{"id":"AZL-96039","summary":"CVE-2026-72315 affecting package kernel 6.6.150.1-1","details":"In the Linux kernel, the following vulnerability has been resolved:\n\nsmb: client: fix busy dentry warning on unmount after DIO\n\nCommit c68337442f03 (\"cifs: Fix busy dentry used after unmounting\") fixed\nthe issue in cifs where deferred close of a file led to a dentry reference\ncount not being released in umount, by flushing deferredclose_wq in\ncifs_kill_sb() to solve it.\n\nHowever, the cifs DIO path suffers from the same busy-dentry problem caused\nby a delayed dentry reference-count release:\n\n\t[dio]\t\t\t[cifsd]\t\t\t[close + umount]\nnetfs_unbuffered_write_iter_locked\n...\n\t\t\t\tcifs_demultiplex_thread\n netfs_unbuffered_write\n  cifs_issue_write\n  netfs_wait_for_in_progress_stream [1]\n\t\t\t\t...\n\t\t\t\t netfs_write_subrequest_terminated\n\t\t\t\t  netfs_subreq_clear_in_progress\n\t\t\t\t   netfs_wake_collector // wake [1]\n\t\t\t\t  netfs_put_subrequest\n netfs_put_request\n  queue_work(system_dfl_wq, xxx) [2]\n // dio write return\t\t\t\t\tcifs_close\n\t\t\t\t\t\t\t _cifsFileInfo_put\n\t\t\t\t\t\t\t  // cfile-\u003ecount 2-\u003e1\n\t\t\t\t\t\t\t  --cfile-\u003ecount [3]\n\n\t\t\t\t\t\t\t// umount\n\t\t\t\t\t\t\tcifs_kill_sb\n\t\t\t\t\t\t\t kill_anon_super\n\t\t\t\t\t\t\t  // warning triggered!\n\t\t\t\t\t\t\t  shrink_dcache_for_umount [4]\n[system_dfl_wq] [5]\nnetfs_free_request\n ...\n _cifsFileInfo_put\n  // cfile-\u003ecount 1-\u003e0\n  --cfile-\u003ecount\n  queue_work(fileinfo_put_wq, xxx)\n\n[fileinfo_put_wq] [6]\ncifsFileInfo_put_work\n cifsFileInfo_put_final\n  dput\n\nIf the umount path is triggered before [5], it results warning:\nBUG: Dentry 00000000eab1f070{i=9a917b66ae404fec,n=test}  still in use (1)\n[unmount of cifs cifs]\n\nThe existing per-inode ictx-\u003eio_count wait in cifs_evict_inode() does not\nhelp: it lives in the inode eviction path, which runs after\nshrink_dcache_for_umount() has already warned about the busy dentries.\n\nFix it by adding a per-superblock outstanding-rreq counter that is\nincremented in cifs_init_request() and decremented in cifs_free_request().\nIn cifs_kill_sb(), before kill_anon_super(), wait for this counter to reach\n0 - which guarantees that all cleanup_work for this sb have run and thus\nall relevant cfile puts are queued on fileinfo_put_wq or serverclose_wq.\nThen drain the workqueue so the dentry refs are dropped.\n\nThis is a targeted wait, not a flush of the system-wide system_dfl_wq.","modified":"2026-08-31T05:26:27Z","published":"2026-08-15T06:22:04Z","upstream":["CVE-2026-72315"],"references":[{"type":"WEB","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-72315"}],"affected":[{"package":{"name":"kernel","ecosystem":"Azure Linux:3","purl":"pkg:rpm/azure-linux/kernel"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"last_affected":"6.6.150.1-1"}]}],"database_specific":{"source":"https://github.com/microsoft/AzureLinuxVulnerabilityData/blob/main/osv/AZL-96039.json"}}],"schema_version":"1.9.0"}