In the Linux kernel’s RAID5 driver, the function retry_aligned_read() contains a logic error that allows a stripe to be prematurely removed from the active list, causing an infinite processing loop that results in a soft lockup. This effect can temporarily halt a CPU, but it does not compromise data integrity or grant privilege escalation. The likely attack vector involves local interaction with a RAID5 array: by provoking a read request on an overlapped stripe, an attacker could trigger the loop and induce the lockup.

It is inferred that all systems that run a Linux kernel containing the unpatched retry_aligned_read() logic are potentially affected, regardless of distribution or kernel release track, because the commit date is not tied to a specific kernel version range. Therefore any kernel build that includes the original code is vulnerable.

The CVSS score is 5.5 and EPSS information is not available. The vulnerability is not listed in CISA’s KEV catalogue. Exploitation requires local access to a system configured to use RAID5; an attacker could trigger the infinite loop by initiating a read on an overlapped stripe, thereby causing a persistent soft lockup. Given the lack of a remote trigger and the need for continuous I/O activity, the exploitation risk is moderate but could be high for critical services running on the affected node.