CVE-2025-21694

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> fs/proc: fix softlockup in __read_vmcore (part 2)<br /> <br /> Since commit 5cbcb62dddf5 ("fs/proc: fix softlockup in __read_vmcore") the<br /> number of softlockups in __read_vmcore at kdump time have gone down, but<br /> they still happen sometimes.<br /> <br /> In a memory constrained environment like the kdump image, a softlockup is<br /> not just a harmless message, but it can interfere with things like RCU<br /> freeing memory, causing the crashdump to get stuck.<br /> <br /> The second loop in __read_vmcore has a lot more opportunities for natural<br /> sleep points, like scheduling out while waiting for a data write to<br /> happen, but apparently that is not always enough.<br /> <br /> Add a cond_resched() to the second loop in __read_vmcore to (hopefully)<br /> get rid of the softlockups.