{"id":"AZL-67247","summary":"CVE-2025-39776 affecting package kernel for versions less than 6.6.104.2-1","details":"In the Linux kernel, the following vulnerability has been resolved:\n\nmm/debug_vm_pgtable: clear page table entries at destroy_args()\n\nThe mm/debug_vm_pagetable test allocates manually page table entries for\nthe tests it runs, using also its manually allocated mm_struct.  That in\nitself is ok, but when it exits, at destroy_args() it fails to clear those\nentries with the *_clear functions.\n\nThe problem is that leaves stale entries.  If another process allocates an\nmm_struct with a pgd at the same address, it may end up running into the\nstale entry.  This is happening in practice on a debug kernel with\nCONFIG_DEBUG_VM_PGTABLE=y, for example this is the output with some extra\ndebugging I added (it prints a warning trace if pgtables_bytes goes\nnegative, in addition to the warning at check_mm() function):\n\n[    2.539353] debug_vm_pgtable: [get_random_vaddr         ]: random_vaddr is 0x7ea247140000\n[    2.539366] kmem_cache info\n[    2.539374] kmem_cachep 0x000000002ce82385 - freelist 0x0000000000000000 - offset 0x508\n[    2.539447] debug_vm_pgtable: [init_args                ]: args-\u003emm is 0x000000002267cc9e\n(...)\n[    2.552800] WARNING: CPU: 5 PID: 116 at include/linux/mm.h:2841 free_pud_range+0x8bc/0x8d0\n[    2.552816] Modules linked in:\n[    2.552843] CPU: 5 UID: 0 PID: 116 Comm: modprobe Not tainted 6.12.0-105.debug_vm2.el10.ppc64le+debug #1 VOLUNTARY\n[    2.552859] Hardware name: IBM,9009-41A POWER9 (architected) 0x4e0202 0xf000005 of:IBM,FW910.00 (VL910_062) hv:phyp pSeries\n[    2.552872] NIP:  c0000000007eef3c LR: c0000000007eef30 CTR: c0000000003d8c90\n[    2.552885] REGS: c0000000622e73b0 TRAP: 0700   Not tainted  (6.12.0-105.debug_vm2.el10.ppc64le+debug)\n[    2.552899] MSR:  800000000282b033 \u003cSF,VEC,VSX,EE,FP,ME,IR,DR,RI,LE\u003e  CR: 24002822  XER: 0000000a\n[    2.552954] CFAR: c0000000008f03f0 IRQMASK: 0\n[    2.552954] GPR00: c0000000007eef30 c0000000622e7650 c000000002b1ac00 0000000000000001\n[    2.552954] GPR04: 0000000000000008 0000000000000000 c0000000007eef30 ffffffffffffffff\n[    2.552954] GPR08: 00000000ffff00f5 0000000000000001 0000000000000048 0000000000004000\n[    2.552954] GPR12: 00000003fa440000 c000000017ffa300 c0000000051d9f80 ffffffffffffffdb\n[    2.552954] GPR16: 0000000000000000 0000000000000008 000000000000000a 60000000000000e0\n[    2.552954] GPR20: 4080000000000000 c0000000113af038 00007fffcf130000 0000700000000000\n[    2.552954] GPR24: c000000062a6a000 0000000000000001 8000000062a68000 0000000000000001\n[    2.552954] GPR28: 000000000000000a c000000062ebc600 0000000000002000 c000000062ebc760\n[    2.553170] NIP [c0000000007eef3c] free_pud_range+0x8bc/0x8d0\n[    2.553185] LR [c0000000007eef30] free_pud_range+0x8b0/0x8d0\n[    2.553199] Call Trace:\n[    2.553207] [c0000000622e7650] [c0000000007eef30] free_pud_range+0x8b0/0x8d0 (unreliable)\n[    2.553229] [c0000000622e7750] [c0000000007f40b4] free_pgd_range+0x284/0x3b0\n[    2.553248] [c0000000622e7800] [c0000000007f4630] free_pgtables+0x450/0x570\n[    2.553274] [c0000000622e78e0] [c0000000008161c0] exit_mmap+0x250/0x650\n[    2.553292] [c0000000622e7a30] [c0000000001b95b8] __mmput+0x98/0x290\n[    2.558344] [c0000000622e7a80] [c0000000001d1018] exit_mm+0x118/0x1b0\n[    2.558361] [c0000000622e7ac0] [c0000000001d141c] do_exit+0x2ec/0x870\n[    2.558376] [c0000000622e7b60] [c0000000001d1ca8] do_group_exit+0x88/0x150\n[    2.558391] [c0000000622e7bb0] [c0000000001d1db8] sys_exit_group+0x48/0x50\n[    2.558407] [c0000000622e7be0] [c00000000003d810] system_call_exception+0x1e0/0x4c0\n[    2.558423] [c0000000622e7e50] [c00000000000d05c] system_call_vectored_common+0x15c/0x2ec\n(...)\n[    2.558892] ---[ end trace 0000000000000000 ]---\n[    2.559022] BUG: Bad rss-counter state mm:000000002267cc9e type:MM_ANONPAGES val:1\n[    2.559037] BUG: non-zero pgtables_bytes on freeing mm: -6144\n\nHere the modprobe process ended up with an allocated mm_struct from the\nmm_struct slab that was used before by the debug_vm_pgtable test.  That is\nnot a problem, since the mm_stru\n---truncated---","modified":"2026-04-01T05:21:10.002223Z","published":"2025-09-11T17:15:43Z","upstream":["CVE-2025-39776"],"references":[{"type":"WEB","url":"https://nvd.nist.gov/vuln/detail/CVE-2025-39776"}],"affected":[{"package":{"name":"kernel","ecosystem":"Azure Linux:3","purl":"pkg:rpm/azure-linux/kernel"},"ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"6.6.104.2-1"}]}],"database_specific":{"source":"https://github.com/microsoft/AzureLinuxVulnerabilityData/blob/main/osv/AZL-67247.json"}}],"schema_version":"1.7.5"}