CVE-2026-89813

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: force complete the KIQ ring fences on reset Like the MES scheduler ring, the KIQ ring sets no_scheduler = true and uses a polling fence, so it is skipped by the force-completion loop in amdgpu_device_pre_asic_reset(). Its hw fence value lives in wb (GTT) memory and survives a MODE1 reset while fence_drv.sync_seq keeps advancing, so after a reset the first KIQ submission can poll forever on a seq that is never written back. Force complete the KIQ ring fences too so their hw fence is realigned to sync_seq.

CVSS

No CVSS.

References

Link	Resource
https://git.kernel.org/stable/c/bdd6a58743e00ebc6d52b9cdf9aafb4e535e0b71
https://git.kernel.org/stable/c/fd65d1742992361fc2201ecb4e43411e6e417fcb

Configurations

No configuration.

History

No history.

Information

Published : 2026-09-16 11:16

Updated : 2026-09-17 10:17

NVD link : CVE-2026-89813

Mitre link : CVE-2026-89813

CVE.ORG link : CVE-2026-89813

JSON object : View

Products Affected

No product.

CWE

No CWE.

{"id": "CVE-2026-89813", "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": "65781c78ad74e4260fbec92c0ecc05738044e177", "lessThan": "bdd6a58743e00ebc6d52b9cdf9aafb4e535e0b71", "versionType": "git"}, {"status": "affected", "version": "65781c78ad74e4260fbec92c0ecc05738044e177", "lessThan": "fd65d1742992361fc2201ecb4e43411e6e417fcb", "versionType": "git"}], "programFiles": ["drivers/gpu/drm/amd/amdgpu/amdgpu_device.c"], "defaultStatus": "unaffected"}, {"repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git", "vendor": "Linux", "product": "Linux", "versions": [{"status": "affected", "version": "4.13"}, {"status": "unaffected", "version": "0", "lessThan": "4.13", "versionType": "semver"}, {"status": "unaffected", "version": "7.2.5", "versionType": "semver", "lessThanOrEqual": "7.2.*"}, {"status": "unaffected", "version": "7.3-rc1", "versionType": "original_commit_for_fix", "lessThanOrEqual": "*"}], "programFiles": ["drivers/gpu/drm/amd/amdgpu/amdgpu_device.c"], "defaultStatus": "affected"}]}], "published": "2026-09-16T11:16:46.503", "references": [{"url": "https://git.kernel.org/stable/c/bdd6a58743e00ebc6d52b9cdf9aafb4e535e0b71", "source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"}, {"url": "https://git.kernel.org/stable/c/fd65d1742992361fc2201ecb4e43411e6e417fcb", "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: force complete the KIQ ring fences on reset\n\nLike the MES scheduler ring, the KIQ ring sets no_scheduler = true and uses a\npolling fence, so it is skipped by the force-completion loop in\namdgpu_device_pre_asic_reset(). Its hw fence value lives in wb (GTT) memory and\nsurvives a MODE1 reset while fence_drv.sync_seq keeps advancing, so after a\nreset the first KIQ submission can poll forever on a seq that is never written\nback.\n\nForce complete the KIQ ring fences too so their hw fence is realigned to\nsync_seq."}], "lastModified": "2026-09-17T10:17:04.490", "sourceIdentifier": "416baaa9-dc9f-4396-8d5f-8c081fb06d67"}