CVE-2025-38322

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> perf/x86/intel: Fix crash in icl_update_topdown_event()<br /> <br /> The perf_fuzzer found a hard-lockup crash on a RaptorLake machine:<br /> <br /> Oops: general protection fault, maybe for address 0xffff89aeceab400: 0000<br /> CPU: 23 UID: 0 PID: 0 Comm: swapper/23<br /> Tainted: [W]=WARN<br /> Hardware name: Dell Inc. Precision 9660/0VJ762<br /> RIP: 0010:native_read_pmc+0x7/0x40<br /> Code: cc e8 8d a9 01 00 48 89 03 5b cd cc cc cc cc 0f 1f ...<br /> RSP: 000:fffb03100273de8 EFLAGS: 00010046<br /> ....<br /> Call Trace:<br /> <br /> icl_update_topdown_event+0x165/0x190<br /> ? ktime_get+0x38/0xd0<br /> intel_pmu_read_event+0xf9/0x210<br /> __perf_event_read+0xf9/0x210<br /> <br /> CPUs 16-23 are E-core CPUs that don&amp;#39;t support the perf metrics feature.<br /> The icl_update_topdown_event() should not be invoked on these CPUs.<br /> <br /> It&amp;#39;s a regression of commit:<br /> <br /> f9bdf1f95339 ("perf/x86/intel: Avoid disable PMU if !cpuc-&gt;enabled in sample read")<br /> <br /> The bug introduced by that commit is that the is_topdown_event() function<br /> is mistakenly used to replace the is_topdown_count() call to check if the<br /> topdown functions for the perf metrics feature should be invoked.<br /> <br /> Fix it.