{"dataType":"CVE_RECORD","dataVersion":"5.2","cveMetadata":{"cveId":"CVE-2024-52319","assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","state":"PUBLISHED","assignerShortName":"Linux","dateReserved":"2025-01-11T12:33:33.694Z","datePublished":"2025-01-11T12:35:39.280Z","dateUpdated":"2026-08-05T11:43:16.556Z"},"containers":{"cna":{"providerMetadata":{"orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux","dateUpdated":"2026-08-05T11:43:16.556Z"},"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\nmm: use aligned address in clear_gigantic_page()\n\nIn current kernel, hugetlb_no_page() calls folio_zero_user() with the\nfault address.  Where the fault address may be not aligned with the huge\npage size.  Then, folio_zero_user() may call clear_gigantic_page() with\nthe address, while clear_gigantic_page() requires the address to be huge\npage size aligned.  So, this may cause memory corruption or information\nleak, addtional, use more obvious naming 'addr_hint' instead of 'addr' for\nclear_gigantic_page()."}],"metrics":[{"cvssV3_1":{"version":"3.1","vectorString":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H","baseScore":7.8,"baseSeverity":"HIGH"},"scenarios":[{"lang":"en","value":"AV:L - The bug is reached through a page fault on a hugetlbfs/MAP_HUGETLB mapping in the faulting task's own address space (hugetlb_no_page() → folio_zero_user()), requiring local execution on the host. There is no network- or adjacent-reachable path into mm/memory.c's page-zeroing helpers.\nAC:L - The attacker fully and deterministically controls the misalignment — vmf->real_address is the raw faulting address, so simply touching any non-huge-page-aligned byte of the mapping on first fault reproduces the wrong cache-alias selection every time, with no race and no memory-layout guessing.\nPR:L - An ordinary unprivileged user can allocate the mapping: mmap(MAP_HUGETLB|MAP_HUGE_*) uses HUGETLB_ANONHUGE_INODE, which bypasses the can_do_hugetlb_shm() capability gate in hugetlb_file_setup(), and hugetlbfs mounts are commonly world-accessible. No capability, and no CAP_IPC_LOCK or hugetlb_shm_group membership, is needed.\nUI:N - The attacker triggers the flaw entirely on their own by faulting a mapping they created themselves; no action by any other user or administrator is required at attack time.\nS:U - The mis-zeroed page and the leaked/corrupted data both live within the same kernel-managed security authority; no VM, IOMMU, or sandbox boundary is crossed.\nC:H - Because the gigantic folio is cleared through the wrong virtual cache alias (sparc64 clear_user_page's TLBTEMP color bit, sh's skipped __flush_purge_region), the user reads stale lines and recovers the prior contents of that physical memory — potentially hundreds of megabytes to gigabytes of another process's or KVM guest's freed hugetlb data. The fix commit explicitly cites \"information leak\".\nI:H - The fix commit explicitly states the unaligned address \"may cause memory corruption\": dirty stale aliased cache lines can be written back over the region the kernel just zeroed, silently corrupting memory in a page the kernel guarantees is zero-filled, including shared hugetlbfs pages consumed by other processes.\nA:H - Silent corruption of hugetlbfs-backed memory that callers assume is zeroed — typically KVM guest RAM or a database's shared segment on these gigantic-page deployments — reliably produces crashes and hangs in the affected workloads, and the higher value is taken given the fix commit's own \"memory corruption\" characterization."}]}],"affected":[{"product":"Linux","vendor":"Linux","defaultStatus":"unaffected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["fs/hugetlbfs/inode.c","mm/memory.c"],"versions":[{"version":"78fefd04c123493bbf28434768fa577b2153c79b","lessThan":"b79b6fe0737f233f0be1465052b7f0e75f324735","status":"affected","versionType":"git"},{"version":"78fefd04c123493bbf28434768fa577b2153c79b","lessThan":"8aca2bc96c833ba695ede7a45ad7784c836a262e","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/hugetlbfs/inode.c","mm/memory.c"],"versions":[{"version":"6.11","status":"affected"},{"version":"0","lessThan":"6.11","status":"unaffected","versionType":"semver"},{"version":"6.12.7","lessThanOrEqual":"6.12.*","status":"unaffected","versionType":"semver"},{"version":"6.13","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":"6.11","versionEndExcluding":"6.12.7"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"6.11","versionEndExcluding":"6.13"}]}]}],"references":[{"url":"https://git.kernel.org/stable/c/b79b6fe0737f233f0be1465052b7f0e75f324735"},{"url":"https://git.kernel.org/stable/c/8aca2bc96c833ba695ede7a45ad7784c836a262e"}],"title":"mm: use aligned address in clear_gigantic_page()","x_generator":{"engine":"bippy-1.2.0"}}}}