{"api_version":"1","generated_at":"2026-10-15T05:15:03+00:00","cve":"CVE-2026-98178","urls":{"html":"https://cve.report/CVE-2026-98178","api":"https://cve.report/api/cve/CVE-2026-98178.json","docs":"https://cve.report/api","cve_org":"https://www.cve.org/CVERecord?id=CVE-2026-98178","nvd":"https://nvd.nist.gov/vuln/detail/CVE-2026-98178"},"summary":{"title":"drm/amdgpu: Skip KFD mapping clear before initialization","description":"In the Linux kernel, the following vulnerability has been resolved:\n\ndrm/amdgpu: Skip KFD mapping clear before initialization\n\namdgpu_amdkfd_clear_kfd_mapping() assumes that a non-NULL kfd_dev\nhas a fully populated node array. This is not true when KFD device\ninitialization fails after probe.\n\nFor example, kgd2kfd_device_init() sets num_nodes before checking\nPCIe atomics support. On Polaris systems without the required atomics,\nit returns before allocating nodes[0], but the kfd_dev remains attached\nto the amdgpu device. A later GPU reset then dereferences nodes[0]->id.\n\nRequire the authoritative KFD initialization flag before walking the\nnode array, matching the existing KFD reset and teardown paths.\n\n(cherry picked from commit 4ac1835823c47903fbb278bbf474773c46f59edc)","state":"PUBLISHED","assigner":"Linux","published_at":"2026-10-06 09:17:59","updated_at":"2026-10-06 09:17:59"},"problem_types":[],"metrics":[],"references":[{"url":"https://git.kernel.org/stable/c/7f9caa70aef0950e06d395ca0035831214d88187","name":"https://git.kernel.org/stable/c/7f9caa70aef0950e06d395ca0035831214d88187","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/157d3f1db7e7e6f320daa221fae84a4c13555b6f","name":"https://git.kernel.org/stable/c/157d3f1db7e7e6f320daa221fae84a4c13555b6f","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://www.cve.org/CVERecord?id=CVE-2026-98178","name":"CVE Program record","refsource":"CVE.ORG","tags":["canonical"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-98178","name":"NVD vulnerability detail","refsource":"NVD","tags":["canonical","analysis"]}],"affected":[{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 70cadefcc6160c575b04f763ada34c20e868d577 157d3f1db7e7e6f320daa221fae84a4c13555b6f git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 70cadefcc6160c575b04f763ada34c20e868d577 7f9caa70aef0950e06d395ca0035831214d88187 git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 7.2","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.2 semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.2.8 7.2.* semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.3-rc4 * original_commit_for_fix","platforms":[]}],"timeline":[],"solutions":[],"workarounds":[],"exploits":[],"credits":[],"nvd_cpes":[],"vendor_comments":[],"enrichments":{"kev":null,"epss":{"cve_year":"2026","cve_id":"98178","cve":"CVE-2026-98178","epss":"0.001980000","percentile":"0.087020000","score_date":"2026-10-06","updated_at":"2026-10-07 00:14:55"},"legacy_qids":[]},"source_records":{"cve_program":{"containers":{"cna":{"affected":[{"defaultStatus":"unaffected","product":"Linux","programFiles":["drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c"],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","vendor":"Linux","versions":[{"lessThan":"157d3f1db7e7e6f320daa221fae84a4c13555b6f","status":"affected","version":"70cadefcc6160c575b04f763ada34c20e868d577","versionType":"git"},{"lessThan":"7f9caa70aef0950e06d395ca0035831214d88187","status":"affected","version":"70cadefcc6160c575b04f763ada34c20e868d577","versionType":"git"}]},{"defaultStatus":"affected","product":"Linux","programFiles":["drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c"],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","vendor":"Linux","versions":[{"status":"affected","version":"7.2"},{"lessThan":"7.2","status":"unaffected","version":"0","versionType":"semver"},{"lessThanOrEqual":"7.2.*","status":"unaffected","version":"7.2.8","versionType":"semver"},{"lessThanOrEqual":"*","status":"unaffected","version":"7.3-rc4","versionType":"original_commit_for_fix"}]}],"cpeApplicability":[{"nodes":[{"cpeMatch":[{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"7.2.8","versionStartIncluding":"7.2","vulnerable":true},{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"7.3-rc4","versionStartIncluding":"7.2","vulnerable":true}],"negate":false,"operator":"OR"}]}],"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\ndrm/amdgpu: Skip KFD mapping clear before initialization\n\namdgpu_amdkfd_clear_kfd_mapping() assumes that a non-NULL kfd_dev\nhas a fully populated node array. This is not true when KFD device\ninitialization fails after probe.\n\nFor example, kgd2kfd_device_init() sets num_nodes before checking\nPCIe atomics support. On Polaris systems without the required atomics,\nit returns before allocating nodes[0], but the kfd_dev remains attached\nto the amdgpu device. A later GPU reset then dereferences nodes[0]->id.\n\nRequire the authoritative KFD initialization flag before walking the\nnode array, matching the existing KFD reset and teardown paths.\n\n(cherry picked from commit 4ac1835823c47903fbb278bbf474773c46f59edc)"}],"providerMetadata":{"dateUpdated":"2026-10-06T08:44:21.602Z","orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux"},"references":[{"url":"https://git.kernel.org/stable/c/157d3f1db7e7e6f320daa221fae84a4c13555b6f"},{"url":"https://git.kernel.org/stable/c/7f9caa70aef0950e06d395ca0035831214d88187"}],"title":"drm/amdgpu: Skip KFD mapping clear before initialization","x_generator":{"engine":"bippy-1.2.0"}}},"cveMetadata":{"assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","assignerShortName":"Linux","cveId":"CVE-2026-98178","datePublished":"2026-10-06T08:44:21.602Z","dateReserved":"2026-09-25T10:25:14.322Z","dateUpdated":"2026-10-06T08:44:21.602Z","state":"PUBLISHED"},"dataType":"CVE_RECORD","dataVersion":"5.2"},"nvd":{"publishedDate":"2026-10-06 09:17:59","lastModifiedDate":"2026-10-06 09:17:59","problem_types":[],"metrics":[],"configurations":[]},"legacy_mitre":{"record":{"CveYear":"2026","CveId":"98178","Ordinal":"1","Title":"drm/amdgpu: Skip KFD mapping clear before initialization","CVE":"CVE-2026-98178","Year":"2026"},"notes":[{"CveYear":"2026","CveId":"98178","Ordinal":"1","NoteData":"In the Linux kernel, the following vulnerability has been resolved:\n\ndrm/amdgpu: Skip KFD mapping clear before initialization\n\namdgpu_amdkfd_clear_kfd_mapping() assumes that a non-NULL kfd_dev\nhas a fully populated node array. This is not true when KFD device\ninitialization fails after probe.\n\nFor example, kgd2kfd_device_init() sets num_nodes before checking\nPCIe atomics support. On Polaris systems without the required atomics,\nit returns before allocating nodes[0], but the kfd_dev remains attached\nto the amdgpu device. A later GPU reset then dereferences nodes[0]->id.\n\nRequire the authoritative KFD initialization flag before walking the\nnode array, matching the existing KFD reset and teardown paths.\n\n(cherry picked from commit 4ac1835823c47903fbb278bbf474773c46f59edc)","Type":"Description","Title":"drm/amdgpu: Skip KFD mapping clear before initialization"}]}}}