Hi Greg, Sasha,
Please consider the attached backport of
f91072ed1b72 ("perf/core: Fix race in the perf_mmap_close() function")
for v4.4.y and v4.9.y.
I have sent separate backport for v4.14.y, v4.19.y and v5.4.y.
--
Regards
Sudip
>From 24ef0388b57352c2be2e51ea5b92eddd625df77e Mon Sep 17 00:00:00 2001
From: Jiri Olsa <jolsa@xxxxxxxxxx>
Date: Wed, 16 Sep 2020 13:53:11 +0200
Subject: [PATCH] perf/core: Fix race in the perf_mmap_close() function
commit f91072ed1b7283b13ca57fcfbece5a3b92726143 upstream
There's a possible race in perf_mmap_close() when checking ring buffer's
mmap_count refcount value. The problem is that the mmap_count check is
not atomic because we call atomic_dec() and atomic_read() separately.
perf_mmap_close:
...
atomic_dec(&rb->mmap_count);
...
if (atomic_read(&rb->mmap_count))
goto out_put;
<ring buffer detach>
free_uid
out_put:
ring_buffer_put(rb); /* could be last */
The race can happen when we have two (or more) events sharing same ring
buffer and they go through atomic_dec() and then they both see 0 as refcount
value later in atomic_read(). Then both will go on and execute code which
is meant to be run just once.
The code that detaches ring buffer is probably fine to be executed more
than once, but the problem is in calling free_uid(), which will later on
demonstrate in related crashes and refcount warnings, like:
refcount_t: addition on 0; use-after-free.
...
RIP: 0010:refcount_warn_saturate+0x6d/0xf
...
Call Trace:
prepare_creds+0x190/0x1e0
copy_creds+0x35/0x172
copy_process+0x471/0x1a80
_do_fork+0x83/0x3a0
__do_sys_wait4+0x83/0x90
__do_sys_clone+0x85/0xa0
do_syscall_64+0x5b/0x1e0
entry_SYSCALL_64_after_hwframe+0x44/0xa9
Using atomic decrease and check instead of separated calls.
Tested-by: Michael Petlan <mpetlan@xxxxxxxxxx>
Signed-off-by: Jiri Olsa <jolsa@xxxxxxxxxx>
Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>
Acked-by: Peter Zijlstra <a.p.zijlstra@xxxxxxxxx>
Acked-by: Namhyung Kim <namhyung@xxxxxxxxxx>
Acked-by: Wade Mealing <wmealing@xxxxxxxxxx>
Fixes: 9bb5d40cd93c ("perf: Fix mmap() accounting hole");
Link: https://lore.kernel.org/r/20200916115311.GE2301783@krava
[sudip: backport to v4.9.y by using ring_buffer]
Signed-off-by: Sudip Mukherjee <sudipm.mukherjee@xxxxxxxxx>
---
kernel/events/core.c | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 7aad4d22b422..38bbec00de2d 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -5069,11 +5069,11 @@ static void perf_pmu_output_stop(struct perf_event *event);
static void perf_mmap_close(struct vm_area_struct *vma)
{
struct perf_event *event = vma->vm_file->private_data;
-
struct ring_buffer *rb = ring_buffer_get(event);
struct user_struct *mmap_user = rb->mmap_user;
int mmap_locked = rb->mmap_locked;
unsigned long size = perf_data_size(rb);
+ bool detach_rest = false;
if (event->pmu->event_unmapped)
event->pmu->event_unmapped(event);
@@ -5104,7 +5104,8 @@ static void perf_mmap_close(struct vm_area_struct *vma)
mutex_unlock(&event->mmap_mutex);
}
- atomic_dec(&rb->mmap_count);
+ if (atomic_dec_and_test(&rb->mmap_count))
+ detach_rest = true;
if (!atomic_dec_and_mutex_lock(&event->mmap_count, &event->mmap_mutex))
goto out_put;
@@ -5113,7 +5114,7 @@ static void perf_mmap_close(struct vm_area_struct *vma)
mutex_unlock(&event->mmap_mutex);
/* If there's still other mmap()s of this buffer, we're done. */
- if (atomic_read(&rb->mmap_count))
+ if (!detach_rest)
goto out_put;
/*
--
2.11.0
