CVE-2026-64373

*** Pendiente de traducción *** In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> cpufreq: Fix hotplug-suspend race during reboot<br /> <br /> During system reboot, cpufreq_suspend() is called via the<br /> kernel_restart() -&gt; device_shutdown() path. Unlike the normal system<br /> suspend path, the reboot path does not call freeze_processes(), so<br /> userspace processes and kernel threads remain active.<br /> <br /> This allows CPU hotplug operations to run concurrently with<br /> cpufreq_suspend(). The original code has no synchronization with CPU<br /> hotplug, leading to a race condition where governor_data can be freed<br /> by the hotplug path while cpufreq_suspend() is still accessing it,<br /> resulting in a null pointer dereference:<br /> <br /> Unable to handle kernel NULL pointer dereference<br /> Call Trace:<br /> do_kernel_fault+0x28/0x3c<br /> cpufreq_suspend+0xdc/0x160<br /> device_shutdown+0x18/0x200<br /> kernel_restart+0x40/0x80<br /> arm64_sys_reboot+0x1b0/0x200<br /> <br /> Fix this by adding cpus_read_lock()/cpus_read_unlock() to<br /> cpufreq_suspend() to block CPU hotplug operations while suspend is in<br /> progress.<br /> <br /> [ rjw: Changelog edits ]