CVE-2026-63845

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/jpeg: set nouserfence for JPEG v4.0 ring JPEG rings do not support 64-bit user fence writes, reject CS submissions with user fences. cherry picked from commit 8d0cac9478a3f046279c657d6a2545de49ae675a...

CVE-2026-63848

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/jpeg: set nouserfence for JPEG v2.0 ring JPEG rings do not support 64-bit user fence writes, reject CS submissions with user fences. cherry picked from commit 96179da0c6b059eb31706a0abe8dd6381c533143...

CVE-2026-63850

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/vcn: set nouserfence for VCN v5.0.0 enc ring VCN encoder and decoder rings do not support 64-bit user fence writes, reject CS submissions with user fences. cherry picked from commit...

CVE-2026-63844

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/jpeg: set nouserfence for JPEG v4.0.3 ring JPEG rings do not support 64-bit user fence writes, reject CS submissions with user fences. cherry picked from commit 2f6afc97d259d530f4f86c7743efbc573a8da927...

CVE-2026-63847

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/jpeg: set nouserfence for JPEG v2.5 ring JPEG rings do not support 64-bit user fence writes, reject CS submissions with user fences. cherry picked from commit 3216a7f4e2642bda5fd14f57586e835ae9202587...

CVE-2026-63846

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/jpeg: set nouserfence for JPEG v3.0 ring JPEG rings do not support 64-bit user fence writes, reject CS submissions with user fences. cherry picked from commit 4d7d774f100efb5089c86a1fb8c5bf47c63fc9ef...

CVE-2026-63849

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/vcn: set nouserfence for VCN v5.0.1 enc ring VCN encoder and decoder rings do not support 64-bit user fence writes, reject CS submissions with user fences. cherry picked from commit...

CVE-2026-63841

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/jpeg: set nouserfence for JPEG v5.0.1 ring JPEG rings do not support 64-bit user fence writes, reject CS submissions with user fences. cherry picked from commit 742a98e2e81702df8fe1b1eccee5223220a03dc2...

CVE-2026-63842

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/jpeg: set nouserfence for JPEG v5.0.0 ring JPEG rings do not support 64-bit user fence writes, reject CS submissions with user fences. cherry picked from commit 0f43893d3cd478fa57836697525b338817c9c23d...

CVE-2026-63840

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/jpeg: set nouserfence for JPEG v5.3.0 ring JPEG rings do not support 64-bit user fence writes, reject CS submissions with user fences. cherry picked from commit 86ac011ae234c03fb872f4945913391ea1d8862e...

CVE-2026-63882 drm/amdkfd: fix NULL pointer bug in svm_range_set_attr

In the Linux kernel, the following vulnerability has been resolved: drm/amdkfd: fix NULL pointer bug in svmrangesetattr The processinfo could be NULL if user doesn't call kfdioctlacquirevm before calling kfdioctlsvm. cherry picked from commit 83a26c812e0529eb040d31a76f73e33e637243d4...

CVE-2026-63880

CVE-2026-63880 affects the Linux kernel AMDGPU driver (drm/amdgpu) in the AMDGPU_GEM_OP_GET_MAPPING_INFO path. A -ENOMEM failure in kvcalloc could skip cleanup and leak three locks (drm_gem_object reference, drm_exec lock on the GEM, and drm_exec lock on the per-process VM root PD), causing a loa...

CVE-2026-63880 drm/amdgpu: fix lock leak on ENOMEM in AMDGPU_GEM_OP_GET_MAPPING_INFO

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: fix lock leak on ENOMEM in AMDGPUGEMOPGETMAPPINGINFO The AMDGPUGEMOPGETMAPPINGINFO branch of amdgpugemopioctl holds three cleanup-tracked resources before calling kvcalloc: the drmgemobject reference from...

CVE-2026-63880

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: fix lock leak on ENOMEM in AMDGPUGEMOPGETMAPPINGINFO The AMDGPUGEMOPGETMAPPINGINFO branch of amdgpugemopioctl holds three cleanup-tracked resources before calling kvcalloc: the drmgemobject reference from...

CVE-2026-63879 drm/amdgpu: fix amdgpu_hmm_range_get_pages

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: fix amdgpuhmmrangegetpages The notifier sequence must only be read once or otherwise we could work with invalid pages. While at it also fix the coding style, e.g. drop the pre-initialized return value and use the comm...

EUVD-2026-45764

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: fix amdgpuhmmrangegetpages The notifier sequence must only be read once or otherwise we could work with invalid pages. While at it also fix the coding style, e.g. drop the pre-initialized return value and use the comm...

CVE-2026-63879

The CVE-2026-63879 entry relates to the Linux kernel DRM/AMDGPU stack, where a race in amdgpu_hmm_range_get_pages could allow operations on invalid pages due to reading the notifier sequence more than once. The issue has been resolved by a patch that fixes the range getter and also tidies coding ...

CVE-2026-63879

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: fix amdgpuhmmrangegetpages The notifier sequence must only be read once or otherwise we could work with invalid pages. While at it also fix the coding style, e.g. drop the pre-initialized return value and use the comm...

CVE-2026-63879

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: fix amdgpuhmmrangegetpages The notifier sequence must only be read once or otherwise we could work with invalid pages. While at it also fix the coding style, e.g. drop the pre-initialized return value and use the comm...

CVE-2026-63879 drm/amdgpu: fix amdgpu_hmm_range_get_pages

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: fix amdgpuhmmrangegetpages The notifier sequence must only be read once or otherwise we could work with invalid pages. While at it also fix the coding style, e.g. drop the pre-initialized return value and use the comm...