[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250205102448.563246589@infradead.org>
Date: Wed, 05 Feb 2025 11:21:23 +0100
From: Peter Zijlstra <peterz@...radead.org>
To: mingo@...nel.org,
ravi.bangoria@....com,
lucas.demarchi@...el.com
Cc: linux-kernel@...r.kernel.org,
peterz@...radead.org,
willy@...radead.org,
acme@...nel.org,
namhyung@...nel.org,
mark.rutland@....com,
alexander.shishkin@...ux.intel.com,
jolsa@...nel.org,
irogers@...gle.com,
adrian.hunter@...el.com,
kan.liang@...ux.intel.com
Subject: [PATCH v2 03/24] perf: Simplify child event tear-down
Currently perf_event_release_kernel() will iterate the child events and attempt
tear-down. However, it removes them from the child_list using list_move(),
notably skipping the state management done by perf_child_detach().
Crucially, it fails to clear PERF_ATTACH_CHILD, which opens the door for a
concurrent perf_remove_from_context() to race.
This way child_list management stays fully serialized using child_mutex.
Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
---
kernel/events/core.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -5526,8 +5526,8 @@ int perf_event_release_kernel(struct per
tmp = list_first_entry_or_null(&event->child_list,
struct perf_event, child_list);
if (tmp == child) {
- perf_remove_from_context(child, DETACH_GROUP);
- list_move(&child->child_list, &free_list);
+ perf_remove_from_context(child, DETACH_GROUP | DETACH_CHILD);
+ list_add(&child->child_list, &free_list);
/*
* This matches the refcount bump in inherit_event();
* this can't be the last reference.
Powered by blists - more mailing lists