In the Linux kernel, the following vulnerability has been resolved:
perf/core: Fix data race between perf_event_set_output() and perf_mmap_close()
Yang Jihing reported a race between perf_event_set_output() and
perf_mmap_close():
CPU1 CPU2
perf_mmap_close(e2)
if (atomic_dec_and_test(&e2->rb->mmap_count)) // 1 - > 0
detach_rest = true
ioctl(e1, IOC_SET_OUTPUT, e2)
perf_event_set_output(e1, e2)
...
list_for_each_entry_rcu(e, &e2->rb->event_list, rb_entry)
ring_buffer_attach(e, NULL);
// e1 isn't yet added and
// therefore not detached
ring_buffer_attach(e1, e2->rb)
list_add_rcu(&e1->rb_entry,
&e2->rb->event_list)
After this; e1 is attached to an unmapped rb and a subsequent
perf_mmap() will loop forever more:
again:
mutex_lock(&e->mmap_mutex);
if (event->rb) {
...
if (!atomic_inc_not_zero(&e->rb->mmap_count)) {
...
mutex_unlock(&e->mmap_mutex);
goto again;
}
}
The loop in perf_mmap_close() holds e2->mmap_mutex, while the attach
in perf_event_set_output() holds e1->mmap_mutex. As such there is no
serialization to avoid this race.
Change perf_event_set_output() to take both e1->mmap_mutex and
e2->mmap_mutex to alleviate that problem. Additionally, have the loop
in perf_mmap() detach the rb directly, this avoids having to wait for
the concurrent perf_mmap_close() to get around to doing it to make
progress.
Metrics
Affected Vendors & Products
| Vendors | Products |
|---|---|
| Linux |
|
Configuration 1 [-]
|
No data.
References
History
Thu, 13 Mar 2025 22:15:00 +0000
| Type | Values Removed | Values Added |
|---|---|---|
| First Time appeared |
Linux
Linux linux Kernel |
|
| Weaknesses | CWE-362 | |
| CPEs | cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:* cpe:2.3:o:linux:linux_kernel:5.19:rc1:*:*:*:*:*:* cpe:2.3:o:linux:linux_kernel:5.19:rc2:*:*:*:*:*:* cpe:2.3:o:linux:linux_kernel:5.19:rc3:*:*:*:*:*:* cpe:2.3:o:linux:linux_kernel:5.19:rc4:*:*:*:*:*:* cpe:2.3:o:linux:linux_kernel:5.19:rc5:*:*:*:*:*:* cpe:2.3:o:linux:linux_kernel:5.19:rc6:*:*:*:*:*:* |
|
| Vendors & Products |
Linux
Linux linux Kernel |
|
| Metrics |
cvssV3_1
|
cvssV3_1
|
Thu, 27 Feb 2025 01:45:00 +0000
| Type | Values Removed | Values Added |
|---|---|---|
| References |
| |
| Metrics |
threat_severity
|
cvssV3_1
|
Wed, 26 Feb 2025 02:45:00 +0000
| Type | Values Removed | Values Added |
|---|---|---|
| Description | In the Linux kernel, the following vulnerability has been resolved: perf/core: Fix data race between perf_event_set_output() and perf_mmap_close() Yang Jihing reported a race between perf_event_set_output() and perf_mmap_close(): CPU1 CPU2 perf_mmap_close(e2) if (atomic_dec_and_test(&e2->rb->mmap_count)) // 1 - > 0 detach_rest = true ioctl(e1, IOC_SET_OUTPUT, e2) perf_event_set_output(e1, e2) ... list_for_each_entry_rcu(e, &e2->rb->event_list, rb_entry) ring_buffer_attach(e, NULL); // e1 isn't yet added and // therefore not detached ring_buffer_attach(e1, e2->rb) list_add_rcu(&e1->rb_entry, &e2->rb->event_list) After this; e1 is attached to an unmapped rb and a subsequent perf_mmap() will loop forever more: again: mutex_lock(&e->mmap_mutex); if (event->rb) { ... if (!atomic_inc_not_zero(&e->rb->mmap_count)) { ... mutex_unlock(&e->mmap_mutex); goto again; } } The loop in perf_mmap_close() holds e2->mmap_mutex, while the attach in perf_event_set_output() holds e1->mmap_mutex. As such there is no serialization to avoid this race. Change perf_event_set_output() to take both e1->mmap_mutex and e2->mmap_mutex to alleviate that problem. Additionally, have the loop in perf_mmap() detach the rb directly, this avoids having to wait for the concurrent perf_mmap_close() to get around to doing it to make progress. | |
| Title | perf/core: Fix data race between perf_event_set_output() and perf_mmap_close() | |
| References |
|
|
MITRE
Status: PUBLISHED
Assigner: Linux
Published: 2025-02-26T02:23:31.823Z
Updated: 2025-05-04T12:44:58.322Z
Reserved: 2025-02-26T02:21:30.417Z
Link: CVE-2022-49607
Vulnrichment
No data.
NVD
Status : Analyzed
Published: 2025-02-26T07:01:36.180
Modified: 2025-03-13T21:55:55.090
Link: CVE-2022-49607
Redhat