{"id":"CVE-2026-46290","title":"In the Linux kernel, the following vulnerability has been resolved:\n\nx86/efi: Fix graceful fault handling after FPU softirq changes\n\nSince commit d02198550423 (\"x86/fpu: Improve crypto performance by\nmaking kernel-mode FPU reliably usabl…","summary":"In the Linux kernel, the following vulnerability has been resolved:\n\nx86/efi: Fix graceful fault handling after FPU softirq changes\n\nSince commit d02198550423 (\"x86/fpu: Improve crypto performance by\nmaking kernel-mode FPU reliably usabl…","severity":"none","published":"2026-06-08","updated":"2026-07-07","source":"NVD","sourceUrl":"https://nvd.nist.gov/vuln/detail/CVE-2026-46290","references":[{"url":"https://git.kernel.org/stable/c/088f65e206087bf903743bd18417261d7a4c9644","label":"416baaa9-dc9f-4396-8d5f-8c081fb06d67"},{"url":"https://git.kernel.org/stable/c/22b365ba1af3d8c6036b8e5112fffe80998b85a0","label":"416baaa9-dc9f-4396-8d5f-8c081fb06d67"},{"url":"https://git.kernel.org/stable/c/db155b86d1523e85941f61efd7d7ffb594cc9a29","label":"416baaa9-dc9f-4396-8d5f-8c081fb06d67"}],"tags":["nvd"],"epss":0.00121,"epssPercentile":0.0216,"ingestedAt":"2026-07-07T18:42:24.295Z","slug":"CVE-2026-46290","body":"## Overview\n\nIn the Linux kernel, the following vulnerability has been resolved:\n\nx86/efi: Fix graceful fault handling after FPU softirq changes\n\nSince commit d02198550423 (\"x86/fpu: Improve crypto performance by\nmaking kernel-mode FPU reliably usable in softirqs\"), kernel_fpu_begin()\ncalls fpregs_lock() which uses local_bh_disable() instead of the\nprevious preempt_disable(). This sets SOFTIRQ_OFFSET in preempt_count\nduring the entire EFI runtime service call, causing in_interrupt() to\nreturn true in normal task context.\n\nThe graceful page fault handler efi_crash_gracefully_on_page_fault()\nuses in_interrupt() to bail out for faults in real interrupt context.\nWith SOFTIRQ_OFFSET now set, the handler always bails out, leaving EFI\nfirmware page faults unhandled. This escalates to die() which also sees\nin_interrupt() as true and calls panic(\"Fatal exception in interrupt\"),\nresulting in a hard system freeze. On systems with buggy firmware that\ntriggers page faults during EFI runtime calls (e.g., accessing unmapped\nmemory in GetTime()), this causes an unrecoverable hang instead of the\nexpected graceful EFI_ABORTED recovery.\n\nFix by replacing in_interrupt() with !in_task(). This preserves the\noriginal intent of bailing for interrupts or NMI faults, while no longer\nfalsely triggering from the FPU code path's local_bh_disable().\n\n[ardb: Sashiko spotted that using 'in_hardirq() || in_nmi()' leaves a\n       window where a softirq may be taken before fpregs_lock() is\n       called, but after efi_rts_work.efi_rts_id has been assigned,\n       and any page faults occurring in that window will then be\n       misidentified as having been caused by the firmware. Instead,\n       use !in_task(), which incorporates in_serving_softirq(). ]\n\n## Remediation\n\nRefer to the linked advisories for vendor-supplied fixes and affected version ranges.","depth":"sunlit","depthScore":3,"depthScoreParts":{"impact":2.8,"likelihood":0,"exploitation":0,"ransomware":0},"changes":[]}