CVE-2026-68276

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/gfx: fix cleaner shader IB buffer overflow The cleaner shader sysfs path allocates a 16-dword (64 byte) IB but incorrectly fills (align_mask + 1) dwords. On GFX rings align_mask is 0xff, so the loop wrote 256 dwords into a 64-byte buffer, causing a kernel page fault. The IB only needs to be a minimal NOP shell to schedule the job; the cleaner shader itself is emitted on the ring via emit_cleaner_shader(). Fill 16 dwords to match the allocation. v2: Use ib_size_dw variable (Lijo) (cherry picked from commit bf21af331ebf72d0935fd70c73192414a422c03a)

CVSS

No CVSS.

References

Link	Resource
https://git.kernel.org/stable/c/201633f47b542a99bb7baafdfcda7781249fb3d9
https://git.kernel.org/stable/c/3e864bf2a32a1cbdf1e0f9c5a5a4176e8575f4a3
https://git.kernel.org/stable/c/9cd9a983769a4d0e9cc80a287316ee79685d38b3
https://git.kernel.org/stable/c/e28420e36542ae8b66a5bdcec419525f521bddf8

Configurations

No configuration.

History

No history.

Information

Published : 2026-08-10 13:20

Updated : 2026-08-17 05:18

NVD link : CVE-2026-68276

Mitre link : CVE-2026-68276

CVE.ORG link : CVE-2026-68276

JSON object : View

Products Affected

No product.

CWE

No CWE.

{"id": "CVE-2026-68276", "cveTags": [], "metrics": {}, "affected": [{"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "affectedData": [{"repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git", "vendor": "Linux", "product": "Linux", "versions": [{"status": "affected", "version": "d361ad5d2fc0e4d59d5d538092c9b37889756642", "lessThan": "201633f47b542a99bb7baafdfcda7781249fb3d9", "versionType": "git"}, {"status": "affected", "version": "d361ad5d2fc0e4d59d5d538092c9b37889756642", "lessThan": "e28420e36542ae8b66a5bdcec419525f521bddf8", "versionType": "git"}, {"status": "affected", "version": "d361ad5d2fc0e4d59d5d538092c9b37889756642", "lessThan": "9cd9a983769a4d0e9cc80a287316ee79685d38b3", "versionType": "git"}, {"status": "affected", "version": "d361ad5d2fc0e4d59d5d538092c9b37889756642", "lessThan": "3e864bf2a32a1cbdf1e0f9c5a5a4176e8575f4a3", "versionType": "git"}], "programFiles": ["drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c"], "defaultStatus": "unaffected"}, {"repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git", "vendor": "Linux", "product": "Linux", "versions": [{"status": "affected", "version": "6.12"}, {"status": "unaffected", "version": "0", "lessThan": "6.12", "versionType": "semver"}, {"status": "unaffected", "version": "6.12.103", "versionType": "semver", "lessThanOrEqual": "6.12.*"}, {"status": "unaffected", "version": "6.18.42", "versionType": "semver", "lessThanOrEqual": "6.18.*"}, {"status": "unaffected", "version": "7.1.6", "versionType": "semver", "lessThanOrEqual": "7.1.*"}, {"status": "unaffected", "version": "7.2", "versionType": "original_commit_for_fix", "lessThanOrEqual": "*"}], "programFiles": ["drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c"], "defaultStatus": "affected"}]}], "published": "2026-08-10T13:20:16.693", "references": [{"url": "https://git.kernel.org/stable/c/201633f47b542a99bb7baafdfcda7781249fb3d9", "source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"}, {"url": "https://git.kernel.org/stable/c/3e864bf2a32a1cbdf1e0f9c5a5a4176e8575f4a3", "source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"}, {"url": "https://git.kernel.org/stable/c/9cd9a983769a4d0e9cc80a287316ee79685d38b3", "source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"}, {"url": "https://git.kernel.org/stable/c/e28420e36542ae8b66a5bdcec419525f521bddf8", "source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"}], "vulnStatus": "Received", "descriptions": [{"lang": "en", "value": "In the Linux kernel, the following vulnerability has been resolved:\n\ndrm/amdgpu/gfx: fix cleaner shader IB buffer overflow\n\nThe cleaner shader sysfs path allocates a 16-dword (64 byte) IB but\nincorrectly fills (align_mask + 1) dwords. On GFX rings align_mask is\n0xff, so the loop wrote 256 dwords into a 64-byte buffer, causing a\nkernel page fault.\n\nThe IB only needs to be a minimal NOP shell to schedule the job; the\ncleaner shader itself is emitted on the ring via emit_cleaner_shader().\nFill 16 dwords to match the allocation.\n\nv2: Use ib_size_dw variable (Lijo)\n\n(cherry picked from commit bf21af331ebf72d0935fd70c73192414a422c03a)"}], "lastModified": "2026-08-17T05:18:30.190", "sourceIdentifier": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"}