{"api_version":"1","generated_at":"2026-07-24T19:55:31+00:00","cve":"CVE-2026-63880","urls":{"html":"https://cve.report/CVE-2026-63880","api":"https://cve.report/api/cve/CVE-2026-63880.json","docs":"https://cve.report/api","cve_org":"https://www.cve.org/CVERecord?id=CVE-2026-63880","nvd":"https://nvd.nist.gov/vuln/detail/CVE-2026-63880"},"summary":{"title":"drm/amdgpu: fix lock leak on ENOMEM in AMDGPU_GEM_OP_GET_MAPPING_INFO","description":"In the Linux kernel, the following vulnerability has been resolved:\n\ndrm/amdgpu: fix lock leak on ENOMEM in AMDGPU_GEM_OP_GET_MAPPING_INFO\n\nThe AMDGPU_GEM_OP_GET_MAPPING_INFO branch of amdgpu_gem_op_ioctl()\nholds three cleanup-tracked resources before calling kvcalloc():\nthe drm_gem_object reference from drm_gem_object_lookup(), the\ndrm_exec lock on the looked-up GEM via drm_exec_lock_obj(), and\nthe drm_exec lock on the per-process VM root page directory via\namdgpu_vm_lock_pd().  All three are released by the out_exec\nlabel that every other error path in this function jumps to.\nThe kvcalloc() failure path returns -ENOMEM directly, skipping\nout_exec and leaking all three.\n\nThe leaked per-process VM root PD dma_resv lock is the\nload-bearing leak: any subsequent operation on the same VM\n(further GEM ops, command-submission, eviction, TTM shrinker\ncallbacks) blocks on the held lock.  DRM_IOCTL_AMDGPU_GEM_OP is\nDRM_AUTH | DRM_RENDER_ALLOW, so this is an unprivileged-local\ndenial of service against the caller's GPU context, reachable\nby any process with /dev/dri/renderD* access.\n\nRoute the failure through out_exec so drm_exec_fini() and\ndrm_gem_object_put() run.\n\nReproduced on stock 7.0.0-10, Ryzen 7 5700U / Radeon Vega\n(Lucienne): the failing ioctl returns -ENOMEM and a second\nGET_MAPPING_INFO on the same fd then blocks in\ndrm_exec_lock_obj() on the leaked dma_resv.  SIGKILL on the\ncaller does not reap the task; the fd-release path during\nprocess exit goes through amdgpu_gem_object_close() ->\ndrm_exec_prepare_obj() on the same lock, leaving the task in D\nstate until the box is rebooted.  The patched kernel was not\nrebuilt and re-tested on this hardware; the fix is mechanical.\nTested on a single Lucienne / Vega box only.\n\nZiyi Guo posted an independent INT_MAX-bound check for\nargs->num_entries in the same branch [1]; the two patches are\ncomplementary and can land in either order.\n\n(cherry picked from commit b69d3256d79de15f54c322986ff4da68f1d65b0a)","state":"PUBLISHED","assigner":"Linux","published_at":"2026-07-19 16:17:05","updated_at":"2026-07-19 16:17:05"},"problem_types":[],"metrics":[],"references":[{"url":"https://git.kernel.org/stable/c/2e7f55eb408c3f72ee1957a0d0ad11d8648a6379","name":"https://git.kernel.org/stable/c/2e7f55eb408c3f72ee1957a0d0ad11d8648a6379","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/1eb86334e391695d4a40743b114afc15df4dc506","name":"https://git.kernel.org/stable/c/1eb86334e391695d4a40743b114afc15df4dc506","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/8f643d534ffc6f1b6182e4f3acff8f04890504b9","name":"https://git.kernel.org/stable/c/8f643d534ffc6f1b6182e4f3acff8f04890504b9","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://www.cve.org/CVERecord?id=CVE-2026-63880","name":"CVE Program record","refsource":"CVE.ORG","tags":["canonical"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-63880","name":"NVD vulnerability detail","refsource":"NVD","tags":["canonical","analysis"]}],"affected":[{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 4d82724f7f2b847eb0454b1aab5450545b39abd4 1eb86334e391695d4a40743b114afc15df4dc506 git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 4d82724f7f2b847eb0454b1aab5450545b39abd4 8f643d534ffc6f1b6182e4f3acff8f04890504b9 git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 4d82724f7f2b847eb0454b1aab5450545b39abd4 2e7f55eb408c3f72ee1957a0d0ad11d8648a6379 git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 6.18","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 6.18 semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 6.18.35 6.18.* semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.0.12 7.0.* semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.1 * original_commit_for_fix","platforms":[]}],"timeline":[],"solutions":[],"workarounds":[],"exploits":[],"credits":[],"nvd_cpes":[],"vendor_comments":[],"enrichments":{"kev":null,"epss":{"cve_year":"2026","cve_id":"63880","cve":"CVE-2026-63880","epss":"0.001980000","percentile":"0.097900000","score_date":"2026-07-20","updated_at":"2026-07-21 00:13:14"},"legacy_qids":[]},"source_records":{"cve_program":{"containers":{"cna":{"affected":[{"defaultStatus":"unaffected","product":"Linux","programFiles":["drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c"],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","vendor":"Linux","versions":[{"lessThan":"1eb86334e391695d4a40743b114afc15df4dc506","status":"affected","version":"4d82724f7f2b847eb0454b1aab5450545b39abd4","versionType":"git"},{"lessThan":"8f643d534ffc6f1b6182e4f3acff8f04890504b9","status":"affected","version":"4d82724f7f2b847eb0454b1aab5450545b39abd4","versionType":"git"},{"lessThan":"2e7f55eb408c3f72ee1957a0d0ad11d8648a6379","status":"affected","version":"4d82724f7f2b847eb0454b1aab5450545b39abd4","versionType":"git"}]},{"defaultStatus":"affected","product":"Linux","programFiles":["drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c"],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","vendor":"Linux","versions":[{"status":"affected","version":"6.18"},{"lessThan":"6.18","status":"unaffected","version":"0","versionType":"semver"},{"lessThanOrEqual":"6.18.*","status":"unaffected","version":"6.18.35","versionType":"semver"},{"lessThanOrEqual":"7.0.*","status":"unaffected","version":"7.0.12","versionType":"semver"},{"lessThanOrEqual":"*","status":"unaffected","version":"7.1","versionType":"original_commit_for_fix"}]}],"cpeApplicability":[{"nodes":[{"cpeMatch":[{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"6.18.35","versionStartIncluding":"6.18","vulnerable":true},{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"7.0.12","versionStartIncluding":"6.18","vulnerable":true},{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"7.1","versionStartIncluding":"6.18","vulnerable":true}],"negate":false,"operator":"OR"}]}],"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\ndrm/amdgpu: fix lock leak on ENOMEM in AMDGPU_GEM_OP_GET_MAPPING_INFO\n\nThe AMDGPU_GEM_OP_GET_MAPPING_INFO branch of amdgpu_gem_op_ioctl()\nholds three cleanup-tracked resources before calling kvcalloc():\nthe drm_gem_object reference from drm_gem_object_lookup(), the\ndrm_exec lock on the looked-up GEM via drm_exec_lock_obj(), and\nthe drm_exec lock on the per-process VM root page directory via\namdgpu_vm_lock_pd().  All three are released by the out_exec\nlabel that every other error path in this function jumps to.\nThe kvcalloc() failure path returns -ENOMEM directly, skipping\nout_exec and leaking all three.\n\nThe leaked per-process VM root PD dma_resv lock is the\nload-bearing leak: any subsequent operation on the same VM\n(further GEM ops, command-submission, eviction, TTM shrinker\ncallbacks) blocks on the held lock.  DRM_IOCTL_AMDGPU_GEM_OP is\nDRM_AUTH | DRM_RENDER_ALLOW, so this is an unprivileged-local\ndenial of service against the caller's GPU context, reachable\nby any process with /dev/dri/renderD* access.\n\nRoute the failure through out_exec so drm_exec_fini() and\ndrm_gem_object_put() run.\n\nReproduced on stock 7.0.0-10, Ryzen 7 5700U / Radeon Vega\n(Lucienne): the failing ioctl returns -ENOMEM and a second\nGET_MAPPING_INFO on the same fd then blocks in\ndrm_exec_lock_obj() on the leaked dma_resv.  SIGKILL on the\ncaller does not reap the task; the fd-release path during\nprocess exit goes through amdgpu_gem_object_close() ->\ndrm_exec_prepare_obj() on the same lock, leaving the task in D\nstate until the box is rebooted.  The patched kernel was not\nrebuilt and re-tested on this hardware; the fix is mechanical.\nTested on a single Lucienne / Vega box only.\n\nZiyi Guo posted an independent INT_MAX-bound check for\nargs->num_entries in the same branch [1]; the two patches are\ncomplementary and can land in either order.\n\n(cherry picked from commit b69d3256d79de15f54c322986ff4da68f1d65b0a)"}],"providerMetadata":{"dateUpdated":"2026-07-19T14:54:52.397Z","orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux"},"references":[{"url":"https://git.kernel.org/stable/c/1eb86334e391695d4a40743b114afc15df4dc506"},{"url":"https://git.kernel.org/stable/c/8f643d534ffc6f1b6182e4f3acff8f04890504b9"},{"url":"https://git.kernel.org/stable/c/2e7f55eb408c3f72ee1957a0d0ad11d8648a6379"}],"title":"drm/amdgpu: fix lock leak on ENOMEM in AMDGPU_GEM_OP_GET_MAPPING_INFO","x_generator":{"engine":"bippy-1.2.0"}}},"cveMetadata":{"assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","assignerShortName":"Linux","cveId":"CVE-2026-63880","datePublished":"2026-07-19T14:54:52.397Z","dateReserved":"2026-07-19T07:54:57.018Z","dateUpdated":"2026-07-19T14:54:52.397Z","state":"PUBLISHED"},"dataType":"CVE_RECORD","dataVersion":"5.2"},"nvd":{"publishedDate":"2026-07-19 16:17:05","lastModifiedDate":"2026-07-19 16:17:05","problem_types":[],"metrics":[],"configurations":[]},"legacy_mitre":{"record":{"CveYear":"2026","CveId":"63880","Ordinal":"1","Title":"drm/amdgpu: fix lock leak on ENOMEM in AMDGPU_GEM_OP_GET_MAPPING","CVE":"CVE-2026-63880","Year":"2026"},"notes":[{"CveYear":"2026","CveId":"63880","Ordinal":"1","NoteData":"In the Linux kernel, the following vulnerability has been resolved:\n\ndrm/amdgpu: fix lock leak on ENOMEM in AMDGPU_GEM_OP_GET_MAPPING_INFO\n\nThe AMDGPU_GEM_OP_GET_MAPPING_INFO branch of amdgpu_gem_op_ioctl()\nholds three cleanup-tracked resources before calling kvcalloc():\nthe drm_gem_object reference from drm_gem_object_lookup(), the\ndrm_exec lock on the looked-up GEM via drm_exec_lock_obj(), and\nthe drm_exec lock on the per-process VM root page directory via\namdgpu_vm_lock_pd().  All three are released by the out_exec\nlabel that every other error path in this function jumps to.\nThe kvcalloc() failure path returns -ENOMEM directly, skipping\nout_exec and leaking all three.\n\nThe leaked per-process VM root PD dma_resv lock is the\nload-bearing leak: any subsequent operation on the same VM\n(further GEM ops, command-submission, eviction, TTM shrinker\ncallbacks) blocks on the held lock.  DRM_IOCTL_AMDGPU_GEM_OP is\nDRM_AUTH | DRM_RENDER_ALLOW, so this is an unprivileged-local\ndenial of service against the caller's GPU context, reachable\nby any process with /dev/dri/renderD* access.\n\nRoute the failure through out_exec so drm_exec_fini() and\ndrm_gem_object_put() run.\n\nReproduced on stock 7.0.0-10, Ryzen 7 5700U / Radeon Vega\n(Lucienne): the failing ioctl returns -ENOMEM and a second\nGET_MAPPING_INFO on the same fd then blocks in\ndrm_exec_lock_obj() on the leaked dma_resv.  SIGKILL on the\ncaller does not reap the task; the fd-release path during\nprocess exit goes through amdgpu_gem_object_close() ->\ndrm_exec_prepare_obj() on the same lock, leaving the task in D\nstate until the box is rebooted.  The patched kernel was not\nrebuilt and re-tested on this hardware; the fix is mechanical.\nTested on a single Lucienne / Vega box only.\n\nZiyi Guo posted an independent INT_MAX-bound check for\nargs->num_entries in the same branch [1]; the two patches are\ncomplementary and can land in either order.\n\n(cherry picked from commit b69d3256d79de15f54c322986ff4da68f1d65b0a)","Type":"Description","Title":"drm/amdgpu: fix lock leak on ENOMEM in AMDGPU_GEM_OP_GET_MAPPING"}]}}}