{"api_version":"1","generated_at":"2026-05-28T06:39:14+00:00","cve":"CVE-2026-46025","urls":{"html":"https://cve.report/CVE-2026-46025","api":"https://cve.report/api/cve/CVE-2026-46025.json","docs":"https://cve.report/api","cve_org":"https://www.cve.org/CVERecord?id=CVE-2026-46025","nvd":"https://nvd.nist.gov/vuln/detail/CVE-2026-46025"},"summary":{"title":"mm/damon/core: fix damon_call() vs kdamond_fn() exit race","description":"In the Linux kernel, the following vulnerability has been resolved:\n\nmm/damon/core: fix damon_call() vs kdamond_fn() exit race\n\nPatch series \"mm/damon/core: fix damon_call()/damos_walk() vs kdmond exit\nrace\".\n\ndamon_call() and damos_walk() can leak memory and/or deadlock when they\nrace with kdamond terminations.  Fix those.\n\n\nThis patch (of 2);\n\nWhen kdamond_fn() main loop is finished, the function cancels all\nremaining damon_call() requests and unset the damon_ctx->kdamond so that\nAPI callers and API functions themselves can know the context is\nterminated.  damon_call() adds the caller's request to the queue first. \nAfter that, it shows if the kdamond of the damon_ctx is still running\n(damon_ctx->kdamond is set).  Only if the kdamond is running, damon_call()\nstarts waiting for the kdamond's handling of the newly added request.\n\nThe damon_call() requests registration and damon_ctx->kdamond unset are\nprotected by different mutexes, though.  Hence, damon_call() could race\nwith damon_ctx->kdamond unset, and result in deadlocks.\n\nFor example, let's suppose kdamond successfully finished the damon_call()\nrequests cancelling.  Right after that, damon_call() is called for the\ncontext.  It registers the new request, and shows the context is still\nrunning, because damon_ctx->kdamond unset is not yet done.  Hence the\ndamon_call() caller starts waiting for the handling of the request. \nHowever, the kdamond is already on the termination steps, so it never\nhandles the new request.  As a result, the damon_call() caller threads\ninfinitely waits.\n\nFix this by introducing another damon_ctx field, namely\ncall_controls_obsolete.  It is protected by the\ndamon_ctx->call_controls_lock, which protects damon_call() requests\nregistration.  Initialize (unset) it in kdamond_fn() before letting\ndamon_start() returns and set it just before the cancelling of remaining\ndamon_call() requests is executed.  damon_call() reads the obsolete field\nunder the lock and avoids adding a new request.\n\nAfter this change, only requests that are guaranteed to be handled or\ncancelled are registered.  Hence the after-registration DAMON context\ntermination check is no longer needed.  Remove it together.\n\nNote that the deadlock will not happen when damon_call() is called for\nrepeat mode request.  In tis case, damon_call() returns instead of waiting\nfor the handling when the request registration succeeds and it shows the\nkdamond is running.  However, if the request also has dealloc_on_cancel,\nthe request memory would be leaked.\n\nThe issue is found by sashiko [1].","state":"PUBLISHED","assigner":"Linux","published_at":"2026-05-27 14:17:21","updated_at":"2026-05-27 14:48:03"},"problem_types":[],"metrics":[],"references":[{"url":"https://git.kernel.org/stable/c/2691332ad88b57179c38653e2cd613d5820a52cf","name":"https://git.kernel.org/stable/c/2691332ad88b57179c38653e2cd613d5820a52cf","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/55da81663b9642dd046b26dd6f1baddbcf337c1e","name":"https://git.kernel.org/stable/c/55da81663b9642dd046b26dd6f1baddbcf337c1e","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://git.kernel.org/stable/c/e6a053a6f4b5048746c49432a5cc5b79fe4695fe","name":"https://git.kernel.org/stable/c/e6a053a6f4b5048746c49432a5cc5b79fe4695fe","refsource":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","tags":[],"title":"","mime":"","httpstatus":"","archivestatus":"0"},{"url":"https://www.cve.org/CVERecord?id=CVE-2026-46025","name":"CVE Program record","refsource":"CVE.ORG","tags":["canonical"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-46025","name":"NVD vulnerability detail","refsource":"NVD","tags":["canonical","analysis"]}],"affected":[{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 42b7491af14cbba2393329ce43d508a957bd94fa 2691332ad88b57179c38653e2cd613d5820a52cf git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 42b7491af14cbba2393329ce43d508a957bd94fa e6a053a6f4b5048746c49432a5cc5b79fe4695fe git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 42b7491af14cbba2393329ce43d508a957bd94fa 55da81663b9642dd046b26dd6f1baddbcf337c1e git","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"affected 6.14","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 6.14 semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 6.18.27 6.18.* semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.0.4 7.0.* semver","platforms":[]},{"source":"CNA","vendor":"Linux","product":"Linux","version":"unaffected 7.1-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":["include/linux/damon.h","mm/damon/core.c"],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","vendor":"Linux","versions":[{"lessThan":"2691332ad88b57179c38653e2cd613d5820a52cf","status":"affected","version":"42b7491af14cbba2393329ce43d508a957bd94fa","versionType":"git"},{"lessThan":"e6a053a6f4b5048746c49432a5cc5b79fe4695fe","status":"affected","version":"42b7491af14cbba2393329ce43d508a957bd94fa","versionType":"git"},{"lessThan":"55da81663b9642dd046b26dd6f1baddbcf337c1e","status":"affected","version":"42b7491af14cbba2393329ce43d508a957bd94fa","versionType":"git"}]},{"defaultStatus":"affected","product":"Linux","programFiles":["include/linux/damon.h","mm/damon/core.c"],"repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","vendor":"Linux","versions":[{"status":"affected","version":"6.14"},{"lessThan":"6.14","status":"unaffected","version":"0","versionType":"semver"},{"lessThanOrEqual":"6.18.*","status":"unaffected","version":"6.18.27","versionType":"semver"},{"lessThanOrEqual":"7.0.*","status":"unaffected","version":"7.0.4","versionType":"semver"},{"lessThanOrEqual":"*","status":"unaffected","version":"7.1-rc1","versionType":"original_commit_for_fix"}]}],"cpeApplicability":[{"nodes":[{"cpeMatch":[{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"6.18.27","versionStartIncluding":"6.14","vulnerable":true},{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"7.0.4","versionStartIncluding":"6.14","vulnerable":true},{"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionEndExcluding":"7.1-rc1","versionStartIncluding":"6.14","vulnerable":true}],"negate":false,"operator":"OR"}]}],"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\nmm/damon/core: fix damon_call() vs kdamond_fn() exit race\n\nPatch series \"mm/damon/core: fix damon_call()/damos_walk() vs kdmond exit\nrace\".\n\ndamon_call() and damos_walk() can leak memory and/or deadlock when they\nrace with kdamond terminations.  Fix those.\n\n\nThis patch (of 2);\n\nWhen kdamond_fn() main loop is finished, the function cancels all\nremaining damon_call() requests and unset the damon_ctx->kdamond so that\nAPI callers and API functions themselves can know the context is\nterminated.  damon_call() adds the caller's request to the queue first. \nAfter that, it shows if the kdamond of the damon_ctx is still running\n(damon_ctx->kdamond is set).  Only if the kdamond is running, damon_call()\nstarts waiting for the kdamond's handling of the newly added request.\n\nThe damon_call() requests registration and damon_ctx->kdamond unset are\nprotected by different mutexes, though.  Hence, damon_call() could race\nwith damon_ctx->kdamond unset, and result in deadlocks.\n\nFor example, let's suppose kdamond successfully finished the damon_call()\nrequests cancelling.  Right after that, damon_call() is called for the\ncontext.  It registers the new request, and shows the context is still\nrunning, because damon_ctx->kdamond unset is not yet done.  Hence the\ndamon_call() caller starts waiting for the handling of the request. \nHowever, the kdamond is already on the termination steps, so it never\nhandles the new request.  As a result, the damon_call() caller threads\ninfinitely waits.\n\nFix this by introducing another damon_ctx field, namely\ncall_controls_obsolete.  It is protected by the\ndamon_ctx->call_controls_lock, which protects damon_call() requests\nregistration.  Initialize (unset) it in kdamond_fn() before letting\ndamon_start() returns and set it just before the cancelling of remaining\ndamon_call() requests is executed.  damon_call() reads the obsolete field\nunder the lock and avoids adding a new request.\n\nAfter this change, only requests that are guaranteed to be handled or\ncancelled are registered.  Hence the after-registration DAMON context\ntermination check is no longer needed.  Remove it together.\n\nNote that the deadlock will not happen when damon_call() is called for\nrepeat mode request.  In tis case, damon_call() returns instead of waiting\nfor the handling when the request registration succeeds and it shows the\nkdamond is running.  However, if the request also has dealloc_on_cancel,\nthe request memory would be leaked.\n\nThe issue is found by sashiko [1]."}],"providerMetadata":{"dateUpdated":"2026-05-27T12:56:31.784Z","orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux"},"references":[{"url":"https://git.kernel.org/stable/c/2691332ad88b57179c38653e2cd613d5820a52cf"},{"url":"https://git.kernel.org/stable/c/e6a053a6f4b5048746c49432a5cc5b79fe4695fe"},{"url":"https://git.kernel.org/stable/c/55da81663b9642dd046b26dd6f1baddbcf337c1e"}],"title":"mm/damon/core: fix damon_call() vs kdamond_fn() exit race","x_generator":{"engine":"bippy-1.2.0"}}},"cveMetadata":{"assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","assignerShortName":"Linux","cveId":"CVE-2026-46025","datePublished":"2026-05-27T12:56:31.784Z","dateReserved":"2026-05-13T15:03:33.093Z","dateUpdated":"2026-05-27T12:56:31.784Z","state":"PUBLISHED"},"dataType":"CVE_RECORD","dataVersion":"5.2"},"nvd":{"publishedDate":"2026-05-27 14:17:21","lastModifiedDate":"2026-05-27 14:48:03","problem_types":[],"metrics":[],"configurations":[]},"legacy_mitre":{"record":{"CveYear":"2026","CveId":"46025","Ordinal":"1","Title":"mm/damon/core: fix damon_call() vs kdamond_fn() exit race","CVE":"CVE-2026-46025","Year":"2026"},"notes":[{"CveYear":"2026","CveId":"46025","Ordinal":"1","NoteData":"In the Linux kernel, the following vulnerability has been resolved:\n\nmm/damon/core: fix damon_call() vs kdamond_fn() exit race\n\nPatch series \"mm/damon/core: fix damon_call()/damos_walk() vs kdmond exit\nrace\".\n\ndamon_call() and damos_walk() can leak memory and/or deadlock when they\nrace with kdamond terminations.  Fix those.\n\n\nThis patch (of 2);\n\nWhen kdamond_fn() main loop is finished, the function cancels all\nremaining damon_call() requests and unset the damon_ctx->kdamond so that\nAPI callers and API functions themselves can know the context is\nterminated.  damon_call() adds the caller's request to the queue first. \nAfter that, it shows if the kdamond of the damon_ctx is still running\n(damon_ctx->kdamond is set).  Only if the kdamond is running, damon_call()\nstarts waiting for the kdamond's handling of the newly added request.\n\nThe damon_call() requests registration and damon_ctx->kdamond unset are\nprotected by different mutexes, though.  Hence, damon_call() could race\nwith damon_ctx->kdamond unset, and result in deadlocks.\n\nFor example, let's suppose kdamond successfully finished the damon_call()\nrequests cancelling.  Right after that, damon_call() is called for the\ncontext.  It registers the new request, and shows the context is still\nrunning, because damon_ctx->kdamond unset is not yet done.  Hence the\ndamon_call() caller starts waiting for the handling of the request. \nHowever, the kdamond is already on the termination steps, so it never\nhandles the new request.  As a result, the damon_call() caller threads\ninfinitely waits.\n\nFix this by introducing another damon_ctx field, namely\ncall_controls_obsolete.  It is protected by the\ndamon_ctx->call_controls_lock, which protects damon_call() requests\nregistration.  Initialize (unset) it in kdamond_fn() before letting\ndamon_start() returns and set it just before the cancelling of remaining\ndamon_call() requests is executed.  damon_call() reads the obsolete field\nunder the lock and avoids adding a new request.\n\nAfter this change, only requests that are guaranteed to be handled or\ncancelled are registered.  Hence the after-registration DAMON context\ntermination check is no longer needed.  Remove it together.\n\nNote that the deadlock will not happen when damon_call() is called for\nrepeat mode request.  In tis case, damon_call() returns instead of waiting\nfor the handling when the request registration succeeds and it shows the\nkdamond is running.  However, if the request also has dealloc_on_cancel,\nthe request memory would be leaked.\n\nThe issue is found by sashiko [1].","Type":"Description","Title":"mm/damon/core: fix damon_call() vs kdamond_fn() exit race"}]}}}