From 232153094288afb03edddca5f0f7169c32110853 Mon Sep 17 00:00:00 2001 From: Leo Yan Date: Mon, 1 Sep 2025 06:42:51 +0000 Subject: [PATCH] perf/aux: Fix pending disable flow when the AUX ring buffer overruns mainline inclusion from mainline-v6.16-rc4 commit 1476b218327b89bbb64c14619a2d34f0c320f2c3 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ICVQU8 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=1476b218327b -------------------------------- If an AUX event overruns, the event core layer intends to disable the event by setting the 'pending_disable' flag. Unfortunately, the event is not actually disabled afterwards. In commit: ca6c21327c6a ("perf: Fix missing SIGTRAPs") the 'pending_disable' flag was changed to a boolean. However, the AUX event code was not updated accordingly. The flag ends up holding a CPU number. If this number is zero, the flag is taken as false and the IRQ work is never triggered. Later, with commit: 2b84def990d3 ("perf: Split __perf_pending_irq() out of perf_pending_irq()") a new IRQ work 'pending_disable_irq' was introduced to handle event disabling. The AUX event path was not updated to kick off the work queue. To fix this bug, when an AUX ring buffer overrun is detected, call perf_event_disable_inatomic() to initiate the pending disable flow. Also update the outdated comment for setting the flag, to reflect the boolean values (0 or 1). Fixes: 2b84def990d3 ("perf: Split __perf_pending_irq() out of perf_pending_irq()") Fixes: ca6c21327c6a ("perf: Fix missing SIGTRAPs") Signed-off-by: Leo Yan Signed-off-by: Ingo Molnar Reviewed-by: James Clark Reviewed-by: Yeoreum Yun Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Ian Rogers Cc: Jiri Olsa Cc: Liang Kan Cc: Marco Elver Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Sebastian Andrzej Siewior Cc: linux-perf-users@vger.kernel.org Link: https://lore.kernel.org/r/20250625170737.2918295-1-leo.yan@arm.com Signed-off-by: Pu Lehui --- kernel/events/core.c | 6 +++--- kernel/events/ring_buffer.c | 4 ++-- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index 6b572eb3fede..922389a83c67 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -6821,15 +6821,15 @@ static void __perf_pending_irq(struct perf_event *event) * CPU-A CPU-B * * perf_event_disable_inatomic() - * @pending_disable = CPU-A; + * @pending_disable = 1; * irq_work_queue(); * * sched-out - * @pending_disable = -1; + * @pending_disable = 0; * * sched-in * perf_event_disable_inatomic() - * @pending_disable = CPU-B; + * @pending_disable = 1; * irq_work_queue(); // FAILS * * irq_work_run() diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c index cfd9448ce28f..610e9b22f0c4 100644 --- a/kernel/events/ring_buffer.c +++ b/kernel/events/ring_buffer.c @@ -437,7 +437,7 @@ void *perf_aux_output_begin(struct perf_output_handle *handle, * store that will be enabled on successful return */ if (!handle->size) { /* A, matches D */ - event->pending_disable = smp_processor_id(); + perf_event_disable_inatomic(handle->event); perf_output_wakeup(handle); WRITE_ONCE(rb->aux_nest, 0); goto err_put; @@ -522,7 +522,7 @@ void perf_aux_output_end(struct perf_output_handle *handle, unsigned long size) if (wakeup) { if (handle->aux_flags & PERF_AUX_FLAG_TRUNCATED) - handle->event->pending_disable = smp_processor_id(); + perf_event_disable_inatomic(handle->event); perf_output_wakeup(handle); } -- Gitee