{"id":"CVE-2026-99428","summary":"ocfs2: snapshot dentry names in file and setattr traces","details":"In the Linux kernel, the following vulnerability has been resolved:\n\nocfs2: snapshot dentry names in file and setattr traces\n\nPatch series \"ocfs2: stabilize dentry names in file and setattr traces\",\nv4.\n\nSnapshot names at all six users of ocfs2__file_ops and at the separate\nsetattr event. Patch 1 handles the five ordinary file-operation callers\nand setattr. Patch 2 handles release separately to preserve its locked\nopen-count sample and take the snapshot outside that critical section.\n\nThe release explanation distinguishes its existing RCU lifetime coverage\nfrom name/length consistency. It does not claim that the ordinary\ncallers' external-name UAF interleaving crosses the held inode spinlock.\nThe release source change is unchanged from v3.\n\nPatch 1 also removes the read callback's redundant NULL-inode check.\nThe two patches apply independently and leave trace formats unchanged.\n\n\nThis patch (of 2):\n\nThe open, fsync, read, write and splice-read trace calls pass the live\ndentry name and length to the shared ocfs2__file_ops event class.\nocfs2_setattr() passes the same unprotected arguments to its own event.\nBoth events size and copy a __string field separately. A reference to\nthe dentry does not pin the external storage backing an individual name.\n\nWith an event enabled, a caller can be preempted after loading the name\npointer. A long-to-short rename can then replace the name and reclaim its\nold external allocation before the trace callback sizes or copies it,\nprovided no other snapshot retains that allocation.\n\nThe setattr inode lock does not exclude every rename: a same-directory\nrename of a source directory need not take that directory's inode lock.\nIt can therefore change the dentry name while setattr records its event.\nInline names and separately loaded lengths can also change during trace\nrecord sizing and copying.\n\nTake a dentry name snapshot around each of these six trace calls and\npass both the string and length from the snapshot. Check event enablement\nbefore taking a snapshot. Keep the event classes and formats unchanged.\n\nAlso remove the redundant NULL-inode check in the read callback.\nfile_inode() returns the open file's inode, which is already used by the\ntrace call above the check.\n\nThe historical report below is from the read path.\n\nKASAN report as below:\n\n    BUG: KASAN: slab-use-after-free in strlen+0x85/0xa0\n    Read of size 1 at addr ffff88810a5b2b18 by task python3/496\n\n    [CPU and host details omitted.]\n    Call Trace:\n     \u003cTASK\u003e\n     dump_stack_lvl+0x93/0xd0\n     print_report+0xce/0x630\n     ? strlen+0x85/0xa0\n     ? srso_alias_return_thunk+0x5/0xfbef5\n     ? __virt_addr_valid+0x20e/0x420\n     ? strlen+0x85/0xa0\n     kasan_report+0xe0/0x110\n     ? strlen+0x85/0xa0\n     strlen+0x85/0xa0\n     trace_event_raw_event_ocfs2__file_ops+0xf0/0x480\n     ? __pfx_trace_event_raw_event_ocfs2__file_ops+0x10/0x10\n     ? srso_alias_return_thunk+0x5/0xfbef5\n     ? pmbd_probe_hit_cookie+0xe7/0x1c0\n     ? srso_alias_return_thunk+0x5/0xfbef5\n     ? __pfx_pmbd_probe_hit_cookie+0x10/0x10\n     ocfs2_file_read_iter+0x6cd/0xb30\n     ? __pfx_ocfs2_file_read_iter+0x10/0x10\n     ? selinux_file_permission+0x11d/0x540\n     ? srso_alias_return_thunk+0x5/0xfbef5\n     ? security_file_permission+0x26/0x80\n     vfs_read+0x7d2/0xc20\n     ? __pfx_vfs_read+0x10/0x10\n     ? srso_alias_return_thunk+0x5/0xfbef5\n     ? __pfx_do_sys_openat2+0x10/0x10\n     ksys_read+0x111/0x200\n     ? __pfx_ksys_read+0x10/0x10\n     ? irqentry_exit+0xbe/0x820\n     ? srso_alias_return_thunk+0x5/0xfbef5\n     ? trace_hardirqs_on_prepare+0x11d/0x150\n     ? srso_alias_return_thunk+0x5/0xfbef5\n     do_syscall_64+0x114/0x620\n     entry_SYSCALL_64_after_hwframe+0x77/0x7f\n    [User register dump omitted.]\n     \u003c/TASK\u003e\n\n    Allocated by task 489:\n     kasan_save_stack+0x33/0x60\n     kasan_save_track+0x14/0x30\n     __kasan_kmalloc+0xaa/0xb0\n     __kmalloc_noprof+0x2de/0x780\n     __d_alloc+0x5fc/0x8c0\n     d_alloc_parallel+0x106/0x11d0\n     lookup_open+0x3ff/0\n---truncated---","modified":"2026-11-19T17:01:18.115980707Z","published":"2026-11-19T14:40:45.853Z","database_specific":{"osv_generated_from":"https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/99xxx/CVE-2026-99428.json","cna_assigner":"Linux"},"references":[{"type":"PACKAGE","url":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git"},{"type":"WEB","url":"https://git.kernel.org/stable/c/84908981fbda3c133b60e75568d853d3191f9fb2"},{"type":"WEB","url":"https://git.kernel.org/stable/c/8cd3ff30659775a6831c66687a69d2b83d163591"},{"type":"WEB","url":"https://git.kernel.org/stable/c/d2052367cbdad321d131e6280f41c4be5a1b5c96"},{"type":"WEB","url":"https://git.kernel.org/stable/c/f976692490bbdf670d4cc241b8951c8cfddb1583"},{"type":"ADVISORY","url":"https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/99xxx/CVE-2026-99428.json"},{"type":"ADVISORY","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-99428"}],"affected":[{"ranges":[{"type":"GIT","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","events":[{"introduced":"468eedde23d6c9335935773f4f5764267d5a7763"},{"fixed":"84908981fbda3c133b60e75568d853d3191f9fb2"},{"fixed":"f976692490bbdf670d4cc241b8951c8cfddb1583"},{"fixed":"d2052367cbdad321d131e6280f41c4be5a1b5c96"},{"fixed":"8cd3ff30659775a6831c66687a69d2b83d163591"}]}],"database_specific":{"source":"https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2026-99428.json"}},{"package":{"name":"Kernel","ecosystem":"Linux"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"2.6.39"},{"fixed":"6.12.117"}]},{"type":"ECOSYSTEM","events":[{"introduced":"6.13.0"},{"fixed":"6.18.61"}]},{"type":"ECOSYSTEM","events":[{"introduced":"6.19.0"},{"fixed":"7.3.2"}]}],"database_specific":{"source":"https://storage.googleapis.com/cve-osv-conversion/osv-output/CVE-2026-99428.json"}}],"schema_version":"1.9.2"}