - Add NULL check for task_info - Do not duplicate the debug messages (ti vs no ti) - Get first reference of task_info in vm_init(), put last in vm_fini() V4: Fixed review comments from Felix - fix double reference increment in create_task_info - change amdgpu_vm_get_task_info_pasid - additional changes in amdgpu_gem.c while porting Cc: Christian Koenig <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Shashank Sharma <shashank.sharma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-11-24drm/sched: Reverse run-queue priority enumerationGravatar Luben Tuikov 1-1/+1
Reverse run-queue priority enumeration such that the higest priority is now 0, and for each consecutive integer the prioirty diminishes. Run-queues correspond to priorities. To an external observer a scheduler created with a single run-queue, and another created with DRM_SCHED_PRIORITY_COUNT number of run-queues, should always schedule sched->sched_rq[0] with the same "priority", as that index run-queue exists in both schedulers, i.e. a scheduler with one run-queue or many. This patch makes it so. In other words, the "priority" of sched->sched_rq[n], n >= 0, is the same for any scheduler created with any allowable number of run-queues (priorities), 0 to DRM_SCHED_PRIORITY_COUNT. Cc: Rob Clark <robdclark@gmail.com> Cc: Abhinav Kumar <quic_abhinavk@quicinc.com> Cc: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> Cc: Danilo Krummrich <dakr@redhat.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: linux-arm-msm@vger.kernel.org Cc: freedreno@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org Signed-off-by: Luben Tuikov <ltuikov89@gmail.com> Reviewed-by: Christian König <christian.koenig@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20231124052752.6915-6-ltuikov89@gmail.com
2023-11-24drm/sched: Rename priority MIN to LOWGravatar Luben Tuikov 1-1/+1
Rename DRM_SCHED_PRIORITY_MIN to DRM_SCHED_PRIORITY_LOW. This mirrors DRM_SCHED_PRIORITY_HIGH, for a list of DRM scheduler priorities in ascending order, DRM_SCHED_PRIORITY_LOW, DRM_SCHED_PRIORITY_NORMAL, DRM_SCHED_PRIORITY_HIGH, DRM_SCHED_PRIORITY_KERNEL. Cc: Rob Clark <robdclark@gmail.com> Cc: Abhinav Kumar <quic_abhinavk@quicinc.com> Cc: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> Cc: Danilo Krummrich <dakr@redhat.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: linux-arm-msm@vger.kernel.org Cc: freedreno@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org Signed-off-by: Luben Tuikov <ltuikov89@gmail.com> Reviewed-by: Christian König <christian.koenig@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20231124052752.6915-5-ltuikov89@gmail.com
2023-11-10drm/sched: implement dynamic job-flow controlGravatar Danilo Krummrich 1-1/+1
Currently, job flow control is implemented simply by limiting the number of jobs in flight. Therefore, a scheduler is initialized with a credit limit that corresponds to the number of jobs which can be sent to the hardware. This implies that for each job, drivers need to account for the maximum job size possible in order to not overflow the ring buffer. However, there are drivers, such as Nouveau, where the job size has a rather large range. For such drivers it can easily happen that job submissions not even filling the ring by 1% can block subsequent submissions, which, in the worst case, can lead to the ring run dry. In order to overcome this issue, allow for tracking the actual job size instead of the number of jobs. Therefore, add a field to track a job's credit count, which represents the number of credits a job contributes to the scheduler's credit limit. Signed-off-by: Danilo Krummrich <dakr@redhat.com> Reviewed-by: Luben Tuikov <ltuikov89@gmail.com> Link: https://patchwork.freedesktop.org/patch/msgid/20231110001638.71750-1-dakr@redhat.com
2023-10-26drm/sched: Convert the GPU scheduler to variable number of run-queuesGravatar Luben Tuikov 1-2/+2
The GPU scheduler has now a variable number of run-queues, which are set up at drm_sched_init() time. This way, each driver announces how many run-queues it requires (supports) per each GPU scheduler it creates. Note, that run-queues correspond to scheduler "priorities", thus if the number of run-queues is set to 1 at drm_sched_init(), then that scheduler supports a single run-queue, i.e. single "priority". If a driver further sets a single entity per run-queue, then this creates a 1-to-1 correspondence between a scheduler and a scheduled entity. Cc: Lucas Stach <l.stach@pengutronix.de> Cc: Russell King <linux+etnaviv@armlinux.org.uk> Cc: Qiang Yu <yuq825@gmail.com> Cc: Rob Clark <robdclark@gmail.com> Cc: Abhinav Kumar <quic_abhinavk@quicinc.com> Cc: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> Cc: Danilo Krummrich <dakr@redhat.com> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Boris Brezillon <boris.brezillon@collabora.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: Emma Anholt <emma@anholt.net> Cc: etnaviv@lists.freedesktop.org Cc: lima@lists.freedesktop.org Cc: linux-arm-msm@vger.kernel.org Cc: freedreno@lists.freedesktop.org Cc: nouveau@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org Signed-off-by: Luben Tuikov <luben.tuikov@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Link: https://lore.kernel.org/r/20231023032251.164775-1-luben.tuikov@amd.com
2023-06-15drm/amdgpu: add VM generation tokenGravatar Christian König 1-2/+2
Instead of using the VRAM lost counter add a 64bit token which indicates if a context or job is still valid to use. Should the VRAM be lost or the page tables need re-creation the token will change indicating that userspace needs to act and re-create the contexts and re-submit the work. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Luben Tuikov <luben.tuikov@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-15drm/amdgpu: abort submissions during prepare on errorGravatar Christian König 1-1/+12
Forward errors from previous submissions to this one. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Luben Tuikov <luben.tuikov@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu: set finished fence error if job timedoutGravatar ZhenGuo Yin 1-0/+2
Set finished fence to ETIME error if job timedout. Signed-off-by: ZhenGuo Yin <zhenguo.yin@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-01-18drm/amdgpu: fix amdgpu_job_free_resources v2Gravatar Christian König 1-2/+8
It can be that neither fence were initialized when we run out of UVD streams for example. v2: fix typo breaking compile Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/2324 Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org # 6.1.x
2022-12-13Merge tag 'drm-next-2022-12-13' of git://anongit.freedesktop.org/drm/drmGravatar Linus Torvalds 1-47/+33
Pull drm updates from Dave Airlie: "The biggest highlight is that the accel subsystem framework is merged. Hopefully for 6.3 we will be able to line up a driver to use it. In drivers land, i915 enables DG2 support by default now, and nouveau has a big stability refactoring and initial ampere support, AMD includes new hw IP support and should build on ARM again. There is also an ofdrm driver to take over offb on platforms it's used. Stuff outside my tree, the dma-buf patches hit a few places, the vc4 firmware changes also do, and i915 has some interactions with MEI for discrete GPUs. I think all of those should have been acked/reviewed by relevant parties. New driver: - ofdrm - replacement for offb fbdev: - add support for nomodeset fourcc: - add Vivante tiled modifier core: - atomic-helpers: CRTC primary plane test fixes, fb access hooks - connector: TV API consistency, cmdline parser improvements - send connector hotplug on cleanup - sort makefile objects tests: - sort kunit tests - improve DP-MST tests - add kunit helpers to create a device sched: - module param for scheduling policy - refcounting fix buddy: - add back random seed log ttm: - convert ttm_resource to size_t - optimize pool allocations edid: - HFVSDB parsing support fixes - logging/debug improvements - DSC quirks dma-buf: - Add unlocked vmap and attachment mapping - move drivers to common locking convention - locking improvements firmware: - new API for rPI firmware and vc4 xilinx: - zynqmp: displayport bridge support - dpsub fix bridge: - adv7533: Remove dynamic lane switching - it6505: Runtime PM support, sync improvements - ps8640: Handle AUX defer messages - tc358775: Drop soft-reset over I2C panel: - panel-edp: Add INX N116BGE-EA2 C2 and C4 support. - Jadard JD9365DA-H3 - NewVision NV3051D amdgpu: - DCN support on ARM - DCN 2.1 secure display - Sienna Cichlid mode2 reset fixes - new GC 11.x firmware versions - drop AMD specific DSC workarounds in favour of drm code - clang warning fixes - scheduler rework - SR-IOV fixes - GPUVM locking fixes - fix memory leak in CS IOCTL error path - flexible array updates - enable new GC/PSP/SMU/NBIO IP - GFX preemption support for gfx9 amdkfd: - cache size fixes - userptr fixes - enable cooperative launch on gfx 10.3 - enable GC 11.0.4 KFD support radeon: - replace kmap with kmap_local_page - ACPI ref count fix - HDA audio notifier support i915: - DG2 enabled by default - MTL enablement work - hotplug refactoring - VBT improvements - Display and watermark refactoring - ADL-P workaround - temp disable runtime_pm for discrete- - fix for A380 as a secondary GPU - Wa_18017747507 for DG2 - CS timestamp support fixes for gen5 and earlier - never purge busy TTM objects - use i915_sg_dma_sizes for all backends - demote GuC kernel contexts to normal priority - gvt: refactor for new MDEV interface - enable DC power states on eDP ports - fix gen 2/3 workarounds nouveau: - fix page fault handling - Ampere acceleration support - driver stability improvements - nva3 backlight support msm: - MSM_INFO_GET_FLAGS support - DPU: XR30 and P010 image formats - Qualcomm SM6115 support - DSI PHY support for QCM2290 - HDMI: refactored dev init path - remove exclusive-fence hack - fix speed-bin detection - enable clamp to idle on 7c3 - improved hangcheck detection vmwgfx: - fb and cursor refactoring - convert to generic hashtable - cursor improvements etnaviv: - hw workarounds - softpin MMU fixes ast: - atomic gamma LUT support - convert to SHMEM lcdif: - support YUV planes - Increase DMA burst size - FIFO threshold tuning meson: - fix return type of cvbs mode_valid mgag200: - fix PLL setup on some revisions sun4i: - A100 and D1 support udl: - modesetting improvements - hot unplug support vc4: - support PAL-M - fix regression preventing 4K @ 60Hz - fix NULL ptr deref v3d: - switch to drm managed resources renesas: - RZ/G2L DSI support - DU Kconfig cleanup mediatek: - fixup dpi and hdmi - MT8188 dpi support - MT8195 AFBC support tegra: - NVDEC hardware on Tegra234 SoC hdlcd: - switch to drm managed resources ingenic: - fix registration error path hisilicon: - convert to drm_mode_init maildp: - use managed resources mtk: - use drm_mode_init rockchip: - use drm_mode_copy" * tag 'drm-next-2022-12-13' of git://anongit.freedesktop.org/drm/drm: (1397 commits) drm/amdgpu: fix mmhub register base coding error drm/amdgpu: add tmz support for GC IP v11.0.4 drm/amdgpu: enable GFX Clock Gating control for GC IP v11.0.4 drm/amdgpu: enable GFX Power Gating for GC IP v11.0.4 drm/amdgpu: enable GFX IP v11.0.4 CG support drm/amdgpu: Make amdgpu_ring_mux functions as static drm/amdgpu: generally allow over-commit during BO allocation drm/amd/display: fix array index out of bound error in DCN32 DML drm/amd/display: 3.2.215 drm/amd/display: set optimized required for comp buf changes drm/amd/display: Add debug option to skip PSR CRTC disable drm/amd/display: correct DML calc error of UrgentLatency drm/amd/display: correct static_screen_event_mask drm/amd/display: Ensure commit_streams returns the DC return code drm/amd/display: read invalid ddc pin status cause engine busy drm/amd/display: Bypass DET swath fill check for max clocks drm/amd/display: Disable uclk pstate for subvp pipes drm/amd/display: Fix DCN2.1 default DSC clocks drm/amd/display: Enable dp_hdmi21_pcon support drm/amd/display: prevent seamless boot on displays that don't have the preferred dig ...
2022-11-25Merge tag 'amd-drm-fixes-6.1-2022-11-23' of ↵Gravatar Dave Airlie 1-1/+5
https://gitlab.freedesktop.org/agd5f/linux into drm-fixes amd-drm-fixes-6.1-2022-11-23: amdgpu: - DCN 3.1.4 fixes - DP MST DSC deadlock fixes - HMM userptr fixes - Fix Aldebaran CU occupancy reporting - GFX11 fixes - PSP suspend/resume fix - DCE12 KASAN fix - DCN 3.2.x fixes - Rotated cursor fix - SMU 13.x fix - DELL platform suspend/resume fixes - VCN4 SR-IOV fix - Display regression fix for polled connectors Signed-off-by: Dave Airlie <airlied@redhat.com> From: Alex Deucher <alexander.deucher@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20221123143453.8977-1-alexander.deucher@amd.com
2022-11-23drm/amdgpu: fix use-after-free during gpu recoveryGravatar Stanley.Yang 1-1/+6
[Why] [ 754.862560] refcount_t: underflow; use-after-free. [ 754.862898] Call Trace: [ 754.862903] <TASK> [ 754.862913] amdgpu_job_free_cb+0xc2/0xe1 [amdgpu] [ 754.863543] drm_sched_main.cold+0x34/0x39 [amd_sched] [How] The fw_fence may be not init, check whether dma_fence_init is performed before job free Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-11-23drm/amdgpu: fix use-after-free during gpu recoveryGravatar Stanley.Yang 1-1/+5
[Why] [ 754.862560] refcount_t: underflow; use-after-free. [ 754.862898] Call Trace: [ 754.862903] <TASK> [ 754.862913] amdgpu_job_free_cb+0xc2/0xe1 [amdgpu] [ 754.863543] drm_sched_main.cold+0x34/0x39 [amd_sched] [How] The fw_fence may be not init, check whether dma_fence_init is performed before job free Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-11-18drm/amdgpu: handle gang submit before VMIDGravatar Christian König 1-3/+3
Otherwise it can happen that not all gang members can get a VMID assigned and we deadlock. Signed-off-by: Christian König <christian.koenig@amd.com> Tested-by: Timur Kristóf <timur.kristof@gmail.com> Acked-by: Timur Kristóf <timur.kristof@gmail.com> Fixes: 68ce8b242242 ("drm/amdgpu: add gang submit backend v2") Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20221118153023.312582-1-christian.koenig@amd.com
2022-11-05Merge drm/drm-next into drm-misc-nextGravatar Thomas Zimmermann 1-1/+0
Backmerging drm/drm-next to get the latest changes in the xlnx driver. Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
2022-11-03drm/scheduler: rename dependency callback into prepare_jobGravatar Christian König 1-2/+2
This now matches much better what this is doing. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Luben Tuikov <luben.tuikov@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20221014084641.128280-14-christian.koenig@amd.com
2022-11-03drm/amdgpu: use scheduler dependencies for CSGravatar Christian König 1-8/+1
Entirely remove the sync obj in the job. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Luben Tuikov <luben.tuikov@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20221014084641.128280-11-christian.koenig@amd.com
2022-11-03drm/amdgpu: move explicit sync check into the CSGravatar Christian König 1-9/+3
This moves the memory allocation out of the critical code path. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Luben Tuikov <luben.tuikov@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20221014084641.128280-8-christian.koenig@amd.com
2022-11-03drm/amdgpu: cleanup scheduler job initialization v2Gravatar Christian König 1-21/+23
Init the DRM scheduler base class while allocating the job. This makes the whole handling much more cleaner. v2: fix coding style Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Luben Tuikov <luben.tuikov@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20221014084641.128280-7-christian.koenig@amd.com
2022-11-03drm/amdgpu: drop amdgpu_sync from amdgpu_vmid_grab v2Gravatar Christian König 1-7/+5
Instead return the fence directly. Avoids memory allocation to store the fence. v2: cleanup coding style as well Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Luben Tuikov <luben.tuikov@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20221014084641.128280-6-christian.koenig@amd.com
2022-11-03drm/amdgpu: drop the fence argument from amdgpu_vmid_grabGravatar Christian König 1-3/+1
This is always the job anyway. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Luben Tuikov <luben.tuikov@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20221014084641.128280-5-christian.koenig@amd.com
2022-10-18Revert "drm/amdgpu: let mode2 reset fallback to default when failure"Gravatar Victor Zhao 1-1/+0
This reverts commit dac6b80818ac2353631c5a33d140d8d5508e2957. This commit reverted the AMDGPU_SKIP_MODE2_RESET as it conflicts with the original design of reset handler. Will redesign it. Fixes: dac6b80818ac23 ("drm/amdgpu: let mode2 reset fallback to default when failure") Signed-off-by: Victor Zhao <Victor.Zhao@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-09-20drm/amdgpu: add gang submit frontend v6Gravatar Christian König 1-1/+1
Allows submitting jobs as gang which needs to run on multiple engines at the same time. All members of the gang get the same implicit, explicit and VM dependencies. So no gang member will start running until everything else is ready. The last job is considered the gang leader (usually a submission to the GFX ring) and used for signaling output dependencies. Each job is remembered individually as user of a buffer object, so there is no joining of work at the end. v2: rebase and fix review comments from Andrey and Yogesh v3: use READ instead of BOOKKEEP for now because of VM unmaps, set gang leader only when necessary v4: fix order of pushing jobs and adding fences found by Trigger. v5: fix job index calculation and adding IBs to jobs v6: fix typo found by Alex Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-09-20drm/amdgpu: add gang submit backend v2Gravatar Christian König 1-2/+26
Allows submitting jobs as gang which needs to run on multiple engines at the same time. Basic idea is that we have a global gang submit fence representing when the gang leader is finally pushed to run on the hardware last. Jobs submitted as gang are never re-submitted in case of a GPU reset since this won't work and will just deadlock the hardware immediately again. v2: fix logic inversion, improve documentation, fix rcu Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-09-13drm/amdgpu: move setting the job resourcesGravatar Christian König 1-0/+17
Move setting the job resources into amdgpu_job.c Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Reviewed-by: Luben Tuikov <luben.tuikov@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-08-29drm/amdgpu: Fix use-after-free in amdgpu_cs_ioctlGravatar YuBiao Wang 1-1/+4
[Why] In amdgpu_cs_ioctl, amdgpu_job_free could be performed ealier if there is -ERESTARTSYS error. In this case, job->hw_fence could be not initialized yet. Putting hw_fence during amdgpu_job_free could lead to a use-after-free warning. [How] Check if drm_sched_job_init is performed before job_free by checking s_fence. v2: Check hw_fence.ops instead since it could be NULL if fence is not initialized. Reverse the condition since !=NULL check is discouraged in kernel. Signed-off-by: YuBiao Wang <YuBiao.Wang@amd.com> Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-08-16drm/amdgpu: reduce reset timeGravatar Victor Zhao 1-0/+2
In multi container use case, reset time is important, so skip ring tests and cp halt wait during ip suspending for reset as they are going to fail and cost more time on reset v2: add a hang flag to indicate the reset comes from a job timeout, skip ring test and cp halt wait in this case v3: move hang flag to adev Signed-off-by: Victor Zhao <Victor.Zhao@amd.com> Acked-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-08-16drm/amdgpu: let mode2 reset fallback to default when failureGravatar Victor Zhao 1-0/+1
- introduce AMDGPU_SKIP_MODE2_RESET flag - let mode2 reset fallback to default reset method if failed v2: move this part out from the asic specific part Signed-off-by: Victor Zhao <Victor.Zhao@amd.com> Acked-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-08-16drm/amdgpu: remove useless condition in amdgpu_job_stop_all_jobs_on_sched()Gravatar Andrey Strachuk 1-4/+0
Local variable 'rq' is initialized by an address of field of drm_sched_job, so it does not make sense to compare 'rq' with NULL. Found by Linux Verification Center (linuxtesting.org) with SVACE. Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Signed-off-by: Andrey Strachuk <strochuk@ispras.ru> Fixes: 7c6e68c777f1 ("drm/amdgpu: Avoid HW GPU reset for RAS.") Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-07-18drm/amdgpu: Get rid of amdgpu_job->external_hw_fenceGravatar Andrey Grodzovsky 1-22/+5
This is a follow-up cleanup to [1]. See bellow refcount balancing for calling amdgpu_job_submit_direct after this cleanup as far as I calculated. amdgpu_fence_emit dma_fence_init 1 dma_fence_get(fence) 2 rcu_assign_pointer(*ptr, dma_fence_get(fence) 3 ---> amdgpu_job_submit_direct completes before fence signaled amdgpu_sa_bo_free (*sa_bo)->fence = dma_fence_get(fence) 4 amdgpu_job_free dma_fence_put 3 amdgpu_vcn_enc_get_destroy_msg *fence = dma_fence_get(f) 4 dma_fence_put(f); 3 amdgpu_vcn_enc_ring_test_ib dma_fence_put(fence) 2 amdgpu_fence_process dma_fence_put 1 amdgpu_sa_bo_remove_locked dma_fence_put 0 ---> amdgpu_job_submit_direct completes after fence signaled amdgpu_fence_process dma_fence_put 2 amdgpu_job_free dma_fence_put 1 amdgpu_vcn_enc_get_destroy_msg *fence = dma_fence_get(f) 2 dma_fence_put(f); 1 amdgpu_vcn_enc_ring_test_ib dma_fence_put(fence) 0 [1] - https://patchwork.kernel.org/project/dri-devel/cover/20220624180955.485440-1-andrey.grodzovsky@amd.com/ Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Suggested-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-07-13drm/amdgpu: support reset flag set for gpu resetGravatar Likun Gao 1-1/+9
Move reset_context out of gpu recover function to make it configurable for different reset purpose. For the reset way of call gpu_recovery sysfs, force to use full reset method. Otherwise, try soft reset by default if the related ASIC supportted, if soft reset failed, will use full reset. Signed-off-by: Likun Gao <Likun.Gao@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-06-28drm/amdgpu: Follow up change to previous drm scheduler change.Gravatar Andrey Grodzovsky 1-4/+0
Align refcount behaviour for amdgpu_job embedded HW fence with classic pointer style HW fences by increasing refcount each time emit is called so amdgpu code doesn't need to make workarounds using amdgpu_job.job_run_counter to keep the HW fence refcount balanced. Also since in the previous patch we resumed setting s_fence->parent to NULL in drm_sched_stop switch to directly checking if job->hw_fence is signaled to short circuit reset if already signed. Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Tested-by: Yiqing Yao <yiqing.yao@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-06-10drm/amdgpu: Rename amdgpu_device_gpu_recover_imp back to ↵Gravatar Andrey Grodzovsky 1-1/+1
amdgpu_device_gpu_recover We removed the wrapper that was queueing the recover function into reset domain queue who was using this name. Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-03-04drm/amdgpu: properly embed the IBs into the jobGravatar Christian König 1-6/+1
We now have standard macros for that. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-02-25Merge tag 'drm-misc-next-2022-02-23' of ↵Gravatar Dave Airlie 1-2/+1
git://anongit.freedesktop.org/drm/drm-misc into drm-next drm-misc-next for v5.18: UAPI Changes: Cross-subsystem Changes: - Split out panel-lvds and lvds dt bindings . - Put yes/no on/off disabled/enabled strings in linux/string_helpers.h and use it in drivers and tomoyo. - Clarify dma_fence_chain and dma_fence_array should never include eachother. - Flatten chains in syncobj's. - Don't double add in fbdev/defio when page is already enlisted. - Don't sort deferred-I/O pages by default in fbdev. Core Changes: - Fix missing pm_runtime_put_sync in bridge. - Set modifier support to only linear fb modifier if drivers don't advertise support. - As a result, we remove allow_fb_modifiers. - Add missing clear for EDID Deep Color Modes in drm_reset_display_info. - Assorted documentation updates. - Warn once in drm_clflush if there is no arch support. - Add missing select for dp helper in drm_panel_edp. - Assorted small fixes. - Improve fb-helper's clipping handling. - Don't dump shmem mmaps in a core dump. - Add accounting to ttm resource manager, and use it in amdgpu. - Allow querying the detected eDP panel through debugfs. - Add helpers for xrgb8888 to 8 and 1 bits gray. - Improve drm's buddy allocator. - Add selftests for the buddy allocator. Driver Changes: - Add support for nomodeset to a lot of drm drivers. - Use drm_module_*_driver in a lot of drm drivers. - Assorted small fixes to bridge/lt9611, v3d, vc4, vmwgfx, mxsfb, nouveau, bridge/dw-hdmi, panfrost, lima, ingenic, sprd, bridge/anx7625, ti-sn65dsi86. - Add bridge/it6505. - Create DP and DVI-I connectors in ast. - Assorted nouveau backlight fixes. - Rework amdgpu reset handling. - Add dt bindings for ingenic,jz4780-dw-hdmi. - Support reading edid through aux channel in ingenic. - Add a drm driver for Solomon SSD130x OLED displays. - Add simple support for sharp LQ140M1JW46. - Add more panels to nt35560. Signed-off-by: Dave Airlie <airlied@redhat.com> From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/686ec871-e77f-c230-22e5-9e3bb80f064a@linux.intel.com
2022-02-14drm/amdgpu: Handle the GPU recovery failure in SRIOV environment.Gravatar Surbhi Kakarya 1-1/+5
This patch handles the GPU recovery failure in sriov environment by retrying the reset if the first reset fails. To determine the condition of retry, a new macro AMDGPU_RETRY_SRIOV_RESET is added which returns true if failure is due to ETIMEDOUT, EINVAL or EBUSY, otherwise return false.A new macro AMDGPU_MAX_RETRY_LIMIT is used to limit the retry to 2. It also handles the return status in Post Asic Reset by updating the return code with asic_reset_res and eventually return the return code in amdgpu_job_timedout(). Signed-off-by: Surbhi Kakarya <surbhi.kakarya@amd.com> Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2022-02-09drm/amdgpu: Serialize non TDR gpu recovery with TDRsGravatar Andrey Grodzovsky 1-1/+1
Use reset domain wq also for non TDR gpu recovery trigers such as sysfs and RAS. We must serialize all possible GPU recoveries to gurantee no concurrency there. For TDR call the original recovery function directly since it's already executed from within the wq. For others just use a wrapper to qeueue work and wait on it to finish. v2: Rename to amdgpu_recover_work_struct Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Link: https://www.spinics.net/lists/amd-gfx/msg74113.html
2021-10-08drm/amdgpu: use adev_to_drm for consistency when accessing drm_deviceGravatar Guchun Chen 1-1/+1
adev_to_drm is used everywhere, so improve recent changes when accessing drm_device pointer from amdgpu_device. Signed-off-by: Guchun Chen <guchun.chen@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-09-14Merge drm/drm-next into drm-misc-nextGravatar Maxime Ripard 1-11/+28
Kickstart new drm-misc-next cycle. Signed-off-by: Maxime Ripard <maxime@cerno.tech>
2021-08-30drm/sched: drop entity parameter from drm_sched_push_jobGravatar Daniel Vetter 1-1/+1
Originally a job was only bound to the queue when we pushed this, but now that's done in drm_sched_job_init, making that parameter entirely redundant. Remove it. The same applies to the context parameter in lima_sched_context_queue_task, simplify that too. v2: Rebase on top of msm adopting drm/sched Reviewed-by: Christian König <christian.koenig@amd.com> Acked-by: Emma Anholt <emma@anholt.net> Acked-by: Melissa Wen <mwen@igalia.com> Reviewed-by: Steven Price <steven.price@arm.com> (v1) Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com> (v1) Signed-off-by: Daniel Vetter <daniel.vetter@intel.com> Cc: Lucas Stach <l.stach@pengutronix.de> Cc: Russell King <linux+etnaviv@armlinux.org.uk> Cc: Christian Gmeiner <christian.gmeiner@gmail.com> Cc: Qiang Yu <yuq825@gmail.com> Cc: Rob Herring <robh@kernel.org> Cc: Tomeu Vizoso <tomeu.vizoso@collabora.com> Cc: Steven Price <steven.price@arm.com> Cc: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com> Cc: Emma Anholt <emma@anholt.net> Cc: David Airlie <airlied@linux.ie> Cc: Daniel Vetter <daniel@ffwll.ch> Cc: Sumit Semwal <sumit.semwal@linaro.org> Cc: "Christian König" <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Nirmoy Das <nirmoy.das@amd.com> Cc: Dave Airlie <airlied@redhat.com> Cc: Chen Li <chenli@uniontech.com> Cc: Lee Jones <lee.jones@linaro.org> Cc: Deepak R Varma <mh12gx2825@gmail.com> Cc: Kevin Wang <kevin1.wang@amd.com> Cc: Luben Tuikov <luben.tuikov@amd.com> Cc: "Marek Olšák" <marek.olsak@amd.com> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Cc: Dennis Li <Dennis.Li@amd.com> Cc: Boris Brezillon <boris.brezillon@collabora.com> Cc: etnaviv@lists.freedesktop.org Cc: lima@lists.freedesktop.org Cc: linux-media@vger.kernel.org Cc: linaro-mm-sig@lists.linaro.org Cc: Rob Clark <robdclark@gmail.com> Cc: Sean Paul <sean@poorly.run> Cc: Melissa Wen <mwen@igalia.com> Cc: linux-arm-msm@vger.kernel.org Cc: freedreno@lists.freedesktop.org Link: https://patchwork.freedesktop.org/patch/msgid/20210805104705.862416-6-daniel.vetter@ffwll.ch
2021-08-30drm/sched: Split drm_sched_job_initGravatar Daniel Vetter 1-0/+2
This is a very confusingly named function, because not just does it init an object, it arms it and provides a point of no return for pushing a job into the scheduler. It would be nice if that's a bit clearer in the interface. But the real reason is that I want to push the dependency tracking helpers into the scheduler code, and that means drm_sched_job_init must be called a lot earlier, without arming the job. v2: - don't change .gitignore (Steven) - don't forget v3d (Emma) v3: Emma noticed that I leak the memory allocated in drm_sched_job_init if we bail out before the point of no return in subsequent driver patches. To be able to fix this change drm_sched_job_cleanup() so it can handle being called both before and after drm_sched_job_arm(). Also improve the kerneldoc for this. v4: - Fix the drm_sched_job_cleanup logic, I inverted the booleans, as usual (Melissa) - Christian pointed out that drm_sched_entity_select_rq() also needs to be moved into drm_sched_job_arm, which made me realize that the job->id definitely needs to be moved too. Shuffle things to fit between job_init and job_arm. v5: Reshuffle the split between init/arm once more, amdgpu abuses drm_sched.ready to signal gpu reset failures. Also document this somewhat. (Christian) v6: Rebase on top of the msm drm/sched support. Note that the drm_sched_job_init() call is completely misplaced, and hence also the split-out drm_sched_entity_push_job(). I've put in a FIXME which the next patch will address. v7: Drop the FIXME in msm, after discussions with Rob I agree it shouldn't be a problem where it is now. Acked-by: Christian König <christian.koenig@amd.com> Acked-by: Melissa Wen <mwen@igalia.com> Cc: Melissa Wen <melissa.srw@gmail.com> Acked-by: Emma Anholt <emma@anholt.net> Acked-by: Steven Price <steven.price@arm.com> (v2) Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com> (v5) Signed-off-by: Daniel Vetter <daniel.vetter@intel.com> Cc: Lucas Stach <l.stach@pengutronix.de> Cc: Russell King <linux+etnaviv@armlinux.org.uk> Cc: Christian Gmeiner <christian.gmeiner@gmail.com> Cc: Qiang Yu <yuq825@gmail.com> Cc: Rob Herring <robh@kernel.org> Cc: Tomeu Vizoso <tomeu.vizoso@collabora.com> Cc: Steven Price <steven.price@arm.com> Cc: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com> Cc: David Airlie <airlied@linux.ie> Cc: Daniel Vetter <daniel@ffwll.ch> Cc: Sumit Semwal <sumit.semwal@linaro.org> Cc: "Christian König" <christian.koenig@amd.com> Cc: Masahiro Yamada <masahiroy@kernel.org> Cc: Kees Cook <keescook@chromium.org> Cc: Adam Borowski <kilobyte@angband.pl> Cc: Nick Terrell <terrelln@fb.com> Cc: Mauro Carvalho Chehab <mchehab+huawei@kernel.org> Cc: Paul Menzel <pmenzel@molgen.mpg.de> Cc: Sami Tolvanen <samitolvanen@google.com> Cc: Viresh Kumar <viresh.kumar@linaro.org> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Dave Airlie <airlied@redhat.com> Cc: Nirmoy Das <nirmoy.das@amd.com> Cc: Deepak R Varma <mh12gx2825@gmail.com> Cc: Lee Jones <lee.jones@linaro.org> Cc: Kevin Wang <kevin1.wang@amd.com> Cc: Chen Li <chenli@uniontech.com> Cc: Luben Tuikov <luben.tuikov@amd.com> Cc: "Marek Olšák" <marek.olsak@amd.com> Cc: Dennis Li <Dennis.Li@amd.com> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Cc: Sonny Jiang <sonny.jiang@amd.com> Cc: Boris Brezillon <boris.brezillon@collabora.com> Cc: Tian Tao <tiantao6@hisilicon.com> Cc: etnaviv@lists.freedesktop.org Cc: lima@lists.freedesktop.org Cc: linux-media@vger.kernel.org Cc: linaro-mm-sig@lists.linaro.org Cc: Emma Anholt <emma@anholt.net> Cc: Rob Clark <robdclark@gmail.com> Cc: Sean Paul <sean@poorly.run> Cc: linux-arm-msm@vger.kernel.org Cc: freedreno@lists.freedesktop.org Link: https://patchwork.freedesktop.org/patch/msgid/20210817084917.3555822-1-daniel.vetter@ffwll.ch
2021-08-16drm/amd/amdgpu embed hw_fence into amdgpu_jobGravatar Jack Zhang 1-11/+28
Why: Previously hw fence is alloced separately with job. It caused historical lifetime issues and corner cases. The ideal situation is to take fence to manage both job and fence's lifetime, and simplify the design of gpu-scheduler. How: We propose to embed hw_fence into amdgpu_job. 1. We cover the normal job submission by this method. 2. For ib_test, and submit without a parent job keep the legacy way to create a hw fence separately. v2: use AMDGPU_FENCE_FLAG_EMBED_IN_JOB_BIT to show that the fence is embedded in a job. v3: remove redundant variable ring in amdgpu_job v4: add tdr sequence support for this feature. Add a job_run_counter to indicate whether this job is a resubmit job. v5 add missing handling in amdgpu_fence_enable_signaling Signed-off-by: Jingwen Chen <Jingwen.Chen2@amd.com> Signed-off-by: Jack Zhang <Jack.Zhang7@hotmail.com> Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Reviewed by: Monk Liu <monk.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-05-19drm/amdgpu: Prevent any job recoveries after device is unplugged.Gravatar Andrey Grodzovsky 1-3/+16
Return DRM_TASK_STATUS_ENODEV back to the scheduler when device is not present so they timeout timer will not be rearmed. v5: Update to match updated return values in enum drm_gpu_sched_stat Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20210512142648.666476-12-andrey.grodzovsky@amd.com
2021-01-29drm/scheduler: Job timeout handler returns status (v3)Gravatar Luben Tuikov 1-2/+4
This patch does not change current behaviour. The driver's job timeout handler now returns status indicating back to the DRM layer whether the device (GPU) is no longer available, such as after it's been unplugged, or whether all is normal, i.e. current behaviour. All drivers which make use of the drm_sched_backend_ops' .timedout_job() callback have been accordingly renamed and return the would've-been default value of DRM_GPU_SCHED_STAT_NOMINAL to restart the task's timeout timer--this is the old behaviour, and is preserved by this patch. v2: Use enum as the status of a driver's job timeout callback method. v3: Return scheduler/device information, rather than task information. Cc: Alexander Deucher <Alexander.Deucher@amd.com> Cc: Andrey Grodzovsky <Andrey.Grodzovsky@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: Lucas Stach <l.stach@pengutronix.de> Cc: Russell King <linux+etnaviv@armlinux.org.uk> Cc: Christian Gmeiner <christian.gmeiner@gmail.com> Cc: Qiang Yu <yuq825@gmail.com> Cc: Rob Herring <robh@kernel.org> Cc: Tomeu Vizoso <tomeu.vizoso@collabora.com> Cc: Steven Price <steven.price@arm.com> Cc: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com> Cc: Eric Anholt <eric@anholt.net> Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Luben Tuikov <luben.tuikov@amd.com> Acked-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com> Acked-by: Christian König <christian.koenig@amd.com> Acked-by: Steven Price <steven.price@arm.com> Signed-off-by: Christian König <christian.koenig@amd.com> Link: https://patchwork.freedesktop.org/patch/415095/
2020-12-08gpu/drm: ring_mirror_list --> pending_listGravatar Luben Tuikov 1-1/+1
Rename "ring_mirror_list" to "pending_list", to describe what something is, not what it does, how it's used, or how the hardware implements it. This also abstracts the actual hardware implementation, i.e. how the low-level driver communicates with the device it drives, ring, CAM, etc., shouldn't be exposed to DRM. The pending_list keeps jobs submitted, which are out of our control. Usually this means they are pending execution status in hardware, but the latter definition is a more general (inclusive) definition. Signed-off-by: Luben Tuikov <luben.tuikov@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Link: https://patchwork.freedesktop.org/patch/405573/ Cc: Alexander Deucher <Alexander.Deucher@amd.com> Cc: Andrey Grodzovsky <Andrey.Grodzovsky@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Christian König <christian.koenig@amd.com>
2020-12-08drm/scheduler: "node" --> "list"Gravatar Luben Tuikov 1-1/+1
Rename "node" to "list" in struct drm_sched_job, in order to make it consistent with what we see being used throughout gpu_scheduler.h, for instance in struct drm_sched_entity, as well as the rest of DRM and the kernel. Signed-off-by: Luben Tuikov <luben.tuikov@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Link: https://patchwork.freedesktop.org/patch/403515/ Cc: Alexander Deucher <Alexander.Deucher@amd.com> Cc: Andrey Grodzovsky <Andrey.Grodzovsky@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Christian König <christian.koenig@amd.com>
2020-08-18drm/scheduler: Scheduler priority fixes (v2)Gravatar Luben Tuikov 1-1/+1
Remove DRM_SCHED_PRIORITY_LOW, as it was used in only one place. Rename and separate by a line DRM_SCHED_PRIORITY_MAX to DRM_SCHED_PRIORITY_COUNT as it represents a (total) count of said priorities and it is used as such in loops throughout the code. (0-based indexing is the the count number.) Remove redundant word HIGH in priority names, and rename *KERNEL* to *HIGH*, as it really means that, high. v2: Add back KERNEL and remove SW and HW, in lieu of a single HIGH between NORMAL and KERNEL. Signed-off-by: Luben Tuikov <luben.tuikov@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2020-08-14drm/amdgpu: revert "fix system hang issue during GPU reset"Gravatar Christian König 1-7/+7
The whole approach wasn't thought through till the end. We already had a reset lock like this in the past and it caused the same problems like this one. Completely revert the patch for now and add individual trylock protection to the hardware access functions as necessary. This reverts commit df9c8d1aa278c435c30a69b8f2418b4a52fcb929. Signed-off-by: Christian König <christian.koenig@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2020-07-27drm/amdgpu: fix system hang issue during GPU resetGravatar Dennis Li 1-7/+7
when GPU hang, driver has multi-paths to enter amdgpu_device_gpu_recover, the atomic adev->in_gpu_reset and hive->in_reset are used to avoid re-entering GPU recovery. During GPU reset and resume, it is unsafe that other threads access GPU, which maybe cause GPU reset failed. Therefore the new rw_semaphore adev->reset_sem is introduced, which protect GPU from being accessed by external threads during recovery. v2: 1. add rwlock for some ioctls, debugfs and file-close function. 2. change to use dqm->is_resetting and dqm_lock for protection in kfd driver. 3. remove try_lock and change adev->in_gpu_reset as atomic, to avoid re-enter GPU recovery for the same GPU hang. v3: 1. change back to use adev->reset_sem to protect kfd callback functions, because dqm_lock couldn't protect all codes, for example: free_mqd must be called outside of dqm_lock; [ 1230.176199] Hardware name: Supermicro SYS-7049GP-TRT/X11DPG-QT, BIOS 3.1 05/23/2019 [ 1230.177221] Call Trace: [ 1230.178249] dump_stack+0x98/0xd5 [ 1230.179443] amdgpu_virt_kiq_reg_write_reg_wait+0x181/0x190 [amdgpu] [ 1230.180673] gmc_v9_0_flush_gpu_tlb+0xcc/0x310 [amdgpu] [ 1230.181882] amdgpu_gart_unbind+0xa9/0xe0 [amdgpu] [ 1230.183098] amdgpu_ttm_backend_unbind+0x46/0x180 [amdgpu] [ 1230.184239] ? ttm_bo_put+0x171/0x5f0 [ttm] [ 1230.185394] ttm_tt_unbind+0x21/0x40 [ttm] [ 1230.186558] ttm_tt_destroy.part.12+0x12/0x60 [ttm] [ 1230.187707] ttm_tt_destroy+0x13/0x20 [ttm] [ 1230.188832] ttm_bo_cleanup_memtype_use+0x36/0x80 [ttm] [ 1230.189979] ttm_bo_put+0x1be/0x5f0 [ttm] [ 1230.191230] amdgpu_bo_unref+0x1e/0x30 [amdgpu] [ 1230.192522] amdgpu_amdkfd_free_gtt_mem+0xaf/0x140 [amdgpu] [ 1230.193833] free_mqd+0x25/0x40 [amdgpu] [ 1230.195143] destroy_queue_cpsch+0x1a7/0x270 [amdgpu] [ 1230.196475] pqm_destroy_queue+0x105/0x260 [amdgpu] [ 1230.197819] kfd_ioctl_destroy_queue+0x37/0x70 [amdgpu] [ 1230.199154] kfd_ioctl+0x277/0x500 [amdgpu] [ 1230.200458] ? kfd_ioctl_get_clock_counters+0x60/0x60 [amdgpu] [ 1230.201656] ? tomoyo_file_ioctl+0x19/0x20 [ 1230.202831] ksys_ioctl+0x98/0xb0 [ 1230.204004] __x64_sys_ioctl+0x1a/0x20 [ 1230.205174] do_syscall_64+0x5f/0x250 [ 1230.206339] entry_SYSCALL_64_after_hwframe+0x49/0xbe 2. remove try_lock and introduce atomic hive->in_reset, to avoid re-enter GPU recovery. v4: 1. remove an unnecessary whitespace change in kfd_chardev.c 2. remove comment codes in amdgpu_device.c 3. add more detailed comment in commit message 4. define a wrap function amdgpu_in_reset v5: 1. Fix some style issues. Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Suggested-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com> Suggested-by: Christian König <christian.koenig@amd.com> Suggested-by: Felix Kuehling <Felix.Kuehling@amd.com> Suggested-by: Lijo Lazar <Lijo.Lazar@amd.com> Suggested-by: Luben Tukov <luben.tuikov@amd.com> Signed-off-by: Dennis Li <Dennis.Li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Copyright © 1990 – 2022 jet tsang zeon-git.
All Rights Reverse Engineered.