* Re: [PATCH 0/4] perf: Fix leaked events when sigtrap = 1 [not found] ` <ZhRI7WfjgkCTK9aK@pavilion.home> @ 2024-04-08 20:20 ` Arnaldo Carvalho de Melo 2024-04-08 20:58 ` Frederic Weisbecker 0 siblings, 1 reply; 2+ messages in thread From: Arnaldo Carvalho de Melo @ 2024-04-08 20:20 UTC (permalink / raw) To: Frederic Weisbecker, Sebastian Andrzej Siewior Cc: Ian Rogers, LKML, Peter Zijlstra, Ingo Molnar, Namhyung Kim, Mark Rutland, Alexander Shishkin, Jiri Olsa, Adrian Hunter, Thomas Gleixner, linux-perf-users On Mon, Apr 08, 2024 at 09:43:41PM +0200, Frederic Weisbecker wrote: > Le Fri, Mar 29, 2024 at 08:23:49PM -0700, Ian Rogers a écrit : > > On Fri, Mar 29, 2024 at 4:58 PM Frederic Weisbecker <frederic@kernel.org> wrote: > > > > > > While looking at task_work users I just noticed that perf doesn't flush > > > its own upon event exiting. This looks especially problematic with child > > > events. Please have a thourough look at the last patch, I may easily > > > have missed something within the maze. > > > > > > Frederic Weisbecker (4): > > > task_work: s/task_work_cancel()/task_work_cancel_func()/ > > > task_work: Introduce task_work_cancel() again > > > perf: Fix event leak upon exit > > > perf: Fix event leak upon exec and file release > > > > > > include/linux/perf_event.h | 1 + > > > include/linux/task_work.h | 3 ++- > > > kernel/events/core.c | 40 +++++++++++++++++++++++++++++++------- > > > kernel/irq/manage.c | 2 +- > > > kernel/task_work.c | 34 +++++++++++++++++++++++++++----- > > > security/keys/keyctl.c | 2 +- > > > 6 files changed, 67 insertions(+), 15 deletions(-) > > > > Thanks for this! I wonder if this relates to fuzzing failures like: > > https://lore.kernel.org/linux-perf-users/CAP-5=fUa+-Tj2b_hxk96Qg5=Qu7jYHgHREbsmBa2ZmuF-X9QaA@mail.gmail.com/ > > "[ 2519.138665] unexpected event refcount: 2; ptr=000000009c56b097" > > Probably not since those seem to happen on perf_event_open() failures. This > looks different. Probably clashes with this one? "[PATCH v3 0/4] perf: Make SIGTRAP and __perf_pending_irq() work on RT." https://lore.kernel.org/all/20240322065208.60456-1-bigeasy@linutronix.de/T/#u - Arnaldo ^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [PATCH 0/4] perf: Fix leaked events when sigtrap = 1 2024-04-08 20:20 ` [PATCH 0/4] perf: Fix leaked events when sigtrap = 1 Arnaldo Carvalho de Melo @ 2024-04-08 20:58 ` Frederic Weisbecker 0 siblings, 0 replies; 2+ messages in thread From: Frederic Weisbecker @ 2024-04-08 20:58 UTC (permalink / raw) To: Arnaldo Carvalho de Melo Cc: Sebastian Andrzej Siewior, Ian Rogers, LKML, Peter Zijlstra, Ingo Molnar, Namhyung Kim, Mark Rutland, Alexander Shishkin, Jiri Olsa, Adrian Hunter, Thomas Gleixner, linux-perf-users Le Mon, Apr 08, 2024 at 05:20:27PM -0300, Arnaldo Carvalho de Melo a écrit : > On Mon, Apr 08, 2024 at 09:43:41PM +0200, Frederic Weisbecker wrote: > > Le Fri, Mar 29, 2024 at 08:23:49PM -0700, Ian Rogers a écrit : > > > On Fri, Mar 29, 2024 at 4:58 PM Frederic Weisbecker <frederic@kernel.org> wrote: > > > > > > > > While looking at task_work users I just noticed that perf doesn't flush > > > > its own upon event exiting. This looks especially problematic with child > > > > events. Please have a thourough look at the last patch, I may easily > > > > have missed something within the maze. > > > > > > > > Frederic Weisbecker (4): > > > > task_work: s/task_work_cancel()/task_work_cancel_func()/ > > > > task_work: Introduce task_work_cancel() again > > > > perf: Fix event leak upon exit > > > > perf: Fix event leak upon exec and file release > > > > > > > > include/linux/perf_event.h | 1 + > > > > include/linux/task_work.h | 3 ++- > > > > kernel/events/core.c | 40 +++++++++++++++++++++++++++++++------- > > > > kernel/irq/manage.c | 2 +- > > > > kernel/task_work.c | 34 +++++++++++++++++++++++++++----- > > > > security/keys/keyctl.c | 2 +- > > > > 6 files changed, 67 insertions(+), 15 deletions(-) > > > > > > Thanks for this! I wonder if this relates to fuzzing failures like: > > > https://lore.kernel.org/linux-perf-users/CAP-5=fUa+-Tj2b_hxk96Qg5=Qu7jYHgHREbsmBa2ZmuF-X9QaA@mail.gmail.com/ > > > "[ 2519.138665] unexpected event refcount: 2; ptr=000000009c56b097" > > > > Probably not since those seem to happen on perf_event_open() failures. This > > looks different. > > Probably clashes with this one? > > "[PATCH v3 0/4] perf: Make SIGTRAP and __perf_pending_irq() work on RT." > > https://lore.kernel.org/all/20240322065208.60456-1-bigeasy@linutronix.de/T/#u Nice, I think it partially fixes the issues I've seen. Lemme review that. Thanks. > > - Arnaldo ^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2024-04-08 20:58 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <20240329235812.18917-1-frederic@kernel.org>
[not found] ` <CAP-5=fVwT1FPTps=i=hG2O0sWr9DH92VSt66Co=Rhf9dyZX7ag@mail.gmail.com>
[not found] ` <ZhRI7WfjgkCTK9aK@pavilion.home>
2024-04-08 20:20 ` [PATCH 0/4] perf: Fix leaked events when sigtrap = 1 Arnaldo Carvalho de Melo
2024-04-08 20:58 ` Frederic Weisbecker
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).