CVE-2026-100077

The Linux kernel contains a vulnerability in the drm/msm driver where retiring a hung submit before GPU recovery is complete is unsafe. This action triggers a Buffer Object (BO) free , potentially causing GPU pagefaults because the GPU might still be actively accessing those BOs. The fix involves...

Astra Linux – Vulnerability in Linux 6.1

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: The calltrace warning in amddrmbuddyfini has been fixed. The following call trace was observed when the amdgpu driver was removed. This issue arises because the BOs allocated for psp are not freed until the driver is...

CVE-2026-53377

A flaw was found in the Linux kernel's Direct Rendering Manager DRM driver for Qualcomm Snapdragon Mobile MSM graphics processing units GPUs. The GPU recovery mechanism, intended to reset the GPU after a hang, did not always activate when no new tasks were submitted. This oversight could allow th...

CVE-2026-53377

In the Linux kernel, the following vulnerability has been resolved: drm/msm: always recover the gpu Previously, in case there was no more work to do, recover worker wouldn't trigger recovery and would instead rely on the gpu going to sleep and then resuming when more work is submitted...

AZL-92558 CVE-2026-53377 affecting package kernel 6.6.157.1-1

In the Linux kernel, the following vulnerability has been resolved: drm/msm: always recover the gpu Previously, in case there was no more work to do, recover worker wouldn't trigger recovery and would instead rely on the gpu going to sleep and then resuming when more work is submitted...

CVE-2026-53377

CVE-2026-53377 affects the Linux kernel DRM MSM driver. The GPU recovery path (Recover_worker) could fail to trigger recovery when no new work is submitted, leaving the GPU hung and potentially causing timeouts or unavailability. The fix changes Recover_worker to always recover the GPU (increment...

CVE-2026-53377 drm/msm: always recover the gpu

In the Linux kernel, the following vulnerability has been resolved: drm/msm: always recover the gpu Previously, in case there was no more work to do, recover worker wouldn't trigger recovery and would instead rely on the gpu going to sleep and then resuming when more work is submitted...

EUVD-2026-45450

In the Linux kernel, the following vulnerability has been resolved: drm/msm: always recover the gpu Previously, in case there was no more work to do, recover worker wouldn't trigger recovery and would instead rely on the gpu going to sleep and then resuming when more work is submitted...

CVE-2026-53377

In the Linux kernel, the following vulnerability has been resolved: drm/msm: always recover the gpu Previously, in case there was no more work to do, recover worker wouldn't trigger recovery and would instead rely on the gpu going to sleep and then resuming when more work is submitted...

CVE-2026-53377 drm/msm: always recover the gpu

In the Linux kernel, the following vulnerability has been resolved: drm/msm: always recover the gpu Previously, in case there was no more work to do, recover worker wouldn't trigger recovery and would instead rely on the gpu going to sleep and then resuming when more work is submitted...

Astra Linux – Vulnerability in Linux 5.10

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: fixed an access race in the gpu recovery process. A potential UAF Uninitialized Address Fault during GPU recovery was avoided due to a race between the sched timeout callback and the tdr work queue. The gpu recovery...

SUSE CVE-2025-68793

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: fix a job-pasid access race in gpu recovery Avoid a possible UAF in GPU recovery due to a race between the sched timeout callback and the tdr work queue. The gpu recovery function calls drmschedstop and later...

CVE-2025-68793

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: fix a job-pasid access race in gpu recovery Avoid a possible UAF in GPU recovery due to a race between the sched timeout callback and the tdr work queue. The gpu recovery function calls drmschedstop and later...

UBUNTU-CVE-2025-68793

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: fix a job-pasid access race in gpu recovery Avoid a possible UAF in GPU recovery due to a race between the sched timeout callback and the tdr work queue. The gpu recovery function calls drmschedstop and later...

CVE-2025-68793

CVE-2025-68793 affects the Linux kernel’s AMDGPU GPU recovery path. The issue is a race between the sched timeout callback and the tdr work queue during gpu recovery, where drm_sched_start() can resume the tdr queue and later free a job, risking a use-after-free on the pasid if the timeout comple...

CVE-2025-68793 drm/amdgpu: fix a job->pasid access race in gpu recovery

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: fix a job-pasid access race in gpu recovery Avoid a possible UAF in GPU recovery due to a race between the sched timeout callback and the tdr work queue. The gpu recovery function calls drmschedstop and later...

CVE-2025-68793: Undefined Security Weakness

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: fix a job-pasid access race in gpu recovery Avoid a possible UAF in GPU recovery due to a race between the sched timeout callback and the tdr work queue. The gpu recovery function calls drmschedstop and later...

PT-2026-2525

Name of the Vulnerable Software and Affected Versions Linux kernel versions prior to 6.16.0-1289896.2.zuul.bf4f11df81c1410bbe901c4373305a31 Description The Linux kernel contains a flaw in the DRM/amdgpu subsystem related to a use-after-free UAF condition during GPU recovery. This issue arises fro...

drm/amdgpu: fix use-after-free during gpu recovery

...

CVE-2025-37854 drm/amdkfd: Fix mode1 reset crash issue

In the Linux kernel, the following vulnerability has been resolved: drm/amdkfd: Fix mode1 reset crash issue If HW scheduler hangs and mode1 reset is used to recover GPU, KFD signal user space to abort the processes. After process abort exit, user queues still use the GPU to access system memory...