diff options
Diffstat (limited to 'kernel')
77 files changed, 2524 insertions, 1475 deletions
diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c index 8a33e8747a0e..286ab3db0fde 100644 --- a/kernel/bpf/cpumap.c +++ b/kernel/bpf/cpumap.c @@ -28,6 +28,7 @@ #include <linux/sched.h> #include <linux/workqueue.h> #include <linux/kthread.h> +#include <linux/completion.h> #include <trace/events/xdp.h> #include <linux/btf_ids.h> @@ -73,6 +74,7 @@ struct bpf_cpu_map_entry { struct rcu_head rcu; struct work_struct kthread_stop_wq; + struct completion kthread_running; }; struct bpf_cpu_map { @@ -122,22 +124,6 @@ static void get_cpu_map_entry(struct bpf_cpu_map_entry *rcpu) atomic_inc(&rcpu->refcnt); } -/* called from workqueue, to workaround syscall using preempt_disable */ -static void cpu_map_kthread_stop(struct work_struct *work) -{ - struct bpf_cpu_map_entry *rcpu; - - rcpu = container_of(work, struct bpf_cpu_map_entry, kthread_stop_wq); - - /* Wait for flush in __cpu_map_entry_free(), via full RCU barrier, - * as it waits until all in-flight call_rcu() callbacks complete. - */ - rcu_barrier(); - - /* kthread_stop will wake_up_process and wait for it to complete */ - kthread_stop(rcpu->kthread); -} - static void __cpu_map_ring_cleanup(struct ptr_ring *ring) { /* The tear-down procedure should have made sure that queue is @@ -145,11 +131,17 @@ static void __cpu_map_ring_cleanup(struct ptr_ring *ring) * invoked cpu_map_kthread_stop(). Catch any broken behaviour * gracefully and warn once. */ - struct xdp_frame *xdpf; + void *ptr; - while ((xdpf = ptr_ring_consume(ring))) - if (WARN_ON_ONCE(xdpf)) - xdp_return_frame(xdpf); + while ((ptr = ptr_ring_consume(ring))) { + WARN_ON_ONCE(1); + if (unlikely(__ptr_test_bit(0, &ptr))) { + __ptr_clear_bit(0, &ptr); + kfree_skb(ptr); + continue; + } + xdp_return_frame(ptr); + } } static void put_cpu_map_entry(struct bpf_cpu_map_entry *rcpu) @@ -165,6 +157,22 @@ static void put_cpu_map_entry(struct bpf_cpu_map_entry *rcpu) } } +/* called from workqueue, to workaround syscall using preempt_disable */ +static void cpu_map_kthread_stop(struct work_struct *work) +{ + struct bpf_cpu_map_entry *rcpu; + + rcpu = container_of(work, struct bpf_cpu_map_entry, kthread_stop_wq); + + /* Wait for flush in __cpu_map_entry_free(), via full RCU barrier, + * as it waits until all in-flight call_rcu() callbacks complete. + */ + rcu_barrier(); + + /* kthread_stop will wake_up_process and wait for it to complete */ + kthread_stop(rcpu->kthread); +} + static void cpu_map_bpf_prog_run_skb(struct bpf_cpu_map_entry *rcpu, struct list_head *listp, struct xdp_cpumap_stats *stats) @@ -290,11 +298,11 @@ static int cpu_map_bpf_prog_run(struct bpf_cpu_map_entry *rcpu, void **frames, return nframes; } - static int cpu_map_kthread_run(void *data) { struct bpf_cpu_map_entry *rcpu = data; + complete(&rcpu->kthread_running); set_current_state(TASK_INTERRUPTIBLE); /* When kthread gives stop order, then rcpu have been disconnected @@ -459,6 +467,7 @@ __cpu_map_entry_alloc(struct bpf_map *map, struct bpf_cpumap_val *value, goto free_ptr_ring; /* Setup kthread */ + init_completion(&rcpu->kthread_running); rcpu->kthread = kthread_create_on_node(cpu_map_kthread_run, rcpu, numa, "cpumap/%d/map:%d", cpu, map->id); @@ -472,6 +481,12 @@ __cpu_map_entry_alloc(struct bpf_map *map, struct bpf_cpumap_val *value, kthread_bind(rcpu->kthread, cpu); wake_up_process(rcpu->kthread); + /* Make sure kthread has been running, so kthread_stop() will not + * stop the kthread prematurely and all pending frames or skbs + * will be handled by the kthread before kthread_stop() returns. + */ + wait_for_completion(&rcpu->kthread_running); + return rcpu; free_prog: diff --git a/kernel/bpf/inode.c b/kernel/bpf/inode.c index 4174f76133df..99d0625b6c82 100644 --- a/kernel/bpf/inode.c +++ b/kernel/bpf/inode.c @@ -118,9 +118,8 @@ static struct inode *bpf_get_inode(struct super_block *sb, return ERR_PTR(-ENOSPC); inode->i_ino = get_next_ino(); - inode->i_atime = current_time(inode); + inode->i_atime = inode_set_ctime_current(inode); inode->i_mtime = inode->i_atime; - inode->i_ctime = inode->i_atime; inode_init_owner(&nop_mnt_idmap, inode, dir, mode); @@ -148,8 +147,7 @@ static void bpf_dentry_finalize(struct dentry *dentry, struct inode *inode, d_instantiate(dentry, inode); dget(dentry); - dir->i_mtime = current_time(dir); - dir->i_ctime = dir->i_mtime; + dir->i_mtime = inode_set_ctime_current(dir); } static int bpf_mkdir(struct mnt_idmap *idmap, struct inode *dir, diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 11e54dd8b6dd..02a021c524ab 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -5573,16 +5573,17 @@ static int update_stack_depth(struct bpf_verifier_env *env, * Since recursion is prevented by check_cfg() this algorithm * only needs a local stack of MAX_CALL_FRAMES to remember callsites */ -static int check_max_stack_depth(struct bpf_verifier_env *env) +static int check_max_stack_depth_subprog(struct bpf_verifier_env *env, int idx) { - int depth = 0, frame = 0, idx = 0, i = 0, subprog_end; struct bpf_subprog_info *subprog = env->subprog_info; struct bpf_insn *insn = env->prog->insnsi; + int depth = 0, frame = 0, i, subprog_end; bool tail_call_reachable = false; int ret_insn[MAX_CALL_FRAMES]; int ret_prog[MAX_CALL_FRAMES]; int j; + i = subprog[idx].start; process_func: /* protect against potential stack overflow that might happen when * bpf2bpf calls get combined with tailcalls. Limit the caller's stack @@ -5621,7 +5622,7 @@ process_func: continue_func: subprog_end = subprog[idx + 1].start; for (; i < subprog_end; i++) { - int next_insn; + int next_insn, sidx; if (!bpf_pseudo_call(insn + i) && !bpf_pseudo_func(insn + i)) continue; @@ -5631,21 +5632,23 @@ continue_func: /* find the callee */ next_insn = i + insn[i].imm + 1; - idx = find_subprog(env, next_insn); - if (idx < 0) { + sidx = find_subprog(env, next_insn); + if (sidx < 0) { WARN_ONCE(1, "verifier bug. No program starts at insn %d\n", next_insn); return -EFAULT; } - if (subprog[idx].is_async_cb) { - if (subprog[idx].has_tail_call) { + if (subprog[sidx].is_async_cb) { + if (subprog[sidx].has_tail_call) { verbose(env, "verifier bug. subprog has tail_call and async cb\n"); return -EFAULT; } - /* async callbacks don't increase bpf prog stack size */ - continue; + /* async callbacks don't increase bpf prog stack size unless called directly */ + if (!bpf_pseudo_call(insn + i)) + continue; } i = next_insn; + idx = sidx; if (subprog[idx].has_tail_call) tail_call_reachable = true; @@ -5681,6 +5684,22 @@ continue_func: goto continue_func; } +static int check_max_stack_depth(struct bpf_verifier_env *env) +{ + struct bpf_subprog_info *si = env->subprog_info; + int ret; + + for (int i = 0; i < env->subprog_cnt; i++) { + if (!i || si[i].is_async_cb) { + ret = check_max_stack_depth_subprog(env, i); + if (ret < 0) + return ret; + } + continue; + } + return 0; +} + #ifndef CONFIG_BPF_JIT_ALWAYS_ON static int get_callee_stack_depth(struct bpf_verifier_env *env, const struct bpf_insn *insn, int idx) diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index bfe3cd8ccf36..5fa95f86cb4d 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -3685,6 +3685,36 @@ static int cpu_stat_show(struct seq_file *seq, void *v) return ret; } +static int __maybe_unused cgroup_local_stat_show(struct seq_file *seq, + struct cgroup *cgrp, int ssid) +{ + struct cgroup_subsys *ss = cgroup_subsys[ssid]; + struct cgroup_subsys_state *css; + int ret; + + if (!ss->css_local_stat_show) + return 0; + + css = cgroup_tryget_css(cgrp, ss); + if (!css) + return 0; + + ret = ss->css_local_stat_show(seq, css); + css_put(css); + return ret; +} + +static int cpu_local_stat_show(struct seq_file *seq, void *v) +{ + struct cgroup __maybe_unused *cgrp = seq_css(seq)->cgroup; + int ret = 0; + +#ifdef CONFIG_CGROUP_SCHED + ret = cgroup_local_stat_show(seq, cgrp, cpu_cgrp_id); +#endif + return ret; +} + #ifdef CONFIG_PSI static int cgroup_io_pressure_show(struct seq_file *seq, void *v) { @@ -3730,7 +3760,7 @@ static ssize_t pressure_write(struct kernfs_open_file *of, char *buf, } psi = cgroup_psi(cgrp); - new = psi_trigger_create(psi, buf, res, of->file); + new = psi_trigger_create(psi, buf, res, of->file, of); if (IS_ERR(new)) { cgroup_put(cgrp); return PTR_ERR(new); @@ -5235,6 +5265,10 @@ static struct cftype cgroup_base_files[] = { .name = "cpu.stat", .seq_show = cpu_stat_show, }, + { + .name = "cpu.stat.local", + .seq_show = cpu_local_stat_show, + }, { } /* terminate */ }; diff --git a/kernel/cpu.c b/kernel/cpu.c index 88a7ede322bd..f6811c857102 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -592,7 +592,10 @@ static void lockdep_release_cpus_lock(void) void __weak arch_smt_update(void) { } #ifdef CONFIG_HOTPLUG_SMT + enum cpuhp_smt_control cpu_smt_control __read_mostly = CPU_SMT_ENABLED; +static unsigned int cpu_smt_max_threads __ro_after_init; +unsigned int cpu_smt_num_threads __read_mostly = UINT_MAX; void __init cpu_smt_disable(bool force) { @@ -606,16 +609,33 @@ void __init cpu_smt_disable(bool force) pr_info("SMT: disabled\n"); cpu_smt_control = CPU_SMT_DISABLED; } + cpu_smt_num_threads = 1; } /* * The decision whether SMT is supported can only be done after the full * CPU identification. Called from architecture code. */ -void __init cpu_smt_check_topology(void) +void __init cpu_smt_set_num_threads(unsigned int num_threads, + unsigned int max_threads) { - if (!topology_smt_supported()) + WARN_ON(!num_threads || (num_threads > max_threads)); + + if (max_threads == 1) cpu_smt_control = CPU_SMT_NOT_SUPPORTED; + + cpu_smt_max_threads = max_threads; + + /* + * If SMT has been disabled via the kernel command line or SMT is + * not supported, set cpu_smt_num_threads to 1 for consistency. + * If enabled, take the architecture requested number of threads + * to bring up into account. + */ + if (cpu_smt_control != CPU_SMT_ENABLED) + cpu_smt_num_threads = 1; + else if (num_threads < cpu_smt_num_threads) + cpu_smt_num_threads = num_threads; } static int __init smt_cmdline_disable(char *str) @@ -625,9 +645,23 @@ static int __init smt_cmdline_disable(char *str) } early_param("nosmt", smt_cmdline_disable); +/* + * For Archicture supporting partial SMT states check if the thread is allowed. + * Otherwise this has already been checked through cpu_smt_max_threads when + * setting the SMT level. + */ +static inline bool cpu_smt_thread_allowed(unsigned int cpu) +{ +#ifdef CONFIG_SMT_NUM_THREADS_DYNAMIC + return topology_smt_thread_allowed(cpu); +#else + return true; +#endif +} + static inline bool cpu_smt_allowed(unsigned int cpu) { - if (cpu_smt_control == CPU_SMT_ENABLED) + if (cpu_smt_control == CPU_SMT_ENABLED && cpu_smt_thread_allowed(cpu)) return true; if (topology_is_primary_thread(cpu)) @@ -642,7 +676,7 @@ static inline bool cpu_smt_allowed(unsigned int cpu) return !cpumask_test_cpu(cpu, &cpus_booted_once_mask); } -/* Returns true if SMT is not supported of forcefully (irreversibly) disabled */ +/* Returns true if SMT is supported and not forcefully (irreversibly) disabled */ bool cpu_smt_possible(void) { return cpu_smt_control != CPU_SMT_FORCE_DISABLED && @@ -650,22 +684,8 @@ bool cpu_smt_possible(void) } EXPORT_SYMBOL_GPL(cpu_smt_possible); -static inline bool cpuhp_smt_aware(void) -{ - return topology_smt_supported(); -} - -static inline const struct cpumask *cpuhp_get_primary_thread_mask(void) -{ - return cpu_primary_thread_mask; -} #else static inline bool cpu_smt_allowed(unsigned int cpu) { return true; } -static inline bool cpuhp_smt_aware(void) { return false; } -static inline const struct cpumask *cpuhp_get_primary_thread_mask(void) -{ - return cpu_present_mask; -} #endif static inline enum cpuhp_state @@ -1793,6 +1813,16 @@ static int __init parallel_bringup_parse_param(char *arg) } early_param("cpuhp.parallel", parallel_bringup_parse_param); +static inline bool cpuhp_smt_aware(void) +{ + return cpu_smt_max_threads > 1; +} + +static inline const struct cpumask *cpuhp_get_primary_thread_mask(void) +{ + return cpu_primary_thread_mask; +} + /* * On architectures which have enabled parallel bringup this invokes all BP * prepare states for each of the to be onlined APs first. The last state @@ -2626,6 +2656,12 @@ int cpuhp_smt_disable(enum cpuhp_smt_control ctrlval) for_each_online_cpu(cpu) { if (topology_is_primary_thread(cpu)) continue; + /* + * Disable can be called with CPU_SMT_ENABLED when changing + * from a higher to lower number of SMT threads per core. + */ + if (ctrlval == CPU_SMT_ENABLED && cpu_smt_thread_allowed(cpu)) + continue; ret = cpu_down_maps_locked(cpu, CPUHP_OFFLINE); if (ret) break; @@ -2660,6 +2696,8 @@ int cpuhp_smt_enable(void) /* Skip online CPUs and CPUs on offline nodes */ if (cpu_online(cpu) || !node_online(cpu_to_node(cpu))) continue; + if (!cpu_smt_thread_allowed(cpu)) + continue; ret = _cpu_up(cpu, 0, CPUHP_ONLINE); if (ret) break; @@ -2838,20 +2876,19 @@ static const struct attribute_group cpuhp_cpu_root_attr_group = { #ifdef CONFIG_HOTPLUG_SMT +static bool cpu_smt_num_threads_valid(unsigned int threads) +{ + if (IS_ENABLED(CONFIG_SMT_NUM_THREADS_DYNAMIC)) + return threads >= 1 && threads <= cpu_smt_max_threads; + return threads == 1 || threads == cpu_smt_max_threads; +} + static ssize_t __store_smt_control(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { - int ctrlval, ret; - - if (sysfs_streq(buf, "on")) - ctrlval = CPU_SMT_ENABLED; - else if (sysfs_streq(buf, "off")) - ctrlval = CPU_SMT_DISABLED; - else if (sysfs_streq(buf, "forceoff")) - ctrlval = CPU_SMT_FORCE_DISABLED; - else - return -EINVAL; + int ctrlval, ret, num_threads, orig_threads; + bool force_off; if (cpu_smt_control == CPU_SMT_FORCE_DISABLED) return -EPERM; @@ -2859,21 +2896,39 @@ __store_smt_control(struct device *dev, struct device_attribute *attr, if (cpu_smt_control == CPU_SMT_NOT_SUPPORTED) return -ENODEV; + if (sysfs_streq(buf, "on")) { + ctrlval = CPU_SMT_ENABLED; + num_threads = cpu_smt_max_threads; + } else if (sysfs_streq(buf, "off")) { + ctrlval = CPU_SMT_DISABLED; + num_threads = 1; + } else if (sysfs_streq(buf, "forceoff")) { + ctrlval = CPU_SMT_FORCE_DISABLED; + num_threads = 1; + } else if (kstrtoint(buf, 10, &num_threads) == 0) { + if (num_threads == 1) + ctrlval = CPU_SMT_DISABLED; + else if (cpu_smt_num_threads_valid(num_threads)) + ctrlval = CPU_SMT_ENABLED; + else + return -EINVAL; + } else { + return -EINVAL; + } + ret = lock_device_hotplug_sysfs(); if (ret) return ret; - if (ctrlval != cpu_smt_control) { - switch (ctrlval) { - case CPU_SMT_ENABLED: - ret = cpuhp_smt_enable(); - break; - case CPU_SMT_DISABLED: - case CPU_SMT_FORCE_DISABLED: - ret = cpuhp_smt_disable(ctrlval); - break; - } - } + orig_threads = cpu_smt_num_threads; + cpu_smt_num_threads = num_threads; + + force_off = ctrlval != cpu_smt_control && ctrlval == CPU_SMT_FORCE_DISABLED; + + if (num_threads > orig_threads) + ret = cpuhp_smt_enable(); + else if (num_threads < orig_threads || force_off) + ret = cpuhp_smt_disable(ctrlval); unlock_device_hotplug(); return ret ? ret : count; @@ -2901,6 +2956,17 @@ static ssize_t control_show(struct device *dev, { const char *state = smt_states[cpu_smt_control]; +#ifdef CONFIG_HOTPLUG_SMT + /* + * If SMT is enabled but not all threads are enabled then show the + * number of threads. If all threads are enabled show "on". Otherwise + * show the state name. + */ + if (cpu_smt_control == CPU_SMT_ENABLED && + cpu_smt_num_threads != cpu_smt_max_threads) + return sysfs_emit(buf, "%d\n", cpu_smt_num_threads); +#endif + return snprintf(buf, PAGE_SIZE - 2, "%s\n", state); } diff --git a/kernel/entry/common.c b/kernel/entry/common.c index be61332c66b5..d7ee4bc3f2ba 100644 --- a/kernel/entry/common.c +++ b/kernel/entry/common.c @@ -205,8 +205,7 @@ static void exit_to_user_mode_prepare(struct pt_regs *regs) arch_exit_to_user_mode_prepare(regs, ti_work); - /* Ensure that the address limit is intact and no locks are held */ - addr_limit_user_check(); + /* Ensure that kernel state is sane for a return to userspace */ kmap_assert_nomap(); lockdep_assert_irqs_disabled(); lockdep_sys_exit(); diff --git a/kernel/events/core.c b/kernel/events/core.c index 78ae7b6f90fd..93015cb64d4e 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -8249,7 +8249,7 @@ static void perf_event_comm_event(struct perf_comm_event *comm_event) unsigned int size; memset(comm, 0, sizeof(comm)); - strlcpy(comm, comm_event->task->comm, sizeof(comm)); + strscpy(comm, comm_event->task->comm, sizeof(comm)); size = ALIGN(strlen(comm)+1, sizeof(u64)); comm_event->comm = comm; @@ -8704,7 +8704,7 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event) } cpy_name: - strlcpy(tmp, name, sizeof(tmp)); + strscpy(tmp, name, sizeof(tmp)); name = tmp; got_name: /* @@ -9128,7 +9128,7 @@ void perf_event_ksymbol(u16 ksym_type, u64 addr, u32 len, bool unregister, ksym_type == PERF_RECORD_KSYMBOL_TYPE_UNKNOWN) goto err; - strlcpy(name, sym, KSYM_NAME_LEN); + strscpy(name, sym, KSYM_NAME_LEN); name_len = strlen(name) + 1; while (!IS_ALIGNED(name_len, sizeof(u64))) name[name_len++] = '\0'; @@ -9595,16 +9595,16 @@ u64 perf_swevent_set_period(struct perf_event *event) hwc->last_period = hwc->sample_period; -again: - old = val = local64_read(&hwc->period_left); - if (val < 0) - return 0; + old = local64_read(&hwc->period_left); + do { + val = old; + if (val < 0) + return 0; - nr = div64_u64(period + val, period); - offset = nr * period; - val -= offset; - if (local64_cmpxchg(&hwc->period_left, old, val) != old) - goto again; + nr = div64_u64(period + val, period); + offset = nr * period; + val -= offset; + } while (!local64_try_cmpxchg(&hwc->period_left, &old, val)); return nr; } diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c index a0433f37b024..fb1e180b5f0a 100644 --- a/kernel/events/ring_buffer.c +++ b/kernel/events/ring_buffer.c @@ -191,9 +191,10 @@ __perf_output_begin(struct perf_output_handle *handle, perf_output_get_handle(handle); + offset = local_read(&rb->head); do { + head = offset; tail = READ_ONCE(rb->user_page->data_tail); - offset = head = local_read(&rb->head); if (!rb->overwrite) { if (unlikely(!ring_buffer_has_space(head, tail, perf_data_size(rb), @@ -217,7 +218,7 @@ __perf_output_begin(struct perf_output_handle *handle, head += size; else head -= size; - } while (local_cmpxchg(&rb->head, offset, head) != offset); + } while (!local_try_cmpxchg(&rb->head, &offset, head)); if (backward) { offset = head; diff --git a/kernel/fork.c b/kernel/fork.c index d2e12b6d2b18..f81149739eb9 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -985,6 +985,14 @@ void __put_task_struct(struct task_struct *tsk) } EXPORT_SYMBOL_GPL(__put_task_struct); +void __put_task_struct_rcu_cb(struct rcu_head *rhp) +{ + struct task_struct *task = container_of(rhp, struct task_struct, rcu); + + __put_task_struct(task); +} +EXPORT_SYMBOL_GPL(__put_task_struct_rcu_cb); + void __init __weak arch_task_cache_init(void) { } /* diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index ee8c0acf39df..dc94e0bf2c94 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -473,11 +473,12 @@ void handle_nested_irq(unsigned int irq) action = desc->action; if (unlikely(!action || irqd_irq_disabled(&desc->irq_data))) { desc->istate |= IRQS_PENDING; - goto out_unlock; + raw_spin_unlock_irq(&desc->lock); + return; } kstat_incr_irqs_this_cpu(desc); - irqd_set(&desc->irq_data, IRQD_IRQ_INPROGRESS); + atomic_inc(&desc->threads_active); raw_spin_unlock_irq(&desc->lock); action_ret = IRQ_NONE; @@ -487,11 +488,7 @@ void handle_nested_irq(unsigned int irq) if (!irq_settings_no_debug(desc)) note_interrupt(desc, action_ret); - raw_spin_lock_irq(&desc->lock); - irqd_clear(&desc->irq_data, IRQD_IRQ_INPROGRESS); - -out_unlock: - raw_spin_unlock_irq(&desc->lock); + wake_threads_waitq(desc); } EXPORT_SYMBOL_GPL(handle_nested_irq); diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h index bdd35bb9c735..bcc7f21db9ee 100644 --- a/kernel/irq/internals.h +++ b/kernel/irq/internals.h @@ -108,8 +108,6 @@ extern int __irq_get_irqchip_state(struct irq_data *data, enum irqchip_irq_state which, bool *state); -extern void init_kstat_irqs(struct irq_desc *desc, int node, int nr); - irqreturn_t __handle_irq_event_percpu(struct irq_desc *desc); irqreturn_t handle_irq_event_percpu(struct irq_desc *desc); irqreturn_t handle_irq_event(struct irq_desc *desc); @@ -121,6 +119,8 @@ void irq_resend_init(struct irq_desc *desc); bool irq_wait_for_poll(struct irq_desc *desc); void __irq_wake_thread(struct irq_desc *desc, struct irqaction *action); +void wake_threads_waitq(struct irq_desc *desc); + #ifdef CONFIG_PROC_FS extern void register_irq_proc(unsigned int irq, struct irq_desc *desc); extern void unregister_irq_proc(unsigned int irq, struct irq_desc *desc); diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index d2742af0f0fd..d309ba84e08a 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -108,6 +108,16 @@ bool synchronize_hardirq(unsigned int irq) } EXPORT_SYMBOL(synchronize_hardirq); +static void __synchronize_irq(struct irq_desc *desc) +{ + __synchronize_hardirq(desc, true); + /* + * We made sure that no hardirq handler is running. Now verify that no + * threaded handlers are active. + */ + wait_event(desc->wait_for_threads, !atomic_read(&desc->threads_active)); +} + /** * synchronize_irq - wait for pending IRQ handlers (on other CPUs) * @irq: interrupt number to wait for @@ -127,16 +137,8 @@ void synchronize_irq(unsigned int irq) { struct irq_desc *desc = irq_to_desc(irq); - if (desc) { - __synchronize_hardirq(desc, true); - /* - * We made sure that no hardirq handler is - * running. Now verify that no threaded handlers are - * active. - */ - wait_event(desc->wait_for_threads, - !atomic_read(&desc->threads_active)); - } + if (desc) + __synchronize_irq(desc); } EXPORT_SYMBOL(synchronize_irq); @@ -1216,7 +1218,7 @@ static irqreturn_t irq_thread_fn(struct irq_desc *desc, return ret; } -static void wake_threads_waitq(struct irq_desc *desc) +void wake_threads_waitq(struct irq_desc *desc) { if (atomic_dec_and_test(&desc->threads_active)) wake_up(&desc->wait_for_threads); @@ -1944,7 +1946,7 @@ static struct irqaction *__free_irq(struct irq_desc *desc, void *dev_id) * supports it also make sure that there is no (not yet serviced) * interrupt in flight at the hardware level. */ - __synchronize_hardirq(desc, true); + __synchronize_irq(desc); #ifdef CONFIG_DEBUG_SHIRQ /* diff --git a/kernel/irq/resend.c b/kernel/irq/resend.c index edec335c0a7a..5f2c66860ac6 100644 --- a/kernel/irq/resend.c +++ b/kernel/irq/resend.c @@ -68,11 +68,16 @@ static int irq_sw_resend(struct irq_desc *desc) */ if (!desc->parent_irq) return -EINVAL; + + desc = irq_to_desc(desc->parent_irq); + if (!desc) + return -EINVAL; } /* Add to resend_list and activate the softirq: */ raw_spin_lock(&irq_resend_lock); - hlist_add_head(&desc->resend_node, &irq_resend_list); + if (hlist_unhashed(&desc->resend_node)) + hlist_add_head(&desc->resend_node, &irq_resend_list); raw_spin_unlock(&irq_resend_lock); tasklet_schedule(&resend_tasklet); return 0; diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c index 7982cc9d497c..18edd57b5fe8 100644 --- a/kernel/kallsyms.c +++ b/kernel/kallsyms.c @@ -163,42 +163,36 @@ unsigned long kallsyms_sym_address(int idx) return kallsyms_relative_base - 1 - kallsyms_offsets[idx]; } -static bool cleanup_symbol_name(char *s) +static void cleanup_symbol_name(char *s) { char *res; if (!IS_ENABLED(CONFIG_LTO_CLANG)) - return false; + return; /* * LLVM appends various suffixes for local functions and variables that * must be promoted to global scope as part of LTO. This can break * hooking of static functions with kprobes. '.' is not a valid - * character in an identifier in C. Suffixes observed: + * character in an identifier in C. Suffixes only in LLVM LTO observed: * - foo.llvm.[0-9a-f]+ - * - foo.[0-9a-f]+ */ - res = strchr(s, '.'); - if (res) { + res = strstr(s, ".llvm."); + if (res) *res = '\0'; - return true; - } - return false; + return; } static int compare_symbol_name(const char *name, char *namebuf) { - int ret; - - ret = strcmp(name, namebuf); - if (!ret) - return ret; - - if (cleanup_symbol_name(namebuf) && !strcmp(name, namebuf)) - return 0; - - return ret; + /* The kallsyms_seqs_of_names is sorted based on names after + * cleanup_symbol_name() (see scripts/kallsyms.c) if clang lto is enabled. + * To ensure correct bisection in kallsyms_lookup_names(), do + * cleanup_symbol_name(namebuf) before comparing name and namebuf. + */ + cleanup_symbol_name(namebuf); + return strcmp(name, namebuf); } static unsigned int get_symbol_seq(int index) diff --git a/kernel/kallsyms_selftest.c b/kernel/kallsyms_selftest.c index a2e3745d15c4..e05ddc33a752 100644 --- a/kernel/kallsyms_selftest.c +++ b/kernel/kallsyms_selftest.c @@ -196,7 +196,7 @@ static bool match_cleanup_name(const char *s, const char *name) if (!IS_ENABLED(CONFIG_LTO_CLANG)) return false; - p = strchr(s, '.'); + p = strstr(s, ".llvm."); if (!p) return false; @@ -344,27 +344,6 @@ static int test_kallsyms_basic_function(void) goto failed; } - /* - * The first '.' may be the initial letter, in which case the - * entire symbol name will be truncated to an empty string in - * cleanup_symbol_name(). Do not test these symbols. - * - * For example: - * cat /proc/kallsyms | awk '{print $3}' | grep -E "^\." | head - * .E_read_words - * .E_leading_bytes - * .E_trailing_bytes - * .E_write_words - * .E_copy - * .str.292.llvm.12122243386960820698 - * .str.24.llvm.12122243386960820698 - * .str.29.llvm.12122243386960820698 - * .str.75.llvm.12122243386960820698 - * .str.99.llvm.12122243386960820698 - */ - if (IS_ENABLED(CONFIG_LTO_CLANG) && !namebuf[0]) - continue; - lookup_addr = kallsyms_lookup_name(namebuf); memset(stat, 0, sizeof(*stat)); diff --git a/kernel/kprobes.c b/kernel/kprobes.c index ce13f1a35251..ca385b61d546 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -1072,7 +1072,7 @@ static int kprobe_ftrace_enabled; static int __arm_kprobe_ftrace(struct kprobe *p, struct ftrace_ops *ops, int *cnt) { - int ret = 0; + int ret; lockdep_assert_held(&kprobe_mutex); @@ -1110,7 +1110,7 @@ static int arm_kprobe_ftrace(struct kprobe *p) static int __disarm_kprobe_ftrace(struct kprobe *p, struct ftrace_ops *ops, int *cnt) { - int ret = 0; + int ret; lockdep_assert_held(&kprobe_mutex); @@ -1545,6 +1545,17 @@ static int check_ftrace_location(struct kprobe *p) return 0; } +static bool is_cfi_preamble_symbol(unsigned long addr) +{ + char symbuf[KSYM_NAME_LEN]; + + if (lookup_symbol_name(addr, symbuf)) + return false; + + return str_has_prefix("__cfi_", symbuf) || + str_has_prefix("__pfx_", symbuf); +} + static int check_kprobe_address_safe(struct kprobe *p, struct module **probed_mod) { @@ -1563,7 +1574,8 @@ static int check_kprobe_address_safe(struct kprobe *p, within_kprobe_blacklist((unsigned long) p->addr) || jump_label_text_reserved(p->addr, p->addr) || static_call_text_reserved(p->addr, p->addr) || - find_bug((unsigned long)p->addr)) { + find_bug((unsigned long)p->addr) || + is_cfi_preamble_symbol((unsigned long)p->addr)) { ret = -EINVAL; goto out; } @@ -2007,9 +2019,9 @@ void __weak arch_kretprobe_fixup_return(struct pt_regs *regs, unsigned long __kretprobe_trampoline_handler(struct pt_regs *regs, void *frame_pointer) { - kprobe_opcode_t *correct_ret_addr = NULL; struct kretprobe_instance *ri = NULL; struct llist_node *first, *node = NULL; + kprobe_opcode_t *correct_ret_addr; struct kretprobe *rp; /* Find correct address and all nodes for this frame. */ @@ -2693,7 +2705,7 @@ void kprobe_free_init_mem(void) static int __init init_kprobes(void) { - int i, err = 0; + int i, err; /* FIXME allocate the probe table, currently defined statically */ /* initialize all list heads */ diff --git a/kernel/locking/locktorture.c b/kernel/locking/locktorture.c index 949d3deae506..270c7f80ce84 100644 --- a/kernel/locking/locktorture.c +++ b/kernel/locking/locktorture.c @@ -45,6 +45,7 @@ torture_param(int, stutter, 5, "Number of jiffies to run/halt test, 0=disable"); torture_param(int, rt_boost, 2, "Do periodic rt-boost. 0=Disable, 1=Only for rt_mutex, 2=For all lock types."); torture_param(int, rt_boost_factor, 50, "A factor determining how often rt-boost happens."); +torture_param(int, writer_fifo, 0, "Run writers at sched_set_fifo() priority"); torture_param(int, verbose, 1, "Enable verbose debugging printk()s"); torture_param(int, nested_locks, 0, "Number of nested locks (max = 8)"); /* Going much higher trips "BUG: MAX_LOCKDEP_CHAIN_HLOCKS too low!" errors */ @@ -809,7 +810,8 @@ static int lock_torture_writer(void *arg) bool skip_main_lock; VERBOSE_TOROUT_STRING("lock_torture_writer task started"); - set_user_nice(current, MAX_NICE); + if (!rt_task(current)) + set_user_nice(current, MAX_NICE); do { if ((torture_random(&rand) & 0xfffff) == 0) @@ -1015,8 +1017,7 @@ static void lock_torture_cleanup(void) if (writer_tasks) { for (i = 0; i < cxt.nrealwriters_stress; i++) - torture_stop_kthread(lock_torture_writer, - writer_tasks[i]); + torture_stop_kthread(lock_torture_writer, writer_tasks[i]); kfree(writer_tasks); writer_tasks = NULL; } @@ -1244,8 +1245,9 @@ static int __init lock_torture_init(void) goto create_reader; /* Create writer. */ - firsterr = torture_create_kthread(lock_torture_writer, &cxt.lwsa[i], - writer_tasks[i]); + firsterr = torture_create_kthread_cb(lock_torture_writer, &cxt.lwsa[i], + writer_tasks[i], + writer_fifo ? sched_set_fifo : NULL); if (torture_init_error(firsterr)) goto unwind; diff --git a/kernel/locking/qspinlock_paravirt.h b/kernel/locking/qspinlock_paravirt.h index 6afc249ce697..6a0184e9c234 100644 --- a/kernel/locking/qspinlock_paravirt.h +++ b/kernel/locking/qspinlock_paravirt.h @@ -486,6 +486,16 @@ gotlock: } /* + * Include the architecture specific callee-save thunk of the + * __pv_queued_spin_unlock(). This thunk is put together with + * __pv_queued_spin_unlock() to make the callee-save thunk and the real unlock + * function close to each other sharing consecutive instruction cachelines. + * Alternatively, architecture specific version of __pv_queued_spin_unlock() + * can be defined. + */ +#include <asm/qspinlock_paravirt.h> + +/* * PV versions of the unlock fastpath and slowpath functions to be used * instead of queued_spin_unlock(). */ @@ -533,16 +543,6 @@ __pv_queued_spin_unlock_slowpath(struct qspinlock *lock, u8 locked) pv_kick(node->cpu); } -/* - * Include the architecture specific callee-save thunk of the - * __pv_queued_spin_unlock(). This thunk is put together with - * __pv_queued_spin_unlock() to make the callee-save thunk and the real unlock - * function close to each other sharing consecutive instruction cachelines. - * Alternatively, architecture specific version of __pv_queued_spin_unlock() - * can be defined. - */ -#include <asm/qspinlock_paravirt.h> - #ifndef __pv_queued_spin_unlock __visible __lockfunc void __pv_queued_spin_unlock(struct qspinlock *lock) { diff --git a/kernel/locking/rtmutex.c b/kernel/locking/rtmutex.c index 728f434de2bb..21db0df0eb00 100644 --- a/kernel/locking/rtmutex.c +++ b/kernel/locking/rtmutex.c @@ -333,21 +333,43 @@ static __always_inline int __waiter_prio(struct task_struct *task) return prio; } +/* + * Update the waiter->tree copy of the sort keys. + */ static __always_inline void waiter_update_prio(struct rt_mutex_waiter *waiter, struct task_struct *task) { - waiter->prio = __waiter_prio(task); - waiter->deadline = task->dl.deadline; + lockdep_assert_held(&waiter->lock->wait_lock); + lockdep_assert(RB_EMPTY_NODE(&waiter->tree.entry)); + + waiter->tree.prio = __waiter_prio(task); + waiter->tree.deadline = task->dl.deadline; +} + +/* + * Update the waiter->pi_tree copy of the sort keys (from the tree copy). + */ +static __always_inline void +waiter_clone_prio(struct rt_mutex_waiter *waiter, struct task_struct *task) +{ + lockdep_assert_held(&waiter->lock->wait_lock); + lockdep_assert_held(&task->pi_lock); + lockdep_assert(RB_EMPTY_NODE(&waiter->pi_tree.entry)); + + waiter->pi_tree.prio = waiter->tree.prio; + waiter->pi_tree.deadline = waiter->tree.deadline; } /* - * Only use with rt_mutex_waiter_{less,equal}() + * Only use with rt_waiter_node_{less,equal}() */ +#define task_to_waiter_node(p) \ + &(struct rt_waiter_node){ .prio = __waiter_prio(p), .deadline = (p)->dl.deadline } #define task_to_waiter(p) \ - &(struct rt_mutex_waiter){ .prio = __waiter_prio(p), .deadline = (p)->dl.deadline } + &(struct rt_mutex_waiter){ .tree = *task_to_waiter_node(p) } -static __always_inline int rt_mutex_waiter_less(struct rt_mutex_waiter *left, - struct rt_mutex_waiter *right) +static __always_inline int rt_waiter_node_less(struct rt_waiter_node *left, + struct rt_waiter_node *right) { if (left->prio < right->prio) return 1; @@ -364,8 +386,8 @@ static __always_inline int rt_mutex_waiter_less(struct rt_mutex_waiter *left, return 0; } -static __always_inline int rt_mutex_waiter_equal(struct rt_mutex_waiter *left, - struct rt_mutex_waiter *right) +static __always_inline int rt_waiter_node_equal(struct rt_waiter_node *left, + struct rt_waiter_node *right) { if (left->prio != right->prio) return 0; @@ -385,7 +407,7 @@ static __always_inline int rt_mutex_waiter_equal(struct rt_mutex_waiter *left, static inline bool rt_mutex_steal(struct rt_mutex_waiter *waiter, struct rt_mutex_waiter *top_waiter) { - if (rt_mutex_waiter_less(waiter, top_waiter)) + if (rt_waiter_node_less(&waiter->tree, &top_waiter->tree)) return true; #ifdef RT_MUTEX_BUILD_SPINLOCKS @@ -393,30 +415,30 @@ static inline bool rt_mutex_steal(struct rt_mutex_waiter *waiter, * Note that RT tasks are excluded from same priority (lateral) * steals to prevent the introduction of an unbounded latency. */ - if (rt_prio(waiter->prio) || dl_prio(waiter->prio)) + if (rt_prio(waiter->tree.prio) || dl_prio(waiter->tree.prio)) return false; - return rt_mutex_waiter_equal(waiter, top_waiter); + return rt_waiter_node_equal(&waiter->tree, &top_waiter->tree); #else return false; #endif } #define __node_2_waiter(node) \ - rb_entry((node), struct rt_mutex_waiter, tree_entry) + rb_entry((node), struct rt_mutex_waiter, tree.entry) static __always_inline bool __waiter_less(struct rb_node *a, const struct rb_node *b) { struct rt_mutex_waiter *aw = __node_2_waiter(a); struct rt_mutex_waiter *bw = __node_2_waiter(b); - if (rt_mutex_waiter_less(aw, bw)) + if (rt_waiter_node_less(&aw->tree, &bw->tree)) return 1; if (!build_ww_mutex()) return 0; - if (rt_mutex_waiter_less(bw, aw)) + if (rt_waiter_node_less(&bw->tree, &aw->tree)) return 0; /* NOTE: relies on waiter->ww_ctx being set before insertion */ @@ -434,48 +456,58 @@ static __always_inline bool __waiter_less(struct rb_node *a, const struct rb_nod static __always_inline void rt_mutex_enqueue(struct rt_mutex_base *lock, struct rt_mutex_waiter *waiter) { - rb_add_cached(&waiter->tree_entry, &lock->waiters, __waiter_less); + lockdep_assert_held(&lock->wait_lock); + + rb_add_cached(&waiter->tree.entry, &lock->waiters, __waiter_less); } static __always_inline void rt_mutex_dequeue(struct rt_mutex_base *lock, struct rt_mutex_waiter *waiter) { - if (RB_EMPTY_NODE(&waiter->tree_entry)) + lockdep_assert_held(&lock->wait_lock); + + if (RB_EMPTY_NODE(&waiter->tree.entry)) return; - rb_erase_cached(&waiter->tree_entry, &lock->waiters); - RB_CLEAR_NODE(&waiter->tree_entry); + rb_erase_cached(&waiter->tree.entry, &lock->waiters); + RB_CLEAR_NODE(&waiter->tree.entry); } -#define __node_2_pi_waiter(node) \ - rb_entry((node), struct rt_mutex_waiter, pi_tree_entry) +#define __node_2_rt_node(node) \ + rb_entry((node), struct rt_waiter_node, entry) -static __always_inline bool -__pi_waiter_less(struct rb_node *a, const struct rb_node *b) +static __always_inline bool __pi_waiter_less(struct rb_node *a, const struct rb_node *b) { - return rt_mutex_waiter_less(__node_2_pi_waiter(a), __node_2_pi_waiter(b)); + return rt_waiter_node_less(__node_2_rt_node(a), __node_2_rt_node(b)); } static __always_inline void rt_mutex_enqueue_pi(struct task_struct *task, struct rt_mutex_waiter *waiter) { - rb_add_cached(&waiter->pi_tree_entry, &task->pi_waiters, __pi_waiter_less); + lockdep_assert_held(&task->pi_lock); + + rb_add_cached(&waiter->pi_tree.entry, &task->pi_waiters, __pi_waiter_less); } static __always_inline void rt_mutex_dequeue_pi(struct task_struct *task, struct rt_mutex_waiter *waiter) { - if (RB_EMPTY_NODE(&waiter->pi_tree_entry)) + lockdep_assert_held(&task->pi_lock); + + if (RB_EMPTY_NODE(&waiter->pi_tree.entry)) return; - rb_erase_cached(&waiter->pi_tree_entry, &task->pi_waiters); - RB_CLEAR_NODE(&waiter->pi_tree_entry); + rb_erase_cached(&waiter->pi_tree.entry, &task->pi_waiters); + RB_CLEAR_NODE(&waiter->pi_tree.entry); } -static __always_inline void rt_mutex_adjust_prio(struct task_struct *p) +static __always_inline void rt_mutex_adjust_prio(struct rt_mutex_base *lock, + struct task_struct *p) { struct task_struct *pi_task = NULL; + lockdep_assert_held(&lock->wait_lock); + lockdep_assert(rt_mutex_owner(lock) == p); lockdep_assert_held(&p->pi_lock); if (task_has_pi_waiters(p)) @@ -571,9 +603,14 @@ static __always_inline struct rt_mutex_base *task_blocked_on_lock(struct task_st * Chain walk basics and protection scope * * [R] refcount on task - * [P] task->pi_lock held + * [Pn] task->pi_lock held * [L] rtmutex->wait_lock held * + * Normal locking order: + * + * rtmutex->wait_lock + * task->pi_lock + * * Step Description Protected by * function arguments: * @task [R] @@ -588,27 +625,32 @@ static __always_inline struct rt_mutex_base *task_blocked_on_lock(struct task_st * again: * loop_sanity_check(); * retry: - * [1] lock(task->pi_lock); [R] acquire [P] - * [2] waiter = task->pi_blocked_on; [P] - * [3] check_exit_conditions_1(); [P] - * [4] lock = waiter->lock; [P] - * [5] if (!try_lock(lock->wait_lock)) { [P] try to acquire [L] - * unlock(task->pi_lock); release [P] + * [1] lock(task->pi_lock); [R] acquire [P1] + * [2] waiter = task->pi_blocked_on; [P1] + * [3] check_exit_conditions_1(); [P1] + * [4] lock = waiter->lock; [P1] + * [5] if (!try_lock(lock->wait_lock)) { [P1] try to acquire [L] + * unlock(task->pi_lock); release [P1] * goto retry; * } - * [6] check_exit_conditions_2(); [P] + [L] - * [7] requeue_lock_waiter(lock, waiter); [P] + [L] - * [8] unlock(task->pi_lock); release [P] + * [6] check_exit_conditions_2(); [P1] + [L] + * [7] requeue_lock_waiter(lock, waiter); [P1] + [L] + * [8] unlock(task->pi_lock); release [P1] * put_task_struct(task); release [R] * [9] check_exit_conditions_3(); [L] * [10] task = owner(lock); [L] * get_task_struct(task); [L] acquire [R] - * lock(task->pi_lock); [L] acquire [P] - * [11] requeue_pi_waiter(tsk, waiters(lock));[P] + [L] - * [12] check_exit_conditions_4(); [P] + [L] - * [13] unlock(task->pi_lock); release [P] + * lock(task->pi_lock); [L] acquire [P2] + * [11] requeue_pi_waiter(tsk, waiters(lock));[P2] + [L] + * [12] check_exit_conditions_4(); [P2] + [L] + * [13] unlock(task->pi_lock); release [P2] * unlock(lock->wait_lock); release [L] * goto again; + * + * Where P1 is the blocking task and P2 is the lock owner; going up one step + * the owner becomes the next blocked task etc.. + * +* */ static int __sched rt_mutex_adjust_prio_chain(struct task_struct *task, enum rtmutex_chainwalk chwalk, @@ -756,7 +798,7 @@ static int __sched rt_mutex_adjust_prio_chain(struct task_struct *task, * enabled we continue, but stop the requeueing in the chain * walk. */ - if (rt_mutex_waiter_equal(waiter, task_to_waiter(task))) { + if (rt_waiter_node_equal(&waiter->tree, task_to_waiter_node(task))) { if (!detect_deadlock) goto out_unlock_pi; else @@ -764,13 +806,18 @@ static int __sched rt_mutex_adjust_prio_chain(struct task_struct *task, } /* - * [4] Get the next lock + * [4] Get the next lock; per holding task->pi_lock we can't unblock + * and guarantee @lock's existence. */ lock = waiter->lock; /* * [5] We need to trylock here as we are holding task->pi_lock, * which is the reverse lock order versus the other rtmutex * operations. + * + * Per the above, holding task->pi_lock guarantees lock exists, so + * inverting this lock order is infeasible from a life-time + * perspective. */ if (!raw_spin_trylock(&lock->wait_lock)) { raw_spin_unlock_irq(&task->pi_lock); @@ -874,17 +921,18 @@ static int __sched rt_mutex_adjust_prio_chain(struct task_struct *task, * or * * DL CBS enforcement advancing the effective deadline. - * - * Even though pi_waiters also uses these fields, and that tree is only - * updated in [11], we can do this here, since we hold [L], which - * serializes all pi_waiters access and rb_erase() does not care about - * the values of the node being removed. */ waiter_update_prio(waiter, task); rt_mutex_enqueue(lock, waiter); - /* [8] Release the task */ + /* + * [8] Release the (blocking) task in preparation for + * taking the owner task in [10]. + * + * Since we hold lock->waiter_lock, task cannot unblock, even if we + * release task->pi_lock. + */ raw_spin_unlock(&task->pi_lock); put_task_struct(task); @@ -908,7 +956,12 @@ static int __sched rt_mutex_adjust_prio_chain(struct task_struct *task, return 0; } - /* [10] Grab the next task, i.e. the owner of @lock */ + /* + * [10] Grab the next task, i.e. the owner of @lock + * + * Per holding lock->wait_lock and checking for !owner above, there + * must be an owner and it cannot go away. + */ task = get_task_struct(rt_mutex_owner(lock)); raw_spin_lock(&task->pi_lock); @@ -921,8 +974,9 @@ static int __sched rt_mutex_adjust_prio_chain(struct task_struct *task, * and adjust the priority of the owner. */ rt_mutex_dequeue_pi(task, prerequeue_top_waiter); + waiter_clone_prio(waiter, task); rt_mutex_enqueue_pi(task, waiter); - rt_mutex_adjust_prio(task); + rt_mutex_adjust_prio(lock, task); } else if (prerequeue_top_waiter == waiter) { /* @@ -937,8 +991,9 @@ static int __sched rt_mutex_adjust_prio_chain(struct task_struct *task, */ rt_mutex_dequeue_pi(task, waiter); waiter = rt_mutex_top_waiter(lock); + waiter_clone_prio(waiter, task); rt_mutex_enqueue_pi(task, waiter); - rt_mutex_adjust_prio(task); + rt_mutex_adjust_prio(lock, task); } else { /* * Nothing changed. No need to do any priority @@ -1154,6 +1209,7 @@ static int __sched task_blocks_on_rt_mutex(struct rt_mutex_base *lock, waiter->task = task; waiter->lock = lock; waiter_update_prio(waiter, task); + waiter_clone_prio(waiter, task); /* Get the top priority waiter on the lock */ if (rt_mutex_has_waiters(lock)) @@ -1187,7 +1243,7 @@ static int __sched task_blocks_on_rt_mutex(struct rt_mutex_base *lock, rt_mutex_dequeue_pi(owner, top_waiter); rt_mutex_enqueue_pi(owner, waiter); - rt_mutex_adjust_prio(owner); + rt_mutex_adjust_prio(lock, owner); if (owner->pi_blocked_on) chain_walk = 1; } else if (rt_mutex_cond_detect_deadlock(waiter, chwalk)) { @@ -1234,6 +1290,8 @@ static void __sched mark_wakeup_next_waiter(struct rt_wake_q_head *wqh, { struct rt_mutex_waiter *waiter; + lockdep_assert_held(&lock->wait_lock); + raw_spin_lock(¤t->pi_lock); waiter = rt_mutex_top_waiter(lock); @@ -1246,7 +1304,7 @@ static void __sched mark_wakeup_next_waiter(struct rt_wake_q_head *wqh, * task unblocks. */ rt_mutex_dequeue_pi(current, waiter); - rt_mutex_adjust_prio(current); + rt_mutex_adjust_prio(lock, current); /* * As we are waking up the top waiter, and the waiter stays @@ -1482,7 +1540,7 @@ static void __sched remove_waiter(struct rt_mutex_base *lock, if (rt_mutex_has_waiters(lock)) rt_mutex_enqueue_pi(owner, rt_mutex_top_waiter(lock)); - rt_mutex_adjust_prio(owner); + rt_mutex_adjust_prio(lock, owner); /* Store the lock on which owner is blocked or NULL */ next_lock = task_blocked_on_lock(owner); diff --git a/kernel/locking/rtmutex_api.c b/kernel/locking/rtmutex_api.c index cb9fdff76a8a..a6974d044593 100644 --- a/kernel/locking/rtmutex_api.c +++ b/kernel/locking/rtmutex_api.c @@ -459,7 +459,7 @@ void __sched rt_mutex_adjust_pi(struct task_struct *task) raw_spin_lock_irqsave(&task->pi_lock, flags); waiter = task->pi_blocked_on; - if (!waiter || rt_mutex_waiter_equal(waiter, task_to_waiter(task))) { + if (!waiter || rt_waiter_node_equal(&waiter->tree, task_to_waiter_node(task))) { raw_spin_unlock_irqrestore(&task->pi_lock, flags); return; } diff --git a/kernel/locking/rtmutex_common.h b/kernel/locking/rtmutex_common.h index c47e8361bfb5..1162e07cdaea 100644 --- a/kernel/locking/rtmutex_common.h +++ b/kernel/locking/rtmutex_common.h @@ -17,27 +17,44 @@ #include <linux/rtmutex.h> #include <linux/sched/wake_q.h> + +/* + * This is a helper for the struct rt_mutex_waiter below. A waiter goes in two + * separate trees and they need their own copy of the sort keys because of + * different locking requirements. + * + * @entry: rbtree node to enqueue into the waiters tree + * @prio: Priority of the waiter + * @deadline: Deadline of the waiter if applicable + * + * See rt_waiter_node_less() and waiter_*_prio(). + */ +struct rt_waiter_node { + struct rb_node entry; + int prio; + u64 deadline; +}; + /* * This is the control structure for tasks blocked on a rt_mutex, * which is allocated on the kernel stack on of the blocked task. * - * @tree_entry: pi node to enqueue into the mutex waiters tree - * @pi_tree_entry: pi node to enqueue into the mutex owner waiters tree + * @tree: node to enqueue into the mutex waiters tree + * @pi_tree: node to enqueue into the mutex owner waiters tree * @task: task reference to the blocked task * @lock: Pointer to the rt_mutex on which the waiter blocks * @wake_state: Wakeup state to use (TASK_NORMAL or TASK_RTLOCK_WAIT) - * @prio: Priority of the waiter - * @deadline: Deadline of the waiter if applicable * @ww_ctx: WW context pointer + * + * @tree is ordered by @lock->wait_lock + * @pi_tree is ordered by rt_mutex_owner(@lock)->pi_lock */ struct rt_mutex_waiter { - struct rb_node tree_entry; - struct rb_node pi_tree_entry; + struct rt_waiter_node tree; + struct rt_waiter_node pi_tree; struct task_struct *task; struct rt_mutex_base *lock; unsigned int wake_state; - int prio; - u64 deadline; struct ww_acquire_ctx *ww_ctx; }; @@ -105,7 +122,7 @@ static inline bool rt_mutex_waiter_is_top_waiter(struct rt_mutex_base *lock, { struct rb_node *leftmost = rb_first_cached(&lock->waiters); - return rb_entry(leftmost, struct rt_mutex_waiter, tree_entry) == waiter; + return rb_entry(leftmost, struct rt_mutex_waiter, tree.entry) == waiter; } static inline struct rt_mutex_waiter *rt_mutex_top_waiter(struct rt_mutex_base *lock) @@ -113,8 +130,10 @@ static inline struct rt_mutex_waiter *rt_mutex_top_waiter(struct rt_mutex_base * struct rb_node *leftmost = rb_first_cached(&lock->waiters); struct rt_mutex_waiter *w = NULL; + lockdep_assert_held(&lock->wait_lock); + if (leftmost) { - w = rb_entry(leftmost, struct rt_mutex_waiter, tree_entry); + w = rb_entry(leftmost, struct rt_mutex_waiter, tree.entry); BUG_ON(w->lock != lock); } return w; @@ -127,8 +146,10 @@ static inline int task_has_pi_waiters(struct task_struct *p) static inline struct rt_mutex_waiter *task_top_pi_waiter(struct task_struct *p) { + lockdep_assert_held(&p->pi_lock); + return rb_entry(p->pi_waiters.rb_leftmost, struct rt_mutex_waiter, - pi_tree_entry); + pi_tree.entry); } #define RT_MUTEX_HAS_WAITERS 1UL @@ -190,8 +211,8 @@ static inline void debug_rt_mutex_free_waiter(struct rt_mutex_waiter *waiter) static inline void rt_mutex_init_waiter(struct rt_mutex_waiter *waiter) { debug_rt_mutex_init_waiter(waiter); - RB_CLEAR_NODE(&waiter->pi_tree_entry); - RB_CLEAR_NODE(&waiter->tree_entry); + RB_CLEAR_NODE(&waiter->pi_tree.entry); + RB_CLEAR_NODE(&waiter->tree.entry); waiter->wake_state = TASK_NORMAL; waiter->task = NULL; } diff --git a/kernel/locking/ww_mutex.h b/kernel/locking/ww_mutex.h index 56f139201f24..3ad2cc4823e5 100644 --- a/kernel/locking/ww_mutex.h +++ b/kernel/locking/ww_mutex.h @@ -96,25 +96,25 @@ __ww_waiter_first(struct rt_mutex *lock) struct rb_node *n = rb_first(&lock->rtmutex.waiters.rb_root); if (!n) return NULL; - return rb_entry(n, struct rt_mutex_waiter, tree_entry); + return rb_entry(n, struct rt_mutex_waiter, tree.entry); } static inline struct rt_mutex_waiter * __ww_waiter_next(struct rt_mutex *lock, struct rt_mutex_waiter *w) { - struct rb_node *n = rb_next(&w->tree_entry); + struct rb_node *n = rb_next(&w->tree.entry); if (!n) return NULL; - return rb_entry(n, struct rt_mutex_waiter, tree_entry); + return rb_entry(n, struct rt_mutex_waiter, tree.entry); } static inline struct rt_mutex_waiter * __ww_waiter_prev(struct rt_mutex *lock, struct rt_mutex_waiter *w) { - struct rb_node *n = rb_prev(&w->tree_entry); + struct rb_node *n = rb_prev(&w->tree.entry); if (!n) return NULL; - return rb_entry(n, struct rt_mutex_waiter, tree_entry); + return rb_entry(n, struct rt_mutex_waiter, tree.entry); } static inline struct rt_mutex_waiter * @@ -123,7 +123,7 @@ __ww_waiter_last(struct rt_mutex *lock) struct rb_node *n = rb_last(&lock->rtmutex.waiters.rb_root); if (!n) return NULL; - return rb_entry(n, struct rt_mutex_waiter, tree_entry); + return rb_entry(n, struct rt_mutex_waiter, tree.entry); } static inline void diff --git a/kernel/nsproxy.c b/kernel/nsproxy.c index 80d9c6d77a45..15781acaac1c 100644 --- a/kernel/nsproxy.c +++ b/kernel/nsproxy.c @@ -30,7 +30,7 @@ static struct kmem_cache *nsproxy_cachep; struct nsproxy init_nsproxy = { - .count = ATOMIC_INIT(1), + .count = REFCOUNT_INIT(1), .uts_ns = &init_uts_ns, #if defined(CONFIG_POSIX_MQUEUE) || defined(CONFIG_SYSVIPC) .ipc_ns = &init_ipc_ns, @@ -55,7 +55,7 @@ static inline struct nsproxy *create_nsproxy(void) nsproxy = kmem_cache_alloc(nsproxy_cachep, GFP_KERNEL); if (nsproxy) - atomic_set(&nsproxy->count, 1); + refcount_set(&nsproxy->count, 1); return nsproxy; } diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index f62e89d0d906..2b4a946a6ff5 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -1166,7 +1166,7 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr, int error; if (!hibernation_available()) - return 0; + return n; if (len && buf[len-1] == '\n') len--; @@ -1179,6 +1179,7 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr, unsigned maj, min, offset; char *p, dummy; + error = 0; if (sscanf(name, "%u:%u%c", &maj, &min, &dummy) == 2 || sscanf(name, "%u:%u:%u:%c", &maj, &min, &offset, &dummy) == 3) { diff --git a/kernel/power/qos.c b/kernel/power/qos.c index af51ed6d45ef..4244b069442e 100644 --- a/kernel/power/qos.c +++ b/kernel/power/qos.c @@ -220,6 +220,11 @@ static struct pm_qos_constraints cpu_latency_constraints = { .type = PM_QOS_MIN, }; +static inline bool cpu_latency_qos_value_invalid(s32 value) +{ + return value < 0 && value != PM_QOS_DEFAULT_VALUE; +} + /** * cpu_latency_qos_limit - Return current system-wide CPU latency QoS limit. */ @@ -263,7 +268,7 @@ static void cpu_latency_qos_apply(struct pm_qos_request *req, */ void cpu_latency_qos_add_request(struct pm_qos_request *req, s32 value) { - if (!req) + if (!req || cpu_latency_qos_value_invalid(value)) return; if (cpu_latency_qos_request_active(req)) { @@ -289,7 +294,7 @@ EXPORT_SYMBOL_GPL(cpu_latency_qos_add_request); */ void cpu_latency_qos_update_request(struct pm_qos_request *req, s32 new_value) { - if (!req) + if (!req || cpu_latency_qos_value_invalid(new_value)) return; if (!cpu_latency_qos_request_active(req)) { @@ -426,6 +431,11 @@ late_initcall(cpu_latency_qos_init); /* Definitions related to the frequency QoS below. */ +static inline bool freq_qos_value_invalid(s32 value) +{ + return value < 0 && value != PM_QOS_DEFAULT_VALUE; +} + /** * freq_constraints_init - Initialize frequency QoS constraints. * @qos: Frequency QoS constraints to initialize. @@ -531,7 +541,7 @@ int freq_qos_add_request(struct freq_constraints *qos, { int ret; - if (IS_ERR_OR_NULL(qos) || !req || value < 0) + if (IS_ERR_OR_NULL(qos) || !req || freq_qos_value_invalid(value)) return -EINVAL; if (WARN(freq_qos_request_active(req), @@ -563,7 +573,7 @@ EXPORT_SYMBOL_GPL(freq_qos_add_request); */ int freq_qos_update_request(struct freq_qos_request *req, s32 new_value) { - if (!req || new_value < 0) + if (!req || freq_qos_value_invalid(new_value)) return -EINVAL; if (WARN(!freq_qos_request_active(req), diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 0415d5ecb977..87e9f7e2bdc0 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -404,6 +404,7 @@ struct bm_position { struct mem_zone_bm_rtree *zone; struct rtree_node *node; unsigned long node_pfn; + unsigned long cur_pfn; int node_bit; }; @@ -589,6 +590,7 @@ static void memory_bm_position_reset(struct memory_bitmap *bm) bm->cur.node = list_entry(bm->cur.zone->leaves.next, struct rtree_node, list); bm->cur.node_pfn = 0; + bm->cur.cur_pfn = BM_END_OF_MAP; bm->cur.node_bit = 0; } @@ -799,6 +801,7 @@ node_found: bm->cur.zone = zone; bm->cur.node = node; bm->cur.node_pfn = (pfn - zone->start_pfn) & ~BM_BLOCK_MASK; + bm->cur.cur_pfn = pfn; /* Set return values */ *addr = node->data; @@ -850,6 +853,11 @@ static void memory_bm_clear_current(struct memory_bitmap *bm) clear_bit(bit, bm->cur.node->data); } +static unsigned long memory_bm_get_current(struct memory_bitmap *bm) +{ + return bm->cur.cur_pfn; +} + static int memory_bm_test_bit(struct memory_bitmap *bm, unsigned long pfn) { void *addr; @@ -929,10 +937,12 @@ static unsigned long memory_bm_next_pfn(struct memory_bitmap *bm) if (bit < bits) { pfn = bm->cur.zone->start_pfn + bm->cur.node_pfn + bit; bm->cur.node_bit = bit + 1; + bm->cur.cur_pfn = pfn; return pfn; } } while (rtree_next_node(bm)); + bm->cur.cur_pfn = BM_END_OF_MAP; return BM_END_OF_MAP; } @@ -1423,14 +1433,19 @@ static unsigned int count_data_pages(void) /* * This is needed, because copy_page and memcpy are not usable for copying - * task structs. + * task structs. Returns true if the page was filled with only zeros, + * otherwise false. */ -static inline void do_copy_page(long *dst, long *src) +static inline bool do_copy_page(long *dst, long *src) { + long z = 0; int n; - for (n = PAGE_SIZE / sizeof(long); n; n--) + for (n = PAGE_SIZE / sizeof(long); n; n--) { + z |= *src; *dst++ = *src++; + } + return !z; } /** @@ -1439,17 +1454,21 @@ static inline void do_copy_page(long *dst, long *src) * Check if the page we are going to copy is marked as present in the kernel * page tables. This always is the case if CONFIG_DEBUG_PAGEALLOC or * CONFIG_ARCH_HAS_SET_DIRECT_MAP is not set. In that case kernel_page_present() - * always returns 'true'. + * always returns 'true'. Returns true if the page was entirely composed of + * zeros, otherwise it will return false. */ -static void safe_copy_page(void *dst, struct page *s_page) +static bool safe_copy_page(void *dst, struct page *s_page) { + bool zeros_only; + if (kernel_page_present(s_page)) { - do_copy_page(dst, page_address(s_page)); + zeros_only = do_copy_page(dst, page_address(s_page)); } else { hibernate_map_page(s_page); - do_copy_page(dst, page_address(s_page)); + zeros_only = do_copy_page(dst, page_address(s_page)); hibernate_unmap_page(s_page); } + return zeros_only; } #ifdef CONFIG_HIGHMEM @@ -1459,17 +1478,18 @@ static inline struct page *page_is_saveable(struct zone *zone, unsigned long pfn saveable_highmem_page(zone, pfn) : saveable_page(zone, pfn); } -static void copy_data_page(unsigned long dst_pfn, unsigned long src_pfn) +static bool copy_data_page(unsigned long dst_pfn, unsigned long src_pfn) { struct page *s_page, *d_page; void *src, *dst; + bool zeros_only; s_page = pfn_to_page(src_pfn); d_page = pfn_to_page(dst_pfn); if (PageHighMem(s_page)) { src = kmap_atomic(s_page); dst = kmap_atomic(d_page); - do_copy_page(dst, src); + zeros_only = do_copy_page(dst, src); kunmap_atomic(dst); kunmap_atomic(src); } else { @@ -1478,30 +1498,39 @@ static void copy_data_page(unsigned long dst_pfn, unsigned long src_pfn) * The page pointed to by src may contain some kernel * data modified by kmap_atomic() */ - safe_copy_page(buffer, s_page); + zeros_only = safe_copy_page(buffer, s_page); dst = kmap_atomic(d_page); copy_page(dst, buffer); kunmap_atomic(dst); } else { - safe_copy_page(page_address(d_page), s_page); + zeros_only = safe_copy_page(page_address(d_page), s_page); } } + return zeros_only; } #else #define page_is_saveable(zone, pfn) saveable_page(zone, pfn) -static inline void copy_data_page(unsigned long dst_pfn, unsigned long src_pfn) +static inline int copy_data_page(unsigned long dst_pfn, unsigned long src_pfn) { - safe_copy_page(page_address(pfn_to_page(dst_pfn)), + return safe_copy_page(page_address(pfn_to_page(dst_pfn)), pfn_to_page(src_pfn)); } #endif /* CONFIG_HIGHMEM */ -static void copy_data_pages(struct memory_bitmap *copy_bm, - struct memory_bitmap *orig_bm) +/* + * Copy data pages will copy all pages into pages pulled from the copy_bm. + * If a page was entirely filled with zeros it will be marked in the zero_bm. + * + * Returns the number of pages copied. + */ +static unsigned long copy_data_pages(struct memory_bitmap *copy_bm, + struct memory_bitmap *orig_bm, + struct memory_bitmap *zero_bm) { + unsigned long copied_pages = 0; struct zone *zone; - unsigned long pfn; + unsigned long pfn, copy_pfn; for_each_populated_zone(zone) { unsigned long max_zone_pfn; @@ -1514,18 +1543,29 @@ static void copy_data_pages(struct memory_bitmap *copy_bm, } memory_bm_position_reset(orig_bm); memory_bm_position_reset(copy_bm); + copy_pfn = memory_bm_next_pfn(copy_bm); for(;;) { pfn = memory_bm_next_pfn(orig_bm); if (unlikely(pfn == BM_END_OF_MAP)) break; - copy_data_page(memory_bm_next_pfn(copy_bm), pfn); + if (copy_data_page(copy_pfn, pfn)) { + memory_bm_set_bit(zero_bm, pfn); + /* Use this copy_pfn for a page that is not full of zeros */ + continue; + } + copied_pages++; + copy_pfn = memory_bm_next_pfn(copy_bm); } + return copied_pages; } /* Total number of image pages */ static unsigned int nr_copy_pages; /* Number of pages needed for saving the original pfns of the image pages */ static unsigned int nr_meta_pages; +/* Number of zero pages */ +static unsigned int nr_zero_pages; + /* * Numbers of normal and highmem page frames allocated for hibernation image * before suspending devices. @@ -1546,6 +1586,9 @@ static struct memory_bitmap orig_bm; */ static struct memory_bitmap copy_bm; +/* Memory bitmap which tracks which saveable pages were zero filled. */ +static struct memory_bitmap zero_bm; + /** * swsusp_free - Free pages allocated for hibernation image. * @@ -1590,6 +1633,7 @@ loop: out: nr_copy_pages = 0; nr_meta_pages = 0; + nr_zero_pages = 0; restore_pblist = NULL; buffer = NULL; alloc_normal = 0; @@ -1808,8 +1852,15 @@ int hibernate_preallocate_memory(void) goto err_out; } + error = memory_bm_create(&zero_bm, GFP_IMAGE, PG_ANY); + if (error) { + pr_err("Cannot allocate zero bitmap\n"); + goto err_out; + } + alloc_normal = 0; alloc_highmem = 0; + nr_zero_pages = 0; /* Count the number of saveable data pages. */ save_highmem = count_highmem_pages(); @@ -2089,19 +2140,19 @@ asmlinkage __visible int swsusp_save(void) * Kill them. */ drain_local_pages(NULL); - copy_data_pages(©_bm, &orig_bm); + nr_copy_pages = copy_data_pages(©_bm, &orig_bm, &zero_bm); /* * End of critical section. From now on, we can write to memory, * but we should not touch disk. This specially means we must _not_ * touch swap space! Except we must write out our image of course. */ - nr_pages += nr_highmem; - nr_copy_pages = nr_pages; + /* We don't actually copy the zero pages */ + nr_zero_pages = nr_pages - nr_copy_pages; nr_meta_pages = DIV_ROUND_UP(nr_pages * sizeof(long), PAGE_SIZE); - pr_info("Image created (%d pages copied)\n", nr_pages); + pr_info("Image created (%d pages copied, %d zero pages)\n", nr_copy_pages, nr_zero_pages); return 0; } @@ -2146,15 +2197,22 @@ static int init_header(struct swsusp_info *info) return init_header_complete(info); } +#define ENCODED_PFN_ZERO_FLAG ((unsigned long)1 << (BITS_PER_LONG - 1)) +#define ENCODED_PFN_MASK (~ENCODED_PFN_ZERO_FLAG) + /** * pack_pfns - Prepare PFNs for saving. * @bm: Memory bitmap. * @buf: Memory buffer to store the PFNs in. + * @zero_bm: Memory bitmap containing PFNs of zero pages. * * PFNs corresponding to set bits in @bm are stored in the area of memory - * pointed to by @buf (1 page at a time). + * pointed to by @buf (1 page at a time). Pages which were filled with only + * zeros will have the highest bit set in the packed format to distinguish + * them from PFNs which will be contained in the image file. */ -static inline void pack_pfns(unsigned long *buf, struct memory_bitmap *bm) +static inline void pack_pfns(unsigned long *buf, struct memory_bitmap *bm, + struct memory_bitmap *zero_bm) { int j; @@ -2162,6 +2220,8 @@ static inline void pack_pfns(unsigned long *buf, struct memory_bitmap *bm) buf[j] = memory_bm_next_pfn(bm); if (unlikely(buf[j] == BM_END_OF_MAP)) break; + if (memory_bm_test_bit(zero_bm, buf[j])) + buf[j] |= ENCODED_PFN_ZERO_FLAG; } } @@ -2203,7 +2263,7 @@ int snapshot_read_next(struct snapshot_handle *handle) memory_bm_position_reset(©_bm); } else if (handle->cur <= nr_meta_pages) { clear_page(buffer); - pack_pfns(buffer, &orig_bm); + pack_pfns(buffer, &orig_bm, &zero_bm); } else { struct page *page; @@ -2299,24 +2359,35 @@ static int load_header(struct swsusp_info *info) * unpack_orig_pfns - Set bits corresponding to given PFNs in a memory bitmap. * @bm: Memory bitmap. * @buf: Area of memory containing the PFNs. + * @zero_bm: Memory bitmap with the zero PFNs marked. * * For each element of the array pointed to by @buf (1 page at a time), set the - * corresponding bit in @bm. + * corresponding bit in @bm. If the page was originally populated with only + * zeros then a corresponding bit will also be set in @zero_bm. */ -static int unpack_orig_pfns(unsigned long *buf, struct memory_bitmap *bm) +static int unpack_orig_pfns(unsigned long *buf, struct memory_bitmap *bm, + struct memory_bitmap *zero_bm) { + unsigned long decoded_pfn; + bool zero; int j; for (j = 0; j < PAGE_SIZE / sizeof(long); j++) { if (unlikely(buf[j] == BM_END_OF_MAP)) break; - if (pfn_valid(buf[j]) && memory_bm_pfn_present(bm, buf[j])) { - memory_bm_set_bit(bm, buf[j]); + zero = !!(buf[j] & ENCODED_PFN_ZERO_FLAG); + decoded_pfn = buf[j] & ENCODED_PFN_MASK; + if (pfn_valid(decoded_pfn) && memory_bm_pfn_present(bm, decoded_pfn)) { + memory_bm_set_bit(bm, decoded_pfn); + if (zero) { + memory_bm_set_bit(zero_bm, decoded_pfn); + nr_zero_pages++; + } } else { - if (!pfn_valid(buf[j])) + if (!pfn_valid(decoded_pfn)) pr_err(FW_BUG "Memory map mismatch at 0x%llx after hibernation\n", - (unsigned long long)PFN_PHYS(buf[j])); + (unsigned long long)PFN_PHYS(decoded_pfn)); return -EFAULT; } } @@ -2538,6 +2609,7 @@ static inline void free_highmem_data(void) {} * prepare_image - Make room for loading hibernation image. * @new_bm: Uninitialized memory bitmap structure. * @bm: Memory bitmap with unsafe pages marked. + * @zero_bm: Memory bitmap containing the zero pages. * * Use @bm to mark the pages that will be overwritten in the process of * restoring the system memory state from the suspend image ("unsafe" pages) @@ -2548,10 +2620,15 @@ static inline void free_highmem_data(void) {} * pages will be used for just yet. Instead, we mark them all as allocated and * create a lists of "safe" pages to be used later. On systems with high * memory a list of "safe" highmem pages is created too. + * + * Because it was not known which pages were unsafe when @zero_bm was created, + * make a copy of it and recreate it within safe pages. */ -static int prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm) +static int prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm, + struct memory_bitmap *zero_bm) { unsigned int nr_pages, nr_highmem; + struct memory_bitmap tmp; struct linked_page *lp; int error; @@ -2568,6 +2645,24 @@ static int prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm) duplicate_memory_bitmap(new_bm, bm); memory_bm_free(bm, PG_UNSAFE_KEEP); + + /* Make a copy of zero_bm so it can be created in safe pages */ + error = memory_bm_create(&tmp, GFP_ATOMIC, PG_ANY); + if (error) + goto Free; + + duplicate_memory_bitmap(&tmp, zero_bm); + memory_bm_free(zero_bm, PG_UNSAFE_KEEP); + + /* Recreate zero_bm in safe pages */ + error = memory_bm_create(zero_bm, GFP_ATOMIC, PG_SAFE); + if (error) + goto Free; + + duplicate_memory_bitmap(zero_bm, &tmp); + memory_bm_free(&tmp, PG_UNSAFE_KEEP); + /* At this point zero_bm is in safe pages and it can be used for restoring. */ + if (nr_highmem > 0) { error = prepare_highmem_image(bm, &nr_highmem); if (error) @@ -2582,7 +2677,7 @@ static int prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm) * * nr_copy_pages cannot be less than allocated_unsafe_pages too. */ - nr_pages = nr_copy_pages - nr_highmem - allocated_unsafe_pages; + nr_pages = (nr_zero_pages + nr_copy_pages) - nr_highmem - allocated_unsafe_pages; nr_pages = DIV_ROUND_UP(nr_pages, PBES_PER_LINKED_PAGE); while (nr_pages > 0) { lp = get_image_page(GFP_ATOMIC, PG_SAFE); @@ -2595,7 +2690,7 @@ static int prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm) nr_pages--; } /* Preallocate memory for the image */ - nr_pages = nr_copy_pages - nr_highmem - allocated_unsafe_pages; + nr_pages = (nr_zero_pages + nr_copy_pages) - nr_highmem - allocated_unsafe_pages; while (nr_pages > 0) { lp = (struct linked_page *)get_zeroed_page(GFP_ATOMIC); if (!lp) { @@ -2683,8 +2778,9 @@ int snapshot_write_next(struct snapshot_handle *handle) static struct chain_allocator ca; int error = 0; +next: /* Check if we have already loaded the entire image */ - if (handle->cur > 1 && handle->cur > nr_meta_pages + nr_copy_pages) + if (handle->cur > 1 && handle->cur > nr_meta_pages + nr_copy_pages + nr_zero_pages) return 0; handle->sync_read = 1; @@ -2709,19 +2805,26 @@ int snapshot_write_next(struct snapshot_handle *handle) if (error) return error; + error = memory_bm_create(&zero_bm, GFP_ATOMIC, PG_ANY); + if (error) + return error; + + nr_zero_pages = 0; + hibernate_restore_protection_begin(); } else if (handle->cur <= nr_meta_pages + 1) { - error = unpack_orig_pfns(buffer, ©_bm); + error = unpack_orig_pfns(buffer, ©_bm, &zero_bm); if (error) return error; if (handle->cur == nr_meta_pages + 1) { - error = prepare_image(&orig_bm, ©_bm); + error = prepare_image(&orig_bm, ©_bm, &zero_bm); if (error) return error; chain_init(&ca, GFP_ATOMIC, PG_SAFE); memory_bm_position_reset(&orig_bm); + memory_bm_position_reset(&zero_bm); restore_pblist = NULL; handle->buffer = get_buffer(&orig_bm, &ca); handle->sync_read = 0; @@ -2738,6 +2841,14 @@ int snapshot_write_next(struct snapshot_handle *handle) handle->sync_read = 0; } handle->cur++; + + /* Zero pages were not included in the image, memset it and move on. */ + if (handle->cur > nr_meta_pages + 1 && + memory_bm_test_bit(&zero_bm, memory_bm_get_current(&orig_bm))) { + memset(handle->buffer, 0, PAGE_SIZE); + goto next; + } + return PAGE_SIZE; } @@ -2754,7 +2865,7 @@ void snapshot_write_finalize(struct snapshot_handle *handle) copy_last_highmem_page(); hibernate_restore_protect_page(handle->buffer); /* Do that only if we have loaded the image entirely */ - if (handle->cur > 1 && handle->cur > nr_meta_pages + nr_copy_pages) { + if (handle->cur > 1 && handle->cur > nr_meta_pages + nr_copy_pages + nr_zero_pages) { memory_bm_recycle(&orig_bm); free_highmem_data(); } @@ -2763,7 +2874,7 @@ void snapshot_write_finalize(struct snapshot_handle *handle) int snapshot_image_loaded(struct snapshot_handle *handle) { return !(!nr_copy_pages || !last_highmem_page_copied() || - handle->cur <= nr_meta_pages + nr_copy_pages); + handle->cur <= nr_meta_pages + nr_copy_pages + nr_zero_pages); } #ifdef CONFIG_HIGHMEM diff --git a/kernel/rcu/rcu.h b/kernel/rcu/rcu.h index 98c1544cf572..5befd8780dcd 100644 --- a/kernel/rcu/rcu.h +++ b/kernel/rcu/rcu.h @@ -511,6 +511,14 @@ static inline void show_rcu_tasks_gp_kthreads(void) {} void rcu_request_urgent_qs_task(struct task_struct *t); #endif /* #else #ifdef CONFIG_TINY_RCU */ +#ifdef CONFIG_TASKS_RCU +struct task_struct *get_rcu_tasks_gp_kthread(void); +#endif // # ifdef CONFIG_TASKS_RCU + +#ifdef CONFIG_TASKS_RUDE_RCU +struct task_struct *get_rcu_tasks_rude_gp_kthread(void); +#endif // # ifdef CONFIG_TASKS_RUDE_RCU + #define RCU_SCHEDULER_INACTIVE 0 #define RCU_SCHEDULER_INIT 1 #define RCU_SCHEDULER_RUNNING 2 diff --git a/kernel/rcu/rcuscale.c b/kernel/rcu/rcuscale.c index d1221731c7cf..ffdb30495e3c 100644 --- a/kernel/rcu/rcuscale.c +++ b/kernel/rcu/rcuscale.c @@ -84,15 +84,17 @@ MODULE_AUTHOR("Paul E. McKenney <paulmck@linux.ibm.com>"); #endif torture_param(bool, gp_async, false, "Use asynchronous GP wait primitives"); -torture_param(int, gp_async_max, 1000, "Max # outstanding waits per reader"); +torture_param(int, gp_async_max, 1000, "Max # outstanding waits per writer"); torture_param(bool, gp_exp, false, "Use expedited GP wait primitives"); torture_param(int, holdoff, 10, "Holdoff time before test start (s)"); +torture_param(int, minruntime, 0, "Minimum run time (s)"); torture_param(int, nreaders, -1, "Number of RCU reader threads"); torture_param(int, nwriters, -1, "Number of RCU updater threads"); torture_param(bool, shutdown, RCUSCALE_SHUTDOWN, "Shutdown at end of scalability tests."); torture_param(int, verbose, 1, "Enable verbose debugging printk()s"); torture_param(int, writer_holdoff, 0, "Holdoff (us) between GPs, zero to disable"); +torture_param(int, writer_holdoff_jiffies, 0, "Holdoff (jiffies) between GPs, zero to disable"); torture_param(int, kfree_rcu_test, 0, "Do we run a kfree_rcu() scale test?"); torture_param(int, kfree_mult, 1, "Multiple of kfree_obj size to allocate."); torture_param(int, kfree_by_call_rcu, 0, "Use call_rcu() to emulate kfree_rcu()?"); @@ -139,6 +141,7 @@ struct rcu_scale_ops { void (*gp_barrier)(void); void (*sync)(void); void (*exp_sync)(void); + struct task_struct *(*rso_gp_kthread)(void); const char *name; }; @@ -295,6 +298,7 @@ static struct rcu_scale_ops tasks_ops = { .gp_barrier = rcu_barrier_tasks, .sync = synchronize_rcu_tasks, .exp_sync = synchronize_rcu_tasks, + .rso_gp_kthread = get_rcu_tasks_gp_kthread, .name = "tasks" }; @@ -306,6 +310,44 @@ static struct rcu_scale_ops tasks_ops = { #endif // #else // #ifdef CONFIG_TASKS_RCU +#ifdef CONFIG_TASKS_RUDE_RCU + +/* + * Definitions for RCU-tasks-rude scalability testing. + */ + +static int tasks_rude_scale_read_lock(void) +{ + return 0; +} + +static void tasks_rude_scale_read_unlock(int idx) +{ +} + +static struct rcu_scale_ops tasks_rude_ops = { + .ptype = RCU_TASKS_RUDE_FLAVOR, + .init = rcu_sync_scale_init, + .readlock = tasks_rude_scale_read_lock, + .readunlock = tasks_rude_scale_read_unlock, + .get_gp_seq = rcu_no_completed, + .gp_diff = rcu_seq_diff, + .async = call_rcu_tasks_rude, + .gp_barrier = rcu_barrier_tasks_rude, + .sync = synchronize_rcu_tasks_rude, + .exp_sync = synchronize_rcu_tasks_rude, + .rso_gp_kthread = get_rcu_tasks_rude_gp_kthread, + .name = "tasks-rude" +}; + +#define TASKS_RUDE_OPS &tasks_rude_ops, + +#else // #ifdef CONFIG_TASKS_RUDE_RCU + +#define TASKS_RUDE_OPS + +#endif // #else // #ifdef CONFIG_TASKS_RUDE_RCU + #ifdef CONFIG_TASKS_TRACE_RCU /* @@ -334,6 +376,7 @@ static struct rcu_scale_ops tasks_tracing_ops = { .gp_barrier = rcu_barrier_tasks_trace, .sync = synchronize_rcu_tasks_trace, .exp_sync = synchronize_rcu_tasks_trace, + .rso_gp_kthread = get_rcu_tasks_trace_gp_kthread, .name = "tasks-tracing" }; @@ -410,10 +453,12 @@ rcu_scale_writer(void *arg) { int i = 0; int i_max; + unsigned long jdone; long me = (long)arg; struct rcu_head *rhp = NULL; bool started = false, done = false, alldone = false; u64 t; + DEFINE_TORTURE_RANDOM(tr); u64 *wdp; u64 *wdpp = writer_durations[me]; @@ -424,7 +469,7 @@ rcu_scale_writer(void *arg) sched_set_fifo_low(current); if (holdoff) - schedule_timeout_uninterruptible(holdoff * HZ); + schedule_timeout_idle(holdoff * HZ); /* * Wait until rcu_end_inkernel_boot() is called for normal GP tests @@ -445,9 +490,12 @@ rcu_scale_writer(void *arg) } } + jdone = jiffies + minruntime * HZ; do { if (writer_holdoff) udelay(writer_holdoff); + if (writer_holdoff_jiffies) + schedule_timeout_idle(torture_random(&tr) % writer_holdoff_jiffies + 1); wdp = &wdpp[i]; *wdp = ktime_get_mono_fast_ns(); if (gp_async) { @@ -475,7 +523,7 @@ retry: if (!started && atomic_read(&n_rcu_scale_writer_started) >= nrealwriters) started = true; - if (!done && i >= MIN_MEAS) { + if (!done && i >= MIN_MEAS && time_after(jiffies, jdone)) { done = true; sched_set_normal(current, 0); pr_alert("%s%s rcu_scale_writer %ld has %d measurements\n", @@ -518,8 +566,8 @@ static void rcu_scale_print_module_parms(struct rcu_scale_ops *cur_ops, const char *tag) { pr_alert("%s" SCALE_FLAG - "--- %s: nreaders=%d nwriters=%d verbose=%d shutdown=%d\n", - scale_type, tag, nrealreaders, nrealwriters, verbose, shutdown); + "--- %s: gp_async=%d gp_async_max=%d gp_exp=%d holdoff=%d minruntime=%d nreaders=%d nwriters=%d writer_holdoff=%d writer_holdoff_jiffies=%d verbose=%d shutdown=%d\n", + scale_type, tag, gp_async, gp_async_max, gp_exp, holdoff, minruntime, nrealreaders, nrealwriters, writer_holdoff, writer_holdoff_jiffies, verbose, shutdown); } /* @@ -556,6 +604,8 @@ static struct task_struct **kfree_reader_tasks; static int kfree_nrealthreads; static atomic_t n_kfree_scale_thread_started; static atomic_t n_kfree_scale_thread_ended; +static struct task_struct *kthread_tp; +static u64 kthread_stime; struct kfree_obj { char kfree_obj[8]; @@ -701,6 +751,10 @@ kfree_scale_init(void) unsigned long jif_start; unsigned long orig_jif; + pr_alert("%s" SCALE_FLAG + "--- kfree_rcu_test: kfree_mult=%d kfree_by_call_rcu=%d kfree_nthreads=%d kfree_alloc_num=%d kfree_loops=%d kfree_rcu_test_double=%d kfree_rcu_test_single=%d\n", + scale_type, kfree_mult, kfree_by_call_rcu, kfree_nthreads, kfree_alloc_num, kfree_loops, kfree_rcu_test_double, kfree_rcu_test_single); + // Also, do a quick self-test to ensure laziness is as much as // expected. if (kfree_by_call_rcu && !IS_ENABLED(CONFIG_RCU_LAZY)) { @@ -797,6 +851,18 @@ rcu_scale_cleanup(void) if (gp_exp && gp_async) SCALEOUT_ERRSTRING("No expedited async GPs, so went with async!"); + // If built-in, just report all of the GP kthread's CPU time. + if (IS_BUILTIN(CONFIG_RCU_SCALE_TEST) && !kthread_tp && cur_ops->rso_gp_kthread) + kthread_tp = cur_ops->rso_gp_kthread(); + if (kthread_tp) { + u32 ns; + u64 us; + + kthread_stime = kthread_tp->stime - kthread_stime; + us = div_u64_rem(kthread_stime, 1000, &ns); + pr_info("rcu_scale: Grace-period kthread CPU time: %llu.%03u us\n", us, ns); + show_rcu_gp_kthreads(); + } if (kfree_rcu_test) { kfree_scale_cleanup(); return; @@ -885,7 +951,7 @@ rcu_scale_init(void) long i; int firsterr = 0; static struct rcu_scale_ops *scale_ops[] = { - &rcu_ops, &srcu_ops, &srcud_ops, TASKS_OPS TASKS_TRACING_OPS + &rcu_ops, &srcu_ops, &srcud_ops, TASKS_OPS TASKS_RUDE_OPS TASKS_TRACING_OPS }; if (!torture_init_begin(scale_type, verbose)) @@ -910,6 +976,11 @@ rcu_scale_init(void) if (cur_ops->init) cur_ops->init(); + if (cur_ops->rso_gp_kthread) { + kthread_tp = cur_ops->rso_gp_kthread(); + if (kthread_tp) + kthread_stime = kthread_tp->stime; + } if (kfree_rcu_test) return kfree_scale_init(); diff --git a/kernel/rcu/rcutorture.c b/kernel/rcu/rcutorture.c index 147551c23baf..ade42d6a9d9b 100644 --- a/kernel/rcu/rcutorture.c +++ b/kernel/rcu/rcutorture.c @@ -1581,6 +1581,7 @@ rcu_torture_writer(void *arg) rcu_access_pointer(rcu_torture_current) != &rcu_tortures[i]) { tracing_off(); + show_rcu_gp_kthreads(); WARN(1, "%s: rtort_pipe_count: %d\n", __func__, rcu_tortures[i].rtort_pipe_count); rcu_ftrace_dump(DUMP_ALL); } @@ -1876,7 +1877,7 @@ static int rcutorture_extend_mask(int oldmask, struct torture_random_state *trsp) { int mask = rcutorture_extend_mask_max(); - unsigned long randmask1 = torture_random(trsp) >> 8; + unsigned long randmask1 = torture_random(trsp); unsigned long randmask2 = randmask1 >> 3; unsigned long preempts = RCUTORTURE_RDR_PREEMPT | RCUTORTURE_RDR_SCHED; unsigned long preempts_irq = preempts | RCUTORTURE_RDR_IRQ; @@ -1935,7 +1936,7 @@ rcutorture_loop_extend(int *readstate, struct torture_random_state *trsp, if (!((mask - 1) & mask)) return rtrsp; /* Current RCU reader not extendable. */ /* Bias towards larger numbers of loops. */ - i = (torture_random(trsp) >> 3); + i = torture_random(trsp); i = ((i | (i >> 3)) & RCUTORTURE_RDR_MAX_LOOPS) + 1; for (j = 0; j < i; j++) { mask = rcutorture_extend_mask(*readstate, trsp); @@ -2136,7 +2137,7 @@ static int rcu_nocb_toggle(void *arg) toggle_fuzz = NSEC_PER_USEC; do { r = torture_random(&rand); - cpu = (r >> 4) % (maxcpu + 1); + cpu = (r >> 1) % (maxcpu + 1); if (r & 0x1) { rcu_nocb_cpu_offload(cpu); atomic_long_inc(&n_nocb_offload); diff --git a/kernel/rcu/refscale.c b/kernel/rcu/refscale.c index 1970ce5f22d4..91a0fd0d4d9a 100644 --- a/kernel/rcu/refscale.c +++ b/kernel/rcu/refscale.c @@ -528,6 +528,38 @@ static struct ref_scale_ops clock_ops = { .name = "clock" }; +static void ref_jiffies_section(const int nloops) +{ + u64 x = 0; + int i; + + preempt_disable(); + for (i = nloops; i >= 0; i--) + x += jiffies; + preempt_enable(); + stopopts = x; +} + +static void ref_jiffies_delay_section(const int nloops, const int udl, const int ndl) +{ + u64 x = 0; + int i; + + preempt_disable(); + for (i = nloops; i >= 0; i--) { + x += jiffies; + un_delay(udl, ndl); + } + preempt_enable(); + stopopts = x; +} + +static struct ref_scale_ops jiffies_ops = { + .readsection = ref_jiffies_section, + .delaysection = ref_jiffies_delay_section, + .name = "jiffies" +}; + //////////////////////////////////////////////////////////////////////// // // Methods leveraging SLAB_TYPESAFE_BY_RCU. @@ -1047,7 +1079,7 @@ ref_scale_init(void) int firsterr = 0; static struct ref_scale_ops *scale_ops[] = { &rcu_ops, &srcu_ops, RCU_TRACE_OPS RCU_TASKS_OPS &refcnt_ops, &rwlock_ops, - &rwsem_ops, &lock_ops, &lock_irq_ops, &acqrel_ops, &clock_ops, + &rwsem_ops, &lock_ops, &lock_irq_ops, &acqrel_ops, &clock_ops, &jiffies_ops, &typesafe_ref_ops, &typesafe_lock_ops, &typesafe_seqlock_ops, }; @@ -1107,12 +1139,11 @@ ref_scale_init(void) VERBOSE_SCALEOUT("Starting %d reader threads", nreaders); for (i = 0; i < nreaders; i++) { + init_waitqueue_head(&reader_tasks[i].wq); firsterr = torture_create_kthread(ref_scale_reader, (void *)i, reader_tasks[i].task); if (torture_init_error(firsterr)) goto unwind; - - init_waitqueue_head(&(reader_tasks[i].wq)); } // Main Task diff --git a/kernel/rcu/tasks.h b/kernel/rcu/tasks.h index b770add3f843..8d65f7d576a3 100644 --- a/kernel/rcu/tasks.h +++ b/kernel/rcu/tasks.h @@ -25,6 +25,8 @@ typedef void (*postgp_func_t)(struct rcu_tasks *rtp); * @cblist: Callback list. * @lock: Lock protecting per-CPU callback list. * @rtp_jiffies: Jiffies counter value for statistics. + * @lazy_timer: Timer to unlazify callbacks. + * @urgent_gp: Number of additional non-lazy grace periods. * @rtp_n_lock_retries: Rough lock-contention statistic. * @rtp_work: Work queue for invoking callbacks. * @rtp_irq_work: IRQ work queue for deferred wakeups. @@ -38,6 +40,8 @@ struct rcu_tasks_percpu { raw_spinlock_t __private lock; unsigned long rtp_jiffies; unsigned long rtp_n_lock_retries; + struct timer_list lazy_timer; + unsigned int urgent_gp; struct work_struct rtp_work; struct irq_work rtp_irq_work; struct rcu_head barrier_q_head; @@ -51,7 +55,6 @@ struct rcu_tasks_percpu { * @cbs_wait: RCU wait allowing a new callback to get kthread's attention. * @cbs_gbl_lock: Lock protecting callback list. * @tasks_gp_mutex: Mutex protecting grace period, needed during mid-boot dead zone. - * @kthread_ptr: This flavor's grace-period/callback-invocation kthread. * @gp_func: This flavor's grace-period-wait function. * @gp_state: Grace period's most recent state transition (debugging). * @gp_sleep: Per-grace-period sleep to prevent CPU-bound looping. @@ -61,6 +64,8 @@ struct rcu_tasks_percpu { * @tasks_gp_seq: Number of grace periods completed since boot. * @n_ipis: Number of IPIs sent to encourage grace periods to end. * @n_ipis_fails: Number of IPI-send failures. + * @kthread_ptr: This flavor's grace-period/callback-invocation kthread. + * @lazy_jiffies: Number of jiffies to allow callbacks to be lazy. * @pregp_func: This flavor's pre-grace-period function (optional). * @pertask_func: This flavor's per-task scan function (optional). * @postscan_func: This flavor's post-task scan function (optional). @@ -92,6 +97,7 @@ struct rcu_tasks { unsigned long n_ipis; unsigned long n_ipis_fails; struct task_struct *kthread_ptr; + unsigned long lazy_jiffies; rcu_tasks_gp_func_t gp_func; pregp_func_t pregp_func; pertask_func_t pertask_func; @@ -127,6 +133,7 @@ static struct rcu_tasks rt_name = \ .gp_func = gp, \ .call_func = call, \ .rtpcpu = &rt_name ## __percpu, \ + .lazy_jiffies = DIV_ROUND_UP(HZ, 4), \ .name = n, \ .percpu_enqueue_shift = order_base_2(CONFIG_NR_CPUS), \ .percpu_enqueue_lim = 1, \ @@ -139,9 +146,7 @@ static struct rcu_tasks rt_name = \ #ifdef CONFIG_TASKS_RCU /* Track exiting tasks in order to allow them to be waited for. */ DEFINE_STATIC_SRCU(tasks_rcu_exit_srcu); -#endif -#ifdef CONFIG_TASKS_RCU /* Report delay in synchronize_srcu() completion in rcu_tasks_postscan(). */ static void tasks_rcu_exit_srcu_stall(struct timer_list *unused); static DEFINE_TIMER(tasks_rcu_exit_srcu_stall_timer, tasks_rcu_exit_srcu_stall); @@ -171,6 +176,8 @@ static int rcu_task_contend_lim __read_mostly = 100; module_param(rcu_task_contend_lim, int, 0444); static int rcu_task_collapse_lim __read_mostly = 10; module_param(rcu_task_collapse_lim, int, 0444); +static int rcu_task_lazy_lim __read_mostly = 32; +module_param(rcu_task_lazy_lim, int, 0444); /* RCU tasks grace-period state for debugging. */ #define RTGS_INIT 0 @@ -229,7 +236,7 @@ static const char *tasks_gp_state_getname(struct rcu_tasks *rtp) #endif /* #ifndef CONFIG_TINY_RCU */ // Initialize per-CPU callback lists for the specified flavor of -// Tasks RCU. +// Tasks RCU. Do not enqueue callbacks before this function is invoked. static void cblist_init_generic(struct rcu_tasks *rtp) { int cpu; @@ -237,7 +244,6 @@ static void cblist_init_generic(struct rcu_tasks *rtp) int lim; int shift; - raw_spin_lock_irqsave(&rtp->cbs_gbl_lock, flags); if (rcu_task_enqueue_lim < 0) { rcu_task_enqueue_lim = 1; rcu_task_cb_adjust = true; @@ -260,22 +266,48 @@ static void cblist_init_generic(struct rcu_tasks *rtp) WARN_ON_ONCE(!rtpcp); if (cpu) raw_spin_lock_init(&ACCESS_PRIVATE(rtpcp, lock)); - raw_spin_lock_rcu_node(rtpcp); // irqs already disabled. + local_irq_save(flags); // serialize initialization if (rcu_segcblist_empty(&rtpcp->cblist)) rcu_segcblist_init(&rtpcp->cblist); + local_irq_restore(flags); INIT_WORK(&rtpcp->rtp_work, rcu_tasks_invoke_cbs_wq); rtpcp->cpu = cpu; rtpcp->rtpp = rtp; if (!rtpcp->rtp_blkd_tasks.next) INIT_LIST_HEAD(&rtpcp->rtp_blkd_tasks); - raw_spin_unlock_rcu_node(rtpcp); // irqs remain disabled. } - raw_spin_unlock_irqrestore(&rtp->cbs_gbl_lock, flags); pr_info("%s: Setting shift to %d and lim to %d rcu_task_cb_adjust=%d.\n", rtp->name, data_race(rtp->percpu_enqueue_shift), data_race(rtp->percpu_enqueue_lim), rcu_task_cb_adjust); } +// Compute wakeup time for lazy callback timer. +static unsigned long rcu_tasks_lazy_time(struct rcu_tasks *rtp) +{ + return jiffies + rtp->lazy_jiffies; +} + +// Timer handler that unlazifies lazy callbacks. +static void call_rcu_tasks_generic_timer(struct timer_list *tlp) +{ + unsigned long flags; + bool needwake = false; + struct rcu_tasks *rtp; + struct rcu_tasks_percpu *rtpcp = from_timer(rtpcp, tlp, lazy_timer); + + rtp = rtpcp->rtpp; + raw_spin_lock_irqsave_rcu_node(rtpcp, flags); + if (!rcu_segcblist_empty(&rtpcp->cblist) && rtp->lazy_jiffies) { + if (!rtpcp->urgent_gp) + rtpcp->urgent_gp = 1; + needwake = true; + mod_timer(&rtpcp->lazy_timer, rcu_tasks_lazy_time(rtp)); + } + raw_spin_unlock_irqrestore_rcu_node(rtpcp, flags); + if (needwake) + rcuwait_wake_up(&rtp->cbs_wait); +} + // IRQ-work handler that does deferred wakeup for call_rcu_tasks_generic(). static void call_rcu_tasks_iw_wakeup(struct irq_work *iwp) { @@ -292,6 +324,7 @@ static void call_rcu_tasks_generic(struct rcu_head *rhp, rcu_callback_t func, { int chosen_cpu; unsigned long flags; + bool havekthread = smp_load_acquire(&rtp->kthread_ptr); int ideal_cpu; unsigned long j; bool needadjust = false; @@ -316,12 +349,19 @@ static void call_rcu_tasks_generic(struct rcu_head *rhp, rcu_callback_t func, READ_ONCE(rtp->percpu_enqueue_lim) != nr_cpu_ids) needadjust = true; // Defer adjustment to avoid deadlock. } - if (!rcu_segcblist_is_enabled(&rtpcp->cblist)) { - raw_spin_unlock_rcu_node(rtpcp); // irqs remain disabled. - cblist_init_generic(rtp); - raw_spin_lock_rcu_node(rtpcp); // irqs already disabled. + // Queuing callbacks before initialization not yet supported. + if (WARN_ON_ONCE(!rcu_segcblist_is_enabled(&rtpcp->cblist))) + rcu_segcblist_init(&rtpcp->cblist); + needwake = (func == wakeme_after_rcu) || + (rcu_segcblist_n_cbs(&rtpcp->cblist) == rcu_task_lazy_lim); + if (havekthread && !needwake && !timer_pending(&rtpcp->lazy_timer)) { + if (rtp->lazy_jiffies) + mod_timer(&rtpcp->lazy_timer, rcu_tasks_lazy_time(rtp)); + else + needwake = rcu_segcblist_empty(&rtpcp->cblist); } - needwake = rcu_segcblist_empty(&rtpcp->cblist); + if (needwake) + rtpcp->urgent_gp = 3; rcu_segcblist_enqueue(&rtpcp->cblist, rhp); raw_spin_unlock_irqrestore_rcu_node(rtpcp, flags); if (unlikely(needadjust)) { @@ -415,9 +455,14 @@ static int rcu_tasks_need_gpcb(struct rcu_tasks *rtp) } rcu_segcblist_advance(&rtpcp->cblist, rcu_seq_current(&rtp->tasks_gp_seq)); (void)rcu_segcblist_accelerate(&rtpcp->cblist, rcu_seq_snap(&rtp->tasks_gp_seq)); - if (rcu_segcblist_pend_cbs(&rtpcp->cblist)) + if (rtpcp->urgent_gp > 0 && rcu_segcblist_pend_cbs(&rtpcp->cblist)) { + if (rtp->lazy_jiffies) + rtpcp->urgent_gp--; needgpcb |= 0x3; - if (!rcu_segcblist_empty(&rtpcp->cblist)) + } else if (rcu_segcblist_empty(&rtpcp->cblist)) { + rtpcp->urgent_gp = 0; + } + if (rcu_segcblist_ready_cbs(&rtpcp->cblist)) needgpcb |= 0x1; raw_spin_unlock_irqrestore_rcu_node(rtpcp, flags); } @@ -525,10 +570,12 @@ static void rcu_tasks_one_gp(struct rcu_tasks *rtp, bool midboot) if (unlikely(midboot)) { needgpcb = 0x2; } else { + mutex_unlock(&rtp->tasks_gp_mutex); set_tasks_gp_state(rtp, RTGS_WAIT_CBS); rcuwait_wait_event(&rtp->cbs_wait, (needgpcb = rcu_tasks_need_gpcb(rtp)), TASK_IDLE); + mutex_lock(&rtp->tasks_gp_mutex); } if (needgpcb & 0x2) { @@ -549,11 +596,19 @@ static void rcu_tasks_one_gp(struct rcu_tasks *rtp, bool midboot) // RCU-tasks kthread that detects grace periods and invokes callbacks. static int __noreturn rcu_tasks_kthread(void *arg) { + int cpu; struct rcu_tasks *rtp = arg; + for_each_possible_cpu(cpu) { + struct rcu_tasks_percpu *rtpcp = per_cpu_ptr(rtp->rtpcpu, cpu); + + timer_setup(&rtpcp->lazy_timer, call_rcu_tasks_generic_timer, 0); + rtpcp->urgent_gp = 1; + } + /* Run on housekeeping CPUs by default. Sysadm can move if desired. */ housekeeping_affine(current, HK_TYPE_RCU); - WRITE_ONCE(rtp->kthread_ptr, current); // Let GPs start! + smp_store_release(&rtp->kthread_ptr, current); // Let GPs start! /* * Each pass through the following loop makes one check for @@ -635,16 +690,22 @@ static void show_rcu_tasks_generic_gp_kthread(struct rcu_tasks *rtp, char *s) { int cpu; bool havecbs = false; + bool haveurgent = false; + bool haveurgentcbs = false; for_each_possible_cpu(cpu) { struct rcu_tasks_percpu *rtpcp = per_cpu_ptr(rtp->rtpcpu, cpu); - if (!data_race(rcu_segcblist_empty(&rtpcp->cblist))) { + if (!data_race(rcu_segcblist_empty(&rtpcp->cblist))) havecbs = true; + if (data_race(rtpcp->urgent_gp)) + haveurgent = true; + if (!data_race(rcu_segcblist_empty(&rtpcp->cblist)) && data_race(rtpcp->urgent_gp)) + haveurgentcbs = true; + if (havecbs && haveurgent && haveurgentcbs) break; - } } - pr_info("%s: %s(%d) since %lu g:%lu i:%lu/%lu %c%c %s\n", + pr_info("%s: %s(%d) since %lu g:%lu i:%lu/%lu %c%c%c%c l:%lu %s\n", rtp->kname, tasks_gp_state_getname(rtp), data_race(rtp->gp_state), jiffies - data_race(rtp->gp_jiffies), @@ -652,6 +713,9 @@ static void show_rcu_tasks_generic_gp_kthread(struct rcu_tasks *rtp, char *s) data_race(rtp->n_ipis_fails), data_race(rtp->n_ipis), ".k"[!!data_race(rtp->kthread_ptr)], ".C"[havecbs], + ".u"[haveurgent], + ".U"[haveurgentcbs], + rtp->lazy_jiffies, s); } #endif // #ifndef CONFIG_TINY_RCU @@ -1020,11 +1084,16 @@ void rcu_barrier_tasks(void) } EXPORT_SYMBOL_GPL(rcu_barrier_tasks); +int rcu_tasks_lazy_ms = -1; +module_param(rcu_tasks_lazy_ms, int, 0444); + static int __init rcu_spawn_tasks_kthread(void) { cblist_init_generic(&rcu_tasks); rcu_tasks.gp_sleep = HZ / 10; rcu_tasks.init_fract = HZ / 10; + if (rcu_tasks_lazy_ms >= 0) + rcu_tasks.lazy_jiffies = msecs_to_jiffies(rcu_tasks_lazy_ms); rcu_tasks.pregp_func = rcu_tasks_pregp_step; rcu_tasks.pertask_func = rcu_tasks_pertask; rcu_tasks.postscan_func = rcu_tasks_postscan; @@ -1042,6 +1111,12 @@ void show_rcu_tasks_classic_gp_kthread(void) EXPORT_SYMBOL_GPL(show_rcu_tasks_classic_gp_kthread); #endif // !defined(CONFIG_TINY_RCU) +struct task_struct *get_rcu_tasks_gp_kthread(void) +{ + return rcu_tasks.kthread_ptr; +} +EXPORT_SYMBOL_GPL(get_rcu_tasks_gp_kthread); + /* * Contribute to protect against tasklist scan blind spot while the * task is exiting and may be removed from the tasklist. See @@ -1173,10 +1248,15 @@ void rcu_barrier_tasks_rude(void) } EXPORT_SYMBOL_GPL(rcu_barrier_tasks_rude); +int rcu_tasks_rude_lazy_ms = -1; +module_param(rcu_tasks_rude_lazy_ms, int, 0444); + static int __init rcu_spawn_tasks_rude_kthread(void) { cblist_init_generic(&rcu_tasks_rude); rcu_tasks_rude.gp_sleep = HZ / 10; + if (rcu_tasks_rude_lazy_ms >= 0) + rcu_tasks_rude.lazy_jiffies = msecs_to_jiffies(rcu_tasks_rude_lazy_ms); rcu_spawn_tasks_kthread_generic(&rcu_tasks_rude); return 0; } @@ -1188,6 +1268,13 @@ void show_rcu_tasks_rude_gp_kthread(void) } EXPORT_SYMBOL_GPL(show_rcu_tasks_rude_gp_kthread); #endif // !defined(CONFIG_TINY_RCU) + +struct task_struct *get_rcu_tasks_rude_gp_kthread(void) +{ + return rcu_tasks_rude.kthread_ptr; +} +EXPORT_SYMBOL_GPL(get_rcu_tasks_rude_gp_kthread); + #endif /* #ifdef CONFIG_TASKS_RUDE_RCU */ //////////////////////////////////////////////////////////////////////// @@ -1793,6 +1880,9 @@ void rcu_barrier_tasks_trace(void) } EXPORT_SYMBOL_GPL(rcu_barrier_tasks_trace); +int rcu_tasks_trace_lazy_ms = -1; +module_param(rcu_tasks_trace_lazy_ms, int, 0444); + static int __init rcu_spawn_tasks_trace_kthread(void) { cblist_init_generic(&rcu_tasks_trace); @@ -1807,6 +1897,8 @@ static int __init rcu_spawn_tasks_trace_kthread(void) if (rcu_tasks_trace.init_fract <= 0) rcu_tasks_trace.init_fract = 1; } + if (rcu_tasks_trace_lazy_ms >= 0) + rcu_tasks_trace.lazy_jiffies = msecs_to_jiffies(rcu_tasks_trace_lazy_ms); rcu_tasks_trace.pregp_func = rcu_tasks_trace_pregp_step; rcu_tasks_trace.postscan_func = rcu_tasks_trace_postscan; rcu_tasks_trace.holdouts_func = check_all_holdout_tasks_trace; @@ -1830,6 +1922,12 @@ void show_rcu_tasks_trace_gp_kthread(void) EXPORT_SYMBOL_GPL(show_rcu_tasks_trace_gp_kthread); #endif // !defined(CONFIG_TINY_RCU) +struct task_struct *get_rcu_tasks_trace_gp_kthread(void) +{ + return rcu_tasks_trace.kthread_ptr; +} +EXPORT_SYMBOL_GPL(get_rcu_tasks_trace_gp_kthread); + #else /* #ifdef CONFIG_TASKS_TRACE_RCU */ static void exit_tasks_rcu_finish_trace(struct task_struct *t) { } #endif /* #else #ifdef CONFIG_TASKS_TRACE_RCU */ diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 1449cb69a0e0..cb1caefa8bd0 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -632,7 +632,7 @@ void __rcu_irq_enter_check_tick(void) // prevents self-deadlock. So we can safely recheck under the lock. // Note that the nohz_full state currently cannot change. raw_spin_lock_rcu_node(rdp->mynode); - if (rdp->rcu_urgent_qs && !rdp->rcu_forced_tick) { + if (READ_ONCE(rdp->rcu_urgent_qs) && !rdp->rcu_forced_tick) { // A nohz_full CPU is in the kernel and RCU needs a // quiescent state. Turn on the tick! WRITE_ONCE(rdp->rcu_forced_tick, true); @@ -677,12 +677,16 @@ static void rcu_disable_urgency_upon_qs(struct rcu_data *rdp) } /** - * rcu_is_watching - see if RCU thinks that the current CPU is not idle + * rcu_is_watching - RCU read-side critical sections permitted on current CPU? * - * Return true if RCU is watching the running CPU, which means that this - * CPU can safely enter RCU read-side critical sections. In other words, - * if the current CPU is not in its idle loop or is in an interrupt or - * NMI handler, return true. + * Return @true if RCU is watching the running CPU and @false otherwise. + * An @true return means that this CPU can safely enter RCU read-side + * critical sections. + * + * Although calls to rcu_is_watching() from most parts of the kernel + * will return @true, there are important exceptions. For example, if the + * current CPU is deep within its idle loop, in kernel entry/exit code, + * or offline, rcu_is_watching() will return @false. * * Make notrace because it can be called by the internal functions of * ftrace, and making this notrace removes unnecessary recursion calls. diff --git a/kernel/rcu/tree_nocb.h b/kernel/rcu/tree_nocb.h index 43229d2b0c44..5598212d1f27 100644 --- a/kernel/rcu/tree_nocb.h +++ b/kernel/rcu/tree_nocb.h @@ -77,9 +77,9 @@ __setup("rcu_nocbs", rcu_nocb_setup); static int __init parse_rcu_nocb_poll(char *arg) { rcu_nocb_poll = true; - return 0; + return 1; } -early_param("rcu_nocb_poll", parse_rcu_nocb_poll); +__setup("rcu_nocb_poll", parse_rcu_nocb_poll); /* * Don't bother bypassing ->cblist if the call_rcu() rate is low. diff --git a/kernel/scftorture.c b/kernel/scftorture.c index 5d113aa59e77..59032aaccd18 100644 --- a/kernel/scftorture.c +++ b/kernel/scftorture.c @@ -171,7 +171,8 @@ static void scf_torture_stats_print(void) scfs.n_all_wait += scf_stats_p[i].n_all_wait; } if (atomic_read(&n_errs) || atomic_read(&n_mb_in_errs) || - atomic_read(&n_mb_out_errs) || atomic_read(&n_alloc_errs)) + atomic_read(&n_mb_out_errs) || + (!IS_ENABLED(CONFIG_KASAN) && atomic_read(&n_alloc_errs))) bangstr = "!!! "; pr_alert("%s %sscf_invoked_count %s: %lld resched: %lld single: %lld/%lld single_ofl: %lld/%lld single_rpc: %lld single_rpc_ofl: %lld many: %lld/%lld all: %lld/%lld ", SCFTORT_FLAG, bangstr, isdone ? "VER" : "ver", invoked_count, scfs.n_resched, @@ -312,6 +313,7 @@ static void scf_handler_1(void *scfc_in) // Randomly do an smp_call_function*() invocation. static void scftorture_invoke_one(struct scf_statistics *scfp, struct torture_random_state *trsp) { + bool allocfail = false; uintptr_t cpu; int ret = 0; struct scf_check *scfcp = NULL; @@ -323,8 +325,10 @@ static void scftorture_invoke_one(struct scf_statistics *scfp, struct torture_ra preempt_disable(); if (scfsp->scfs_prim == SCF_PRIM_SINGLE || scfsp->scfs_wait) { scfcp = kmalloc(sizeof(*scfcp), GFP_ATOMIC); - if (WARN_ON_ONCE(!scfcp)) { + if (!scfcp) { + WARN_ON_ONCE(!IS_ENABLED(CONFIG_KASAN)); atomic_inc(&n_alloc_errs); + allocfail = true; } else { scfcp->scfc_cpu = -1; scfcp->scfc_wait = scfsp->scfs_wait; @@ -431,7 +435,9 @@ static void scftorture_invoke_one(struct scf_statistics *scfp, struct torture_ra cpus_read_unlock(); else preempt_enable(); - if (!(torture_random(trsp) & 0xfff)) + if (allocfail) + schedule_timeout_idle((1 + longwait) * HZ); // Let no-wait handlers complete. + else if (!(torture_random(trsp) & 0xfff)) schedule_timeout_uninterruptible(1); } diff --git a/kernel/sched/completion.c b/kernel/sched/completion.c index d57a5c1c1cd9..3561ab533dd4 100644 --- a/kernel/sched/completion.c +++ b/kernel/sched/completion.c @@ -13,6 +13,23 @@ * Waiting for completion is a typically sync point, but not an exclusion point. */ +static void complete_with_flags(struct completion *x, int wake_flags) +{ + unsigned long flags; + + raw_spin_lock_irqsave(&x->wait.lock, flags); + + if (x->done != UINT_MAX) + x->done++; + swake_up_locked(&x->wait, wake_flags); + raw_spin_unlock_irqrestore(&x->wait.lock, flags); +} + +void complete_on_current_cpu(struct completion *x) +{ + return complete_with_flags(x, WF_CURRENT_CPU); +} + /** * complete: - signals a single thread waiting on this completion * @x: holds the state of this particular completion @@ -27,14 +44,7 @@ */ void complete(struct completion *x) { - unsigned long flags; - - raw_spin_lock_irqsave(&x->wait.lock, flags); - - if (x->done != UINT_MAX) - x->done++; - swake_up_locked(&x->wait); - raw_spin_unlock_irqrestore(&x->wait.lock, flags); + complete_with_flags(x, 0); } EXPORT_SYMBOL(complete); diff --git a/kernel/sched/core.c b/kernel/sched/core.c index c52c2eba7c73..2299a5cfbfb9 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -1097,25 +1097,22 @@ int get_nohz_timer_target(void) hk_mask = housekeeping_cpumask(HK_TYPE_TIMER); - rcu_read_lock(); + guard(rcu)(); + for_each_domain(cpu, sd) { for_each_cpu_and(i, sched_domain_span(sd), hk_mask) { if (cpu == i) continue; - if (!idle_cpu(i)) { - cpu = i; - goto unlock; - } + if (!idle_cpu(i)) + return i; } } if (default_cpu == -1) default_cpu = housekeeping_any_cpu(HK_TYPE_TIMER); - cpu = default_cpu; -unlock: - rcu_read_unlock(); - return cpu; + + return default_cpu; } /* @@ -1194,6 +1191,20 @@ static void nohz_csd_func(void *info) #endif /* CONFIG_NO_HZ_COMMON */ #ifdef CONFIG_NO_HZ_FULL +static inline bool __need_bw_check(struct rq *rq, struct task_struct *p) +{ + if (rq->nr_running != 1) + return false; + + if (p->sched_class != &fair_sched_class) + return false; + + if (!task_on_rq_queued(p)) + return false; + + return true; +} + bool sched_can_stop_tick(struct rq *rq) { int fifo_nr_running; @@ -1229,6 +1240,18 @@ bool sched_can_stop_tick(struct rq *rq) if (rq->nr_running > 1) return false; + /* + * If there is one task and it has CFS runtime bandwidth constraints + * and it's on the cpu now we don't want to stop the tick. + * This check prevents clearing the bit if a newly enqueued task here is + * dequeued by migrating while the constrained task continues to run. + * E.g. going from 2->1 without going through pick_next_task(). + */ + if (sched_feat(HZ_BW) && __need_bw_check(rq, rq->curr)) { + if (cfs_task_bw_constrained(rq->curr)) + return false; + } + return true; } #endif /* CONFIG_NO_HZ_FULL */ @@ -1804,7 +1827,8 @@ static int sysctl_sched_uclamp_handler(struct ctl_table *table, int write, int old_min, old_max, old_min_rt; int result; - mutex_lock(&uclamp_mutex); + guard(mutex)(&uclamp_mutex); + old_min = sysctl_sched_uclamp_util_min; old_max = sysctl_sched_uclamp_util_max; old_min_rt = sysctl_sched_uclamp_util_min_rt_default; @@ -1813,7 +1837,7 @@ static int sysctl_sched_uclamp_handler(struct ctl_table *table, int write, if (result) goto undo; if (!write) - goto done; + return 0; if (sysctl_sched_uclamp_util_min > sysctl_sched_uclamp_util_max || sysctl_sched_uclamp_util_max > SCHED_CAPACITY_SCALE || @@ -1849,16 +1873,12 @@ static int sysctl_sched_uclamp_handler(struct ctl_table *table, int write, * Otherwise, keep it simple and do just a lazy update at each next * task enqueue time. */ - - goto done; + return 0; undo: sysctl_sched_uclamp_util_min = old_min; sysctl_sched_uclamp_util_max = old_max; sysctl_sched_uclamp_util_min_rt_default = old_min_rt; -done: - mutex_unlock(&uclamp_mutex); - return result; } #endif @@ -3413,7 +3433,6 @@ static int migrate_swap_stop(void *data) { struct migration_swap_arg *arg = data; struct rq *src_rq, *dst_rq; - int ret = -EAGAIN; if (!cpu_active(arg->src_cpu) || !cpu_active(arg->dst_cpu)) return -EAGAIN; @@ -3421,33 +3440,25 @@ static int migrate_swap_stop(void *data) src_rq = cpu_rq(arg->src_cpu); dst_rq = cpu_rq(arg->dst_cpu); - double_raw_lock(&arg->src_task->pi_lock, - &arg->dst_task->pi_lock); - double_rq_lock(src_rq, dst_rq); + guard(double_raw_spinlock)(&arg->src_task->pi_lock, &arg->dst_task->pi_lock); + guard(double_rq_lock)(src_rq, dst_rq); if (task_cpu(arg->dst_task) != arg->dst_cpu) - goto unlock; + return -EAGAIN; if (task_cpu(arg->src_task) != arg->src_cpu) - goto unlock; + return -EAGAIN; if (!cpumask_test_cpu(arg->dst_cpu, arg->src_task->cpus_ptr)) - goto unlock; + return -EAGAIN; if (!cpumask_test_cpu(arg->src_cpu, arg->dst_task->cpus_ptr)) - goto unlock; + return -EAGAIN; __migrate_swap_task(arg->src_task, arg->dst_cpu); __migrate_swap_task(arg->dst_task, arg->src_cpu); - ret = 0; - -unlock: - double_rq_unlock(src_rq, dst_rq); - raw_spin_unlock(&arg->dst_task->pi_lock); - raw_spin_unlock(&arg->src_task->pi_lock); - - return ret; + return 0; } /* @@ -3722,14 +3733,14 @@ ttwu_stat(struct task_struct *p, int cpu, int wake_flags) struct sched_domain *sd; __schedstat_inc(p->stats.nr_wakeups_remote); - rcu_read_lock(); + + guard(rcu)(); for_each_domain(rq->cpu, sd) { if (cpumask_test_cpu(cpu, sched_domain_span(sd))) { __schedstat_inc(sd->ttwu_wake_remote); break; } } - rcu_read_unlock(); } if (wake_flags & WF_MIGRATED) @@ -3928,21 +3939,13 @@ static void __ttwu_queue_wakelist(struct task_struct *p, int cpu, int wake_flags void wake_up_if_idle(int cpu) { struct rq *rq = cpu_rq(cpu); - struct rq_flags rf; - - rcu_read_lock(); - if (!is_idle_task(rcu_dereference(rq->curr))) - goto out; - - rq_lock_irqsave(rq, &rf); - if (is_idle_task(rq->curr)) - resched_curr(rq); - /* Else CPU is not idle, do nothing here: */ - rq_unlock_irqrestore(rq, &rf); - -out: - rcu_read_unlock(); + guard(rcu)(); + if (is_idle_task(rcu_dereference(rq->curr))) { + guard(rq_lock_irqsave)(rq); + if (is_idle_task(rq->curr)) + resched_curr(rq); + } } bool cpus_share_cache(int this_cpu, int that_cpu) @@ -4193,13 +4196,11 @@ bool ttwu_state_match(struct task_struct *p, unsigned int state, int *success) * Return: %true if @p->state changes (an actual wakeup was done), * %false otherwise. */ -static int -try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) +int try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) { - unsigned long flags; + guard(preempt)(); int cpu, success = 0; - preempt_disable(); if (p == current) { /* * We're waking current, this means 'p->on_rq' and 'task_cpu(p) @@ -4226,129 +4227,127 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) * reordered with p->state check below. This pairs with smp_store_mb() * in set_current_state() that the waiting thread does. */ - raw_spin_lock_irqsave(&p->pi_lock, flags); - smp_mb__after_spinlock(); - if (!ttwu_state_match(p, state, &success)) - goto unlock; + scoped_guard (raw_spinlock_irqsave, &p->pi_lock) { + smp_mb__after_spinlock(); + if (!ttwu_state_match(p, state, &success)) + break; - trace_sched_waking(p); + trace_sched_waking(p); - /* - * Ensure we load p->on_rq _after_ p->state, otherwise it would - * be possible to, falsely, observe p->on_rq == 0 and get stuck - * in smp_cond_load_acquire() below. - * - * sched_ttwu_pending() try_to_wake_up() - * STORE p->on_rq = 1 LOAD p->state - * UNLOCK rq->lock - * - * __schedule() (switch to task 'p') - * LOCK rq->lock smp_rmb(); - * smp_mb__after_spinlock(); - * UNLOCK rq->lock - * - * [task p] - * STORE p->state = UNINTERRUPTIBLE LOAD p->on_rq - * - * Pairs with the LOCK+smp_mb__after_spinlock() on rq->lock in - * __schedule(). See the comment for smp_mb__after_spinlock(). - * - * A similar smb_rmb() lives in try_invoke_on_locked_down_task(). - */ - smp_rmb(); - if (READ_ONCE(p->on_rq) && ttwu_runnable(p, wake_flags)) - goto unlock; + /* + * Ensure we load p->on_rq _after_ p->state, otherwise it would + * be possible to, falsely, observe p->on_rq == 0 and get stuck + * in smp_cond_load_acquire() below. + * + * sched_ttwu_pending() try_to_wake_up() + * STORE p->on_rq = 1 LOAD p->state + * UNLOCK rq->lock + * + * __schedule() (switch to task 'p') + * LOCK rq->lock smp_rmb(); + * smp_mb__after_spinlock(); + * UNLOCK rq->lock + * + * [task p] + * STORE p->state = UNINTERRUPTIBLE LOAD p->on_rq + * + * Pairs with the LOCK+smp_mb__after_spinlock() on rq->lock in + * __schedule(). See the comment for smp_mb__after_spinlock(). + * + * A similar smb_rmb() lives in try_invoke_on_locked_down_task(). + */ + smp_rmb(); + if (READ_ONCE(p->on_rq) && ttwu_runnable(p, wake_flags)) + break; #ifdef CONFIG_SMP - /* - * Ensure we load p->on_cpu _after_ p->on_rq, otherwise it would be - * possible to, falsely, observe p->on_cpu == 0. - * - * One must be running (->on_cpu == 1) in order to remove oneself - * from the runqueue. - * - * __schedule() (switch to task 'p') try_to_wake_up() - * STORE p->on_cpu = 1 LOAD p->on_rq - * UNLOCK rq->lock - * - * __schedule() (put 'p' to sleep) - * LOCK rq->lock smp_rmb(); - * smp_mb__after_spinlock(); - * STORE p->on_rq = 0 LOAD p->on_cpu - * - * Pairs with the LOCK+smp_mb__after_spinlock() on rq->lock in - * __schedule(). See the comment for smp_mb__after_spinlock(). - * - * Form a control-dep-acquire with p->on_rq == 0 above, to ensure - * schedule()'s deactivate_task() has 'happened' and p will no longer - * care about it's own p->state. See the comment in __schedule(). - */ - smp_acquire__after_ctrl_dep(); + /* + * Ensure we load p->on_cpu _after_ p->on_rq, otherwise it would be + * possible to, falsely, observe p->on_cpu == 0. + * + * One must be running (->on_cpu == 1) in order to remove oneself + * from the runqueue. + * + * __schedule() (switch to task 'p') try_to_wake_up() + * STORE p->on_cpu = 1 LOAD p->on_rq + * UNLOCK rq->lock + * + * __schedule() (put 'p' to sleep) + * LOCK rq->lock smp_rmb(); + * smp_mb__after_spinlock(); + * STORE p->on_rq = 0 LOAD p->on_cpu + * + * Pairs with the LOCK+smp_mb__after_spinlock() on rq->lock in + * __schedule(). See the comment for smp_mb__after_spinlock(). + * + * Form a control-dep-acquire with p->on_rq == 0 above, to ensure + * schedule()'s deactivate_task() has 'happened' and p will no longer + * care about it's own p->state. See the comment in __schedule(). + */ + smp_acquire__after_ctrl_dep(); - /* - * We're doing the wakeup (@success == 1), they did a dequeue (p->on_rq - * == 0), which means we need to do an enqueue, change p->state to - * TASK_WAKING such that we can unlock p->pi_lock before doing the - * enqueue, such as ttwu_queue_wakelist(). - */ - WRITE_ONCE(p->__state, TASK_WAKING); + /* + * We're doing the wakeup (@success == 1), they did a dequeue (p->on_rq + * == 0), which means we need to do an enqueue, change p->state to + * TASK_WAKING such that we can unlock p->pi_lock before doing the + * enqueue, such as ttwu_queue_wakelist(). + */ + WRITE_ONCE(p->__state, TASK_WAKING); - /* - * If the owning (remote) CPU is still in the middle of schedule() with - * this task as prev, considering queueing p on the remote CPUs wake_list - * which potentially sends an IPI instead of spinning on p->on_cpu to - * let the waker make forward progress. This is safe because IRQs are - * disabled and the IPI will deliver after on_cpu is cleared. - * - * Ensure we load task_cpu(p) after p->on_cpu: - * - * set_task_cpu(p, cpu); - * STORE p->cpu = @cpu - * __schedule() (switch to task 'p') - * LOCK rq->lock - * smp_mb__after_spin_lock() smp_cond_load_acquire(&p->on_cpu) - * STORE p->on_cpu = 1 LOAD p->cpu - * - * to ensure we observe the correct CPU on which the task is currently - * scheduling. - */ - if (smp_load_acquire(&p->on_cpu) && - ttwu_queue_wakelist(p, task_cpu(p), wake_flags)) - goto unlock; + /* + * If the owning (remote) CPU is still in the middle of schedule() with + * this task as prev, considering queueing p on the remote CPUs wake_list + * which potentially sends an IPI instead of spinning on p->on_cpu to + * let the waker make forward progress. This is safe because IRQs are + * disabled and the IPI will deliver after on_cpu is cleared. + * + * Ensure we load task_cpu(p) after p->on_cpu: + * + * set_task_cpu(p, cpu); + * STORE p->cpu = @cpu + * __schedule() (switch to task 'p') + * LOCK rq->lock + * smp_mb__after_spin_lock() smp_cond_load_acquire(&p->on_cpu) + * STORE p->on_cpu = 1 LOAD p->cpu + * + * to ensure we observe the correct CPU on which the task is currently + * scheduling. + */ + if (smp_load_acquire(&p->on_cpu) && + ttwu_queue_wakelist(p, task_cpu(p), wake_flags)) + break; - /* - * If the owning (remote) CPU is still in the middle of schedule() with - * this task as prev, wait until it's done referencing the task. - * - * Pairs with the smp_store_release() in finish_task(). - * - * This ensures that tasks getting woken will be fully ordered against - * their previous state and preserve Program Order. - */ - smp_cond_load_acquire(&p->on_cpu, !VAL); + /* + * If the owning (remote) CPU is still in the middle of schedule() with + * this task as prev, wait until it's done referencing the task. + * + * Pairs with the smp_store_release() in finish_task(). + * + * This ensures that tasks getting woken will be fully ordered against + * their previous state and preserve Program Order. + */ + smp_cond_load_acquire(&p->on_cpu, !VAL); - cpu = select_task_rq(p, p->wake_cpu, wake_flags | WF_TTWU); - if (task_cpu(p) != cpu) { - if (p->in_iowait) { - delayacct_blkio_end(p); - atomic_dec(&task_rq(p)->nr_iowait); - } + cpu = select_task_rq(p, p->wake_cpu, wake_flags | WF_TTWU); + if (task_cpu(p) != cpu) { + if (p->in_iowait) { + delayacct_blkio_end(p); + atomic_dec(&task_rq(p)->nr_iowait); + } - wake_flags |= WF_MIGRATED; - psi_ttwu_dequeue(p); - set_task_cpu(p, cpu); - } + wake_flags |= WF_MIGRATED; + psi_ttwu_dequeue(p); + set_task_cpu(p, cpu); + } #else - cpu = task_cpu(p); + cpu = task_cpu(p); #endif /* CONFIG_SMP */ - ttwu_queue(p, cpu, wake_flags); -unlock: - raw_spin_unlock_irqrestore(&p->pi_lock, flags); + ttwu_queue(p, cpu, wake_flags); + } out: if (success) ttwu_stat(p, task_cpu(p), wake_flags); - preempt_enable(); return success; } @@ -4501,6 +4500,8 @@ static void __sched_fork(unsigned long clone_flags, struct task_struct *p) p->se.prev_sum_exec_runtime = 0; p->se.nr_migrations = 0; p->se.vruntime = 0; + p->se.vlag = 0; + p->se.slice = sysctl_sched_base_slice; INIT_LIST_HEAD(&p->se.group_node); #ifdef CONFIG_FAIR_GROUP_SCHED @@ -5496,23 +5497,20 @@ unsigned int nr_iowait(void) void sched_exec(void) { struct task_struct *p = current; - unsigned long flags; + struct migration_arg arg; int dest_cpu; - raw_spin_lock_irqsave(&p->pi_lock, flags); - dest_cpu = p->sched_class->select_task_rq(p, task_cpu(p), WF_EXEC); - if (dest_cpu == smp_processor_id()) - goto unlock; + scoped_guard (raw_spinlock_irqsave, &p->pi_lock) { + dest_cpu = p->sched_class->select_task_rq(p, task_cpu(p), WF_EXEC); + if (dest_cpu == smp_processor_id()) + return; - if (likely(cpu_active(dest_cpu))) { - struct migration_arg arg = { p, dest_cpu }; + if (unlikely(!cpu_active(dest_cpu))) + return; - raw_spin_unlock_irqrestore(&p->pi_lock, flags); - stop_one_cpu(task_cpu(p), migration_cpu_stop, &arg); - return; + arg = (struct migration_arg){ p, dest_cpu }; } -unlock: - raw_spin_unlock_irqrestore(&p->pi_lock, flags); + stop_one_cpu(task_cpu(p), migration_cpu_stop, &arg); } #endif @@ -5722,9 +5720,6 @@ static void sched_tick_remote(struct work_struct *work) struct tick_work *twork = container_of(dwork, struct tick_work, work); int cpu = twork->cpu; struct rq *rq = cpu_rq(cpu); - struct task_struct *curr; - struct rq_flags rf; - u64 delta; int os; /* @@ -5734,30 +5729,26 @@ static void sched_tick_remote(struct work_struct *work) * statistics and checks timeslices in a time-independent way, regardless * of when exactly it is running. */ - if (!tick_nohz_tick_stopped_cpu(cpu)) - goto out_requeue; + if (tick_nohz_tick_stopped_cpu(cpu)) { + guard(rq_lock_irq)(rq); + struct task_struct *curr = rq->curr; - rq_lock_irq(rq, &rf); - curr = rq->curr; - if (cpu_is_offline(cpu)) - goto out_unlock; + if (cpu_online(cpu)) { + update_rq_clock(rq); - update_rq_clock(rq); + if (!is_idle_task(curr)) { + /* + * Make sure the next tick runs within a + * reasonable amount of time. + */ + u64 delta = rq_clock_task(rq) - curr->se.exec_start; + WARN_ON_ONCE(delta > (u64)NSEC_PER_SEC * 3); + } + curr->sched_class->task_tick(rq, curr, 0); - if (!is_idle_task(curr)) { - /* - * Make sure the next tick runs within a reasonable - * amount of time. - */ - delta = rq_clock_task(rq) - curr->se.exec_start; - WARN_ON_ONCE(delta > (u64)NSEC_PER_SEC * 3); + calc_load_nohz_remote(rq); + } } - curr->sched_class->task_tick(rq, curr, 0); - - calc_load_nohz_remote(rq); -out_unlock: - rq_unlock_irq(rq, &rf); -out_requeue: /* * Run the remote tick once per second (1Hz). This arbitrary @@ -6306,19 +6297,19 @@ static bool try_steal_cookie(int this, int that) unsigned long cookie; bool success = false; - local_irq_disable(); - double_rq_lock(dst, src); + guard(irq)(); + guard(double_rq_lock)(dst, src); cookie = dst->core->core_cookie; if (!cookie) - goto unlock; + return false; if (dst->curr != dst->idle) - goto unlock; + return false; p = sched_core_find(src, cookie); if (!p) - goto unlock; + return false; do { if (p == src->core_pick || p == src->curr) @@ -6330,9 +6321,10 @@ static bool try_steal_cookie(int this, int that) if (p->core_occupation > dst->idle->core_occupation) goto next; /* - * sched_core_find() and sched_core_next() will ensure that task @p - * is not throttled now, we also need to check whether the runqueue - * of the destination CPU is being throttled. + * sched_core_find() and sched_core_next() will ensure + * that task @p is not throttled now, we also need to + * check whether the runqueue of the destination CPU is + * being throttled. */ if (sched_task_is_throttled(p, this)) goto next; @@ -6350,10 +6342,6 @@ next: p = sched_core_next(p, cookie); } while (p); -unlock: - double_rq_unlock(dst, src); - local_irq_enable(); - return success; } @@ -6411,20 +6399,24 @@ static void queue_core_balance(struct rq *rq) queue_balance_callback(rq, &per_cpu(core_balance_head, rq->cpu), sched_core_balance); } +DEFINE_LOCK_GUARD_1(core_lock, int, + sched_core_lock(*_T->lock, &_T->flags), + sched_core_unlock(*_T->lock, &_T->flags), + unsigned long flags) + static void sched_core_cpu_starting(unsigned int cpu) { const struct cpumask *smt_mask = cpu_smt_mask(cpu); struct rq *rq = cpu_rq(cpu), *core_rq = NULL; - unsigned long flags; int t; - sched_core_lock(cpu, &flags); + guard(core_lock)(&cpu); WARN_ON_ONCE(rq->core != rq); /* if we're the first, we'll be our own leader */ if (cpumask_weight(smt_mask) == 1) - goto unlock; + return; /* find the leader */ for_each_cpu(t, smt_mask) { @@ -6438,7 +6430,7 @@ static void sched_core_cpu_starting(unsigned int cpu) } if (WARN_ON_ONCE(!core_rq)) /* whoopsie */ - goto unlock; + return; /* install and validate core_rq */ for_each_cpu(t, smt_mask) { @@ -6449,29 +6441,25 @@ static void sched_core_cpu_starting(unsigned int cpu) WARN_ON_ONCE(rq->core != core_rq); } - -unlock: - sched_core_unlock(cpu, &flags); } static void sched_core_cpu_deactivate(unsigned int cpu) { const struct cpumask *smt_mask = cpu_smt_mask(cpu); struct rq *rq = cpu_rq(cpu), *core_rq = NULL; - unsigned long flags; int t; - sched_core_lock(cpu, &flags); + guard(core_lock)(&cpu); /* if we're the last man standing, nothing to do */ if (cpumask_weight(smt_mask) == 1) { WARN_ON_ONCE(rq->core != rq); - goto unlock; + return; } /* if we're not the leader, nothing to do */ if (rq->core != rq) - goto unlock; + return; /* find a new leader */ for_each_cpu(t, smt_mask) { @@ -6482,7 +6470,7 @@ static void sched_core_cpu_deactivate(unsigned int cpu) } if (WARN_ON_ONCE(!core_rq)) /* impossible */ - goto unlock; + return; /* copy the shared state to the new leader */ core_rq->core_task_seq = rq->core_task_seq; @@ -6504,9 +6492,6 @@ static void sched_core_cpu_deactivate(unsigned int cpu) rq = cpu_rq(t); rq->core = core_rq; } - -unlock: - sched_core_unlock(cpu, &flags); } static inline void sched_core_cpu_dying(unsigned int cpu) @@ -7030,7 +7015,7 @@ asmlinkage __visible void __sched preempt_schedule_irq(void) int default_wake_function(wait_queue_entry_t *curr, unsigned mode, int wake_flags, void *key) { - WARN_ON_ONCE(IS_ENABLED(CONFIG_SCHED_DEBUG) && wake_flags & ~WF_SYNC); + WARN_ON_ONCE(IS_ENABLED(CONFIG_SCHED_DEBUG) && wake_flags & ~(WF_SYNC|WF_CURRENT_CPU)); return try_to_wake_up(curr->private, mode, wake_flags); } EXPORT_SYMBOL(default_wake_function); @@ -7383,6 +7368,19 @@ struct task_struct *idle_task(int cpu) return cpu_rq(cpu)->idle; } +#ifdef CONFIG_SCHED_CORE +int sched_core_idle_cpu(int cpu) +{ + struct rq *rq = cpu_rq(cpu); + + if (sched_core_enabled(rq) && rq->curr == rq->idle) + return 1; + + return idle_cpu(cpu); +} + +#endif + #ifdef CONFIG_SMP /* * This function computes an effective utilization for the given CPU, to be @@ -9940,7 +9938,7 @@ void __init sched_init(void) ptr += nr_cpu_ids * sizeof(void **); root_task_group.shares = ROOT_TASK_GROUP_LOAD; - init_cfs_bandwidth(&root_task_group.cfs_bandwidth); + init_cfs_bandwidth(&root_task_group.cfs_bandwidth, NULL); #endif /* CONFIG_FAIR_GROUP_SCHED */ #ifdef CONFIG_RT_GROUP_SCHED root_task_group.rt_se = (struct sched_rt_entity **)ptr; @@ -11074,11 +11072,16 @@ static int tg_cfs_schedulable_down(struct task_group *tg, void *data) /* * Ensure max(child_quota) <= parent_quota. On cgroup2, - * always take the min. On cgroup1, only inherit when no - * limit is set: + * always take the non-RUNTIME_INF min. On cgroup1, only + * inherit when no limit is set. In both cases this is used + * by the scheduler to determine if a given CFS task has a + * bandwidth constraint at some higher level. */ if (cgroup_subsys_on_dfl(cpu_cgrp_subsys)) { - quota = min(quota, parent_quota); + if (quota == RUNTIME_INF) + quota = parent_quota; + else if (parent_quota != RUNTIME_INF) + quota = min(quota, parent_quota); } else { if (quota == RUNTIME_INF) quota = parent_quota; @@ -11139,6 +11142,27 @@ static int cpu_cfs_stat_show(struct seq_file *sf, void *v) return 0; } + +static u64 throttled_time_self(struct task_group *tg) +{ + int i; + u64 total = 0; + + for_each_possible_cpu(i) { + total += READ_ONCE(tg->cfs_rq[i]->throttled_clock_self_time); + } + + return total; +} + +static int cpu_cfs_local_stat_show(struct seq_file *sf, void *v) +{ + struct task_group *tg = css_tg(seq_css(sf)); + + seq_printf(sf, "throttled_time %llu\n", throttled_time_self(tg)); + + return 0; +} #endif /* CONFIG_CFS_BANDWIDTH */ #endif /* CONFIG_FAIR_GROUP_SCHED */ @@ -11215,6 +11239,10 @@ static struct cftype cpu_legacy_files[] = { .name = "stat", .seq_show = cpu_cfs_stat_show, }, + { + .name = "stat.local", + .seq_show = cpu_cfs_local_stat_show, + }, #endif #ifdef CONFIG_RT_GROUP_SCHED { @@ -11271,6 +11299,24 @@ static int cpu_extra_stat_show(struct seq_file *sf, return 0; } +static int cpu_local_stat_show(struct seq_file *sf, + struct cgroup_subsys_state *css) +{ +#ifdef CONFIG_CFS_BANDWIDTH + { + struct task_group *tg = css_tg(css); + u64 throttled_self_usec; + + throttled_self_usec = throttled_time_self(tg); + do_div(throttled_self_usec, NSEC_PER_USEC); + + seq_printf(sf, "throttled_usec %llu\n", + throttled_self_usec); + } +#endif + return 0; +} + #ifdef CONFIG_FAIR_GROUP_SCHED static u64 cpu_weight_read_u64(struct cgroup_subsys_state *css, struct cftype *cft) @@ -11449,6 +11495,7 @@ struct cgroup_subsys cpu_cgrp_subsys = { .css_released = cpu_cgroup_css_released, .css_free = cpu_cgroup_css_free, .css_extra_stat_show = cpu_extra_stat_show, + .css_local_stat_show = cpu_local_stat_show, #ifdef CONFIG_RT_GROUP_SCHED .can_attach = cpu_cgroup_can_attach, #endif diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index 066ff1c8ae4e..4c3d0d9f3db6 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -347,10 +347,7 @@ static __init int sched_init_debug(void) debugfs_create_file("preempt", 0644, debugfs_sched, NULL, &sched_dynamic_fops); #endif - debugfs_create_u32("latency_ns", 0644, debugfs_sched, &sysctl_sched_latency); - debugfs_create_u32("min_granularity_ns", 0644, debugfs_sched, &sysctl_sched_min_granularity); - debugfs_create_u32("idle_min_granularity_ns", 0644, debugfs_sched, &sysctl_sched_idle_min_granularity); - debugfs_create_u32("wakeup_granularity_ns", 0644, debugfs_sched, &sysctl_sched_wakeup_granularity); + debugfs_create_u32("base_slice_ns", 0644, debugfs_sched, &sysctl_sched_base_slice); debugfs_create_u32("latency_warn_ms", 0644, debugfs_sched, &sysctl_resched_latency_warn_ms); debugfs_create_u32("latency_warn_once", 0644, debugfs_sched, &sysctl_resched_latency_warn_once); @@ -427,6 +424,7 @@ static void register_sd(struct sched_domain *sd, struct dentry *parent) #undef SDM debugfs_create_file("flags", 0444, parent, &sd->flags, &sd_flags_fops); + debugfs_create_file("groups_flags", 0444, parent, &sd->groups->flags, &sd_flags_fops); } void update_sched_domain_debugfs(void) @@ -581,9 +579,13 @@ print_task(struct seq_file *m, struct rq *rq, struct task_struct *p) else SEQ_printf(m, " %c", task_state_to_char(p)); - SEQ_printf(m, " %15s %5d %9Ld.%06ld %9Ld %5d ", + SEQ_printf(m, "%15s %5d %9Ld.%06ld %c %9Ld.%06ld %9Ld.%06ld %9Ld.%06ld %9Ld %5d ", p->comm, task_pid_nr(p), SPLIT_NS(p->se.vruntime), + entity_eligible(cfs_rq_of(&p->se), &p->se) ? 'E' : 'N', + SPLIT_NS(p->se.deadline), + SPLIT_NS(p->se.slice), + SPLIT_NS(p->se.sum_exec_runtime), (long long)(p->nvcsw + p->nivcsw), p->prio); @@ -626,10 +628,9 @@ static void print_rq(struct seq_file *m, struct rq *rq, int rq_cpu) void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) { - s64 MIN_vruntime = -1, min_vruntime, max_vruntime = -1, - spread, rq0_min_vruntime, spread0; + s64 left_vruntime = -1, min_vruntime, right_vruntime = -1, spread; + struct sched_entity *last, *first; struct rq *rq = cpu_rq(cpu); - struct sched_entity *last; unsigned long flags; #ifdef CONFIG_FAIR_GROUP_SCHED @@ -643,26 +644,25 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) SPLIT_NS(cfs_rq->exec_clock)); raw_spin_rq_lock_irqsave(rq, flags); - if (rb_first_cached(&cfs_rq->tasks_timeline)) - MIN_vruntime = (__pick_first_entity(cfs_rq))->vruntime; + first = __pick_first_entity(cfs_rq); + if (first) + left_vruntime = first->vruntime; last = __pick_last_entity(cfs_rq); if (last) - max_vruntime = last->vruntime; + right_vruntime = last->vruntime; min_vruntime = cfs_rq->min_vruntime; - rq0_min_vruntime = cpu_rq(0)->cfs.min_vruntime; raw_spin_rq_unlock_irqrestore(rq, flags); - SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "MIN_vruntime", - SPLIT_NS(MIN_vruntime)); + + SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "left_vruntime", + SPLIT_NS(left_vruntime)); SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "min_vruntime", SPLIT_NS(min_vruntime)); - SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "max_vruntime", - SPLIT_NS(max_vruntime)); - spread = max_vruntime - MIN_vruntime; - SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "spread", - SPLIT_NS(spread)); - spread0 = min_vruntime - rq0_min_vruntime; - SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "spread0", - SPLIT_NS(spread0)); + SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "avg_vruntime", + SPLIT_NS(avg_vruntime(cfs_rq))); + SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "right_vruntime", + SPLIT_NS(right_vruntime)); + spread = right_vruntime - left_vruntime; + SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "spread", SPLIT_NS(spread)); SEQ_printf(m, " .%-30s: %d\n", "nr_spread_over", cfs_rq->nr_spread_over); SEQ_printf(m, " .%-30s: %d\n", "nr_running", cfs_rq->nr_running); @@ -863,10 +863,7 @@ static void sched_debug_header(struct seq_file *m) SEQ_printf(m, " .%-40s: %Ld\n", #x, (long long)(x)) #define PN(x) \ SEQ_printf(m, " .%-40s: %Ld.%06ld\n", #x, SPLIT_NS(x)) - PN(sysctl_sched_latency); - PN(sysctl_sched_min_granularity); - PN(sysctl_sched_idle_min_granularity); - PN(sysctl_sched_wakeup_granularity); + PN(sysctl_sched_base_slice); P(sysctl_sched_child_runs_first); P(sysctl_sched_features); #undef PN diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index a80a73909dc2..911d0063763c 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -47,6 +47,7 @@ #include <linux/psi.h> #include <linux/ratelimit.h> #include <linux/task_work.h> +#include <linux/rbtree_augmented.h> #include <asm/switch_to.h> @@ -57,22 +58,6 @@ #include "autogroup.h" /* - * Targeted preemption latency for CPU-bound tasks: - * - * NOTE: this latency value is not the same as the concept of - * 'timeslice length' - timeslices in CFS are of variable length - * and have no persistent notion like in traditional, time-slice - * based scheduling concepts. - * - * (to see the precise effective timeslice length of your workload, - * run vmstat and monitor the context-switches (cs) field) - * - * (default: 6ms * (1 + ilog(ncpus)), units: nanoseconds) - */ -unsigned int sysctl_sched_latency = 6000000ULL; -static unsigned int normalized_sysctl_sched_latency = 6000000ULL; - -/* * The initial- and re-scaling of tunables is configurable * * Options are: @@ -90,21 +75,8 @@ unsigned int sysctl_sched_tunable_scaling = SCHED_TUNABLESCALING_LOG; * * (default: 0.75 msec * (1 + ilog(ncpus)), units: nanoseconds) */ -unsigned int sysctl_sched_min_granularity = 750000ULL; -static unsigned int normalized_sysctl_sched_min_granularity = 750000ULL; - -/* - * Minimal preemption granularity for CPU-bound SCHED_IDLE tasks. - * Applies only when SCHED_IDLE tasks compete with normal tasks. - * - * (default: 0.75 msec) - */ -unsigned int sysctl_sched_idle_min_granularity = 750000ULL; - -/* - * This value is kept at sysctl_sched_latency/sysctl_sched_min_granularity - */ -static unsigned int sched_nr_latency = 8; +unsigned int sysctl_sched_base_slice = 750000ULL; +static unsigned int normalized_sysctl_sched_base_slice = 750000ULL; /* * After fork, child runs first. If set to 0 (default) then @@ -112,18 +84,6 @@ static unsigned int sched_nr_latency = 8; */ unsigned int sysctl_sched_child_runs_first __read_mostly; -/* - * SCHED_OTHER wake-up granularity. - * - * This option delays the preemption effects of decoupled workloads - * and reduces their over-scheduling. Synchronous workloads will still - * have immediate wakeup/sleep latencies. - * - * (default: 1 msec * (1 + ilog(ncpus)), units: nanoseconds) - */ -unsigned int sysctl_sched_wakeup_granularity = 1000000UL; -static unsigned int normalized_sysctl_sched_wakeup_granularity = 1000000UL; - const_debug unsigned int sysctl_sched_migration_cost = 500000UL; int sched_thermal_decay_shift; @@ -277,9 +237,7 @@ static void update_sysctl(void) #define SET_SYSCTL(name) \ (sysctl_##name = (factor) * normalized_sysctl_##name) - SET_SYSCTL(sched_min_granularity); - SET_SYSCTL(sched_latency); - SET_SYSCTL(sched_wakeup_granularity); + SET_SYSCTL(sched_base_slice); #undef SET_SYSCTL } @@ -347,6 +305,16 @@ static u64 __calc_delta(u64 delta_exec, unsigned long weight, struct load_weight return mul_u64_u32_shr(delta_exec, fact, shift); } +/* + * delta /= w + */ +static inline u64 calc_delta_fair(u64 delta, struct sched_entity *se) +{ + if (unlikely(se->load.weight != NICE_0_LOAD)) + delta = __calc_delta(delta, NICE_0_LOAD, &se->load); + + return delta; +} const struct sched_class fair_sched_class; @@ -601,13 +569,198 @@ static inline bool entity_before(const struct sched_entity *a, return (s64)(a->vruntime - b->vruntime) < 0; } +static inline s64 entity_key(struct cfs_rq *cfs_rq, struct sched_entity *se) +{ + return (s64)(se->vruntime - cfs_rq->min_vruntime); +} + #define __node_2_se(node) \ rb_entry((node), struct sched_entity, run_node) +/* + * Compute virtual time from the per-task service numbers: + * + * Fair schedulers conserve lag: + * + * \Sum lag_i = 0 + * + * Where lag_i is given by: + * + * lag_i = S - s_i = w_i * (V - v_i) + * + * Where S is the ideal service time and V is it's virtual time counterpart. + * Therefore: + * + * \Sum lag_i = 0 + * \Sum w_i * (V - v_i) = 0 + * \Sum w_i * V - w_i * v_i = 0 + * + * From which we can solve an expression for V in v_i (which we have in + * se->vruntime): + * + * \Sum v_i * w_i \Sum v_i * w_i + * V = -------------- = -------------- + * \Sum w_i W + * + * Specifically, this is the weighted average of all entity virtual runtimes. + * + * [[ NOTE: this is only equal to the ideal scheduler under the condition + * that join/leave operations happen at lag_i = 0, otherwise the + * virtual time has non-continguous motion equivalent to: + * + * V +-= lag_i / W + * + * Also see the comment in place_entity() that deals with this. ]] + * + * However, since v_i is u64, and the multiplcation could easily overflow + * transform it into a relative form that uses smaller quantities: + * + * Substitute: v_i == (v_i - v0) + v0 + * + * \Sum ((v_i - v0) + v0) * w_i \Sum (v_i - v0) * w_i + * V = ---------------------------- = --------------------- + v0 + * W W + * + * Which we track using: + * + * v0 := cfs_rq->min_vruntime + * \Sum (v_i - v0) * w_i := cfs_rq->avg_vruntime + * \Sum w_i := cfs_rq->avg_load + * + * Since min_vruntime is a monotonic increasing variable that closely tracks + * the per-task service, these deltas: (v_i - v), will be in the order of the + * maximal (virtual) lag induced in the system due to quantisation. + * + * Also, we use scale_load_down() to reduce the size. + * + * As measured, the max (key * weight) value was ~44 bits for a kernel build. + */ +static void +avg_vruntime_add(struct cfs_rq *cfs_rq, struct sched_entity *se) +{ + unsigned long weight = scale_load_down(se->load.weight); + s64 key = entity_key(cfs_rq, se); + + cfs_rq->avg_vruntime += key * weight; + cfs_rq->avg_load += weight; +} + +static void +avg_vruntime_sub(struct cfs_rq *cfs_rq, struct sched_entity *se) +{ + unsigned long weight = scale_load_down(se->load.weight); + s64 key = entity_key(cfs_rq, se); + + cfs_rq->avg_vruntime -= key * weight; + cfs_rq->avg_load -= weight; +} + +static inline +void avg_vruntime_update(struct cfs_rq *cfs_rq, s64 delta) +{ + /* + * v' = v + d ==> avg_vruntime' = avg_runtime - d*avg_load + */ + cfs_rq->avg_vruntime -= cfs_rq->avg_load * delta; +} + +u64 avg_vruntime(struct cfs_rq *cfs_rq) +{ + struct sched_entity *curr = cfs_rq->curr; + s64 avg = cfs_rq->avg_vruntime; + long load = cfs_rq->avg_load; + + if (curr && curr->on_rq) { + unsigned long weight = scale_load_down(curr->load.weight); + + avg += entity_key(cfs_rq, curr) * weight; + load += weight; + } + + if (load) + avg = div_s64(avg, load); + + return cfs_rq->min_vruntime + avg; +} + +/* + * lag_i = S - s_i = w_i * (V - v_i) + * + * However, since V is approximated by the weighted average of all entities it + * is possible -- by addition/removal/reweight to the tree -- to move V around + * and end up with a larger lag than we started with. + * + * Limit this to either double the slice length with a minimum of TICK_NSEC + * since that is the timing granularity. + * + * EEVDF gives the following limit for a steady state system: + * + * -r_max < lag < max(r_max, q) + * + * XXX could add max_slice to the augmented data to track this. + */ +void update_entity_lag(struct cfs_rq *cfs_rq, struct sched_entity *se) +{ + s64 lag, limit; + + SCHED_WARN_ON(!se->on_rq); + lag = avg_vruntime(cfs_rq) - se->vruntime; + + limit = calc_delta_fair(max_t(u64, 2*se->slice, TICK_NSEC), se); + se->vlag = clamp(lag, -limit, limit); +} + +/* + * Entity is eligible once it received less service than it ought to have, + * eg. lag >= 0. + * + * lag_i = S - s_i = w_i*(V - v_i) + * + * lag_i >= 0 -> V >= v_i + * + * \Sum (v_i - v)*w_i + * V = ------------------ + v + * \Sum w_i + * + * lag_i >= 0 -> \Sum (v_i - v)*w_i >= (v_i - v)*(\Sum w_i) + * + * Note: using 'avg_vruntime() > se->vruntime' is inacurate due + * to the loss in precision caused by the division. + */ +int entity_eligible(struct cfs_rq *cfs_rq, struct sched_entity *se) +{ + struct sched_entity *curr = cfs_rq->curr; + s64 avg = cfs_rq->avg_vruntime; + long load = cfs_rq->avg_load; + + if (curr && curr->on_rq) { + unsigned long weight = scale_load_down(curr->load.weight); + + avg += entity_key(cfs_rq, curr) * weight; + load += weight; + } + + return avg >= entity_key(cfs_rq, se) * load; +} + +static u64 __update_min_vruntime(struct cfs_rq *cfs_rq, u64 vruntime) +{ + u64 min_vruntime = cfs_rq->min_vruntime; + /* + * open coded max_vruntime() to allow updating avg_vruntime + */ + s64 delta = (s64)(vruntime - min_vruntime); + if (delta > 0) { + avg_vruntime_update(cfs_rq, delta); + min_vruntime = vruntime; + } + return min_vruntime; +} + static void update_min_vruntime(struct cfs_rq *cfs_rq) { + struct sched_entity *se = __pick_first_entity(cfs_rq); struct sched_entity *curr = cfs_rq->curr; - struct rb_node *leftmost = rb_first_cached(&cfs_rq->tasks_timeline); u64 vruntime = cfs_rq->min_vruntime; @@ -618,9 +771,7 @@ static void update_min_vruntime(struct cfs_rq *cfs_rq) curr = NULL; } - if (leftmost) { /* non-empty tree */ - struct sched_entity *se = __node_2_se(leftmost); - + if (se) { if (!curr) vruntime = se->vruntime; else @@ -629,7 +780,7 @@ static void update_min_vruntime(struct cfs_rq *cfs_rq) /* ensure we never gain time by being placed backwards. */ u64_u32_store(cfs_rq->min_vruntime, - max_vruntime(cfs_rq->min_vruntime, vruntime)); + __update_min_vruntime(cfs_rq, vruntime)); } static inline bool __entity_less(struct rb_node *a, const struct rb_node *b) @@ -637,17 +788,51 @@ static inline bool __entity_less(struct rb_node *a, const struct rb_node *b) return entity_before(__node_2_se(a), __node_2_se(b)); } +#define deadline_gt(field, lse, rse) ({ (s64)((lse)->field - (rse)->field) > 0; }) + +static inline void __update_min_deadline(struct sched_entity *se, struct rb_node *node) +{ + if (node) { + struct sched_entity *rse = __node_2_se(node); + if (deadline_gt(min_deadline, se, rse)) + se->min_deadline = rse->min_deadline; + } +} + +/* + * se->min_deadline = min(se->deadline, left->min_deadline, right->min_deadline) + */ +static inline bool min_deadline_update(struct sched_entity *se, bool exit) +{ + u64 old_min_deadline = se->min_deadline; + struct rb_node *node = &se->run_node; + + se->min_deadline = se->deadline; + __update_min_deadline(se, node->rb_right); + __update_min_deadline(se, node->rb_left); + + return se->min_deadline == old_min_deadline; +} + +RB_DECLARE_CALLBACKS(static, min_deadline_cb, struct sched_entity, + run_node, min_deadline, min_deadline_update); + /* * Enqueue an entity into the rb-tree: */ static void __enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) { - rb_add_cached(&se->run_node, &cfs_rq->tasks_timeline, __entity_less); + avg_vruntime_add(cfs_rq, se); + se->min_deadline = se->deadline; + rb_add_augmented_cached(&se->run_node, &cfs_rq->tasks_timeline, + __entity_less, &min_deadline_cb); } static void __dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) { - rb_erase_cached(&se->run_node, &cfs_rq->tasks_timeline); + rb_erase_augmented_cached(&se->run_node, &cfs_rq->tasks_timeline, + &min_deadline_cb); + avg_vruntime_sub(cfs_rq, se); } struct sched_entity *__pick_first_entity(struct cfs_rq *cfs_rq) @@ -660,14 +845,88 @@ struct sched_entity *__pick_first_entity(struct cfs_rq *cfs_rq) return __node_2_se(left); } -static struct sched_entity *__pick_next_entity(struct sched_entity *se) +/* + * Earliest Eligible Virtual Deadline First + * + * In order to provide latency guarantees for different request sizes + * EEVDF selects the best runnable task from two criteria: + * + * 1) the task must be eligible (must be owed service) + * + * 2) from those tasks that meet 1), we select the one + * with the earliest virtual deadline. + * + * We can do this in O(log n) time due to an augmented RB-tree. The + * tree keeps the entries sorted on service, but also functions as a + * heap based on the deadline by keeping: + * + * se->min_deadline = min(se->deadline, se->{left,right}->min_deadline) + * + * Which allows an EDF like search on (sub)trees. + */ +static struct sched_entity *pick_eevdf(struct cfs_rq *cfs_rq) { - struct rb_node *next = rb_next(&se->run_node); + struct rb_node *node = cfs_rq->tasks_timeline.rb_root.rb_node; + struct sched_entity *curr = cfs_rq->curr; + struct sched_entity *best = NULL; - if (!next) - return NULL; + if (curr && (!curr->on_rq || !entity_eligible(cfs_rq, curr))) + curr = NULL; + + /* + * Once selected, run a task until it either becomes non-eligible or + * until it gets a new slice. See the HACK in set_next_entity(). + */ + if (sched_feat(RUN_TO_PARITY) && curr && curr->vlag == curr->deadline) + return curr; + + while (node) { + struct sched_entity *se = __node_2_se(node); + + /* + * If this entity is not eligible, try the left subtree. + */ + if (!entity_eligible(cfs_rq, se)) { + node = node->rb_left; + continue; + } + + /* + * If this entity has an earlier deadline than the previous + * best, take this one. If it also has the earliest deadline + * of its subtree, we're done. + */ + if (!best || deadline_gt(deadline, best, se)) { + best = se; + if (best->deadline == best->min_deadline) + break; + } - return __node_2_se(next); + /* + * If the earlest deadline in this subtree is in the fully + * eligible left half of our space, go there. + */ + if (node->rb_left && + __node_2_se(node->rb_left)->min_deadline == se->min_deadline) { + node = node->rb_left; + continue; + } + + node = node->rb_right; + } + + if (!best || (curr && deadline_gt(deadline, best, curr))) + best = curr; + + if (unlikely(!best)) { + struct sched_entity *left = __pick_first_entity(cfs_rq); + if (left) { + pr_err("EEVDF scheduling fail, picking leftmost\n"); + return left; + } + } + + return best; } #ifdef CONFIG_SCHED_DEBUG @@ -684,109 +943,51 @@ struct sched_entity *__pick_last_entity(struct cfs_rq *cfs_rq) /************************************************************** * Scheduling class statistics methods: */ - +#ifdef CONFIG_SMP int sched_update_scaling(void) { unsigned int factor = get_update_sysctl_factor(); - sched_nr_latency = DIV_ROUND_UP(sysctl_sched_latency, - sysctl_sched_min_granularity); - #define WRT_SYSCTL(name) \ (normalized_sysctl_##name = sysctl_##name / (factor)) - WRT_SYSCTL(sched_min_granularity); - WRT_SYSCTL(sched_latency); - WRT_SYSCTL(sched_wakeup_granularity); + WRT_SYSCTL(sched_base_slice); #undef WRT_SYSCTL return 0; } #endif +#endif -/* - * delta /= w - */ -static inline u64 calc_delta_fair(u64 delta, struct sched_entity *se) -{ - if (unlikely(se->load.weight != NICE_0_LOAD)) - delta = __calc_delta(delta, NICE_0_LOAD, &se->load); - - return delta; -} - -/* - * The idea is to set a period in which each task runs once. - * - * When there are too many tasks (sched_nr_latency) we have to stretch - * this period because otherwise the slices get too small. - * - * p = (nr <= nl) ? l : l*nr/nl - */ -static u64 __sched_period(unsigned long nr_running) -{ - if (unlikely(nr_running > sched_nr_latency)) - return nr_running * sysctl_sched_min_granularity; - else - return sysctl_sched_latency; -} - -static bool sched_idle_cfs_rq(struct cfs_rq *cfs_rq); +static void clear_buddies(struct cfs_rq *cfs_rq, struct sched_entity *se); /* - * We calculate the wall-time slice from the period by taking a part - * proportional to the weight. - * - * s = p*P[w/rw] + * XXX: strictly: vd_i += N*r_i/w_i such that: vd_i > ve_i + * this is probably good enough. */ -static u64 sched_slice(struct cfs_rq *cfs_rq, struct sched_entity *se) +static void update_deadline(struct cfs_rq *cfs_rq, struct sched_entity *se) { - unsigned int nr_running = cfs_rq->nr_running; - struct sched_entity *init_se = se; - unsigned int min_gran; - u64 slice; - - if (sched_feat(ALT_PERIOD)) - nr_running = rq_of(cfs_rq)->cfs.h_nr_running; - - slice = __sched_period(nr_running + !se->on_rq); - - for_each_sched_entity(se) { - struct load_weight *load; - struct load_weight lw; - struct cfs_rq *qcfs_rq; - - qcfs_rq = cfs_rq_of(se); - load = &qcfs_rq->load; - - if (unlikely(!se->on_rq)) { - lw = qcfs_rq->load; + if ((s64)(se->vruntime - se->deadline) < 0) + return; - update_load_add(&lw, se->load.weight); - load = &lw; - } - slice = __calc_delta(slice, se->load.weight, load); - } + /* + * For EEVDF the virtual time slope is determined by w_i (iow. + * nice) while the request time r_i is determined by + * sysctl_sched_base_slice. + */ + se->slice = sysctl_sched_base_slice; - if (sched_feat(BASE_SLICE)) { - if (se_is_idle(init_se) && !sched_idle_cfs_rq(cfs_rq)) - min_gran = sysctl_sched_idle_min_granularity; - else - min_gran = sysctl_sched_min_granularity; + /* + * EEVDF: vd_i = ve_i + r_i / w_i + */ + se->deadline = se->vruntime + calc_delta_fair(se->slice, se); - slice = max_t(u64, slice, min_gran); + /* + * The task has consumed its request, reschedule. + */ + if (cfs_rq->nr_running > 1) { + resched_curr(rq_of(cfs_rq)); + clear_buddies(cfs_rq, se); } - - return slice; -} - -/* - * We calculate the vruntime slice of a to-be-inserted task. - * - * vs = s/w - */ -static u64 sched_vslice(struct cfs_rq *cfs_rq, struct sched_entity *se) -{ - return calc_delta_fair(sched_slice(cfs_rq, se), se); } #include "pelt.h" @@ -921,6 +1122,7 @@ static void update_curr(struct cfs_rq *cfs_rq) schedstat_add(cfs_rq->exec_clock, delta_exec); curr->vruntime += calc_delta_fair(delta_exec, curr); + update_deadline(cfs_rq, curr); update_min_vruntime(cfs_rq); if (entity_is_task(curr)) { @@ -3375,16 +3577,36 @@ dequeue_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) { } static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, unsigned long weight) { + unsigned long old_weight = se->load.weight; + if (se->on_rq) { /* commit outstanding execution time */ if (cfs_rq->curr == se) update_curr(cfs_rq); + else + avg_vruntime_sub(cfs_rq, se); update_load_sub(&cfs_rq->load, se->load.weight); } dequeue_load_avg(cfs_rq, se); update_load_set(&se->load, weight); + if (!se->on_rq) { + /* + * Because we keep se->vlag = V - v_i, while: lag_i = w_i*(V - v_i), + * we need to scale se->vlag when w_i changes. + */ + se->vlag = div_s64(se->vlag * old_weight, weight); + } else { + s64 deadline = se->deadline - se->vruntime; + /* + * When the weight changes, the virtual time slope changes and + * we should adjust the relative virtual deadline accordingly. + */ + deadline = div_s64(deadline * old_weight, weight); + se->deadline = se->vruntime + deadline; + } + #ifdef CONFIG_SMP do { u32 divider = get_pelt_divider(&se->avg); @@ -3394,9 +3616,11 @@ static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, #endif enqueue_load_avg(cfs_rq, se); - if (se->on_rq) + if (se->on_rq) { update_load_add(&cfs_rq->load, se->load.weight); - + if (cfs_rq->curr != se) + avg_vruntime_add(cfs_rq, se); + } } void reweight_task(struct task_struct *p, int prio) @@ -4692,159 +4916,125 @@ static inline void update_misfit_status(struct task_struct *p, struct rq *rq) {} #endif /* CONFIG_SMP */ -static void check_spread(struct cfs_rq *cfs_rq, struct sched_entity *se) -{ -#ifdef CONFIG_SCHED_DEBUG - s64 d = se->vruntime - cfs_rq->min_vruntime; - - if (d < 0) - d = -d; - - if (d > 3*sysctl_sched_latency) - schedstat_inc(cfs_rq->nr_spread_over); -#endif -} - -static inline bool entity_is_long_sleeper(struct sched_entity *se) +static void +place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { - struct cfs_rq *cfs_rq; - u64 sleep_time; + u64 vslice = calc_delta_fair(se->slice, se); + u64 vruntime = avg_vruntime(cfs_rq); + s64 lag = 0; - if (se->exec_start == 0) - return false; - - cfs_rq = cfs_rq_of(se); - - sleep_time = rq_clock_task(rq_of(cfs_rq)); + /* + * Due to how V is constructed as the weighted average of entities, + * adding tasks with positive lag, or removing tasks with negative lag + * will move 'time' backwards, this can screw around with the lag of + * other tasks. + * + * EEVDF: placement strategy #1 / #2 + */ + if (sched_feat(PLACE_LAG) && cfs_rq->nr_running) { + struct sched_entity *curr = cfs_rq->curr; + unsigned long load; - /* Happen while migrating because of clock task divergence */ - if (sleep_time <= se->exec_start) - return false; + lag = se->vlag; - sleep_time -= se->exec_start; - if (sleep_time > ((1ULL << 63) / scale_load_down(NICE_0_LOAD))) - return true; + /* + * If we want to place a task and preserve lag, we have to + * consider the effect of the new entity on the weighted + * average and compensate for this, otherwise lag can quickly + * evaporate. + * + * Lag is defined as: + * + * lag_i = S - s_i = w_i * (V - v_i) + * + * To avoid the 'w_i' term all over the place, we only track + * the virtual lag: + * + * vl_i = V - v_i <=> v_i = V - vl_i + * + * And we take V to be the weighted average of all v: + * + * V = (\Sum w_j*v_j) / W + * + * Where W is: \Sum w_j + * + * Then, the weighted average after adding an entity with lag + * vl_i is given by: + * + * V' = (\Sum w_j*v_j + w_i*v_i) / (W + w_i) + * = (W*V + w_i*(V - vl_i)) / (W + w_i) + * = (W*V + w_i*V - w_i*vl_i) / (W + w_i) + * = (V*(W + w_i) - w_i*l) / (W + w_i) + * = V - w_i*vl_i / (W + w_i) + * + * And the actual lag after adding an entity with vl_i is: + * + * vl'_i = V' - v_i + * = V - w_i*vl_i / (W + w_i) - (V - vl_i) + * = vl_i - w_i*vl_i / (W + w_i) + * + * Which is strictly less than vl_i. So in order to preserve lag + * we should inflate the lag before placement such that the + * effective lag after placement comes out right. + * + * As such, invert the above relation for vl'_i to get the vl_i + * we need to use such that the lag after placement is the lag + * we computed before dequeue. + * + * vl'_i = vl_i - w_i*vl_i / (W + w_i) + * = ((W + w_i)*vl_i - w_i*vl_i) / (W + w_i) + * + * (W + w_i)*vl'_i = (W + w_i)*vl_i - w_i*vl_i + * = W*vl_i + * + * vl_i = (W + w_i)*vl'_i / W + */ + load = cfs_rq->avg_load; + if (curr && curr->on_rq) + load += scale_load_down(curr->load.weight); - return false; -} + lag *= load + scale_load_down(se->load.weight); + if (WARN_ON_ONCE(!load)) + load = 1; + lag = div_s64(lag, load); + } -static void -place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int initial) -{ - u64 vruntime = cfs_rq->min_vruntime; + se->vruntime = vruntime - lag; /* - * The 'current' period is already promised to the current tasks, - * however the extra weight of the new task will slow them down a - * little, place the new task so that it fits in the slot that - * stays open at the end. + * When joining the competition; the exisiting tasks will be, + * on average, halfway through their slice, as such start tasks + * off with half a slice to ease into the competition. */ - if (initial && sched_feat(START_DEBIT)) - vruntime += sched_vslice(cfs_rq, se); - - /* sleeps up to a single latency don't count. */ - if (!initial) { - unsigned long thresh; + if (sched_feat(PLACE_DEADLINE_INITIAL) && (flags & ENQUEUE_INITIAL)) + vslice /= 2; - if (se_is_idle(se)) - thresh = sysctl_sched_min_granularity; - else - thresh = sysctl_sched_latency; - - /* - * Halve their sleep time's effect, to allow - * for a gentler effect of sleepers: - */ - if (sched_feat(GENTLE_FAIR_SLEEPERS)) - thresh >>= 1; - - vruntime -= thresh; - } - - /* - * Pull vruntime of the entity being placed to the base level of - * cfs_rq, to prevent boosting it if placed backwards. - * However, min_vruntime can advance much faster than real time, with - * the extreme being when an entity with the minimal weight always runs - * on the cfs_rq. If the waking entity slept for a long time, its - * vruntime difference from min_vruntime may overflow s64 and their - * comparison may get inversed, so ignore the entity's original - * vruntime in that case. - * The maximal vruntime speedup is given by the ratio of normal to - * minimal weight: scale_load_down(NICE_0_LOAD) / MIN_SHARES. - * When placing a migrated waking entity, its exec_start has been set - * from a different rq. In order to take into account a possible - * divergence between new and prev rq's clocks task because of irq and - * stolen time, we take an additional margin. - * So, cutting off on the sleep time of - * 2^63 / scale_load_down(NICE_0_LOAD) ~ 104 days - * should be safe. - */ - if (entity_is_long_sleeper(se)) - se->vruntime = vruntime; - else - se->vruntime = max_vruntime(se->vruntime, vruntime); + /* + * EEVDF: vd_i = ve_i + r_i/w_i + */ + se->deadline = se->vruntime + vslice; } static void check_enqueue_throttle(struct cfs_rq *cfs_rq); +static inline int cfs_rq_throttled(struct cfs_rq *cfs_rq); static inline bool cfs_bandwidth_used(void); -/* - * MIGRATION - * - * dequeue - * update_curr() - * update_min_vruntime() - * vruntime -= min_vruntime - * - * enqueue - * update_curr() - * update_min_vruntime() - * vruntime += min_vruntime - * - * this way the vruntime transition between RQs is done when both - * min_vruntime are up-to-date. - * - * WAKEUP (remote) - * - * ->migrate_task_rq_fair() (p->state == TASK_WAKING) - * vruntime -= min_vruntime - * - * enqueue - * update_curr() - * update_min_vruntime() - * vruntime += min_vruntime - * - * this way we don't have the most up-to-date min_vruntime on the originating - * CPU and an up-to-date min_vruntime on the destination CPU. - */ - static void enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { - bool renorm = !(flags & ENQUEUE_WAKEUP) || (flags & ENQUEUE_MIGRATED); bool curr = cfs_rq->curr == se; /* * If we're the current task, we must renormalise before calling * update_curr(). */ - if (renorm && curr) - se->vruntime += cfs_rq->min_vruntime; + if (curr) + place_entity(cfs_rq, se, flags); update_curr(cfs_rq); /* - * Otherwise, renormalise after, such that we're placed at the current - * moment in time, instead of some random moment in the past. Being - * placed in the past could significantly boost this task to the - * fairness detriment of existing tasks. - */ - if (renorm && !curr) - se->vruntime += cfs_rq->min_vruntime; - - /* * When enqueuing a sched_entity, we must: * - Update loads to have both entity and cfs_rq synced with now. * - For group_entity, update its runnable_weight to reflect the new @@ -4855,37 +5045,46 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) */ update_load_avg(cfs_rq, se, UPDATE_TG | DO_ATTACH); se_update_runnable(se); + /* + * XXX update_load_avg() above will have attached us to the pelt sum; + * but update_cfs_group() here will re-adjust the weight and have to + * undo/redo all that. Seems wasteful. + */ update_cfs_group(se); + + /* + * XXX now that the entity has been re-weighted, and it's lag adjusted, + * we can place the entity. + */ + if (!curr) + place_entity(cfs_rq, se, flags); + account_entity_enqueue(cfs_rq, se); - if (flags & ENQUEUE_WAKEUP) - place_entity(cfs_rq, se, 0); /* Entity has migrated, no longer consider this task hot */ if (flags & ENQUEUE_MIGRATED) se->exec_start = 0; check_schedstat_required(); update_stats_enqueue_fair(cfs_rq, se, flags); - check_spread(cfs_rq, se); if (!curr) __enqueue_entity(cfs_rq, se); se->on_rq = 1; if (cfs_rq->nr_running == 1) { check_enqueue_throttle(cfs_rq); - if (!throttled_hierarchy(cfs_rq)) + if (!throttled_hierarchy(cfs_rq)) { list_add_leaf_cfs_rq(cfs_rq); - } -} - -static void __clear_buddies_last(struct sched_entity *se) -{ - for_each_sched_entity(se) { - struct cfs_rq *cfs_rq = cfs_rq_of(se); - if (cfs_rq->last != se) - break; + } else { +#ifdef CONFIG_CFS_BANDWIDTH + struct rq *rq = rq_of(cfs_rq); - cfs_rq->last = NULL; + if (cfs_rq_throttled(cfs_rq) && !cfs_rq->throttled_clock) + cfs_rq->throttled_clock = rq_clock(rq); + if (!cfs_rq->throttled_clock_self) + cfs_rq->throttled_clock_self = rq_clock(rq); +#endif + } } } @@ -4900,27 +5099,10 @@ static void __clear_buddies_next(struct sched_entity *se) } } -static void __clear_buddies_skip(struct sched_entity *se) -{ - for_each_sched_entity(se) { - struct cfs_rq *cfs_rq = cfs_rq_of(se); - if (cfs_rq->skip != se) - break; - - cfs_rq->skip = NULL; - } -} - static void clear_buddies(struct cfs_rq *cfs_rq, struct sched_entity *se) { - if (cfs_rq->last == se) - __clear_buddies_last(se); - if (cfs_rq->next == se) __clear_buddies_next(se); - - if (cfs_rq->skip == se) - __clear_buddies_skip(se); } static __always_inline void return_cfs_rq_runtime(struct cfs_rq *cfs_rq); @@ -4954,20 +5136,12 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) clear_buddies(cfs_rq, se); + update_entity_lag(cfs_rq, se); if (se != cfs_rq->curr) __dequeue_entity(cfs_rq, se); se->on_rq = 0; account_entity_dequeue(cfs_rq, se); - /* - * Normalize after update_curr(); which will also have moved - * min_vruntime if @se is the one holding it back. But before doing - * update_min_vruntime() again, which will discount @se's position and - * can move min_vruntime forward still more. - */ - if (!(flags & DEQUEUE_SLEEP)) - se->vruntime -= cfs_rq->min_vruntime; - /* return excess runtime on last dequeue */ return_cfs_rq_runtime(cfs_rq); @@ -4986,52 +5160,6 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) update_idle_cfs_rq_clock_pelt(cfs_rq); } -/* - * Preempt the current task with a newly woken task if needed: - */ -static void -check_preempt_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr) -{ - unsigned long ideal_runtime, delta_exec; - struct sched_entity *se; - s64 delta; - - /* - * When many tasks blow up the sched_period; it is possible that - * sched_slice() reports unusually large results (when many tasks are - * very light for example). Therefore impose a maximum. - */ - ideal_runtime = min_t(u64, sched_slice(cfs_rq, curr), sysctl_sched_latency); - - delta_exec = curr->sum_exec_runtime - curr->prev_sum_exec_runtime; - if (delta_exec > ideal_runtime) { - resched_curr(rq_of(cfs_rq)); - /* - * The current task ran long enough, ensure it doesn't get - * re-elected due to buddy favours. - */ - clear_buddies(cfs_rq, curr); - return; - } - - /* - * Ensure that a task that missed wakeup preemption by a - * narrow margin doesn't have to wait for a full slice. - * This also mitigates buddy induced latencies under load. - */ - if (delta_exec < sysctl_sched_min_granularity) - return; - - se = __pick_first_entity(cfs_rq); - delta = curr->vruntime - se->vruntime; - - if (delta < 0) - return; - - if (delta > ideal_runtime) - resched_curr(rq_of(cfs_rq)); -} - static void set_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) { @@ -5047,6 +5175,11 @@ set_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) update_stats_wait_end_fair(cfs_rq, se); __dequeue_entity(cfs_rq, se); update_load_avg(cfs_rq, se, UPDATE_TG); + /* + * HACK, stash a copy of deadline at the point of pick in vlag, + * which isn't used until dequeue. + */ + se->vlag = se->deadline; } update_stats_curr_start(cfs_rq, se); @@ -5070,9 +5203,6 @@ set_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) se->prev_sum_exec_runtime = se->sum_exec_runtime; } -static int -wakeup_preempt_entity(struct sched_entity *curr, struct sched_entity *se); - /* * Pick the next process, keeping these things in mind, in this order: * 1) keep things fair between processes/task groups @@ -5083,50 +5213,14 @@ wakeup_preempt_entity(struct sched_entity *curr, struct sched_entity *se); static struct sched_entity * pick_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *curr) { - struct sched_entity *left = __pick_first_entity(cfs_rq); - struct sched_entity *se; - /* - * If curr is set we have to see if its left of the leftmost entity - * still in the tree, provided there was anything in the tree at all. + * Enabling NEXT_BUDDY will affect latency but not fairness. */ - if (!left || (curr && entity_before(curr, left))) - left = curr; - - se = left; /* ideally we run the leftmost entity */ - - /* - * Avoid running the skip buddy, if running something else can - * be done without getting too unfair. - */ - if (cfs_rq->skip && cfs_rq->skip == se) { - struct sched_entity *second; - - if (se == curr) { - second = __pick_first_entity(cfs_rq); - } else { - second = __pick_next_entity(se); - if (!second || (curr && entity_before(curr, second))) - second = curr; - } - - if (second && wakeup_preempt_entity(second, left) < 1) - se = second; - } + if (sched_feat(NEXT_BUDDY) && + cfs_rq->next && entity_eligible(cfs_rq, cfs_rq->next)) + return cfs_rq->next; - if (cfs_rq->next && wakeup_preempt_entity(cfs_rq->next, left) < 1) { - /* - * Someone really wants this to run. If it's not unfair, run it. - */ - se = cfs_rq->next; - } else if (cfs_rq->last && wakeup_preempt_entity(cfs_rq->last, left) < 1) { - /* - * Prefer last buddy, try to return the CPU to a preempted task. - */ - se = cfs_rq->last; - } - - return se; + return pick_eevdf(cfs_rq); } static bool check_cfs_rq_runtime(struct cfs_rq *cfs_rq); @@ -5143,8 +5237,6 @@ static void put_prev_entity(struct cfs_rq *cfs_rq, struct sched_entity *prev) /* throttle cfs_rqs exceeding runtime */ check_cfs_rq_runtime(cfs_rq); - check_spread(cfs_rq, prev); - if (prev->on_rq) { update_stats_wait_start_fair(cfs_rq, prev); /* Put 'current' back into the tree. */ @@ -5185,9 +5277,6 @@ entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr, int queued) hrtimer_active(&rq_of(cfs_rq)->hrtick_timer)) return; #endif - - if (cfs_rq->nr_running > 1) - check_preempt_tick(cfs_rq, curr); } @@ -5377,6 +5466,17 @@ static int tg_unthrottle_up(struct task_group *tg, void *data) /* Add cfs_rq with load or one or more already running entities to the list */ if (!cfs_rq_is_decayed(cfs_rq)) list_add_leaf_cfs_rq(cfs_rq); + + if (cfs_rq->throttled_clock_self) { + u64 delta = rq_clock(rq) - cfs_rq->throttled_clock_self; + + cfs_rq->throttled_clock_self = 0; + + if (SCHED_WARN_ON((s64)delta < 0)) + delta = 0; + + cfs_rq->throttled_clock_self_time += delta; + } } return 0; @@ -5391,6 +5491,10 @@ static int tg_throttle_down(struct task_group *tg, void *data) if (!cfs_rq->throttle_count) { cfs_rq->throttled_clock_pelt = rq_clock_pelt(rq); list_del_leaf_cfs_rq(cfs_rq); + + SCHED_WARN_ON(cfs_rq->throttled_clock_self); + if (cfs_rq->nr_running) + cfs_rq->throttled_clock_self = rq_clock(rq); } cfs_rq->throttle_count++; @@ -5480,7 +5584,9 @@ done: * throttled-list. rq->lock protects completion. */ cfs_rq->throttled = 1; - cfs_rq->throttled_clock = rq_clock(rq); + SCHED_WARN_ON(cfs_rq->throttled_clock); + if (cfs_rq->nr_running) + cfs_rq->throttled_clock = rq_clock(rq); return true; } @@ -5498,7 +5604,10 @@ void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) update_rq_clock(rq); raw_spin_lock(&cfs_b->lock); - cfs_b->throttled_time += rq_clock(rq) - cfs_rq->throttled_clock; + if (cfs_rq->throttled_clock) { + cfs_b->throttled_time += rq_clock(rq) - cfs_rq->throttled_clock; + cfs_rq->throttled_clock = 0; + } list_del_rcu(&cfs_rq->throttled_list); raw_spin_unlock(&cfs_b->lock); @@ -6014,13 +6123,14 @@ static enum hrtimer_restart sched_cfs_period_timer(struct hrtimer *timer) return idle ? HRTIMER_NORESTART : HRTIMER_RESTART; } -void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) +void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b, struct cfs_bandwidth *parent) { raw_spin_lock_init(&cfs_b->lock); cfs_b->runtime = 0; cfs_b->quota = RUNTIME_INF; cfs_b->period = ns_to_ktime(default_cfs_period()); cfs_b->burst = 0; + cfs_b->hierarchical_quota = parent ? parent->hierarchical_quota : RUNTIME_INF; INIT_LIST_HEAD(&cfs_b->throttled_cfs_rq); hrtimer_init(&cfs_b->period_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED); @@ -6157,6 +6267,46 @@ static void __maybe_unused unthrottle_offline_cfs_rqs(struct rq *rq) rq_clock_stop_loop_update(rq); } +bool cfs_task_bw_constrained(struct task_struct *p) +{ + struct cfs_rq *cfs_rq = task_cfs_rq(p); + + if (!cfs_bandwidth_used()) + return false; + + if (cfs_rq->runtime_enabled || + tg_cfs_bandwidth(cfs_rq->tg)->hierarchical_quota != RUNTIME_INF) + return true; + + return false; +} + +#ifdef CONFIG_NO_HZ_FULL +/* called from pick_next_task_fair() */ +static void sched_fair_update_stop_tick(struct rq *rq, struct task_struct *p) +{ + int cpu = cpu_of(rq); + + if (!sched_feat(HZ_BW) || !cfs_bandwidth_used()) + return; + + if (!tick_nohz_full_cpu(cpu)) + return; + + if (rq->nr_running != 1) + return; + + /* + * We know there is only one task runnable and we've just picked it. The + * normal enqueue path will have cleared TICK_DEP_BIT_SCHED if we will + * be otherwise able to stop the tick. Just need to check if we are using + * bandwidth control. + */ + if (cfs_task_bw_constrained(p)) + tick_nohz_dep_set_cpu(cpu, TICK_DEP_BIT_SCHED); +} +#endif + #else /* CONFIG_CFS_BANDWIDTH */ static inline bool cfs_bandwidth_used(void) @@ -6186,9 +6336,8 @@ static inline int throttled_lb_pair(struct task_group *tg, return 0; } -void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) {} - #ifdef CONFIG_FAIR_GROUP_SCHED +void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b, struct cfs_bandwidth *parent) {} static void init_cfs_rq_runtime(struct cfs_rq *cfs_rq) {} #endif @@ -6199,9 +6348,18 @@ static inline struct cfs_bandwidth *tg_cfs_bandwidth(struct task_group *tg) static inline void destroy_cfs_bandwidth(struct cfs_bandwidth *cfs_b) {} static inline void update_runtime_enabled(struct rq *rq) {} static inline void unthrottle_offline_cfs_rqs(struct rq *rq) {} - +#ifdef CONFIG_CGROUP_SCHED +bool cfs_task_bw_constrained(struct task_struct *p) +{ + return false; +} +#endif #endif /* CONFIG_CFS_BANDWIDTH */ +#if !defined(CONFIG_CFS_BANDWIDTH) || !defined(CONFIG_NO_HZ_FULL) +static inline void sched_fair_update_stop_tick(struct rq *rq, struct task_struct *p) {} +#endif + /************************************************** * CFS operations on tasks: */ @@ -6210,13 +6368,12 @@ static inline void unthrottle_offline_cfs_rqs(struct rq *rq) {} static void hrtick_start_fair(struct rq *rq, struct task_struct *p) { struct sched_entity *se = &p->se; - struct cfs_rq *cfs_rq = cfs_rq_of(se); SCHED_WARN_ON(task_rq(p) != rq); if (rq->cfs.h_nr_running > 1) { - u64 slice = sched_slice(cfs_rq, se); u64 ran = se->sum_exec_runtime - se->prev_sum_exec_runtime; + u64 slice = se->slice; s64 delta = slice - ran; if (delta < 0) { @@ -6240,8 +6397,7 @@ static void hrtick_update(struct rq *rq) if (!hrtick_enabled_fair(rq) || curr->sched_class != &fair_sched_class) return; - if (cfs_rq_of(&curr->se)->nr_running < sched_nr_latency) - hrtick_start_fair(rq, curr); + hrtick_start_fair(rq, curr); } #else /* !CONFIG_SCHED_HRTICK */ static inline void @@ -6282,17 +6438,6 @@ static int sched_idle_rq(struct rq *rq) rq->nr_running); } -/* - * Returns true if cfs_rq only has SCHED_IDLE entities enqueued. Note the use - * of idle_nr_running, which does not consider idle descendants of normal - * entities. - */ -static bool sched_idle_cfs_rq(struct cfs_rq *cfs_rq) -{ - return cfs_rq->nr_running && - cfs_rq->nr_running == cfs_rq->idle_nr_running; -} - #ifdef CONFIG_SMP static int sched_idle_cpu(int cpu) { @@ -7065,7 +7210,7 @@ select_idle_capacity(struct task_struct *p, struct sched_domain *sd, int target) util_min = uclamp_eff_value(p, UCLAMP_MIN); util_max = uclamp_eff_value(p, UCLAMP_MAX); - for_each_cpu_wrap(cpu, cpus, target + 1) { + for_each_cpu_wrap(cpu, cpus, target) { unsigned long cpu_cap = capacity_of(cpu); if (!available_idle_cpu(cpu) && !sched_idle_cpu(cpu)) @@ -7174,7 +7319,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) recent_used_cpu != target && cpus_share_cache(recent_used_cpu, target) && (available_idle_cpu(recent_used_cpu) || sched_idle_cpu(recent_used_cpu)) && - cpumask_test_cpu(p->recent_used_cpu, p->cpus_ptr) && + cpumask_test_cpu(recent_used_cpu, p->cpus_ptr) && asym_fits_cpu(task_util, util_min, util_max, recent_used_cpu)) { return recent_used_cpu; } @@ -7289,9 +7434,6 @@ cpu_util(int cpu, struct task_struct *p, int dst_cpu, int boost) util_est = READ_ONCE(cfs_rq->avg.util_est.enqueued); - if (boost) - util_est = max(util_est, runnable); - /* * During wake-up @p isn't enqueued yet and doesn't contribute * to any cpu_rq(cpu)->cfs.avg.util_est.enqueued. @@ -7741,6 +7883,10 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags) if (wake_flags & WF_TTWU) { record_wakee(p); + if ((wake_flags & WF_CURRENT_CPU) && + cpumask_test_cpu(cpu, p->cpus_ptr)) + return cpu; + if (sched_energy_enabled()) { new_cpu = find_energy_efficient_cpu(p, prev_cpu); if (new_cpu >= 0) @@ -7798,18 +7944,6 @@ static void migrate_task_rq_fair(struct task_struct *p, int new_cpu) { struct sched_entity *se = &p->se; - /* - * As blocked tasks retain absolute vruntime the migration needs to - * deal with this by subtracting the old and adding the new - * min_vruntime -- the latter is done by enqueue_entity() when placing - * the task on the new runqueue. - */ - if (READ_ONCE(p->__state) == TASK_WAKING) { - struct cfs_rq *cfs_rq = cfs_rq_of(se); - - se->vruntime -= u64_u32_load(cfs_rq->min_vruntime); - } - if (!task_on_rq_migrating(p)) { remove_entity_load_avg(se); @@ -7847,66 +7981,6 @@ balance_fair(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) } #endif /* CONFIG_SMP */ -static unsigned long wakeup_gran(struct sched_entity *se) -{ - unsigned long gran = sysctl_sched_wakeup_granularity; - - /* - * Since its curr running now, convert the gran from real-time - * to virtual-time in his units. - * - * By using 'se' instead of 'curr' we penalize light tasks, so - * they get preempted easier. That is, if 'se' < 'curr' then - * the resulting gran will be larger, therefore penalizing the - * lighter, if otoh 'se' > 'curr' then the resulting gran will - * be smaller, again penalizing the lighter task. - * - * This is especially important for buddies when the leftmost - * task is higher priority than the buddy. - */ - return calc_delta_fair(gran, se); -} - -/* - * Should 'se' preempt 'curr'. - * - * |s1 - * |s2 - * |s3 - * g - * |<--->|c - * - * w(c, s1) = -1 - * w(c, s2) = 0 - * w(c, s3) = 1 - * - */ -static int -wakeup_preempt_entity(struct sched_entity *curr, struct sched_entity *se) -{ - s64 gran, vdiff = curr->vruntime - se->vruntime; - - if (vdiff <= 0) - return -1; - - gran = wakeup_gran(se); - if (vdiff > gran) - return 1; - - return 0; -} - -static void set_last_buddy(struct sched_entity *se) -{ - for_each_sched_entity(se) { - if (SCHED_WARN_ON(!se->on_rq)) - return; - if (se_is_idle(se)) - return; - cfs_rq_of(se)->last = se; - } -} - static void set_next_buddy(struct sched_entity *se) { for_each_sched_entity(se) { @@ -7918,12 +7992,6 @@ static void set_next_buddy(struct sched_entity *se) } } -static void set_skip_buddy(struct sched_entity *se) -{ - for_each_sched_entity(se) - cfs_rq_of(se)->skip = se; -} - /* * Preempt the current task with a newly woken task if needed: */ @@ -7932,7 +8000,6 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ struct task_struct *curr = rq->curr; struct sched_entity *se = &curr->se, *pse = &p->se; struct cfs_rq *cfs_rq = task_cfs_rq(curr); - int scale = cfs_rq->nr_running >= sched_nr_latency; int next_buddy_marked = 0; int cse_is_idle, pse_is_idle; @@ -7948,7 +8015,7 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ if (unlikely(throttled_hierarchy(cfs_rq_of(pse)))) return; - if (sched_feat(NEXT_BUDDY) && scale && !(wake_flags & WF_FORK)) { + if (sched_feat(NEXT_BUDDY) && !(wake_flags & WF_FORK)) { set_next_buddy(pse); next_buddy_marked = 1; } @@ -7993,35 +8060,19 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ if (cse_is_idle != pse_is_idle) return; - update_curr(cfs_rq_of(se)); - if (wakeup_preempt_entity(se, pse) == 1) { - /* - * Bias pick_next to pick the sched entity that is - * triggering this preemption. - */ - if (!next_buddy_marked) - set_next_buddy(pse); + cfs_rq = cfs_rq_of(se); + update_curr(cfs_rq); + + /* + * XXX pick_eevdf(cfs_rq) != se ? + */ + if (pick_eevdf(cfs_rq) == pse) goto preempt; - } return; preempt: resched_curr(rq); - /* - * Only set the backward buddy when the current task is still - * on the rq. This can happen when a wakeup gets interleaved - * with schedule on the ->pre_schedule() or idle_balance() - * point, either of which can * drop the rq lock. - * - * Also, during early boot the idle thread is in the fair class, - * for obvious reasons its a bad idea to schedule back to it. - */ - if (unlikely(!se->on_rq || curr == rq->idle)) - return; - - if (sched_feat(LAST_BUDDY) && scale && entity_is_task(se)) - set_last_buddy(se); } #ifdef CONFIG_SMP @@ -8172,6 +8223,7 @@ done: __maybe_unused; hrtick_start_fair(rq, p); update_misfit_status(p, rq); + sched_fair_update_stop_tick(rq, p); return p; @@ -8222,8 +8274,6 @@ static void put_prev_task_fair(struct rq *rq, struct task_struct *prev) /* * sched_yield() is very simple - * - * The magic of dealing with the ->skip buddy is in pick_next_entity. */ static void yield_task_fair(struct rq *rq) { @@ -8239,21 +8289,19 @@ static void yield_task_fair(struct rq *rq) clear_buddies(cfs_rq, se); - if (curr->policy != SCHED_BATCH) { - update_rq_clock(rq); - /* - * Update run-time statistics of the 'current'. - */ - update_curr(cfs_rq); - /* - * Tell update_rq_clock() that we've just updated, - * so we don't do microscopic update in schedule() - * and double the fastpath cost. - */ - rq_clock_skip_update(rq); - } + update_rq_clock(rq); + /* + * Update run-time statistics of the 'current'. + */ + update_curr(cfs_rq); + /* + * Tell update_rq_clock() that we've just updated, + * so we don't do microscopic update in schedule() + * and double the fastpath cost. + */ + rq_clock_skip_update(rq); - set_skip_buddy(se); + se->deadline += calc_delta_fair(se->slice, se); } static bool yield_to_task_fair(struct rq *rq, struct task_struct *p) @@ -8416,6 +8464,11 @@ enum group_type { */ group_misfit_task, /* + * Balance SMT group that's fully busy. Can benefit from migration + * a task on SMT with busy sibling to another CPU on idle core. + */ + group_smt_balance, + /* * SD_ASYM_PACKING only: One local CPU with higher capacity is available, * and the task should be migrated to it instead of running on the * current CPU. @@ -8496,8 +8549,7 @@ static int task_hot(struct task_struct *p, struct lb_env *env) * Buddy candidates are cache hot: */ if (sched_feat(CACHE_HOT_BUDDY) && env->dst_rq->nr_running && - (&p->se == cfs_rq_of(&p->se)->next || - &p->se == cfs_rq_of(&p->se)->last)) + (&p->se == cfs_rq_of(&p->se)->next)) return 1; if (sysctl_sched_migration_cost == -1) @@ -9123,6 +9175,7 @@ struct sg_lb_stats { unsigned int group_weight; enum group_type group_type; unsigned int group_asym_packing; /* Tasks should be moved to preferred CPU */ + unsigned int group_smt_balance; /* Task on busy SMT be moved */ unsigned long group_misfit_task_load; /* A CPU has a task too big for its capacity */ #ifdef CONFIG_NUMA_BALANCING unsigned int nr_numa_running; @@ -9396,6 +9449,9 @@ group_type group_classify(unsigned int imbalance_pct, if (sgs->group_asym_packing) return group_asym_packing; + if (sgs->group_smt_balance) + return group_smt_balance; + if (sgs->group_misfit_task_load) return group_misfit_task; @@ -9465,6 +9521,71 @@ sched_asym(struct lb_env *env, struct sd_lb_stats *sds, struct sg_lb_stats *sgs return sched_asym_prefer(env->dst_cpu, group->asym_prefer_cpu); } +/* One group has more than one SMT CPU while the other group does not */ +static inline bool smt_vs_nonsmt_groups(struct sched_group *sg1, + struct sched_group *sg2) +{ + if (!sg1 || !sg2) + return false; + + return (sg1->flags & SD_SHARE_CPUCAPACITY) != + (sg2->flags & SD_SHARE_CPUCAPACITY); +} + +static inline bool smt_balance(struct lb_env *env, struct sg_lb_stats *sgs, + struct sched_group *group) +{ + if (env->idle == CPU_NOT_IDLE) + return false; + + /* + * For SMT source group, it is better to move a task + * to a CPU that doesn't have multiple tasks sharing its CPU capacity. + * Note that if a group has a single SMT, SD_SHARE_CPUCAPACITY + * will not be on. + */ + if (group->flags & SD_SHARE_CPUCAPACITY && + sgs->sum_h_nr_running > 1) + return true; + + return false; +} + +static inline long sibling_imbalance(struct lb_env *env, + struct sd_lb_stats *sds, + struct sg_lb_stats *busiest, + struct sg_lb_stats *local) +{ + int ncores_busiest, ncores_local; + long imbalance; + + if (env->idle == CPU_NOT_IDLE || !busiest->sum_nr_running) + return 0; + + ncores_busiest = sds->busiest->cores; + ncores_local = sds->local->cores; + + if (ncores_busiest == ncores_local) { + imbalance = busiest->sum_nr_running; + lsub_positive(&imbalance, local->sum_nr_running); + return imbalance; + } + + /* Balance such that nr_running/ncores ratio are same on both groups */ + imbalance = ncores_local * busiest->sum_nr_running; + lsub_positive(&imbalance, ncores_busiest * local->sum_nr_running); + /* Normalize imbalance and do rounding on normalization */ + imbalance = 2 * imbalance + ncores_local + ncores_busiest; + imbalance /= ncores_local + ncores_busiest; + + /* Take advantage of resource in an empty sched group */ + if (imbalance == 0 && local->sum_nr_running == 0 && + busiest->sum_nr_running > 1) + imbalance = 2; + + return imbalance; +} + static inline bool sched_reduced_capacity(struct rq *rq, struct sched_domain *sd) { @@ -9557,6 +9678,10 @@ static inline void update_sg_lb_stats(struct lb_env *env, sgs->group_asym_packing = 1; } + /* Check for loaded SMT group to be balanced to dst CPU */ + if (!local_group && smt_balance(env, sgs, group)) + sgs->group_smt_balance = 1; + sgs->group_type = group_classify(env->sd->imbalance_pct, group, sgs); /* Computing avg_load makes sense only when group is overloaded */ @@ -9641,6 +9766,7 @@ static bool update_sd_pick_busiest(struct lb_env *env, return false; break; + case group_smt_balance: case group_fully_busy: /* * Select the fully busy group with highest avg_load. In @@ -9670,6 +9796,18 @@ static bool update_sd_pick_busiest(struct lb_env *env, case group_has_spare: /* + * Do not pick sg with SMT CPUs over sg with pure CPUs, + * as we do not want to pull task off SMT core with one task + * and make the core idle. + */ + if (smt_vs_nonsmt_groups(sds->busiest, sg)) { + if (sg->flags & SD_SHARE_CPUCAPACITY && sgs->sum_h_nr_running <= 1) + return false; + else + return true; + } + + /* * Select not overloaded group with lowest number of idle cpus * and highest number of running tasks. We could also compare * the spare capacity which is more stable but it can end up @@ -9865,6 +10003,7 @@ static bool update_pick_idlest(struct sched_group *idlest, case group_imbalanced: case group_asym_packing: + case group_smt_balance: /* Those types are not used in the slow wakeup path */ return false; @@ -9996,6 +10135,7 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu) case group_imbalanced: case group_asym_packing: + case group_smt_balance: /* Those type are not used in the slow wakeup path */ return NULL; @@ -10250,6 +10390,13 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s return; } + if (busiest->group_type == group_smt_balance) { + /* Reduce number of tasks sharing CPU capacity */ + env->migration_type = migrate_task; + env->imbalance = 1; + return; + } + if (busiest->group_type == group_imbalanced) { /* * In the group_imb case we cannot rely on group-wide averages @@ -10297,14 +10444,12 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s } if (busiest->group_weight == 1 || sds->prefer_sibling) { - unsigned int nr_diff = busiest->sum_nr_running; /* * When prefer sibling, evenly spread running tasks on * groups. */ env->migration_type = migrate_task; - lsub_positive(&nr_diff, local->sum_nr_running); - env->imbalance = nr_diff; + env->imbalance = sibling_imbalance(env, sds, busiest, local); } else { /* @@ -10501,20 +10646,27 @@ static struct sched_group *find_busiest_group(struct lb_env *env) * group's child domain. */ if (sds.prefer_sibling && local->group_type == group_has_spare && - busiest->sum_nr_running > local->sum_nr_running + 1) + sibling_imbalance(env, &sds, busiest, local) > 1) goto force_balance; if (busiest->group_type != group_overloaded) { - if (env->idle == CPU_NOT_IDLE) + if (env->idle == CPU_NOT_IDLE) { /* * If the busiest group is not overloaded (and as a * result the local one too) but this CPU is already * busy, let another idle CPU try to pull task. */ goto out_balanced; + } + + if (busiest->group_type == group_smt_balance && + smt_vs_nonsmt_groups(sds.local, sds.busiest)) { + /* Let non SMT CPU pull from SMT CPU sharing with sibling */ + goto force_balance; + } if (busiest->group_weight > 1 && - local->idle_cpus <= (busiest->idle_cpus + 1)) + local->idle_cpus <= (busiest->idle_cpus + 1)) { /* * If the busiest group is not overloaded * and there is no imbalance between this and busiest @@ -10525,12 +10677,14 @@ static struct sched_group *find_busiest_group(struct lb_env *env) * there is more than 1 CPU per group. */ goto out_balanced; + } - if (busiest->sum_h_nr_running == 1) + if (busiest->sum_h_nr_running == 1) { /* * busiest doesn't have any tasks waiting to run */ goto out_balanced; + } } force_balance: @@ -10764,7 +10918,7 @@ static int active_load_balance_cpu_stop(void *data); static int should_we_balance(struct lb_env *env) { struct sched_group *sg = env->sd->groups; - int cpu; + int cpu, idle_smt = -1; /* * Ensure the balancing environment is consistent; can happen @@ -10791,10 +10945,24 @@ static int should_we_balance(struct lb_env *env) if (!idle_cpu(cpu)) continue; + /* + * Don't balance to idle SMT in busy core right away when + * balancing cores, but remember the first idle SMT CPU for + * later consideration. Find CPU on an idle core first. + */ + if (!(env->sd->flags & SD_SHARE_CPUCAPACITY) && !is_core_idle(cpu)) { + if (idle_smt == -1) + idle_smt = cpu; + continue; + } + /* Are we the first idle CPU? */ return cpu == env->dst_cpu; } + if (idle_smt == env->dst_cpu) + return true; + /* Are we the first CPU of this group ? */ return group_balance_cpu(sg) == env->dst_cpu; } @@ -12007,8 +12175,8 @@ static void rq_offline_fair(struct rq *rq) static inline bool __entity_slice_used(struct sched_entity *se, int min_nr_tasks) { - u64 slice = sched_slice(cfs_rq_of(se), se); u64 rtime = se->sum_exec_runtime - se->prev_sum_exec_runtime; + u64 slice = se->slice; return (rtime * min_nr_tasks > slice); } @@ -12164,8 +12332,8 @@ static void task_tick_fair(struct rq *rq, struct task_struct *curr, int queued) */ static void task_fork_fair(struct task_struct *p) { - struct cfs_rq *cfs_rq; struct sched_entity *se = &p->se, *curr; + struct cfs_rq *cfs_rq; struct rq *rq = this_rq(); struct rq_flags rf; @@ -12174,22 +12342,9 @@ static void task_fork_fair(struct task_struct *p) cfs_rq = task_cfs_rq(current); curr = cfs_rq->curr; - if (curr) { + if (curr) update_curr(cfs_rq); - se->vruntime = curr->vruntime; - } - place_entity(cfs_rq, se, 1); - - if (sysctl_sched_child_runs_first && curr && entity_before(curr, se)) { - /* - * Upon rescheduling, sched_class::put_prev_task() will place - * 'current' within the tree based on its new key value. - */ - swap(curr->vruntime, se->vruntime); - resched_curr(rq); - } - - se->vruntime -= cfs_rq->min_vruntime; + place_entity(cfs_rq, se, ENQUEUE_INITIAL); rq_unlock(rq, &rf); } @@ -12218,34 +12373,6 @@ prio_changed_fair(struct rq *rq, struct task_struct *p, int oldprio) check_preempt_curr(rq, p, 0); } -static inline bool vruntime_normalized(struct task_struct *p) -{ - struct sched_entity *se = &p->se; - - /* - * In both the TASK_ON_RQ_QUEUED and TASK_ON_RQ_MIGRATING cases, - * the dequeue_entity(.flags=0) will already have normalized the - * vruntime. - */ - if (p->on_rq) - return true; - - /* - * When !on_rq, vruntime of the task has usually NOT been normalized. - * But there are some cases where it has already been normalized: - * - * - A forked child which is waiting for being woken up by - * wake_up_new_task(). - * - A task which has been woken up by try_to_wake_up() and - * waiting for actually being woken up by sched_ttwu_pending(). - */ - if (!se->sum_exec_runtime || - (READ_ONCE(p->__state) == TASK_WAKING && p->sched_remote_wakeup)) - return true; - - return false; -} - #ifdef CONFIG_FAIR_GROUP_SCHED /* * Propagate the changes of the sched_entity across the tg tree to make it @@ -12316,16 +12443,6 @@ static void attach_entity_cfs_rq(struct sched_entity *se) static void detach_task_cfs_rq(struct task_struct *p) { struct sched_entity *se = &p->se; - struct cfs_rq *cfs_rq = cfs_rq_of(se); - - if (!vruntime_normalized(p)) { - /* - * Fix up our vruntime so that the current sleep doesn't - * cause 'unlimited' sleep bonus. - */ - place_entity(cfs_rq, se, 0); - se->vruntime -= cfs_rq->min_vruntime; - } detach_entity_cfs_rq(se); } @@ -12333,12 +12450,8 @@ static void detach_task_cfs_rq(struct task_struct *p) static void attach_task_cfs_rq(struct task_struct *p) { struct sched_entity *se = &p->se; - struct cfs_rq *cfs_rq = cfs_rq_of(se); attach_entity_cfs_rq(se); - - if (!vruntime_normalized(p)) - se->vruntime += cfs_rq->min_vruntime; } static void switched_from_fair(struct rq *rq, struct task_struct *p) @@ -12450,7 +12563,7 @@ int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent) tg->shares = NICE_0_LOAD; - init_cfs_bandwidth(tg_cfs_bandwidth(tg)); + init_cfs_bandwidth(tg_cfs_bandwidth(tg), tg_cfs_bandwidth(parent)); for_each_possible_cpu(i) { cfs_rq = kzalloc_node(sizeof(struct cfs_rq), @@ -12703,7 +12816,7 @@ static unsigned int get_rr_interval_fair(struct rq *rq, struct task_struct *task * idle runqueue: */ if (rq->cfs.load.weight) - rr_interval = NS_TO_JIFFIES(sched_slice(cfs_rq_of(se), se)); + rr_interval = NS_TO_JIFFIES(se->slice); return rr_interval; } diff --git a/kernel/sched/features.h b/kernel/sched/features.h index ee7f23c76bd3..f770168230ae 100644 --- a/kernel/sched/features.h +++ b/kernel/sched/features.h @@ -1,16 +1,12 @@ /* SPDX-License-Identifier: GPL-2.0 */ -/* - * Only give sleepers 50% of their service deficit. This allows - * them to run sooner, but does not allow tons of sleepers to - * rip the spread apart. - */ -SCHED_FEAT(GENTLE_FAIR_SLEEPERS, true) /* - * Place new tasks ahead so that they do not starve already running - * tasks + * Using the avg_vruntime, do the right thing and preserve lag across + * sleep+wake cycles. EEVDF placement strategy #1, #2 if disabled. */ -SCHED_FEAT(START_DEBIT, true) +SCHED_FEAT(PLACE_LAG, true) +SCHED_FEAT(PLACE_DEADLINE_INITIAL, true) +SCHED_FEAT(RUN_TO_PARITY, true) /* * Prefer to schedule the task we woke last (assuming it failed @@ -20,13 +16,6 @@ SCHED_FEAT(START_DEBIT, true) SCHED_FEAT(NEXT_BUDDY, false) /* - * Prefer to schedule the task that ran last (when we did - * wake-preempt) as that likely will touch the same data, increases - * cache locality. - */ -SCHED_FEAT(LAST_BUDDY, true) - -/* * Consider buddies to be cache hot, decreases the likeliness of a * cache buddy being migrated away, increases cache locality. */ @@ -99,5 +88,4 @@ SCHED_FEAT(UTIL_EST_FASTUP, true) SCHED_FEAT(LATENCY_WARN, false) -SCHED_FEAT(ALT_PERIOD, true) -SCHED_FEAT(BASE_SLICE, true) +SCHED_FEAT(HZ_BW, true) diff --git a/kernel/sched/psi.c b/kernel/sched/psi.c index 81fca77397f6..1d0f634725a6 100644 --- a/kernel/sched/psi.c +++ b/kernel/sched/psi.c @@ -140,7 +140,7 @@ static int psi_bug __read_mostly; DEFINE_STATIC_KEY_FALSE(psi_disabled); -DEFINE_STATIC_KEY_TRUE(psi_cgroups_enabled); +static DEFINE_STATIC_KEY_TRUE(psi_cgroups_enabled); #ifdef CONFIG_PSI_DEFAULT_DISABLED static bool psi_enable; @@ -493,8 +493,12 @@ static u64 update_triggers(struct psi_group *group, u64 now, bool *update_total, continue; /* Generate an event */ - if (cmpxchg(&t->event, 0, 1) == 0) - wake_up_interruptible(&t->event_wait); + if (cmpxchg(&t->event, 0, 1) == 0) { + if (t->of) + kernfs_notify(t->of->kn); + else + wake_up_interruptible(&t->event_wait); + } t->last_event_time = now; /* Reset threshold breach flag once event got generated */ t->pending_event = false; @@ -1271,8 +1275,9 @@ int psi_show(struct seq_file *m, struct psi_group *group, enum psi_res res) return 0; } -struct psi_trigger *psi_trigger_create(struct psi_group *group, - char *buf, enum psi_res res, struct file *file) +struct psi_trigger *psi_trigger_create(struct psi_group *group, char *buf, + enum psi_res res, struct file *file, + struct kernfs_open_file *of) { struct psi_trigger *t; enum psi_states state; @@ -1331,7 +1336,9 @@ struct psi_trigger *psi_trigger_create(struct psi_group *group, t->event = 0; t->last_event_time = 0; - init_waitqueue_head(&t->event_wait); + t->of = of; + if (!of) + init_waitqueue_head(&t->event_wait); t->pending_event = false; t->aggregator = privileged ? PSI_POLL : PSI_AVGS; @@ -1388,7 +1395,10 @@ void psi_trigger_destroy(struct psi_trigger *t) * being accessed later. Can happen if cgroup is deleted from under a * polling process. */ - wake_up_pollfree(&t->event_wait); + if (t->of) + kernfs_notify(t->of->kn); + else + wake_up_interruptible(&t->event_wait); if (t->aggregator == PSI_AVGS) { mutex_lock(&group->avgs_lock); @@ -1465,7 +1475,10 @@ __poll_t psi_trigger_poll(void **trigger_ptr, if (!t) return DEFAULT_POLLMASK | EPOLLERR | EPOLLPRI; - poll_wait(file, &t->event_wait, wait); + if (t->of) + kernfs_generic_poll(t->of, wait); + else + poll_wait(file, &t->event_wait, wait); if (cmpxchg(&t->event, 1, 0) == 1) ret |= EPOLLPRI; @@ -1535,7 +1548,7 @@ static ssize_t psi_write(struct file *file, const char __user *user_buf, return -EBUSY; } - new = psi_trigger_create(&psi_system, buf, res, file); + new = psi_trigger_create(&psi_system, buf, res, file, NULL); if (IS_ERR(new)) { mutex_unlock(&seq->lock); return PTR_ERR(new); diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index 00e0e5074115..0597ba0f85ff 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -25,7 +25,7 @@ unsigned int sysctl_sched_rt_period = 1000000; int sysctl_sched_rt_runtime = 950000; #ifdef CONFIG_SYSCTL -static int sysctl_sched_rr_timeslice = (MSEC_PER_SEC / HZ) * RR_TIMESLICE; +static int sysctl_sched_rr_timeslice = (MSEC_PER_SEC * RR_TIMESLICE) / HZ; static int sched_rt_handler(struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos); static int sched_rr_handler(struct ctl_table *table, int write, void *buffer, @@ -3062,6 +3062,9 @@ static int sched_rr_handler(struct ctl_table *table, int write, void *buffer, sched_rr_timeslice = sysctl_sched_rr_timeslice <= 0 ? RR_TIMESLICE : msecs_to_jiffies(sysctl_sched_rr_timeslice); + + if (sysctl_sched_rr_timeslice <= 0) + sysctl_sched_rr_timeslice = jiffies_to_msecs(RR_TIMESLICE); } mutex_unlock(&mutex); diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index e93e006a942b..04846272409c 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -454,11 +454,12 @@ extern void unregister_fair_sched_group(struct task_group *tg); extern void init_tg_cfs_entry(struct task_group *tg, struct cfs_rq *cfs_rq, struct sched_entity *se, int cpu, struct sched_entity *parent); -extern void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b); +extern void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b, struct cfs_bandwidth *parent); extern void __refill_cfs_bandwidth_runtime(struct cfs_bandwidth *cfs_b); extern void start_cfs_bandwidth(struct cfs_bandwidth *cfs_b); extern void unthrottle_cfs_rq(struct cfs_rq *cfs_rq); +extern bool cfs_task_bw_constrained(struct task_struct *p); extern void init_tg_rt_entry(struct task_group *tg, struct rt_rq *rt_rq, struct sched_rt_entity *rt_se, int cpu, @@ -494,6 +495,7 @@ static inline void set_task_rq_fair(struct sched_entity *se, #else /* CONFIG_CGROUP_SCHED */ struct cfs_bandwidth { }; +static inline bool cfs_task_bw_constrained(struct task_struct *p) { return false; } #endif /* CONFIG_CGROUP_SCHED */ @@ -548,6 +550,9 @@ struct cfs_rq { unsigned int idle_nr_running; /* SCHED_IDLE */ unsigned int idle_h_nr_running; /* SCHED_IDLE */ + s64 avg_vruntime; + u64 avg_load; + u64 exec_clock; u64 min_vruntime; #ifdef CONFIG_SCHED_CORE @@ -567,8 +572,6 @@ struct cfs_rq { */ struct sched_entity *curr; struct sched_entity *next; - struct sched_entity *last; - struct sched_entity *skip; #ifdef CONFIG_SCHED_DEBUG unsigned int nr_spread_over; @@ -636,6 +639,8 @@ struct cfs_rq { u64 throttled_clock; u64 throttled_clock_pelt; u64 throttled_clock_pelt_time; + u64 throttled_clock_self; + u64 throttled_clock_self_time; int throttled; int throttle_count; struct list_head throttled_list; @@ -1245,6 +1250,7 @@ static inline raw_spinlock_t *__rq_lockp(struct rq *rq) bool cfs_prio_less(const struct task_struct *a, const struct task_struct *b, bool fi); +void task_vruntime_update(struct rq *rq, struct task_struct *p, bool in_fi); /* * Helpers to check if the CPU's core cookie matches with the task's cookie @@ -1700,6 +1706,21 @@ rq_unlock(struct rq *rq, struct rq_flags *rf) raw_spin_rq_unlock(rq); } +DEFINE_LOCK_GUARD_1(rq_lock, struct rq, + rq_lock(_T->lock, &_T->rf), + rq_unlock(_T->lock, &_T->rf), + struct rq_flags rf) + +DEFINE_LOCK_GUARD_1(rq_lock_irq, struct rq, + rq_lock_irq(_T->lock, &_T->rf), + rq_unlock_irq(_T->lock, &_T->rf), + struct rq_flags rf) + +DEFINE_LOCK_GUARD_1(rq_lock_irqsave, struct rq, + rq_lock_irqsave(_T->lock, &_T->rf), + rq_unlock_irqrestore(_T->lock, &_T->rf), + struct rq_flags rf) + static inline struct rq * this_rq_lock_irq(struct rq_flags *rf) __acquires(rq->lock) @@ -1882,6 +1903,7 @@ struct sched_group { atomic_t ref; unsigned int group_weight; + unsigned int cores; struct sched_group_capacity *sgc; int asym_prefer_cpu; /* CPU of highest priority in group */ int flags; @@ -2131,12 +2153,13 @@ static inline int task_on_rq_migrating(struct task_struct *p) } /* Wake flags. The first three directly map to some SD flag value */ -#define WF_EXEC 0x02 /* Wakeup after exec; maps to SD_BALANCE_EXEC */ -#define WF_FORK 0x04 /* Wakeup after fork; maps to SD_BALANCE_FORK */ -#define WF_TTWU 0x08 /* Wakeup; maps to SD_BALANCE_WAKE */ +#define WF_EXEC 0x02 /* Wakeup after exec; maps to SD_BALANCE_EXEC */ +#define WF_FORK 0x04 /* Wakeup after fork; maps to SD_BALANCE_FORK */ +#define WF_TTWU 0x08 /* Wakeup; maps to SD_BALANCE_WAKE */ -#define WF_SYNC 0x10 /* Waker goes to sleep after wakeup */ -#define WF_MIGRATED 0x20 /* Internal use, task got migrated */ +#define WF_SYNC 0x10 /* Waker goes to sleep after wakeup */ +#define WF_MIGRATED 0x20 /* Internal use, task got migrated */ +#define WF_CURRENT_CPU 0x40 /* Prefer to move the wakee to the current CPU. */ #ifdef CONFIG_SMP static_assert(WF_EXEC == SD_BALANCE_EXEC); @@ -2195,6 +2218,7 @@ extern const u32 sched_prio_to_wmult[40]; #else #define ENQUEUE_MIGRATED 0x00 #endif +#define ENQUEUE_INITIAL 0x80 #define RETRY_TASK ((void *)-1UL) @@ -2398,6 +2422,7 @@ static inline struct cpuidle_state *idle_get_state(struct rq *rq) #endif extern void schedule_idle(void); +asmlinkage void schedule_user(void); extern void sysrq_sched_debug_show(void); extern void sched_init_granularity(void); @@ -2499,11 +2524,9 @@ extern void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags); extern const_debug unsigned int sysctl_sched_nr_migrate; extern const_debug unsigned int sysctl_sched_migration_cost; +extern unsigned int sysctl_sched_base_slice; + #ifdef CONFIG_SCHED_DEBUG -extern unsigned int sysctl_sched_latency; -extern unsigned int sysctl_sched_min_granularity; -extern unsigned int sysctl_sched_idle_min_granularity; -extern unsigned int sysctl_sched_wakeup_granularity; extern int sysctl_resched_latency_warn_ms; extern int sysctl_resched_latency_warn_once; @@ -2609,6 +2632,12 @@ static inline void double_rq_clock_clear_update(struct rq *rq1, struct rq *rq2) static inline void double_rq_clock_clear_update(struct rq *rq1, struct rq *rq2) {} #endif +#define DEFINE_LOCK_GUARD_2(name, type, _lock, _unlock, ...) \ +__DEFINE_UNLOCK_GUARD(name, type, _unlock, type *lock2; __VA_ARGS__) \ +static inline class_##name##_t class_##name##_constructor(type *lock, type *lock2) \ +{ class_##name##_t _t = { .lock = lock, .lock2 = lock2 }, *_T = &_t; \ + _lock; return _t; } + #ifdef CONFIG_SMP static inline bool rq_order_less(struct rq *rq1, struct rq *rq2) @@ -2738,6 +2767,16 @@ static inline void double_raw_lock(raw_spinlock_t *l1, raw_spinlock_t *l2) raw_spin_lock_nested(l2, SINGLE_DEPTH_NESTING); } +static inline void double_raw_unlock(raw_spinlock_t *l1, raw_spinlock_t *l2) +{ + raw_spin_unlock(l1); + raw_spin_unlock(l2); +} + +DEFINE_LOCK_GUARD_2(double_raw_spinlock, raw_spinlock_t, + double_raw_lock(_T->lock, _T->lock2), + double_raw_unlock(_T->lock, _T->lock2)) + /* * double_rq_unlock - safely unlock two runqueues * @@ -2795,6 +2834,10 @@ static inline void double_rq_unlock(struct rq *rq1, struct rq *rq2) #endif +DEFINE_LOCK_GUARD_2(double_rq_lock, struct rq, + double_rq_lock(_T->lock, _T->lock2), + double_rq_unlock(_T->lock, _T->lock2)) + extern struct sched_entity *__pick_first_entity(struct cfs_rq *cfs_rq); extern struct sched_entity *__pick_last_entity(struct cfs_rq *cfs_rq); @@ -3229,6 +3272,8 @@ static inline bool is_per_cpu_kthread(struct task_struct *p) extern void swake_up_all_locked(struct swait_queue_head *q); extern void __prepare_to_swait(struct swait_queue_head *q, struct swait_queue *wait); +extern int try_to_wake_up(struct task_struct *tsk, unsigned int state, int wake_flags); + #ifdef CONFIG_PREEMPT_DYNAMIC extern int preempt_dynamic_mode; extern int sched_dynamic_mode(const char *str); @@ -3480,4 +3525,7 @@ static inline void task_tick_mm_cid(struct rq *rq, struct task_struct *curr) { } static inline void init_sched_mm_cid(struct task_struct *t) { } #endif +extern u64 avg_vruntime(struct cfs_rq *cfs_rq); +extern int entity_eligible(struct cfs_rq *cfs_rq, struct sched_entity *se); + #endif /* _KERNEL_SCHED_SCHED_H */ diff --git a/kernel/sched/swait.c b/kernel/sched/swait.c index 76b9b796e695..72505cd3b60a 100644 --- a/kernel/sched/swait.c +++ b/kernel/sched/swait.c @@ -18,7 +18,7 @@ EXPORT_SYMBOL(__init_swait_queue_head); * If for some reason it would return 0, that means the previously waiting * task is already running, so it will observe condition true (or has already). */ -void swake_up_locked(struct swait_queue_head *q) +void swake_up_locked(struct swait_queue_head *q, int wake_flags) { struct swait_queue *curr; @@ -26,7 +26,7 @@ void swake_up_locked(struct swait_queue_head *q) return; curr = list_first_entry(&q->task_list, typeof(*curr), task_list); - wake_up_process(curr->task); + try_to_wake_up(curr->task, TASK_NORMAL, wake_flags); list_del_init(&curr->task_list); } EXPORT_SYMBOL(swake_up_locked); @@ -41,7 +41,7 @@ EXPORT_SYMBOL(swake_up_locked); void swake_up_all_locked(struct swait_queue_head *q) { while (!list_empty(&q->task_list)) - swake_up_locked(q); + swake_up_locked(q, 0); } void swake_up_one(struct swait_queue_head *q) @@ -49,7 +49,7 @@ void swake_up_one(struct swait_queue_head *q) unsigned long flags; raw_spin_lock_irqsave(&q->lock, flags); - swake_up_locked(q); + swake_up_locked(q, 0); raw_spin_unlock_irqrestore(&q->lock, flags); } EXPORT_SYMBOL(swake_up_one); diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c index d3a3b2646ec4..05a5bc678c08 100644 --- a/kernel/sched/topology.c +++ b/kernel/sched/topology.c @@ -722,8 +722,7 @@ cpu_attach_domain(struct sched_domain *sd, struct root_domain *rd, int cpu) if (parent->parent) { parent->parent->child = tmp; - if (tmp->flags & SD_SHARE_CPUCAPACITY) - parent->parent->groups->flags |= SD_SHARE_CPUCAPACITY; + parent->parent->groups->flags = tmp->flags; } /* @@ -1275,14 +1274,24 @@ build_sched_groups(struct sched_domain *sd, int cpu) static void init_sched_groups_capacity(int cpu, struct sched_domain *sd) { struct sched_group *sg = sd->groups; + struct cpumask *mask = sched_domains_tmpmask2; WARN_ON(!sg); do { - int cpu, max_cpu = -1; + int cpu, cores = 0, max_cpu = -1; sg->group_weight = cpumask_weight(sched_group_span(sg)); + cpumask_copy(mask, sched_group_span(sg)); + for_each_cpu(cpu, mask) { + cores++; +#ifdef CONFIG_SCHED_SMT + cpumask_andnot(mask, mask, cpu_smt_mask(cpu)); +#endif + } + sg->cores = cores; + if (!(sd->flags & SD_ASYM_PACKING)) goto next; diff --git a/kernel/sched/wait.c b/kernel/sched/wait.c index 48c53e4739ea..802d98cf2de3 100644 --- a/kernel/sched/wait.c +++ b/kernel/sched/wait.c @@ -161,6 +161,11 @@ int __wake_up(struct wait_queue_head *wq_head, unsigned int mode, } EXPORT_SYMBOL(__wake_up); +void __wake_up_on_current_cpu(struct wait_queue_head *wq_head, unsigned int mode, void *key) +{ + __wake_up_common_lock(wq_head, mode, 1, WF_CURRENT_CPU, key); +} + /* * Same as __wake_up but called with the spinlock in wait_queue_head_t held. */ diff --git a/kernel/seccomp.c b/kernel/seccomp.c index d3e584065c7f..255999ba9190 100644 --- a/kernel/seccomp.c +++ b/kernel/seccomp.c @@ -110,11 +110,13 @@ struct seccomp_knotif { * @flags: The flags for the new file descriptor. At the moment, only O_CLOEXEC * is allowed. * @ioctl_flags: The flags used for the seccomp_addfd ioctl. + * @setfd: whether or not SECCOMP_ADDFD_FLAG_SETFD was set during notify_addfd * @ret: The return value of the installing process. It is set to the fd num * upon success (>= 0). * @completion: Indicates that the installing process has completed fd * installation, or gone away (either due to successful * reply, or signal) + * @list: list_head for chaining seccomp_kaddfd together. * */ struct seccomp_kaddfd { @@ -138,14 +140,17 @@ struct seccomp_kaddfd { * structure is fairly large, we store the notification-specific stuff in a * separate structure. * - * @request: A semaphore that users of this notification can wait on for - * changes. Actual reads and writes are still controlled with - * filter->notify_lock. + * @requests: A semaphore that users of this notification can wait on for + * changes. Actual reads and writes are still controlled with + * filter->notify_lock. + * @flags: A set of SECCOMP_USER_NOTIF_FD_* flags. * @next_id: The id of the next request. * @notifications: A list of struct seccomp_knotif elements. */ + struct notification { - struct semaphore request; + atomic_t requests; + u32 flags; u64 next_id; struct list_head notifications; }; @@ -555,6 +560,8 @@ static void __seccomp_filter_release(struct seccomp_filter *orig) * drop its reference count, and notify * about unused filters * + * @tsk: task the filter should be released from. + * * This function should only be called when the task is exiting as * it detaches it from its filter tree. As such, READ_ONCE() and * barriers are not needed here, as would normally be needed. @@ -574,6 +581,8 @@ void seccomp_filter_release(struct task_struct *tsk) /** * seccomp_sync_threads: sets all threads to use current's filter * + * @flags: SECCOMP_FILTER_FLAG_* flags to set during sync. + * * Expects sighand and cred_guard_mutex locks to be held, and for * seccomp_can_sync_threads() to have returned success already * without dropping the locks. @@ -1116,8 +1125,11 @@ static int seccomp_do_user_notification(int this_syscall, list_add_tail(&n.list, &match->notif->notifications); INIT_LIST_HEAD(&n.addfd); - up(&match->notif->request); - wake_up_poll(&match->wqh, EPOLLIN | EPOLLRDNORM); + atomic_inc(&match->notif->requests); + if (match->notif->flags & SECCOMP_USER_NOTIF_FD_SYNC_WAKE_UP) + wake_up_poll_on_current_cpu(&match->wqh, EPOLLIN | EPOLLRDNORM); + else + wake_up_poll(&match->wqh, EPOLLIN | EPOLLRDNORM); /* * This is where we wait for a reply from userspace. @@ -1450,6 +1462,37 @@ find_notification(struct seccomp_filter *filter, u64 id) return NULL; } +static int recv_wake_function(wait_queue_entry_t *wait, unsigned int mode, int sync, + void *key) +{ + /* Avoid a wakeup if event not interesting for us. */ + if (key && !(key_to_poll(key) & (EPOLLIN | EPOLLERR))) + return 0; + return autoremove_wake_function(wait, mode, sync, key); +} + +static int recv_wait_event(struct seccomp_filter *filter) +{ + DEFINE_WAIT_FUNC(wait, recv_wake_function); + int ret; + + if (atomic_dec_if_positive(&filter->notif->requests) >= 0) + return 0; + + for (;;) { + ret = prepare_to_wait_event(&filter->wqh, &wait, TASK_INTERRUPTIBLE); + + if (atomic_dec_if_positive(&filter->notif->requests) >= 0) + break; + + if (ret) + return ret; + + schedule(); + } + finish_wait(&filter->wqh, &wait); + return 0; +} static long seccomp_notify_recv(struct seccomp_filter *filter, void __user *buf) @@ -1467,7 +1510,7 @@ static long seccomp_notify_recv(struct seccomp_filter *filter, memset(&unotif, 0, sizeof(unotif)); - ret = down_interruptible(&filter->notif->request); + ret = recv_wait_event(filter); if (ret < 0) return ret; @@ -1515,7 +1558,8 @@ out: if (should_sleep_killable(filter, knotif)) complete(&knotif->ready); knotif->state = SECCOMP_NOTIFY_INIT; - up(&filter->notif->request); + atomic_inc(&filter->notif->requests); + wake_up_poll(&filter->wqh, EPOLLIN | EPOLLRDNORM); } mutex_unlock(&filter->notify_lock); } @@ -1561,7 +1605,10 @@ static long seccomp_notify_send(struct seccomp_filter *filter, knotif->error = resp.error; knotif->val = resp.val; knotif->flags = resp.flags; - complete(&knotif->ready); + if (filter->notif->flags & SECCOMP_USER_NOTIF_FD_SYNC_WAKE_UP) + complete_on_current_cpu(&knotif->ready); + else + complete(&knotif->ready); out: mutex_unlock(&filter->notify_lock); return ret; @@ -1591,6 +1638,22 @@ static long seccomp_notify_id_valid(struct seccomp_filter *filter, return ret; } +static long seccomp_notify_set_flags(struct seccomp_filter *filter, + unsigned long flags) +{ + long ret; + + if (flags & ~SECCOMP_USER_NOTIF_FD_SYNC_WAKE_UP) + return -EINVAL; + + ret = mutex_lock_interruptible(&filter->notify_lock); + if (ret < 0) + return ret; + filter->notif->flags = flags; + mutex_unlock(&filter->notify_lock); + return 0; +} + static long seccomp_notify_addfd(struct seccomp_filter *filter, struct seccomp_notif_addfd __user *uaddfd, unsigned int size) @@ -1720,6 +1783,8 @@ static long seccomp_notify_ioctl(struct file *file, unsigned int cmd, case SECCOMP_IOCTL_NOTIF_ID_VALID_WRONG_DIR: case SECCOMP_IOCTL_NOTIF_ID_VALID: return seccomp_notify_id_valid(filter, buf); + case SECCOMP_IOCTL_NOTIF_SET_FLAGS: + return seccomp_notify_set_flags(filter, arg); } /* Extensible Argument ioctls */ @@ -1777,7 +1842,6 @@ static struct file *init_listener(struct seccomp_filter *filter) if (!filter->notif) goto out; - sema_init(&filter->notif->request, 0); filter->notif->next_id = get_random_u64(); INIT_LIST_HEAD(&filter->notif->notifications); diff --git a/kernel/signal.c b/kernel/signal.c index b5370fe5c198..128e9bb3d1a2 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -562,6 +562,10 @@ bool unhandled_signal(struct task_struct *tsk, int sig) if (handler != SIG_IGN && handler != SIG_DFL) return false; + /* If dying, we handle all new signals by ignoring them */ + if (fatal_signal_pending(tsk)) + return false; + /* if ptraced, let the tracer determine */ return !tsk->ptrace; } diff --git a/kernel/smp.c b/kernel/smp.c index 385179dae360..8455a53465af 100644 --- a/kernel/smp.c +++ b/kernel/smp.c @@ -46,6 +46,8 @@ static DEFINE_PER_CPU_ALIGNED(struct call_function_data, cfd_data); static DEFINE_PER_CPU_SHARED_ALIGNED(struct llist_head, call_single_queue); +static DEFINE_PER_CPU(atomic_t, trigger_backtrace) = ATOMIC_INIT(1); + static void __flush_smp_call_function_queue(bool warn_cpu_offline); int smpcfd_prepare_cpu(unsigned int cpu) @@ -253,13 +255,15 @@ static bool csd_lock_wait_toolong(struct __call_single_data *csd, u64 ts0, u64 * *bug_id, !cpu_cur_csd ? "unresponsive" : "handling this request"); } if (cpu >= 0) { - dump_cpu_task(cpu); + if (atomic_cmpxchg_acquire(&per_cpu(trigger_backtrace, cpu), 1, 0)) + dump_cpu_task(cpu); if (!cpu_cur_csd) { pr_alert("csd: Re-sending CSD lock (#%d) IPI from CPU#%02d to CPU#%02d\n", *bug_id, raw_smp_processor_id(), cpu); arch_send_call_function_single_ipi(cpu); } } - dump_stack(); + if (firsttime) + dump_stack(); *ts1 = ts2; return false; @@ -433,9 +437,14 @@ static void __flush_smp_call_function_queue(bool warn_cpu_offline) struct llist_node *entry, *prev; struct llist_head *head; static bool warned; + atomic_t *tbt; lockdep_assert_irqs_disabled(); + /* Allow waiters to send backtrace NMI from here onwards */ + tbt = this_cpu_ptr(&trigger_backtrace); + atomic_set_release(tbt, 1); + head = this_cpu_ptr(&call_single_queue); entry = llist_del_all(head); entry = llist_reverse_order(entry); diff --git a/kernel/softirq.c b/kernel/softirq.c index 807b34ccd797..210cf5f8d92c 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -612,7 +612,7 @@ static inline void tick_irq_exit(void) int cpu = smp_processor_id(); /* Make sure that timer wheel updates are propagated */ - if ((idle_cpu(cpu) && !need_resched()) || tick_nohz_full_cpu(cpu)) { + if ((sched_core_idle_cpu(cpu) && !need_resched()) || tick_nohz_full_cpu(cpu)) { if (!in_hardirq()) tick_nohz_irq_exit(); } diff --git a/kernel/sys.c b/kernel/sys.c index 05f838929e72..2410e3999ebe 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -2535,11 +2535,6 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, else return -EINVAL; break; - case PR_GET_AUXV: - if (arg4 || arg5) - return -EINVAL; - error = prctl_get_auxv((void __user *)arg2, arg3); - break; default: return -EINVAL; } @@ -2694,6 +2689,11 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, case PR_SET_VMA: error = prctl_set_vma(arg2, arg3, arg4, arg5); break; + case PR_GET_AUXV: + if (arg4 || arg5) + return -EINVAL; + error = prctl_get_auxv((void __user *)arg2, arg3); + break; #ifdef CONFIG_KSM case PR_SET_MEMORY_MERGE: if (arg3 || arg4 || arg5) diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index 88cbc1181b23..c108ed8a9804 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -473,8 +473,8 @@ static void clocksource_watchdog(struct timer_list *unused) /* Check the deviation from the watchdog clocksource. */ md = cs->uncertainty_margin + watchdog->uncertainty_margin; if (abs(cs_nsec - wd_nsec) > md) { - u64 cs_wd_msec; - u64 wd_msec; + s64 cs_wd_msec; + s64 wd_msec; u32 wd_rem; pr_warn("timekeeping watchdog on CPU%d: Marking clocksource '%s' as unstable because the skew is too large:\n", @@ -483,8 +483,8 @@ static void clocksource_watchdog(struct timer_list *unused) watchdog->name, wd_nsec, wdnow, wdlast, watchdog->mask); pr_warn(" '%s' cs_nsec: %lld cs_now: %llx cs_last: %llx mask: %llx\n", cs->name, cs_nsec, csnow, cslast, cs->mask); - cs_wd_msec = div_u64_rem(cs_nsec - wd_nsec, 1000U * 1000U, &wd_rem); - wd_msec = div_u64_rem(wd_nsec, 1000U * 1000U, &wd_rem); + cs_wd_msec = div_s64_rem(cs_nsec - wd_nsec, 1000 * 1000, &wd_rem); + wd_msec = div_s64_rem(wd_nsec, 1000 * 1000, &wd_rem); pr_warn(" Clocksource '%s' skewed %lld ns (%lld ms) over watchdog '%s' interval of %lld ns (%lld ms)\n", cs->name, cs_nsec - wd_nsec, cs_wd_msec, watchdog->name, wd_nsec, wd_msec); if (curr_clocksource == cs) diff --git a/kernel/torture.c b/kernel/torture.c index 1a0519b836ac..b28b05bbef02 100644 --- a/kernel/torture.c +++ b/kernel/torture.c @@ -37,6 +37,7 @@ #include <linux/ktime.h> #include <asm/byteorder.h> #include <linux/torture.h> +#include <linux/sched/rt.h> #include "rcu/rcu.h" MODULE_LICENSE("GPL"); @@ -54,6 +55,9 @@ module_param(verbose_sleep_frequency, int, 0444); static int verbose_sleep_duration = 1; module_param(verbose_sleep_duration, int, 0444); +static int random_shuffle; +module_param(random_shuffle, int, 0444); + static char *torture_type; static int verbose; @@ -88,8 +92,8 @@ int torture_hrtimeout_ns(ktime_t baset_ns, u32 fuzzt_ns, struct torture_random_s ktime_t hto = baset_ns; if (trsp) - hto += (torture_random(trsp) >> 3) % fuzzt_ns; - set_current_state(TASK_UNINTERRUPTIBLE); + hto += torture_random(trsp) % fuzzt_ns; + set_current_state(TASK_IDLE); return schedule_hrtimeout(&hto, HRTIMER_MODE_REL); } EXPORT_SYMBOL_GPL(torture_hrtimeout_ns); @@ -350,22 +354,22 @@ torture_onoff(void *arg) if (onoff_holdoff > 0) { VERBOSE_TOROUT_STRING("torture_onoff begin holdoff"); - schedule_timeout_interruptible(onoff_holdoff); + torture_hrtimeout_jiffies(onoff_holdoff, &rand); VERBOSE_TOROUT_STRING("torture_onoff end holdoff"); } while (!torture_must_stop()) { if (disable_onoff_at_boot && !rcu_inkernel_boot_has_ended()) { - schedule_timeout_interruptible(HZ / 10); + torture_hrtimeout_jiffies(HZ / 10, &rand); continue; } - cpu = (torture_random(&rand) >> 4) % (maxcpu + 1); + cpu = torture_random(&rand) % (maxcpu + 1); if (!torture_offline(cpu, &n_offline_attempts, &n_offline_successes, &sum_offline, &min_offline, &max_offline)) torture_online(cpu, &n_online_attempts, &n_online_successes, &sum_online, &min_online, &max_online); - schedule_timeout_interruptible(onoff_interval); + torture_hrtimeout_jiffies(onoff_interval, &rand); } stop: @@ -518,6 +522,7 @@ static void torture_shuffle_task_unregister_all(void) */ static void torture_shuffle_tasks(void) { + DEFINE_TORTURE_RANDOM(rand); struct shuffle_task *stp; cpumask_setall(shuffle_tmp_mask); @@ -537,8 +542,10 @@ static void torture_shuffle_tasks(void) cpumask_clear_cpu(shuffle_idle_cpu, shuffle_tmp_mask); mutex_lock(&shuffle_task_mutex); - list_for_each_entry(stp, &shuffle_task_list, st_l) - set_cpus_allowed_ptr(stp->st_t, shuffle_tmp_mask); + list_for_each_entry(stp, &shuffle_task_list, st_l) { + if (!random_shuffle || torture_random(&rand) & 0x1) + set_cpus_allowed_ptr(stp->st_t, shuffle_tmp_mask); + } mutex_unlock(&shuffle_task_mutex); cpus_read_unlock(); @@ -550,9 +557,11 @@ static void torture_shuffle_tasks(void) */ static int torture_shuffle(void *arg) { + DEFINE_TORTURE_RANDOM(rand); + VERBOSE_TOROUT_STRING("torture_shuffle task started"); do { - schedule_timeout_interruptible(shuffle_interval); + torture_hrtimeout_jiffies(shuffle_interval, &rand); torture_shuffle_tasks(); torture_shutdown_absorb("torture_shuffle"); } while (!torture_must_stop()); @@ -728,12 +737,12 @@ bool stutter_wait(const char *title) cond_resched_tasks_rcu_qs(); spt = READ_ONCE(stutter_pause_test); for (; spt; spt = READ_ONCE(stutter_pause_test)) { - if (!ret) { + if (!ret && !rt_task(current)) { sched_set_normal(current, MAX_NICE); ret = true; } if (spt == 1) { - schedule_timeout_interruptible(1); + torture_hrtimeout_jiffies(1, NULL); } else if (spt == 2) { while (READ_ONCE(stutter_pause_test)) { if (!(i++ & 0xffff)) @@ -741,7 +750,7 @@ bool stutter_wait(const char *title) cond_resched(); } } else { - schedule_timeout_interruptible(round_jiffies_relative(HZ)); + torture_hrtimeout_jiffies(round_jiffies_relative(HZ), NULL); } torture_shutdown_absorb(title); } @@ -926,7 +935,7 @@ EXPORT_SYMBOL_GPL(torture_kthread_stopping); * it starts, you will need to open-code your own. */ int _torture_create_kthread(int (*fn)(void *arg), void *arg, char *s, char *m, - char *f, struct task_struct **tp) + char *f, struct task_struct **tp, void (*cbf)(struct task_struct *tp)) { int ret = 0; @@ -938,6 +947,10 @@ int _torture_create_kthread(int (*fn)(void *arg), void *arg, char *s, char *m, *tp = NULL; return ret; } + + if (cbf) + cbf(*tp); + wake_up_process(*tp); // Process is sleeping, so ordering provided. torture_shuffle_task_register(*tp); return ret; diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 5f2dcabad202..bd1a42b23f3f 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -661,8 +661,7 @@ static DEFINE_PER_CPU(int, bpf_trace_nest_level); BPF_CALL_5(bpf_perf_event_output, struct pt_regs *, regs, struct bpf_map *, map, u64, flags, void *, data, u64, size) { - struct bpf_trace_sample_data *sds = this_cpu_ptr(&bpf_trace_sds); - int nest_level = this_cpu_inc_return(bpf_trace_nest_level); + struct bpf_trace_sample_data *sds; struct perf_raw_record raw = { .frag = { .size = size, @@ -670,7 +669,11 @@ BPF_CALL_5(bpf_perf_event_output, struct pt_regs *, regs, struct bpf_map *, map, }, }; struct perf_sample_data *sd; - int err; + int nest_level, err; + + preempt_disable(); + sds = this_cpu_ptr(&bpf_trace_sds); + nest_level = this_cpu_inc_return(bpf_trace_nest_level); if (WARN_ON_ONCE(nest_level > ARRAY_SIZE(sds->sds))) { err = -EBUSY; @@ -688,9 +691,9 @@ BPF_CALL_5(bpf_perf_event_output, struct pt_regs *, regs, struct bpf_map *, map, perf_sample_save_raw_data(sd, &raw); err = __bpf_perf_event_output(regs, map, flags, sd); - out: this_cpu_dec(bpf_trace_nest_level); + preempt_enable(); return err; } @@ -715,7 +718,6 @@ static DEFINE_PER_CPU(struct bpf_trace_sample_data, bpf_misc_sds); u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, void *ctx, u64 ctx_size, bpf_ctx_copy_t ctx_copy) { - int nest_level = this_cpu_inc_return(bpf_event_output_nest_level); struct perf_raw_frag frag = { .copy = ctx_copy, .size = ctx_size, @@ -732,8 +734,12 @@ u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, }; struct perf_sample_data *sd; struct pt_regs *regs; + int nest_level; u64 ret; + preempt_disable(); + nest_level = this_cpu_inc_return(bpf_event_output_nest_level); + if (WARN_ON_ONCE(nest_level > ARRAY_SIZE(bpf_misc_sds.sds))) { ret = -EBUSY; goto out; @@ -748,6 +754,7 @@ u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, ret = __bpf_perf_event_output(regs, map, flags, sd); out: this_cpu_dec(bpf_event_output_nest_level); + preempt_enable(); return ret; } diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c index cd2c35b1dd8f..c83c005e654e 100644 --- a/kernel/trace/fgraph.c +++ b/kernel/trace/fgraph.c @@ -15,6 +15,7 @@ #include <trace/events/sched.h> #include "ftrace_internal.h" +#include "trace.h" #ifdef CONFIG_DYNAMIC_FTRACE #define ASSIGN_OPS_HASH(opsname, val) \ diff --git a/kernel/trace/fprobe.c b/kernel/trace/fprobe.c index e4704ec26df7..3b21f4063258 100644 --- a/kernel/trace/fprobe.c +++ b/kernel/trace/fprobe.c @@ -100,14 +100,22 @@ static void fprobe_kprobe_handler(unsigned long ip, unsigned long parent_ip, return; } + /* + * This user handler is shared with other kprobes and is not expected to be + * called recursively. So if any other kprobe handler is running, this will + * exit as kprobe does. See the section 'Share the callbacks with kprobes' + * in Documentation/trace/fprobe.rst for more information. + */ if (unlikely(kprobe_running())) { fp->nmissed++; - return; + goto recursion_unlock; } kprobe_busy_begin(); __fprobe_handler(ip, parent_ip, ops, fregs); kprobe_busy_end(); + +recursion_unlock: ftrace_test_recursion_unlock(bit); } @@ -371,19 +379,16 @@ int unregister_fprobe(struct fprobe *fp) if (!fprobe_is_registered(fp)) return -EINVAL; - /* - * rethook_free() starts disabling the rethook, but the rethook handlers - * may be running on other processors at this point. To make sure that all - * current running handlers are finished, call unregister_ftrace_function() - * after this. - */ if (fp->rethook) - rethook_free(fp->rethook); + rethook_stop(fp->rethook); ret = unregister_ftrace_function(&fp->ops); if (ret < 0) return ret; + if (fp->rethook) + rethook_free(fp->rethook); + ftrace_free_filter(&fp->ops); return ret; diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 3740aca79fe7..05c0024815bf 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -3305,6 +3305,22 @@ static int ftrace_allocate_records(struct ftrace_page *pg, int count) return cnt; } +static void ftrace_free_pages(struct ftrace_page *pages) +{ + struct ftrace_page *pg = pages; + + while (pg) { + if (pg->records) { + free_pages((unsigned long)pg->records, pg->order); + ftrace_number_of_pages -= 1 << pg->order; + } + pages = pg->next; + kfree(pg); + pg = pages; + ftrace_number_of_groups--; + } +} + static struct ftrace_page * ftrace_allocate_pages(unsigned long num_to_init) { @@ -3343,17 +3359,7 @@ ftrace_allocate_pages(unsigned long num_to_init) return start_pg; free_pages: - pg = start_pg; - while (pg) { - if (pg->records) { - free_pages((unsigned long)pg->records, pg->order); - ftrace_number_of_pages -= 1 << pg->order; - } - start_pg = pg->next; - kfree(pg); - pg = start_pg; - ftrace_number_of_groups--; - } + ftrace_free_pages(start_pg); pr_info("ftrace: FAILED to allocate memory for functions\n"); return NULL; } @@ -6471,9 +6477,11 @@ static int ftrace_process_locs(struct module *mod, unsigned long *start, unsigned long *end) { + struct ftrace_page *pg_unuse = NULL; struct ftrace_page *start_pg; struct ftrace_page *pg; struct dyn_ftrace *rec; + unsigned long skipped = 0; unsigned long count; unsigned long *p; unsigned long addr; @@ -6536,8 +6544,10 @@ static int ftrace_process_locs(struct module *mod, * object files to satisfy alignments. * Skip any NULL pointers. */ - if (!addr) + if (!addr) { + skipped++; continue; + } end_offset = (pg->index+1) * sizeof(pg->records[0]); if (end_offset > PAGE_SIZE << pg->order) { @@ -6551,8 +6561,10 @@ static int ftrace_process_locs(struct module *mod, rec->ip = addr; } - /* We should have used all pages */ - WARN_ON(pg->next); + if (pg->next) { + pg_unuse = pg->next; + pg->next = NULL; + } /* Assign the last page to ftrace_pages */ ftrace_pages = pg; @@ -6574,6 +6586,11 @@ static int ftrace_process_locs(struct module *mod, out: mutex_unlock(&ftrace_lock); + /* We should have used all pages unless we skipped some */ + if (pg_unuse) { + WARN_ON(!skipped); + ftrace_free_pages(pg_unuse); + } return ret; } diff --git a/kernel/trace/ftrace_internal.h b/kernel/trace/ftrace_internal.h index 382775edf690..5012c04f92c0 100644 --- a/kernel/trace/ftrace_internal.h +++ b/kernel/trace/ftrace_internal.h @@ -2,6 +2,9 @@ #ifndef _LINUX_KERNEL_FTRACE_INTERNAL_H #define _LINUX_KERNEL_FTRACE_INTERNAL_H +int __register_ftrace_function(struct ftrace_ops *ops); +int __unregister_ftrace_function(struct ftrace_ops *ops); + #ifdef CONFIG_FUNCTION_TRACER extern struct mutex ftrace_lock; @@ -15,8 +18,6 @@ int ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip, void *regs); #else /* !CONFIG_DYNAMIC_FTRACE */ -int __register_ftrace_function(struct ftrace_ops *ops); -int __unregister_ftrace_function(struct ftrace_ops *ops); /* Keep as macros so we do not need to define the commands */ # define ftrace_startup(ops, command) \ ({ \ diff --git a/kernel/trace/rethook.c b/kernel/trace/rethook.c index f32ee484391a..5eb9b598f4e9 100644 --- a/kernel/trace/rethook.c +++ b/kernel/trace/rethook.c @@ -54,6 +54,19 @@ static void rethook_free_rcu(struct rcu_head *head) } /** + * rethook_stop() - Stop using a rethook. + * @rh: the struct rethook to stop. + * + * Stop using a rethook to prepare for freeing it. If you want to wait for + * all running rethook handler before calling rethook_free(), you need to + * call this first and wait RCU, and call rethook_free(). + */ +void rethook_stop(struct rethook *rh) +{ + WRITE_ONCE(rh->handler, NULL); +} + +/** * rethook_free() - Free struct rethook. * @rh: the struct rethook to be freed. * diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 834b361a4a66..52dea5dd5362 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -523,6 +523,8 @@ struct ring_buffer_per_cpu { rb_time_t before_stamp; u64 event_stamp[MAX_NEST]; u64 read_stamp; + /* pages removed since last reset */ + unsigned long pages_removed; /* ring buffer pages to update, > 0 to add, < 0 to remove */ long nr_pages_to_update; struct list_head new_pages; /* new pages to add */ @@ -536,6 +538,7 @@ struct trace_buffer { unsigned flags; int cpus; atomic_t record_disabled; + atomic_t resizing; cpumask_var_t cpumask; struct lock_class_key *reader_lock_key; @@ -558,6 +561,7 @@ struct ring_buffer_iter { struct buffer_page *head_page; struct buffer_page *cache_reader_page; unsigned long cache_read; + unsigned long cache_pages_removed; u64 read_stamp; u64 page_stamp; struct ring_buffer_event *event; @@ -946,6 +950,7 @@ static void rb_wake_up_waiters(struct irq_work *work) /** * ring_buffer_wake_waiters - wake up any waiters on this ring buffer * @buffer: The ring buffer to wake waiters on + * @cpu: The CPU buffer to wake waiters on * * In the case of a file that represents a ring buffer is closing, * it is prudent to wake up any waiters that are on this. @@ -1956,6 +1961,8 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned long nr_pages) to_remove = rb_list_head(to_remove)->next; head_bit |= (unsigned long)to_remove & RB_PAGE_HEAD; } + /* Read iterators need to reset themselves when some pages removed */ + cpu_buffer->pages_removed += nr_removed; next_page = rb_list_head(to_remove)->next; @@ -1977,12 +1984,6 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned long nr_pages) cpu_buffer->head_page = list_entry(next_page, struct buffer_page, list); - /* - * change read pointer to make sure any read iterators reset - * themselves - */ - cpu_buffer->read = 0; - /* pages are removed, resume tracing and then free the pages */ atomic_dec(&cpu_buffer->record_disabled); raw_spin_unlock_irq(&cpu_buffer->reader_lock); @@ -2167,7 +2168,7 @@ int ring_buffer_resize(struct trace_buffer *buffer, unsigned long size, /* prevent another thread from changing buffer sizes */ mutex_lock(&buffer->mutex); - + atomic_inc(&buffer->resizing); if (cpu_id == RING_BUFFER_ALL_CPUS) { /* @@ -2322,6 +2323,7 @@ int ring_buffer_resize(struct trace_buffer *buffer, unsigned long size, atomic_dec(&buffer->record_disabled); } + atomic_dec(&buffer->resizing); mutex_unlock(&buffer->mutex); return 0; @@ -2342,6 +2344,7 @@ int ring_buffer_resize(struct trace_buffer *buffer, unsigned long size, } } out_err_unlock: + atomic_dec(&buffer->resizing); mutex_unlock(&buffer->mutex); return err; } @@ -3373,7 +3376,6 @@ void ring_buffer_nest_end(struct trace_buffer *buffer) /** * ring_buffer_unlock_commit - commit a reserved * @buffer: The buffer to commit to - * @event: The event pointer to commit. * * This commits the data to the ring buffer, and releases any locks held. * @@ -4392,6 +4394,7 @@ static void rb_iter_reset(struct ring_buffer_iter *iter) iter->cache_reader_page = iter->head_page; iter->cache_read = cpu_buffer->read; + iter->cache_pages_removed = cpu_buffer->pages_removed; if (iter->head) { iter->read_stamp = cpu_buffer->read_stamp; @@ -4846,12 +4849,13 @@ rb_iter_peek(struct ring_buffer_iter *iter, u64 *ts) buffer = cpu_buffer->buffer; /* - * Check if someone performed a consuming read to - * the buffer. A consuming read invalidates the iterator - * and we need to reset the iterator in this case. + * Check if someone performed a consuming read to the buffer + * or removed some pages from the buffer. In these cases, + * iterator was invalidated and we need to reset it. */ if (unlikely(iter->cache_read != cpu_buffer->read || - iter->cache_reader_page != cpu_buffer->reader_page)) + iter->cache_reader_page != cpu_buffer->reader_page || + iter->cache_pages_removed != cpu_buffer->pages_removed)) rb_iter_reset(iter); again: @@ -5242,28 +5246,34 @@ unsigned long ring_buffer_size(struct trace_buffer *buffer, int cpu) } EXPORT_SYMBOL_GPL(ring_buffer_size); +static void rb_clear_buffer_page(struct buffer_page *page) +{ + local_set(&page->write, 0); + local_set(&page->entries, 0); + rb_init_page(page->page); + page->read = 0; +} + static void rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer) { + struct buffer_page *page; + rb_head_page_deactivate(cpu_buffer); cpu_buffer->head_page = list_entry(cpu_buffer->pages, struct buffer_page, list); - local_set(&cpu_buffer->head_page->write, 0); - local_set(&cpu_buffer->head_page->entries, 0); - local_set(&cpu_buffer->head_page->page->commit, 0); - - cpu_buffer->head_page->read = 0; + rb_clear_buffer_page(cpu_buffer->head_page); + list_for_each_entry(page, cpu_buffer->pages, list) { + rb_clear_buffer_page(page); + } cpu_buffer->tail_page = cpu_buffer->head_page; cpu_buffer->commit_page = cpu_buffer->head_page; INIT_LIST_HEAD(&cpu_buffer->reader_page->list); INIT_LIST_HEAD(&cpu_buffer->new_pages); - local_set(&cpu_buffer->reader_page->write, 0); - local_set(&cpu_buffer->reader_page->entries, 0); - local_set(&cpu_buffer->reader_page->page->commit, 0); - cpu_buffer->reader_page->read = 0; + rb_clear_buffer_page(cpu_buffer->reader_page); local_set(&cpu_buffer->entries_bytes, 0); local_set(&cpu_buffer->overrun, 0); @@ -5289,6 +5299,7 @@ rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer) cpu_buffer->last_overrun = 0; rb_head_page_activate(cpu_buffer); + cpu_buffer->pages_removed = 0; } /* Must have disabled the cpu buffer then done a synchronize_rcu */ @@ -5347,7 +5358,6 @@ EXPORT_SYMBOL_GPL(ring_buffer_reset_cpu); /** * ring_buffer_reset_online_cpus - reset a ring buffer per CPU buffer * @buffer: The ring buffer to reset a per cpu buffer of - * @cpu: The CPU buffer to be reset */ void ring_buffer_reset_online_cpus(struct trace_buffer *buffer) { @@ -5535,6 +5545,15 @@ int ring_buffer_swap_cpu(struct trace_buffer *buffer_a, if (local_read(&cpu_buffer_b->committing)) goto out_dec; + /* + * When resize is in progress, we cannot swap it because + * it will mess the state of the cpu buffer. + */ + if (atomic_read(&buffer_a->resizing)) + goto out_dec; + if (atomic_read(&buffer_b->resizing)) + goto out_dec; + buffer_a->buffers[cpu] = cpu_buffer_b; buffer_b->buffers[cpu] = cpu_buffer_a; diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 4529e264cb86..8e64aaad5361 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1928,9 +1928,10 @@ update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu) * place on this CPU. We fail to record, but we reset * the max trace buffer (no one writes directly to it) * and flag that it failed. + * Another reason is resize is in progress. */ trace_array_printk_buf(tr->max_buffer.buffer, _THIS_IP_, - "Failed to swap buffers due to commit in progress\n"); + "Failed to swap buffers due to commit or resize in progress\n"); } WARN_ON_ONCE(ret && ret != -EAGAIN && ret != -EBUSY); @@ -3118,6 +3119,7 @@ static void __ftrace_trace_stack(struct trace_buffer *buffer, struct ftrace_stack *fstack; struct stack_entry *entry; int stackidx; + void *ptr; /* * Add one, for this function and the call to save_stack_trace() @@ -3161,9 +3163,25 @@ static void __ftrace_trace_stack(struct trace_buffer *buffer, trace_ctx); if (!event) goto out; - entry = ring_buffer_event_data(event); + ptr = ring_buffer_event_data(event); + entry = ptr; + + /* + * For backward compatibility reasons, the entry->caller is an + * array of 8 slots to store the stack. This is also exported + * to user space. The amount allocated on the ring buffer actually + * holds enough for the stack specified by nr_entries. This will + * go into the location of entry->caller. Due to string fortifiers + * checking the size of the destination of memcpy() it triggers + * when it detects that size is greater than 8. To hide this from + * the fortifiers, we use "ptr" and pointer arithmetic to assign caller. + * + * The below is really just: + * memcpy(&entry->caller, fstack->calls, size); + */ + ptr += offsetof(typeof(*entry), caller); + memcpy(ptr, fstack->calls, size); - memcpy(&entry->caller, fstack->calls, size); entry->size = nr_entries; if (!call_filter_check_discard(call, entry, buffer, event)) @@ -4195,8 +4213,15 @@ static void *s_start(struct seq_file *m, loff_t *pos) * will point to the same string as current_trace->name. */ mutex_lock(&trace_types_lock); - if (unlikely(tr->current_trace && iter->trace->name != tr->current_trace->name)) + if (unlikely(tr->current_trace && iter->trace->name != tr->current_trace->name)) { + /* Close iter->trace before switching to the new current tracer */ + if (iter->trace->close) + iter->trace->close(iter); *iter->trace = *tr->current_trace; + /* Reopen the new current tracer */ + if (iter->trace->open) + iter->trace->open(iter); + } mutex_unlock(&trace_types_lock); #ifdef CONFIG_TRACER_MAX_TRACE @@ -5259,11 +5284,17 @@ int tracing_set_cpumask(struct trace_array *tr, !cpumask_test_cpu(cpu, tracing_cpumask_new)) { atomic_inc(&per_cpu_ptr(tr->array_buffer.data, cpu)->disabled); ring_buffer_record_disable_cpu(tr->array_buffer.buffer, cpu); +#ifdef CONFIG_TRACER_MAX_TRACE + ring_buffer_record_disable_cpu(tr->max_buffer.buffer, cpu); +#endif } if (!cpumask_test_cpu(cpu, tr->tracing_cpumask) && cpumask_test_cpu(cpu, tracing_cpumask_new)) { atomic_dec(&per_cpu_ptr(tr->array_buffer.data, cpu)->disabled); ring_buffer_record_enable_cpu(tr->array_buffer.buffer, cpu); +#ifdef CONFIG_TRACER_MAX_TRACE + ring_buffer_record_enable_cpu(tr->max_buffer.buffer, cpu); +#endif } } arch_spin_unlock(&tr->max_lock); @@ -6687,10 +6718,36 @@ tracing_max_lat_write(struct file *filp, const char __user *ubuf, #endif +static int open_pipe_on_cpu(struct trace_array *tr, int cpu) +{ + if (cpu == RING_BUFFER_ALL_CPUS) { + if (cpumask_empty(tr->pipe_cpumask)) { + cpumask_setall(tr->pipe_cpumask); + return 0; + } + } else if (!cpumask_test_cpu(cpu, tr->pipe_cpumask)) { + cpumask_set_cpu(cpu, tr->pipe_cpumask); + return 0; + } + return -EBUSY; +} + +static void close_pipe_on_cpu(struct trace_array *tr, int cpu) +{ + if (cpu == RING_BUFFER_ALL_CPUS) { + WARN_ON(!cpumask_full(tr->pipe_cpumask)); + cpumask_clear(tr->pipe_cpumask); + } else { + WARN_ON(!cpumask_test_cpu(cpu, tr->pipe_cpumask)); + cpumask_clear_cpu(cpu, tr->pipe_cpumask); + } +} + static int tracing_open_pipe(struct inode *inode, struct file *filp) { struct trace_array *tr = inode->i_private; struct trace_iterator *iter; + int cpu; int ret; ret = tracing_check_open_get_tr(tr); @@ -6698,13 +6755,16 @@ static int tracing_open_pipe(struct inode *inode, struct file *filp) return ret; mutex_lock(&trace_types_lock); + cpu = tracing_get_cpu(inode); + ret = open_pipe_on_cpu(tr, cpu); + if (ret) + goto fail_pipe_on_cpu; /* create a buffer to store the information to pass to userspace */ iter = kzalloc(sizeof(*iter), GFP_KERNEL); if (!iter) { ret = -ENOMEM; - __trace_array_put(tr); - goto out; + goto fail_alloc_iter; } trace_seq_init(&iter->seq); @@ -6727,7 +6787,7 @@ static int tracing_open_pipe(struct inode *inode, struct file *filp) iter->tr = tr; iter->array_buffer = &tr->array_buffer; - iter->cpu_file = tracing_get_cpu(inode); + iter->cpu_file = cpu; mutex_init(&iter->mutex); filp->private_data = iter; @@ -6737,12 +6797,15 @@ static int tracing_open_pipe(struct inode *inode, struct file *filp) nonseekable_open(inode, filp); tr->trace_ref++; -out: + mutex_unlock(&trace_types_lock); return ret; fail: kfree(iter); +fail_alloc_iter: + close_pipe_on_cpu(tr, cpu); +fail_pipe_on_cpu: __trace_array_put(tr); mutex_unlock(&trace_types_lock); return ret; @@ -6759,11 +6822,12 @@ static int tracing_release_pipe(struct inode *inode, struct file *file) if (iter->trace->pipe_close) iter->trace->pipe_close(iter); - + close_pipe_on_cpu(tr, iter->cpu_file); mutex_unlock(&trace_types_lock); free_cpumask_var(iter->started); kfree(iter->fmt); + kfree(iter->temp); mutex_destroy(&iter->mutex); kfree(iter); @@ -9422,6 +9486,9 @@ static struct trace_array *trace_array_create(const char *name) if (!alloc_cpumask_var(&tr->tracing_cpumask, GFP_KERNEL)) goto out_free_tr; + if (!alloc_cpumask_var(&tr->pipe_cpumask, GFP_KERNEL)) + goto out_free_tr; + tr->trace_flags = global_trace.trace_flags & ~ZEROED_TRACE_FLAGS; cpumask_copy(tr->tracing_cpumask, cpu_all_mask); @@ -9463,6 +9530,7 @@ static struct trace_array *trace_array_create(const char *name) out_free_tr: ftrace_free_ftrace_ops(tr); free_trace_buffers(tr); + free_cpumask_var(tr->pipe_cpumask); free_cpumask_var(tr->tracing_cpumask); kfree(tr->name); kfree(tr); @@ -9565,6 +9633,7 @@ static int __remove_instance(struct trace_array *tr) } kfree(tr->topts); + free_cpumask_var(tr->pipe_cpumask); free_cpumask_var(tr->tracing_cpumask); kfree(tr->name); kfree(tr); @@ -10362,12 +10431,14 @@ __init static int tracer_alloc_buffers(void) if (trace_create_savedcmd() < 0) goto out_free_temp_buffer; + if (!alloc_cpumask_var(&global_trace.pipe_cpumask, GFP_KERNEL)) + goto out_free_savedcmd; + /* TODO: make the number of buffers hot pluggable with CPUS */ if (allocate_trace_buffers(&global_trace, ring_buf_size) < 0) { MEM_FAIL(1, "tracer: failed to allocate ring buffer!\n"); - goto out_free_savedcmd; + goto out_free_pipe_cpumask; } - if (global_trace.buffer_disabled) tracing_off(); @@ -10420,6 +10491,8 @@ __init static int tracer_alloc_buffers(void) return 0; +out_free_pipe_cpumask: + free_cpumask_var(global_trace.pipe_cpumask); out_free_savedcmd: free_saved_cmdlines_buffer(savedcmd); out_free_temp_buffer: diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index ed7906b13f09..73eaec158473 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -113,6 +113,8 @@ enum trace_type { #define MEM_FAIL(condition, fmt, ...) \ DO_ONCE_LITE_IF(condition, pr_err, "ERROR: " fmt, ##__VA_ARGS__) +#define FAULT_STRING "(fault)" + #define HIST_STACKTRACE_DEPTH 16 #define HIST_STACKTRACE_SIZE (HIST_STACKTRACE_DEPTH * sizeof(unsigned long)) #define HIST_STACKTRACE_SKIP 5 @@ -375,6 +377,8 @@ struct trace_array { struct list_head events; struct trace_event_file *trace_marker_file; cpumask_var_t tracing_cpumask; /* only trace on set CPUs */ + /* one per_cpu trace_pipe can be opened by only one user */ + cpumask_var_t pipe_cpumask; int ref; int trace_ref; #ifdef CONFIG_FUNCTION_TRACER @@ -1293,6 +1297,14 @@ static inline void trace_branch_disable(void) /* set ring buffers to default size if not already done so */ int tracing_update_buffers(void); +union trace_synth_field { + u8 as_u8; + u16 as_u16; + u32 as_u32; + u64 as_u64; + struct trace_dynamic_info as_dynamic; +}; + struct ftrace_event_field { struct list_head link; const char *name; diff --git a/kernel/trace/trace_eprobe.c b/kernel/trace/trace_eprobe.c index cb0077ba2b49..a0a704ba27db 100644 --- a/kernel/trace/trace_eprobe.c +++ b/kernel/trace/trace_eprobe.c @@ -644,6 +644,7 @@ static int enable_trace_eprobe(struct trace_event_call *call, struct trace_eprobe *ep; bool enabled; int ret = 0; + int cnt = 0; tp = trace_probe_primary_from_call(call); if (WARN_ON_ONCE(!tp)) @@ -667,12 +668,25 @@ static int enable_trace_eprobe(struct trace_event_call *call, if (ret) break; enabled = true; + cnt++; } if (ret) { /* Failed to enable one of them. Roll back all */ - if (enabled) - disable_eprobe(ep, file->tr); + if (enabled) { + /* + * It's a bug if one failed for something other than memory + * not being available but another eprobe succeeded. + */ + WARN_ON_ONCE(ret != -ENOMEM); + + list_for_each_entry(pos, trace_probe_probe_list(tp), list) { + ep = container_of(pos, struct trace_eprobe, tp); + disable_eprobe(ep, file->tr); + if (!--cnt) + break; + } + } if (file) trace_probe_remove_file(tp, file); else diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 5d6ae4eae510..578f1f7d49a6 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -611,7 +611,6 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file, { struct trace_event_call *call = file->event_call; struct trace_array *tr = file->tr; - unsigned long file_flags = file->flags; int ret = 0; int disable; @@ -635,6 +634,8 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file, break; disable = file->flags & EVENT_FILE_FL_SOFT_DISABLED; clear_bit(EVENT_FILE_FL_SOFT_MODE_BIT, &file->flags); + /* Disable use of trace_buffered_event */ + trace_buffered_event_disable(); } else disable = !(file->flags & EVENT_FILE_FL_SOFT_MODE); @@ -673,6 +674,8 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file, if (atomic_inc_return(&file->sm_ref) > 1) break; set_bit(EVENT_FILE_FL_SOFT_MODE_BIT, &file->flags); + /* Enable use of trace_buffered_event */ + trace_buffered_event_enable(); } if (!(file->flags & EVENT_FILE_FL_ENABLED)) { @@ -712,15 +715,6 @@ static int __ftrace_event_enable_disable(struct trace_event_file *file, break; } - /* Enable or disable use of trace_buffered_event */ - if ((file_flags & EVENT_FILE_FL_SOFT_DISABLED) != - (file->flags & EVENT_FILE_FL_SOFT_DISABLED)) { - if (file->flags & EVENT_FILE_FL_SOFT_DISABLED) - trace_buffered_event_enable(); - else - trace_buffered_event_disable(); - } - return ret; } diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c index b97d3ad832f1..d06938ae0717 100644 --- a/kernel/trace/trace_events_hist.c +++ b/kernel/trace/trace_events_hist.c @@ -6663,13 +6663,16 @@ static int event_hist_trigger_parse(struct event_command *cmd_ops, if (get_named_trigger_data(trigger_data)) goto enable; - if (has_hist_vars(hist_data)) - save_hist_vars(hist_data); - ret = create_actions(hist_data); if (ret) goto out_unreg; + if (has_hist_vars(hist_data) || hist_data->n_var_refs) { + ret = save_hist_vars(hist_data); + if (ret) + goto out_unreg; + } + ret = tracing_map_init(hist_data->map); if (ret) goto out_unreg; diff --git a/kernel/trace/trace_events_synth.c b/kernel/trace/trace_events_synth.c index d6a70aff2410..9897d0bfcab7 100644 --- a/kernel/trace/trace_events_synth.c +++ b/kernel/trace/trace_events_synth.c @@ -127,7 +127,7 @@ static bool synth_event_match(const char *system, const char *event, struct synth_trace_event { struct trace_entry ent; - u64 fields[]; + union trace_synth_field fields[]; }; static int synth_event_define_fields(struct trace_event_call *call) @@ -321,19 +321,19 @@ static const char *synth_field_fmt(char *type) static void print_synth_event_num_val(struct trace_seq *s, char *print_fmt, char *name, - int size, u64 val, char *space) + int size, union trace_synth_field *val, char *space) { switch (size) { case 1: - trace_seq_printf(s, print_fmt, name, (u8)val, space); + trace_seq_printf(s, print_fmt, name, val->as_u8, space); break; case 2: - trace_seq_printf(s, print_fmt, name, (u16)val, space); + trace_seq_printf(s, print_fmt, name, val->as_u16, space); break; case 4: - trace_seq_printf(s, print_fmt, name, (u32)val, space); + trace_seq_printf(s, print_fmt, name, val->as_u32, space); break; default: @@ -350,7 +350,7 @@ static enum print_line_t print_synth_event(struct trace_iterator *iter, struct trace_seq *s = &iter->seq; struct synth_trace_event *entry; struct synth_event *se; - unsigned int i, n_u64; + unsigned int i, j, n_u64; char print_fmt[32]; const char *fmt; @@ -374,43 +374,28 @@ static enum print_line_t print_synth_event(struct trace_iterator *iter, /* parameter values */ if (se->fields[i]->is_string) { if (se->fields[i]->is_dynamic) { - u32 offset, data_offset; - char *str_field; - - offset = (u32)entry->fields[n_u64]; - data_offset = offset & 0xffff; - - str_field = (char *)entry + data_offset; + union trace_synth_field *data = &entry->fields[n_u64]; trace_seq_printf(s, print_fmt, se->fields[i]->name, STR_VAR_LEN_MAX, - str_field, + (char *)entry + data->as_dynamic.offset, i == se->n_fields - 1 ? "" : " "); n_u64++; } else { trace_seq_printf(s, print_fmt, se->fields[i]->name, STR_VAR_LEN_MAX, - (char *)&entry->fields[n_u64], + (char *)&entry->fields[n_u64].as_u64, i == se->n_fields - 1 ? "" : " "); n_u64 += STR_VAR_LEN_MAX / sizeof(u64); } } else if (se->fields[i]->is_stack) { - u32 offset, data_offset, len; - unsigned long *p, *end; - - offset = (u32)entry->fields[n_u64]; - data_offset = offset & 0xffff; - len = offset >> 16; - - p = (void *)entry + data_offset; - end = (void *)p + len - (sizeof(long) - 1); + union trace_synth_field *data = &entry->fields[n_u64]; + unsigned long *p = (void *)entry + data->as_dynamic.offset; trace_seq_printf(s, "%s=STACK:\n", se->fields[i]->name); - - for (; *p && p < end; p++) - trace_seq_printf(s, "=> %pS\n", (void *)*p); + for (j = 1; j < data->as_dynamic.len / sizeof(long); j++) + trace_seq_printf(s, "=> %pS\n", (void *)p[j]); n_u64++; - } else { struct trace_print_flags __flags[] = { __def_gfpflag_names, {-1, NULL} }; @@ -419,13 +404,13 @@ static enum print_line_t print_synth_event(struct trace_iterator *iter, print_synth_event_num_val(s, print_fmt, se->fields[i]->name, se->fields[i]->size, - entry->fields[n_u64], + &entry->fields[n_u64], space); if (strcmp(se->fields[i]->type, "gfp_t") == 0) { trace_seq_puts(s, " ("); trace_print_flags_seq(s, "|", - entry->fields[n_u64], + entry->fields[n_u64].as_u64, __flags); trace_seq_putc(s, ')'); } @@ -454,21 +439,16 @@ static unsigned int trace_string(struct synth_trace_event *entry, int ret; if (is_dynamic) { - u32 data_offset; + union trace_synth_field *data = &entry->fields[*n_u64]; - data_offset = struct_size(entry, fields, event->n_u64); - data_offset += data_size; - - len = fetch_store_strlen((unsigned long)str_val); - - data_offset |= len << 16; - *(u32 *)&entry->fields[*n_u64] = data_offset; + data->as_dynamic.offset = struct_size(entry, fields, event->n_u64) + data_size; + data->as_dynamic.len = fetch_store_strlen((unsigned long)str_val); ret = fetch_store_string((unsigned long)str_val, &entry->fields[*n_u64], entry); (*n_u64)++; } else { - str_field = (char *)&entry->fields[*n_u64]; + str_field = (char *)&entry->fields[*n_u64].as_u64; #ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE if ((unsigned long)str_val < TASK_SIZE) @@ -492,6 +472,7 @@ static unsigned int trace_stack(struct synth_trace_event *entry, unsigned int data_size, unsigned int *n_u64) { + union trace_synth_field *data = &entry->fields[*n_u64]; unsigned int len; u32 data_offset; void *data_loc; @@ -504,10 +485,6 @@ static unsigned int trace_stack(struct synth_trace_event *entry, break; } - /* Include the zero'd element if it fits */ - if (len < HIST_STACKTRACE_DEPTH) - len++; - len *= sizeof(long); /* Find the dynamic section to copy the stack into. */ @@ -515,8 +492,9 @@ static unsigned int trace_stack(struct synth_trace_event *entry, memcpy(data_loc, stack, len); /* Fill in the field that holds the offset/len combo */ - data_offset |= len << 16; - *(u32 *)&entry->fields[*n_u64] = data_offset; + + data->as_dynamic.offset = data_offset; + data->as_dynamic.len = len; (*n_u64)++; @@ -550,7 +528,8 @@ static notrace void trace_event_raw_event_synth(void *__data, str_val = (char *)(long)var_ref_vals[val_idx]; if (event->dynamic_fields[i]->is_stack) { - len = *((unsigned long *)str_val); + /* reserve one extra element for size */ + len = *((unsigned long *)str_val) + 1; len *= sizeof(unsigned long); } else { len = fetch_store_strlen((unsigned long)str_val); @@ -592,19 +571,19 @@ static notrace void trace_event_raw_event_synth(void *__data, switch (field->size) { case 1: - *(u8 *)&entry->fields[n_u64] = (u8)val; + entry->fields[n_u64].as_u8 = (u8)val; break; case 2: - *(u16 *)&entry->fields[n_u64] = (u16)val; + entry->fields[n_u64].as_u16 = (u16)val; break; case 4: - *(u32 *)&entry->fields[n_u64] = (u32)val; + entry->fields[n_u64].as_u32 = (u32)val; break; default: - entry->fields[n_u64] = val; + entry->fields[n_u64].as_u64 = val; break; } n_u64++; @@ -1230,6 +1209,7 @@ EXPORT_SYMBOL_GPL(__synth_event_gen_cmd_start); * synth_event_gen_cmd_array_start - Start synthetic event command from an array * @cmd: A pointer to the dynevent_cmd struct representing the new event * @name: The name of the synthetic event + * @mod: The module creating the event, NULL if not created from a module * @fields: An array of type/name field descriptions * @n_fields: The number of field descriptions contained in the fields array * @@ -1790,19 +1770,19 @@ int synth_event_trace(struct trace_event_file *file, unsigned int n_vals, ...) switch (field->size) { case 1: - *(u8 *)&state.entry->fields[n_u64] = (u8)val; + state.entry->fields[n_u64].as_u8 = (u8)val; break; case 2: - *(u16 *)&state.entry->fields[n_u64] = (u16)val; + state.entry->fields[n_u64].as_u16 = (u16)val; break; case 4: - *(u32 *)&state.entry->fields[n_u64] = (u32)val; + state.entry->fields[n_u64].as_u32 = (u32)val; break; default: - state.entry->fields[n_u64] = val; + state.entry->fields[n_u64].as_u64 = val; break; } n_u64++; @@ -1883,19 +1863,19 @@ int synth_event_trace_array(struct trace_event_file *file, u64 *vals, switch (field->size) { case 1: - *(u8 *)&state.entry->fields[n_u64] = (u8)val; + state.entry->fields[n_u64].as_u8 = (u8)val; break; case 2: - *(u16 *)&state.entry->fields[n_u64] = (u16)val; + state.entry->fields[n_u64].as_u16 = (u16)val; break; case 4: - *(u32 *)&state.entry->fields[n_u64] = (u32)val; + state.entry->fields[n_u64].as_u32 = (u32)val; break; default: - state.entry->fields[n_u64] = val; + state.entry->fields[n_u64].as_u64 = val; break; } n_u64++; @@ -2030,19 +2010,19 @@ static int __synth_event_add_val(const char *field_name, u64 val, } else { switch (field->size) { case 1: - *(u8 *)&trace_state->entry->fields[field->offset] = (u8)val; + trace_state->entry->fields[field->offset].as_u8 = (u8)val; break; case 2: - *(u16 *)&trace_state->entry->fields[field->offset] = (u16)val; + trace_state->entry->fields[field->offset].as_u16 = (u16)val; break; case 4: - *(u32 *)&trace_state->entry->fields[field->offset] = (u32)val; + trace_state->entry->fields[field->offset].as_u32 = (u32)val; break; default: - trace_state->entry->fields[field->offset] = val; + trace_state->entry->fields[field->offset].as_u64 = val; break; } } diff --git a/kernel/trace/trace_events_trigger.c b/kernel/trace/trace_events_trigger.c index e535959939d3..46439e3bcec4 100644 --- a/kernel/trace/trace_events_trigger.c +++ b/kernel/trace/trace_events_trigger.c @@ -31,7 +31,9 @@ void trigger_data_free(struct event_trigger_data *data) /** * event_triggers_call - Call triggers associated with a trace event * @file: The trace_event_file associated with the event + * @buffer: The ring buffer that the event is being written to * @rec: The trace entry for the event, NULL for unconditional invocation + * @event: The event meta data in the ring buffer * * For each trigger associated with an event, invoke the trigger * function registered with the associated trigger command. If rec is diff --git a/kernel/trace/trace_events_user.c b/kernel/trace/trace_events_user.c index 4f5e74bbdab2..33cb6af31f39 100644 --- a/kernel/trace/trace_events_user.c +++ b/kernel/trace/trace_events_user.c @@ -1317,6 +1317,9 @@ static int user_field_set_string(struct ftrace_event_field *field, pos += snprintf(buf + pos, LEN_OR_ZERO, " "); pos += snprintf(buf + pos, LEN_OR_ZERO, "%s", field->name); + if (str_has_prefix(field->type, "struct ")) + pos += snprintf(buf + pos, LEN_OR_ZERO, " %d", field->size); + if (colon) pos += snprintf(buf + pos, LEN_OR_ZERO, ";"); diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index 590b3d51afae..ba37f768e2f2 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -231,7 +231,8 @@ static void irqsoff_trace_open(struct trace_iterator *iter) { if (is_graph(iter->tr)) graph_trace_open(iter); - + else + iter->private = NULL; } static void irqsoff_trace_close(struct trace_iterator *iter) diff --git a/kernel/trace/trace_kprobe_selftest.c b/kernel/trace/trace_kprobe_selftest.c index 16548ee4c8c6..3851cd1e6a62 100644 --- a/kernel/trace/trace_kprobe_selftest.c +++ b/kernel/trace/trace_kprobe_selftest.c @@ -1,4 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 + +#include "trace_kprobe_selftest.h" + /* * Function used during the kprobe self test. This function is in a separate * compile unit so it can be compile with CC_FLAGS_FTRACE to ensure that it diff --git a/kernel/trace/trace_probe.c b/kernel/trace/trace_probe.c index 7ba371da0926..c68a72707852 100644 --- a/kernel/trace/trace_probe.c +++ b/kernel/trace/trace_probe.c @@ -67,7 +67,7 @@ int PRINT_TYPE_FUNC_NAME(string)(struct trace_seq *s, void *data, void *ent) int len = *(u32 *)data >> 16; if (!len) - trace_seq_puts(s, "(fault)"); + trace_seq_puts(s, FAULT_STRING); else trace_seq_printf(s, "\"%s\"", (const char *)get_loc_data(data, ent)); @@ -386,12 +386,12 @@ static const struct btf_type *find_btf_func_proto(const char *funcname) /* Get BTF_KIND_FUNC type */ t = btf_type_by_id(btf, id); - if (!btf_type_is_func(t)) + if (!t || !btf_type_is_func(t)) return ERR_PTR(-ENOENT); /* The type of BTF_KIND_FUNC is BTF_KIND_FUNC_PROTO */ t = btf_type_by_id(btf, t->type); - if (!btf_type_is_func_proto(t)) + if (!t || !btf_type_is_func_proto(t)) return ERR_PTR(-ENOENT); return t; @@ -443,7 +443,7 @@ static int parse_btf_arg(const char *varname, struct fetch_insn *code, if (!ctx->params) { params = find_btf_func_param(ctx->funcname, &ctx->nr_params, ctx->flags & TPARG_FL_TPOINT); - if (IS_ERR(params)) { + if (IS_ERR_OR_NULL(params)) { trace_probe_log_err(ctx->offset, NO_BTF_ENTRY); return PTR_ERR(params); } @@ -1273,7 +1273,7 @@ const char **traceprobe_expand_meta_args(int argc, const char *argv[], params = find_btf_func_param(ctx->funcname, &nr_params, ctx->flags & TPARG_FL_TPOINT); - if (IS_ERR(params)) { + if (IS_ERR_OR_NULL(params)) { if (args_idx != -1) { /* $arg* requires BTF info */ trace_probe_log_err(0, NOSUP_BTFARG); diff --git a/kernel/trace/trace_probe_kernel.h b/kernel/trace/trace_probe_kernel.h index c4e1d4c03a85..bb723eefd7b7 100644 --- a/kernel/trace/trace_probe_kernel.h +++ b/kernel/trace/trace_probe_kernel.h @@ -2,8 +2,6 @@ #ifndef __TRACE_PROBE_KERNEL_H_ #define __TRACE_PROBE_KERNEL_H_ -#define FAULT_STRING "(fault)" - /* * This depends on trace_probe.h, but can not include it due to * the way trace_probe_tmpl.h is used by trace_kprobe.c and trace_eprobe.c. @@ -15,16 +13,8 @@ static nokprobe_inline int fetch_store_strlen_user(unsigned long addr) { const void __user *uaddr = (__force const void __user *)addr; - int ret; - ret = strnlen_user_nofault(uaddr, MAX_STRING_SIZE); - /* - * strnlen_user_nofault returns zero on fault, insert the - * FAULT_STRING when that occurs. - */ - if (ret <= 0) - return strlen(FAULT_STRING) + 1; - return ret; + return strnlen_user_nofault(uaddr, MAX_STRING_SIZE); } /* Return the length of string -- including null terminal byte */ @@ -44,18 +34,14 @@ fetch_store_strlen(unsigned long addr) len++; } while (c && ret == 0 && len < MAX_STRING_SIZE); - /* For faults, return enough to hold the FAULT_STRING */ - return (ret < 0) ? strlen(FAULT_STRING) + 1 : len; + return (ret < 0) ? ret : len; } -static nokprobe_inline void set_data_loc(int ret, void *dest, void *__dest, void *base, int len) +static nokprobe_inline void set_data_loc(int ret, void *dest, void *__dest, void *base) { - if (ret >= 0) { - *(u32 *)dest = make_data_loc(ret, __dest - base); - } else { - strscpy(__dest, FAULT_STRING, len); - ret = strlen(__dest) + 1; - } + if (ret < 0) + ret = 0; + *(u32 *)dest = make_data_loc(ret, __dest - base); } /* @@ -76,7 +62,7 @@ fetch_store_string_user(unsigned long addr, void *dest, void *base) __dest = get_loc_data(dest, base); ret = strncpy_from_user_nofault(__dest, uaddr, maxlen); - set_data_loc(ret, dest, __dest, base, maxlen); + set_data_loc(ret, dest, __dest, base); return ret; } @@ -107,7 +93,7 @@ fetch_store_string(unsigned long addr, void *dest, void *base) * probing. */ ret = strncpy_from_kernel_nofault(__dest, (void *)addr, maxlen); - set_data_loc(ret, dest, __dest, base, maxlen); + set_data_loc(ret, dest, __dest, base); return ret; } diff --git a/kernel/trace/trace_probe_tmpl.h b/kernel/trace/trace_probe_tmpl.h index 00707630788d..3935b347f874 100644 --- a/kernel/trace/trace_probe_tmpl.h +++ b/kernel/trace/trace_probe_tmpl.h @@ -156,11 +156,11 @@ stage3: code++; goto array; case FETCH_OP_ST_USTRING: - ret += fetch_store_strlen_user(val + code->offset); + ret = fetch_store_strlen_user(val + code->offset); code++; goto array; case FETCH_OP_ST_SYMSTR: - ret += fetch_store_symstrlen(val + code->offset); + ret = fetch_store_symstrlen(val + code->offset); code++; goto array; default: @@ -204,6 +204,8 @@ stage3: array: /* the last stage: Loop on array */ if (code->op == FETCH_OP_LP_ARRAY) { + if (ret < 0) + ret = 0; total += ret; if (++i < code->param) { code = s3; @@ -265,9 +267,7 @@ store_trace_args(void *data, struct trace_probe *tp, void *rec, if (unlikely(arg->dynamic)) *dl = make_data_loc(maxlen, dyndata - base); ret = process_fetch_insn(arg->code, rec, dl, base); - if (unlikely(ret < 0 && arg->dynamic)) { - *dl = make_data_loc(0, dyndata - base); - } else { + if (arg->dynamic && likely(ret > 0)) { dyndata += ret; maxlen -= ret; } diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index 330aee1c1a49..0469a04a355f 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -168,6 +168,8 @@ static void wakeup_trace_open(struct trace_iterator *iter) { if (is_graph(iter->tr)) graph_trace_open(iter); + else + iter->private = NULL; } static void wakeup_trace_close(struct trace_iterator *iter) diff --git a/kernel/trace/trace_seq.c b/kernel/trace/trace_seq.c index e5e299260d0c..bac06ee3b98b 100644 --- a/kernel/trace/trace_seq.c +++ b/kernel/trace/trace_seq.c @@ -131,6 +131,7 @@ EXPORT_SYMBOL_GPL(trace_seq_bitmask); * trace_seq_vprintf - sequence printing of trace information * @s: trace sequence descriptor * @fmt: printf format string + * @args: Arguments for the format string * * The tracer may use either sequence operations or its own * copy to user routines. To simplify formatting of a trace diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c index fa09b33ee731..688bf579f2f1 100644 --- a/kernel/trace/trace_uprobe.c +++ b/kernel/trace/trace_uprobe.c @@ -170,7 +170,8 @@ fetch_store_string(unsigned long addr, void *dest, void *base) */ ret++; *(u32 *)dest = make_data_loc(ret, (void *)dst - base); - } + } else + *(u32 *)dest = make_data_loc(0, (void *)dst - base); return ret; } diff --git a/kernel/trace/tracing_map.h b/kernel/trace/tracing_map.h index 2c765ee2a4d4..99c37eeebc16 100644 --- a/kernel/trace/tracing_map.h +++ b/kernel/trace/tracing_map.h @@ -272,10 +272,6 @@ extern u64 tracing_map_read_sum(struct tracing_map_elt *elt, unsigned int i); extern u64 tracing_map_read_var(struct tracing_map_elt *elt, unsigned int i); extern u64 tracing_map_read_var_once(struct tracing_map_elt *elt, unsigned int i); -extern void tracing_map_set_field_descr(struct tracing_map *map, - unsigned int i, - unsigned int key_offset, - tracing_map_cmp_fn_t cmp_fn); extern int tracing_map_sort_entries(struct tracing_map *map, struct tracing_map_sort_key *sort_keys, diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 02a8f402eeb5..800b4208dba9 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -52,6 +52,7 @@ #include <linux/sched/debug.h> #include <linux/nmi.h> #include <linux/kvm_para.h> +#include <linux/delay.h> #include "workqueue_internal.h" @@ -338,8 +339,10 @@ static cpumask_var_t *wq_numa_possible_cpumask; * Per-cpu work items which run for longer than the following threshold are * automatically considered CPU intensive and excluded from concurrency * management to prevent them from noticeably delaying other per-cpu work items. + * ULONG_MAX indicates that the user hasn't overridden it with a boot parameter. + * The actual value is initialized in wq_cpu_intensive_thresh_init(). */ -static unsigned long wq_cpu_intensive_thresh_us = 10000; +static unsigned long wq_cpu_intensive_thresh_us = ULONG_MAX; module_param_named(cpu_intensive_thresh_us, wq_cpu_intensive_thresh_us, ulong, 0644); static bool wq_disable_numa; @@ -6513,6 +6516,42 @@ void __init workqueue_init_early(void) !system_freezable_power_efficient_wq); } +static void __init wq_cpu_intensive_thresh_init(void) +{ + unsigned long thresh; + unsigned long bogo; + + /* if the user set it to a specific value, keep it */ + if (wq_cpu_intensive_thresh_us != ULONG_MAX) + return; + + /* + * The default of 10ms is derived from the fact that most modern (as of + * 2023) processors can do a lot in 10ms and that it's just below what + * most consider human-perceivable. However, the kernel also runs on a + * lot slower CPUs including microcontrollers where the threshold is way + * too low. + * + * Let's scale up the threshold upto 1 second if BogoMips is below 4000. + * This is by no means accurate but it doesn't have to be. The mechanism + * is still useful even when the threshold is fully scaled up. Also, as + * the reports would usually be applicable to everyone, some machines + * operating on longer thresholds won't significantly diminish their + * usefulness. + */ + thresh = 10 * USEC_PER_MSEC; + + /* see init/calibrate.c for lpj -> BogoMIPS calculation */ + bogo = max_t(unsigned long, loops_per_jiffy / 500000 * HZ, 1); + if (bogo < 4000) + thresh = min_t(unsigned long, thresh * 4000 / bogo, USEC_PER_SEC); + + pr_debug("wq_cpu_intensive_thresh: lpj=%lu BogoMIPS=%lu thresh_us=%lu\n", + loops_per_jiffy, bogo, thresh); + + wq_cpu_intensive_thresh_us = thresh; +} + /** * workqueue_init - bring workqueue subsystem fully online * @@ -6528,6 +6567,8 @@ void __init workqueue_init(void) struct worker_pool *pool; int cpu, bkt; + wq_cpu_intensive_thresh_init(); + /* * It'd be simpler to initialize NUMA in workqueue_init_early() but * CPU to node mapping may not be available that early on some |