A race condition in the Linux kernel’s signal handling logic allows a multi‑threaded process that performs an execve during a pending group stop to leave stale job control flags set on the calling thread. This concurrency flaw is a classic example of a race condition, aligning with CWE‑362. When the execve returns, the kernel re‑examines the thread’s stopped state, sees the bogus STOP_PENDING flag, and attempts to decrement the already‑zero group stop counter, which triggers a kernel warning and can lead to fatal error handling of stopped processes. While the defect does not provide remote code execution, it can cause abnormal kernel behaviour and potentially a crash or denial of service if repeated or combined with other bugs.

All Linux kernel builds released prior to the fix. No specific version list is supplied, so any deployment of the kernel before the patch contains the vulnerability. The issue is present in the common Linux kernel code used by distributions that ship the standard kernel source.

The vulnerability has not been publicly exploited and the EPSS score is unavailable, although the design suggests a non‑zero likelihood of exploitation. An attacker would need to create a carefully timed thread race in a user‑mode process that calls execve while a group stop is pending. The impact is mainly a local denial of service at the kernel level, with no direct remote trigger. The lack of a KEV listing and the absence of a known exploit pathway mean that the overall risk is moderate, but the kernel crash potential warrants patching.