CVE-2026-52946
In the Linux kernel, the following vulnerability has been resolved: fs/fcntl: fix SOFTIRQ-unsafe lock order in fasync signaling A SOFTIRQ-safe to SOFTIRQ-unsafe lock order deadlock can occur in send_sigio() and send_sigurg() when a process group…
Does this matter?
High impact if exploited, but EPSS currently rates exploitation as unlikely (0.64%). Schedule it in the normal patch cycle and watch for a rise in EPSS or a public exploit.
Description
In the Linux kernel, the following vulnerability has been resolved: fs/fcntl: fix SOFTIRQ-unsafe lock order in fasync signaling A SOFTIRQ-safe to SOFTIRQ-unsafe lock order deadlock can occur in send_sigio() and send_sigurg() when a process group receives a signal. When FASYNC is configured for a process group (PIDTYPE_PGID), both functions use read_lock(&tasklist_lock) to traverse the task list. However, they are frequently called from softirq context: - send_sigio() via input_inject_event -> kill_fasync - send_sigurg() via tcp_check_urg -> sk_send_sigurg (NET_RX_SOFTIRQ) The deadlock is caused by the rwlock writer fairness mechanism: 1. CPU 0 (process context) holds read_lock(&tasklist_lock) in do_wait(). 2. CPU 1 (process context) attempts write_lock(&tasklist_lock) in fork() or exit() and spins, which blocks all new readers. 3. CPU 0 is interrupted by a softirq (e.g., TCP URG packet reception). 4. The softirq calls send_sigurg() and attempts to acquire read_lock(&tasklist_lock), deadlocking because CPU 1 is waiting. Since PID hashing and do_each_pid_task() traversals are already RCU-protected, the read_lock on tasklist_lock is no longer strictly required for safe traversal. Fix this by replacing tasklist_lock with rcu_read_lock(), aligning the process group signaling path with the single-PID path. This also mitigates a potential remote denial of service vector via TCP URG packets. Lockdep splat: ===================================================== WARNING: SOFTIRQ-safe -> SOFTIRQ-unsafe lock order detected [...] Chain exists of: &dev->event_lock --> &f_owner->lock --> tasklist_lock Possible interrupt unsafe locking scenario: CPU0 CPU1 ---- ---- lock(tasklist_lock); local_irq_disable(); lock(&dev->event_lock); lock(&f_owner->lock); <Interrupt> lock(&dev->event_lock); *** DEADLOCK ***
- CVSS 3.1
- 7.5 HIGHCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
- EPSS
- 0.64% probability · 49th percentile
- CISA KEV
- Not listed
- Weakness
- CWE-667
- Affected
- linux/linux kernel
- Source
- 416baaa9-dc9f-4396-8d5f-8c081fb06d67
References
- https://git.kernel.org/stable/c/00633c4683828acd5256fa8d5163f440d74bbe71Patch
- https://git.kernel.org/stable/c/1bee417678f1135e35b25a37734db46aa94258d2Patch
- https://git.kernel.org/stable/c/20a93e397abe850c49b6fa0e8cc827b5f634a8f5Patch
- https://git.kernel.org/stable/c/32dbd5ce4be3a3ed7e00f8af18795cc84fc50a33Patch
- https://git.kernel.org/stable/c/36c1b57b2ecf3c61ac93f5f07bd29b6f21e226edPatch
- https://git.kernel.org/stable/c/54626335ea4174ab2d9a183b511d825f6765e47bPatch
- https://git.kernel.org/stable/c/897d6a7247739fb1528f98c575df4f2e5de7f994Patch
- https://git.kernel.org/stable/c/b5fa9e32fb6718f70c986ee14dd5d01b4846f331Patch
- https://git.kernel.org/stable/c/bfcc8e8d8a495bb34cae9e620adfb75fb13a3954Patch
- https://cert-portal.siemens.com/productcert/html/ssa-019113.html
Source: NVD record, EPSS from FIRST.org, KEV from CISA. Refreshed daily.