UBUNTU-CVE-2025-21693
In the Linux kernel, the following vulnerability has been resolved: mm: zswap: properly synchronize freeing resources during CPU hotunplug In zswap_compress() and zswap_decompress(), the per-CPU acomp_ctx of the current CPU at the beginning of the operation is retrieved and used throughout. However, since neither preemption nor migration are disabled, it is possible that the operation continues on a different CPU. If the original CPU is hotunplugged while the acomp_ctx is still in use, we run into a UAF bug as some of the resources attached to the acomp_ctx are freed during hotunplug in zswap_cpu_comp_dead() (i.e. acomp_ctx.buffer, acomp_ctx.req, or acomp_ctx.acomp). The problem was introduced in commit 1ec3b5fe6eec ("mm/zswap: move to use crypto_acomp API for hardware acceleration") when the switch to the crypto_acomp API was made. Prior to that, the per-CPU crypto_comp was retrieved using get_cpu_ptr() which disables preemption and makes sure the CPU cannot go away from under us. Preemption cannot be disabled with the crypto_acomp API as a sleepable context is needed. Use the acomp_ctx.mutex to synchronize CPU hotplug callbacks allocating and freeing resources with compression/decompression paths. Make sure that acomp_ctx.req is NULL when the resources are freed. In the compression/decompression paths, check if acomp_ctx.req is NULL after acquiring the mutex (meaning the CPU was offlined) and retry on the new CPU. The initialization of acomp_ctx.mutex is moved from the CPU hotplug callback to the pool initialization where it belongs (where the mutex is allocated). In addition to adding clarity, this makes sure that CPU hotplug cannot reinitialize a mutex that is already locked by compression/decompression. Previously a fix was attempted by holding cpus_read_lock() [1]. This would have caused a potential deadlock as it is possible for code already holding the lock to fall into reclaim and enter zswap (causing a deadlock). A fix was also attempted using SRCU for synchronization, but Johannes pointed out that synchronize_srcu() cannot be used in CPU hotplug notifiers [2]. Alternative fixes that were considered/attempted and could have worked: - Refcounting the per-CPU acomp_ctx. This involves complexity in handling the race between the refcount dropping to zero in zswap_[de]compress() and the refcount being re-initialized when the CPU is onlined. - Disabling migration before getting the per-CPU acomp_ctx [3], but that's discouraged and is a much bigger hammer than needed, and could result in subtle performance issues. [1]https://lkml.kernel.org/[email protected]/ [2]https://lkml.kernel.org/[email protected]/ [3]https://lkml.kernel.org/[email protected]/ [[email protected]: remove comment]
02 / AFFECTED SOFTWARE
Affected packages
11 explicit affected versions
89 explicit affected versions
50 explicit affected versions
6 explicit affected versions
12 explicit affected versions
19 explicit affected versions
6 explicit affected versions
100 explicit affected versions
24 explicit affected versions
85 explicit affected versions
84 explicit affected versions
86 explicit affected versions
76 explicit affected versions
10 explicit affected versions
39 explicit affected versions
26 explicit affected versions
90 explicit affected versions
7 explicit affected versions
84 explicit affected versions
42 explicit affected versions
13 explicit affected versions
46 explicit affected versions
35 explicit affected versions
51 explicit affected versions
50 explicit affected versions
18 explicit affected versions
42 explicit affected versions
64 explicit affected versions
48 explicit affected versions
13 explicit affected versions
4 explicit affected versions
23 explicit affected versions
55 explicit affected versions
1 explicit affected versions
21 explicit affected versions
9 explicit affected versions
16 explicit affected versions
10 explicit affected versions
51 explicit affected versions
3 explicit affected versions
7 explicit affected versions
12 explicit affected versions
2 explicit affected versions
12 explicit affected versions
14 explicit affected versions
1 explicit affected versions
7 explicit affected versions
19 explicit affected versions
12 explicit affected versions
12 explicit affected versions
44 explicit affected versions
45 explicit affected versions
29 explicit affected versions
37 explicit affected versions
10 explicit affected versions
37 explicit affected versions
10 explicit affected versions
13 explicit affected versions
42 explicit affected versions
10 explicit affected versions
68 explicit affected versions
26 explicit affected versions
12 explicit affected versions
13 explicit affected versions
37 explicit affected versions
29 explicit affected versions
13 explicit affected versions
96 explicit affected versions
43 explicit affected versions
43 explicit affected versions
92 explicit affected versions
55 explicit affected versions
33 explicit affected versions
23 explicit affected versions
60 explicit affected versions
10 explicit affected versions
23 explicit affected versions
22 explicit affected versions
11 explicit affected versions
12 explicit affected versions
7 explicit affected versions
12 explicit affected versions
77 explicit affected versions
93 explicit affected versions
10 explicit affected versions
26 explicit affected versions
42 explicit affected versions
1 explicit affected versions
69 explicit affected versions
37 explicit affected versions
48 explicit affected versions
13 explicit affected versions
48 explicit affected versions
16 explicit affected versions
91 explicit affected versions
53 explicit affected versions
3 explicit affected versions
2 explicit affected versions
2 explicit affected versions
51 explicit affected versions
11 explicit affected versions
43 explicit affected versions
12 explicit affected versions
71 explicit affected versions
80 explicit affected versions
27 explicit affected versions
66 explicit affected versions
45 explicit affected versions
8 explicit affected versions
55 explicit affected versions
5 explicit affected versions
3 explicit affected versions
8 explicit affected versions
10 explicit affected versions
2 explicit affected versions
49 explicit affected versions
13 explicit affected versions
13 explicit affected versions
7 explicit affected versions
2 explicit affected versions
10 explicit affected versions
12 explicit affected versions
39 explicit affected versions
44 explicit affected versions
16 explicit affected versions
3 explicit affected versions
13 explicit affected versions
26 explicit affected versions
45 explicit affected versions
37 explicit affected versions
53 explicit affected versions
14 explicit affected versions
38 explicit affected versions
9 explicit affected versions
11 explicit affected versions
42 explicit affected versions
21 explicit affected versions
92 explicit affected versions
8 explicit affected versions
7 explicit affected versions
4 explicit affected versions
81 explicit affected versions
1 explicit affected versions
1 explicit affected versions
1 explicit affected versions
91 explicit affected versions
80 explicit affected versions
79 explicit affected versions
13 explicit affected versions
43 explicit affected versions
10 explicit affected versions
8 explicit affected versions
48 explicit affected versions
14 explicit affected versions
4 explicit affected versions
78 explicit affected versions
11 explicit affected versions
1 explicit affected versions
22 explicit affected versions
41 explicit affected versions
80 explicit affected versions
03 / CONNECTIONS
Connected vulnerabilities
04 / EVIDENCE
Source records
In the Linux kernel, the following vulnerability has been resolved: mm: zswap: properly synchronize freeing resources during CPU hotunplug In zswap_compress() and zswap_decompress(), the per-CPU acomp_ctx of the current CPU at the beginning of the operation is retrieved and used throughout. However, since neither preemption nor migration are disabled, it is possible that the operation continues on a different CPU. If the original CPU is hotunplugged while the acomp_ctx is still in use, we run into a UAF bug as some of the resources attached to the acomp_ctx are freed during hotunplug in zswap_cpu_comp_dead() (i.e. acomp_ctx.buffer, acomp_ctx.req, or acomp_ctx.acomp). The problem was introduced in commit 1ec3b5fe6eec ("mm/zswap: move to use crypto_acomp API for hardware acceleration") when the switch to the crypto_acomp API was made. Prior to that, the per-CPU crypto_comp was retrieved using get_cpu_ptr() which disables preemption and makes sure the CPU cannot go away from under us. Preemption cannot be disabled with the crypto_acomp API as a sleepable context is needed. Use the acomp_ctx.mutex to synchronize CPU hotplug callbacks allocating and freeing resources with compression/decompression paths. Make sure that acomp_ctx.req is NULL when the resources are freed. In the compression/decompression paths, check if acomp_ctx.req is NULL after acquiring the mutex (meaning the CPU was offlined) and retry on the new CPU. The initialization of acomp_ctx.mutex is moved from the CPU hotplug callback to the pool initialization where it belongs (where the mutex is allocated). In addition to adding clarity, this makes sure that CPU hotplug cannot reinitialize a mutex that is already locked by compression/decompression. Previously a fix was attempted by holding cpus_read_lock() [1]. This would have caused a potential deadlock as it is possible for code already holding the lock to fall into reclaim and enter zswap (causing a deadlock). A fix was also attempted using SRCU for synchronization, but Johannes pointed out that synchronize_srcu() cannot be used in CPU hotplug notifiers [2]. Alternative fixes that were considered/attempted and could have worked: - Refcounting the per-CPU acomp_ctx. This involves complexity in handling the race between the refcount dropping to zero in zswap_[de]compress() and the refcount being re-initialized when the CPU is onlined. - Disabling migration before getting the per-CPU acomp_ctx [3], but that's discouraged and is a much bigger hammer than needed, and could result in subtle performance issues. [1]https://lkml.kernel.org/[email protected]/ [2]https://lkml.kernel.org/[email protected]/ [3]https://lkml.kernel.org/[email protected]/ [[email protected]: remove comment]
05 / REFERENCES
Further evidence
- https://git.kernel.org/linus/12dcb0ef540629a281533f9dedc1b6b8e14cfb65
- https://git.kernel.org/stable/c/12dcb0ef540629a281533f9dedc1b6b8e14cfb65
- https://git.kernel.org/stable/c/8d29ff5d50304daa41dc3cfdda4a9d1e46cf5be1
- https://ubuntu.com/security/CVE-2025-21693
- https://ubuntu.com/security/notices/USN-7445-1
- https://ubuntu.com/security/notices/USN-7448-1
- https://www.cve.org/CVERecord?id=CVE-2025-21693