{"dataType":"CVE_RECORD","dataVersion":"5.2","cveMetadata":{"cveId":"CVE-2026-89647","assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","state":"PUBLISHED","assignerShortName":"Linux","dateReserved":"2026-09-11T19:38:34.741Z","datePublished":"2026-09-11T19:45:38.593Z","dateUpdated":"2026-09-13T06:32:25.355Z"},"containers":{"cna":{"providerMetadata":{"orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux","dateUpdated":"2026-09-13T06:32:25.355Z"},"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\nceph: do not repeat ceph_trim_dentries() if no progress possible\n\nceph_cap_reclaim_work() re-queues itself for as long as\nceph_trim_dentries() returns -EAGAIN, which happens whenever a lease\nwalk exhausts its `nr_to_scan` budget.  This creates a busy loop that\nconsumes CPU without making any progress when there is nothing to\nreclaim: with no cap pressure (`count==0`) and every scanned lease\nstill valid, each pass runs the full scan budget down to zero and\nreturns `-EAGAIN`, only to be queued again immediately.\n\nThe dir-lease walk made this worse.  When `expire_dir_lease` is\n`false` (i.e. we have no intention of reclaiming dir leases),\n__dir_lease_check() returned `TOUCH` for every valid lease.  `TOUCH`\nmoves the dentry to the tail of the list and resets `di->time` via\n__dentry_dir_lease_touch(), so a walk over N valid leases pointlessly\nrewrote the list, refreshed the timestamps (preventing them from ever\naging out) and always drained `nr_to_scan`, guaranteeing the `-EAGAIN`\nrequeue.\n\nFix this in three steps:\n\n - Return `KEEP` instead of `TOUCH` when `expire_dir_lease` is\n   `false`.  If we are not going to reclaim the lease, leave it in\n   place instead of churning the list and resetting its timestamp; the\n   walk then terminates naturally (or via `STOP` at the first fresh\n   lease).\n\n - Only return `-EAGAIN` from the first (dentry-lease) walk when something\n   was actually freed.  A full batch that frees nothing means retrying\n   the same list immediately is futile; fall through to the dir-lease\n   walk instead.\n\n - After both walks, bail out with success (0) when nothing was freed\n   and there is no cap pressure (`count==0`).  There is no reason to\n   keep retrying when we are not over the cap limit and made no\n   progress.\n\nUnder real cap pressure (`count>0`) the reclaim path is unchanged and\nstill retries via `-EAGAIN`.\n\nWithout this patch, I saw 500 ceph_trim_dentries() calls per second on\nour web servers.  This is very visible in `/proc/lock_stat` (5 minute\ncapture):\n\n              class name    con-bounces    contentions   waittime-min   waittime-max waittime-total   waittime-avg    acq-bounces   acquisitions   holdtime-min   holdtime-max holdtime-total   holdtime-avg\n\n &mdsc->dentry_list_lock:        126180         128218           0.04        8063.44    15986965.20         124.69        1573354        5296812           0.04        8291.28    74164526.48          14.00\n -----------------------\n &mdsc->dentry_list_lock         111736          [<000000007b11e319>] __ceph_dentry_dir_lease_touch+0x7c/0xa8\n &mdsc->dentry_list_lock           2631          [<0000000050597999>] __dentry_leases_walk+0x64/0x2c8\n &mdsc->dentry_list_lock           3878          [<00000000c0022f62>] __ceph_dentry_lease_touch+0x5c/0xa8\n &mdsc->dentry_list_lock           9973          [<000000002f27cb6f>] __dentry_lease_unlist+0x50/0xa0\n -----------------------\n &mdsc->dentry_list_lock         123621          [<0000000050597999>] __dentry_leases_walk+0x64/0x2c8\n &mdsc->dentry_list_lock           1822          [<000000007b11e319>] __ceph_dentry_dir_lease_touch+0x7c/0xa8\n &mdsc->dentry_list_lock           2720          [<000000002f27cb6f>] __dentry_lease_unlist+0x50/0xa0\n &mdsc->dentry_list_lock             55          [<00000000c0022f62>] __ceph_dentry_lease_touch+0x5c/0xa8\n\nWith this patch:\n\n              class name    con-bounces    contentions   waittime-min   waittime-max waittime-total   waittime-avg    acq-bounces   acquisitions   holdtime-min   holdtime-max holdtime-total   holdtime-avg\n\n &mdsc->dentry_list_lock:          1203           1215           0.16         408.88       33082.88          27.23        4320501        7357389           0.04         500.64     1961578.00           0.27\n -----------------------\n &mdsc->dentry_list_lock           1029          [<000000003c9aea8a>] __ceph_dentry_dir_lease_touch+0x7c/0xa8\n &mdsc->dentry_list_lock            1\n---truncated---"}],"metrics":[{"cvssV3_1":{"version":"3.1","vectorString":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","baseScore":7.5,"baseSeverity":"HIGH"},"scenarios":[{"lang":"en","value":"AV:N - fs/ceph registers ceph_export_ops, so remote nfsd/ksmbd readdir and lookup on an exported CephFS mount populate dentry and dir-lease lists and drive ceph_trim_dentries() via the cap-reclaim workqueue; a compromised MDS on the existing client TCP session can also keep those leases valid.\nAC:L - Default caps_max leaves count==0, so expire_dir_lease is false and valid dir leases are TOUCHed until nr_to_scan hits zero; ceph_cap_reclaim_work() then immediately requeues on -EAGAIN. An attacker who walks or creates many dentries, or a peer that grants many still-valid leases, triggers this deterministically with no race.\nPR:N - In the network-export scenario the attacker needs no local account or capability on the victim host; guest or anonymous NFS/SMB exports of CephFS, and a remote MDS peer on an already-connected kernel client, are reasonable high-severity deployments matching other Ceph client CVEs.\nUI:N - After CephFS is mounted and optionally exported, delayed_work periodically queues reclaim and the attacker’s own readdir/lookup or MDS lease traffic is enough; no separate victim mount, open, or confirmation is required at exploit time.\nS:U - The busy loop, CPU burn, and dentry_list_lock contention stay inside the host kernel CephFS client and do not cross a VM, IOMMU, or sandbox boundary to another security authority.\nC:N - The defect only rescans and TOUCHes still-valid dentry leases; it does not perform an out-of-bounds read, use-after-free, or any other kernel-memory disclosure.\nI:N - The walk only reshuffles lease-list order and refreshes dentry timestamps; there is no memory corruption, arbitrary write, or control-flow hijack primitive.\nA:H - ceph_cap_reclaim_work() immediately requeues itself on -EAGAIN, forming an unbounded busy loop on the max_active=1 ceph-cap workqueue (observed ~500 iterations/s) that starves cap-release work, burns CPU, and contends dentry_list_lock enough to stall Ceph directory operations."}]}],"affected":[{"product":"Linux","vendor":"Linux","defaultStatus":"unaffected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["fs/ceph/dir.c"],"versions":[{"version":"37c4efc1ddf98ba8b234d116d863a9464445901e","lessThan":"5a541eb401acb89d791da41189d7a79220164b93","status":"affected","versionType":"git"},{"version":"37c4efc1ddf98ba8b234d116d863a9464445901e","lessThan":"37d6edb2f03b29399a3a337fae78674de51e1695","status":"affected","versionType":"git"},{"version":"37c4efc1ddf98ba8b234d116d863a9464445901e","lessThan":"3d122b2feb1dd76bb5041bdea5e1e1b007d8d415","status":"affected","versionType":"git"},{"version":"37c4efc1ddf98ba8b234d116d863a9464445901e","lessThan":"e7d7aa7b730178278109c41fa1b17b06873065d5","status":"affected","versionType":"git"}]},{"product":"Linux","vendor":"Linux","defaultStatus":"affected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["fs/ceph/dir.c"],"versions":[{"version":"5.1","status":"affected"},{"version":"0","lessThan":"5.1","status":"unaffected","versionType":"semver"},{"version":"6.12.109","lessThanOrEqual":"6.12.*","status":"unaffected","versionType":"semver"},{"version":"6.18.50","lessThanOrEqual":"6.18.*","status":"unaffected","versionType":"semver"},{"version":"7.2.4","lessThanOrEqual":"7.2.*","status":"unaffected","versionType":"semver"},{"version":"7.3-rc1","lessThanOrEqual":"*","status":"unaffected","versionType":"original_commit_for_fix"}]}],"cpeApplicability":[{"nodes":[{"operator":"OR","negate":false,"cpeMatch":[{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"5.1","versionEndExcluding":"6.12.109"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"5.1","versionEndExcluding":"6.18.50"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"5.1","versionEndExcluding":"7.2.4"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"5.1","versionEndExcluding":"7.3-rc1"}]}]}],"references":[{"url":"https://git.kernel.org/stable/c/5a541eb401acb89d791da41189d7a79220164b93"},{"url":"https://git.kernel.org/stable/c/37d6edb2f03b29399a3a337fae78674de51e1695"},{"url":"https://git.kernel.org/stable/c/3d122b2feb1dd76bb5041bdea5e1e1b007d8d415"},{"url":"https://git.kernel.org/stable/c/e7d7aa7b730178278109c41fa1b17b06873065d5"}],"title":"ceph: do not repeat ceph_trim_dentries() if no progress possible","x_generator":{"engine":"bippy-1.2.0"}}}}