CVE-2026-23358 PUBLISHED

drm/amdgpu: Fix error handling in slot reset

Assigner: Linux
Reserved: 13.01.2026 Published: 25.03.2026 Updated: 25.03.2026

In the Linux kernel, the following vulnerability has been resolved:

drm/amdgpu: Fix error handling in slot reset

If the device has not recovered after slot reset is called, it goes to out label for error handling. There it could make decision based on uninitialized hive pointer and could result in accessing an uninitialized list.

Initialize the list and hive properly so that it handles the error situation and also releases the reset domain lock which is acquired during error_detected callback.

(cherry picked from commit bb71362182e59caa227e4192da5a612b09349696)

Product Status

Vendor Linux
Product Linux
Versions Default: unaffected
  • affected from 732c6cefc1ecfc8de5d7a2029480798655d979d8 to 73e8bdf14248136459753252a438177df7ed8c7c (excl.)
  • affected from 732c6cefc1ecfc8de5d7a2029480798655d979d8 to baf4e7968911635eb816870af0ea587ac1457052 (excl.)
  • affected from 732c6cefc1ecfc8de5d7a2029480798655d979d8 to b57c4ec98c17789136a4db948aec6daadceb5024 (excl.)
Vendor Linux
Product Linux
Versions Default: affected
  • Version 6.16 is affected
  • unaffected from 0 to 6.16 (excl.)
  • unaffected from 6.18.17 to 6.18.* (incl.)
  • unaffected from 6.19.7 to 6.19.* (incl.)
  • unaffected from 7.0-rc2 to * (incl.)

References