{"api_version":"1","generated_at":"2026-09-12T05:51:29+00:00","cve":"CVE-2026-89681","urls":{"html":"https://cve.report/CVE-2026-89681","api":"https://cve.report/api/cve/CVE-2026-89681.json","docs":"https://cve.report/api","cve_org":"https://www.cve.org/CVERecord?id=CVE-2026-89681","nvd":"https://nvd.nist.gov/vuln/detail/CVE-2026-89681"},"summary":{"title":"nfsd: fix layout fence worker double-reference race","description":"In the Linux kernel, the following vulnerability has been resolved:\n\nnfsd: fix layout fence worker double-reference race\n\nThe workqueue core clears WORK_STRUCT_PENDING before the callback\nis invoked, so delayed_work_pending() in lm_breaker_timedout() can\nreturn false while the fence worker is already running. This lets\nthe breaker take a duplicate sc_count reference and schedule a new\nworker that coalesces with the in-progress one. The extra reference\nis never put, leaking the layout stateid.\n\nReplace the racy delayed_work_pending() check with an\nls_fence_inflight boolean set atomically with\nrefcount_inc_not_zero() under ls_lock, and cleared under ls_lock\nbefore the final nfs4_put_stid() on the dispose path; the retry\npath intentionally retains it.  Remove the self-rearm\nmod_delayed_work() at the top of the worker.","state":"PUBLISHED","assigner":"Linux","published_at":"2026-09-11 20:19:54","updated_at":"2026-09-11 20:19:54"},"problem_types":[],"metrics":[],"references":[{"url":"https://git.kernel.org/stable/c/8580571227451384399b3fa53fcde19848c48e5a","name":"https://git.kernel.org/stable/c/8580571227451384399b3fa53fcde19848c48e5a","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/a278d361e0e8f242211891193d56483a7a9f47a8","name":"https://git.kernel.org/stable/c/a278d361e0e8f242211891193d56483a7a9f47a8","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://www.cve.org/CVERecord?id=CVE-2026-89681","name":"CVE Program record","refsource":"CVE.ORG","tags":["canonical"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-89681","name":"NVD vulnerability detail","refsource":"NVD","tags":["canonical","analysis"]}],"affected":[{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected f52792f484ba2316853736856dde19b7e7458861 a278d361e0e8f242211891193d56483a7a9f47a8 git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected f52792f484ba2316853736856dde19b7e7458861 8580571227451384399b3fa53fcde19848c48e5a git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 7.1","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.1 semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.2.4 7.2.* semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.3-rc1 * original_commit_for_fix","platforms":[]}],"timeline":[],"solutions":[],"workarounds":[],"exploits":[],"credits":[],"nvd_cpes":[],"vendor_comments":[],"enrichments":{"kev":null,"epss":null,"legacy_qids":[]},"source_records":{"cve_program":{"containers":{"cna":{"affected":[{"defaultStatus":"unaffected","product":"Linux","programFiles":["fs/nfsd/nfs4layouts.c","fs/nfsd/state.h"],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","vendor":"Linux","versions":[{"lessThan":"a278d361e0e8f242211891193d56483a7a9f47a8","status":"affected","version":"f52792f484ba2316853736856dde19b7e7458861","versionType":"git"},{"lessThan":"8580571227451384399b3fa53fcde19848c48e5a","status":"affected","version":"f52792f484ba2316853736856dde19b7e7458861","versionType":"git"}]},{"defaultStatus":"affected","product":"Linux","programFiles":["fs/nfsd/nfs4layouts.c","fs/nfsd/state.h"],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","vendor":"Linux","versions":[{"status":"affected","version":"7.1"},{"lessThan":"7.1","status":"unaffected","version":"0","versionType":"semver"},{"lessThanOrEqual":"7.2.*","status":"unaffected","version":"7.2.4","versionType":"semver"},{"lessThanOrEqual":"*","status":"unaffected","version":"7.3-rc1","versionType":"original_commit_for_fix"}]}],"cpeApplicability":[{"nodes":[{"cpeMatch":[{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"7.2.4","versionStartIncluding":"7.1","vulnerable":true},{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"7.3-rc1","versionStartIncluding":"7.1","vulnerable":true}],"negate":false,"operator":"OR"}]}],"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\nnfsd: fix layout fence worker double-reference race\n\nThe workqueue core clears WORK_STRUCT_PENDING before the callback\nis invoked, so delayed_work_pending() in lm_breaker_timedout() can\nreturn false while the fence worker is already running. This lets\nthe breaker take a duplicate sc_count reference and schedule a new\nworker that coalesces with the in-progress one. The extra reference\nis never put, leaking the layout stateid.\n\nReplace the racy delayed_work_pending() check with an\nls_fence_inflight boolean set atomically with\nrefcount_inc_not_zero() under ls_lock, and cleared under ls_lock\nbefore the final nfs4_put_stid() on the dispose path; the retry\npath intentionally retains it.  Remove the self-rearm\nmod_delayed_work() at the top of the worker."}],"providerMetadata":{"dateUpdated":"2026-09-11T19:46:04.121Z","orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux"},"references":[{"url":"https://git.kernel.org/stable/c/a278d361e0e8f242211891193d56483a7a9f47a8"},{"url":"https://git.kernel.org/stable/c/8580571227451384399b3fa53fcde19848c48e5a"}],"title":"nfsd: fix layout fence worker double-reference race","x_generator":{"engine":"bippy-1.2.0"}}},"cveMetadata":{"assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","assignerShortName":"Linux","cveId":"CVE-2026-89681","datePublished":"2026-09-11T19:46:04.121Z","dateReserved":"2026-09-11T19:38:34.748Z","dateUpdated":"2026-09-11T19:46:04.121Z","state":"PUBLISHED"},"dataType":"CVE_RECORD","dataVersion":"5.2"},"nvd":{"publishedDate":"2026-09-11 20:19:54","lastModifiedDate":"2026-09-11 20:19:54","problem_types":[],"metrics":[],"configurations":[]},"legacy_mitre":{"record":{"CveYear":"2026","CveId":"89681","Ordinal":"1","Title":"nfsd: fix layout fence worker double-reference race","CVE":"CVE-2026-89681","Year":"2026"},"notes":[{"CveYear":"2026","CveId":"89681","Ordinal":"1","NoteData":"In the Linux kernel, the following vulnerability has been resolved:\n\nnfsd: fix layout fence worker double-reference race\n\nThe workqueue core clears WORK_STRUCT_PENDING before the callback\nis invoked, so delayed_work_pending() in lm_breaker_timedout() can\nreturn false while the fence worker is already running. This lets\nthe breaker take a duplicate sc_count reference and schedule a new\nworker that coalesces with the in-progress one. The extra reference\nis never put, leaking the layout stateid.\n\nReplace the racy delayed_work_pending() check with an\nls_fence_inflight boolean set atomically with\nrefcount_inc_not_zero() under ls_lock, and cleared under ls_lock\nbefore the final nfs4_put_stid() on the dispose path; the retry\npath intentionally retains it.  Remove the self-rearm\nmod_delayed_work() at the top of the worker.","Type":"Description","Title":"nfsd: fix layout fence worker double-reference race"}]}}}