From 3da6bb419750f3ad834786d6ba7c9d5d062c770b Mon Sep 17 00:00:00 2001 From: Tetsuo Handa Date: Wed, 9 Jul 2025 20:27:52 +0900 Subject: perf/core: Fix WARN in perf_sigtrap() Since exit_task_work() runs after perf_event_exit_task_context() updated ctx->task to TASK_TOMBSTONE, perf_sigtrap() from perf_pending_task() might observe event->ctx->task == TASK_TOMBSTONE. Swap the early exit tests in order not to hit WARN_ON_ONCE(). Closes: https://syzkaller.appspot.com/bug?extid=2fe61cb2a86066be6985 Reported-by: syzbot Signed-off-by: Tetsuo Handa Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/b1c224bd-97f9-462c-a3e3-125d5e19c983@I-love.SAKURA.ne.jp --- kernel/events/core.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/events/core.c b/kernel/events/core.c index 0db36b2b2448..22fdf0c187cd 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -7204,18 +7204,18 @@ void perf_event_wakeup(struct perf_event *event) static void perf_sigtrap(struct perf_event *event) { /* - * We'd expect this to only occur if the irq_work is delayed and either - * ctx->task or current has changed in the meantime. This can be the - * case on architectures that do not implement arch_irq_work_raise(). + * Both perf_pending_task() and perf_pending_irq() can race with the + * task exiting. */ - if (WARN_ON_ONCE(event->ctx->task != current)) + if (current->flags & PF_EXITING) return; /* - * Both perf_pending_task() and perf_pending_irq() can race with the - * task exiting. + * We'd expect this to only occur if the irq_work is delayed and either + * ctx->task or current has changed in the meantime. This can be the + * case on architectures that do not implement arch_irq_work_raise(). */ - if (current->flags & PF_EXITING) + if (WARN_ON_ONCE(event->ctx->task != current)) return; send_sig_perf((void __user *)event->pending_addr, -- cgit v1.2.3 From db6cc3f4ac2e6cdc898fc9cbc8b32ae1bf56bdad Mon Sep 17 00:00:00 2001 From: Chen Yu Date: Fri, 4 Jul 2025 21:56:20 +0800 Subject: Revert "sched/numa: add statistics of numa balance task" This reverts commit ad6b26b6a0a79166b53209df2ca1cf8636296382. This commit introduces per-memcg/task NUMA balance statistics, but unfortunately it introduced a NULL pointer exception due to the following race condition: After a swap task candidate was chosen, its mm_struct pointer was set to NULL due to task exit. Later, when performing the actual task swapping, the p->mm caused the problem. CPU0 CPU1 : ... task_numa_migrate task_numa_find_cpu task_numa_compare # a normal task p is chosen env->best_task = p # p exit: exit_signals(p); p->flags |= PF_EXITING exit_mm p->mm = NULL; migrate_swap_stop __migrate_swap_task((arg->src_task, arg->dst_cpu) count_memcg_event_mm(p->mm, NUMA_TASK_SWAP)# p->mm is NULL task_lock() should be held and the PF_EXITING flag needs to be checked to prevent this from happening. After discussion, the conclusion was that adding a lock is not worthwhile for some statistics calculations. Revert the change and rely on the tracepoint for this purpose. Link: https://lkml.kernel.org/r/20250704135620.685752-1-yu.c.chen@intel.com Link: https://lkml.kernel.org/r/20250708064917.BBD13C4CEED@smtp.kernel.org Fixes: ad6b26b6a0a7 ("sched/numa: add statistics of numa balance task") Signed-off-by: Chen Yu Reported-by: Jirka Hladky Closes: https://lore.kernel.org/all/CAE4VaGBLJxpd=NeRJXpSCuw=REhC5LWJpC29kDy-Zh2ZDyzQZA@mail.gmail.com/ Reported-by: Srikanth Aithal Reported-by: Suneeth D Acked-by: Michal Hocko Cc: Borislav Petkov Cc: Ingo Molnar Cc: Jiri Hladky Cc: Libo Chen Cc: Peter Zijlstra Cc: Thomas Gleixner Signed-off-by: Andrew Morton --- kernel/sched/core.c | 9 ++------- kernel/sched/debug.c | 4 ---- 2 files changed, 2 insertions(+), 11 deletions(-) (limited to 'kernel') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index ec68fc686bd7..81c6df746df1 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -3362,10 +3362,6 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) #ifdef CONFIG_NUMA_BALANCING static void __migrate_swap_task(struct task_struct *p, int cpu) { - __schedstat_inc(p->stats.numa_task_swapped); - count_vm_numa_event(NUMA_TASK_SWAP); - count_memcg_event_mm(p->mm, NUMA_TASK_SWAP); - if (task_on_rq_queued(p)) { struct rq *src_rq, *dst_rq; struct rq_flags srf, drf; @@ -7939,9 +7935,8 @@ int migrate_task_to(struct task_struct *p, int target_cpu) if (!cpumask_test_cpu(target_cpu, p->cpus_ptr)) return -EINVAL; - __schedstat_inc(p->stats.numa_task_migrated); - count_vm_numa_event(NUMA_TASK_MIGRATE); - count_memcg_event_mm(p->mm, NUMA_TASK_MIGRATE); + /* TODO: This is not properly updating schedstats */ + trace_sched_move_numa(p, curr_cpu, target_cpu); return stop_one_cpu(curr_cpu, migration_cpu_stop, &arg); } diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index 9d71baf08075..557246880a7e 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -1210,10 +1210,6 @@ void proc_sched_show_task(struct task_struct *p, struct pid_namespace *ns, P_SCHEDSTAT(nr_failed_migrations_running); P_SCHEDSTAT(nr_failed_migrations_hot); P_SCHEDSTAT(nr_forced_migrations); -#ifdef CONFIG_NUMA_BALANCING - P_SCHEDSTAT(numa_task_migrated); - P_SCHEDSTAT(numa_task_swapped); -#endif P_SCHEDSTAT(nr_wakeups); P_SCHEDSTAT(nr_wakeups_sync); P_SCHEDSTAT(nr_wakeups_migrate); -- cgit v1.2.3 From 75b63ce2c98b41c45ee3ff14c76f27c3238bc6d2 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Thu, 10 Jul 2025 10:43:26 +0200 Subject: PM: suspend: Drop a misplaced pm_restore_gfp_mask() call The pm_restore_gfp_mask() call added by commit 12ffc3b1513e ("PM: Restrict swap use to later in the suspend sequence") to suspend_devices_and_enter() is done too early because it takes place before calling dpm_resume() in dpm_resume_end() and some swap-backing devices may not be ready at that point. Moreover, dpm_resume_end() called subsequently in the same code path invokes pm_restore_gfp_mask() again and calling it twice in a row is pointless. Drop the misplaced pm_restore_gfp_mask() call from suspend_devices_and_enter() to address this issue. Fixes: 12ffc3b1513e ("PM: Restrict swap use to later in the suspend sequence") Signed-off-by: Rafael J. Wysocki Reviewed-by: Mario Limonciello Link: https://patch.msgid.link/2810409.mvXUDI8C0e@rjwysocki.net [ rjw: Changelog edits ] Signed-off-by: Rafael J. Wysocki --- kernel/power/suspend.c | 1 - 1 file changed, 1 deletion(-) (limited to 'kernel') diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index bb608b68fb30..632f83b60d16 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -540,7 +540,6 @@ int suspend_devices_and_enter(suspend_state_t state) return error; Recover_platform: - pm_restore_gfp_mask(); platform_recover(state); goto Resume_devices; } -- cgit v1.2.3 From 228b9deded0011482149cdb474e3ad15aeeb4a97 Mon Sep 17 00:00:00 2001 From: Zihuan Zhang Date: Sat, 12 Jul 2025 11:08:24 +0800 Subject: PM: suspend: clean up redundant filesystems_freeze/thaw() handling The recently introduced support for freezing filesystems during system suspend included calls to filesystems_freeze() in both suspend_prepare() and enter_state(), as well as calls to filesystems_thaw() in both suspend_finish() and the Unlock path in enter_state(). These are redundant. Moreover, calling filesystems_freeze() twice, from both suspend_prepare() and enter_state(), leads to a black screen and makes the system unable to resume in some cases. Address this as follows: - filesystems_freeze() is already called in suspend_prepare(), which is the proper and consistent place to handle pre-suspend operations. The second call in enter_state() is unnecessary and so remove it. - filesystems_thaw() is invoked in suspend_finish(), which covers successful suspend/resume paths. In the failure case, add a call to filesystems_thaw() only when needed, avoiding the duplicate call in the general Unlock path. This change simplifies the suspend code and avoids repeated freeze/thaw calls, while preserving correct ordering and behavior. Fixes: eacfbf74196f ("power: freeze filesystems during suspend/resume") Signed-off-by: Zihuan Zhang Link: https://patch.msgid.link/20250712030824.81474-1-zhangzihuan@kylinos.cn [ rjw: Changelog edits ] Signed-off-by: Rafael J. Wysocki --- kernel/power/suspend.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index 632f83b60d16..b4ca17c2fecf 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -384,6 +384,7 @@ static int suspend_prepare(suspend_state_t state) return 0; dpm_save_failed_step(SUSPEND_FREEZE); + filesystems_thaw(); pm_notifier_call_chain(PM_POST_SUSPEND); Restore: pm_restore_console(); @@ -592,8 +593,6 @@ static int enter_state(suspend_state_t state) ksys_sync_helper(); trace_suspend_resume(TPS("sync_filesystems"), 0, false); } - if (filesystem_freeze_enabled) - filesystems_freeze(); pm_pr_dbg("Preparing system for sleep (%s)\n", mem_sleep_labels[state]); pm_suspend_clear_flags(); @@ -613,7 +612,6 @@ static int enter_state(suspend_state_t state) pm_pr_dbg("Finishing wakeup.\n"); suspend_finish(); Unlock: - filesystems_thaw(); mutex_unlock(&system_transition_mutex); return error; } -- cgit v1.2.3 From 1ed171a3afe81531b3ace96bd151a372dda3ee25 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Tue, 15 Jul 2025 20:19:44 -0700 Subject: tracing/probes: Avoid using params uninitialized in parse_btf_arg() After a recent change in clang to strengthen uninitialized warnings [1], it points out that in one of the error paths in parse_btf_arg(), params is used uninitialized: kernel/trace/trace_probe.c:660:19: warning: variable 'params' is uninitialized when used here [-Wuninitialized] 660 | return PTR_ERR(params); | ^~~~~~ Match many other NO_BTF_ENTRY error cases and return -ENOENT, clearing up the warning. Link: https://lore.kernel.org/all/20250715-trace_probe-fix-const-uninit-warning-v1-1-98960f91dd04@kernel.org/ Cc: stable@vger.kernel.org Closes: https://github.com/ClangBuiltLinux/linux/issues/2110 Fixes: d157d7694460 ("tracing/probes: Support BTF field access from $retval") Link: https://github.com/llvm/llvm-project/commit/2464313eef01c5b1edf0eccf57a32cdee01472c7 [1] Signed-off-by: Nathan Chancellor Signed-off-by: Masami Hiramatsu (Google) --- kernel/trace/trace_probe.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/trace_probe.c b/kernel/trace/trace_probe.c index 424751cdf31f..40830a3ecd96 100644 --- a/kernel/trace/trace_probe.c +++ b/kernel/trace/trace_probe.c @@ -657,7 +657,7 @@ static int parse_btf_arg(char *varname, ret = query_btf_context(ctx); if (ret < 0 || ctx->nr_params == 0) { trace_probe_log_err(ctx->offset, NO_BTF_ENTRY); - return PTR_ERR(params); + return -ENOENT; } } params = ctx->params; -- cgit v1.2.3