{"id":"CVE-2026-93248","published":"2026-09-24T16:17:20.963","lastModified":"2026-09-24T16:17:20.963","description":"In the Linux kernel, the following vulnerability has been resolved:\n\ndrm/xe: don't WARN on kernel job timeout when device already wedged\n\nigt@xe_wedged@wedged-at-any-timeout wedges the device in mode 2\n(UPON_ANY_HANG_NO_RESET) and then rebinds the driver. During unbind,\na GSC proxy kernel submission can still time out; with the device wedged\nand the GuC CT stopped it can never complete, so its kernel job times out.\n\n  Tile0: GT1: Kernel-submitted job timed out\n  WARNING: drivers/gpu/drm/xe/xe_guc_submit.c:...\n\t   at guc_exec_queue_timedout_job()\n  Workqueue: gt-ordered-wq drm_sched_job_timedout\n\nKilled queues skip guc_submit_hint_wedged(), leaving 'wedged' false even\nthough the device is already wedged. The timeout handler then treats the\nkernel queue timeout as unexpected and taints the kernel.\n\nHonour an already-wedged device even for killed queues so the expected\nteardown timeout no longer trips the WARN.\n\n(cherry picked from commit a1c1dbd0f047bb05de6aaf6abe9103031179bf19)","cvssScore":null,"cvssSeverity":null,"cvssVector":null,"cwes":[],"vendors":[],"products":[],"references":[{"url":"https://git.kernel.org/stable/c/13087ad7817e6e5f210064518bfc4d6d58a9c2f3","tags":[]},{"url":"https://git.kernel.org/stable/c/24d81b72992006f9c390f1fc85c7f991e6d06fd6","tags":[]}],"exploitRefs":[],"hasPoc":false,"ai":null}