CVE-2026-99700
Severity unknownNo known exploitationFix availablemm/hugetlb_vmemmap: map the shared tail page writable
Record last updated
CVSS score
No score published yet
EPSS, chance of exploitation in 30 days
Higher than 6% of scored CVEs. A prediction, not evidence.
CISA remediation deadline
Not in the KEV catalog
Description
In the Linux kernel, the following vulnerability has been resolved: mm/hugetlb_vmemmap: map the shared tail page writable page_ref_add_unless_zero() reads the refcount and then updates it with a cmpxchg. A lockless get, such as a page cache lookup or a PFN scanner, can read a non-zero refcount from a page that is then freed and reused as a hugetlb tail page. If HVO remaps its struct page to the shared tail page in between, the cmpxchg writes to read-only vmemmap: CPU0 (lockless get) CPU1 c = atomic_read(&page->_refcount) page freed page allocated as a hugetlb tail HVO remaps the struct page RO try_cmpxchg(&page->_refcount, &c, c + 1) -> write fault on read-only vmemmap The refcount is zero by then and the compare fails, but a compare-and-swap instruction needs write access whether or not it succeeds. The kernel oopses when hugetlb pages are allocated at runtime with HVO enabled while a lockless get races with it. Commit bd225530a4c7 ("mm/hugetlb_vmemmap: fix race with speculative PFN walkers") closed the window with RCU on both sides. The RCU was removed together with fake heads, on the wrong belief that it was only needed for them. Map the shared tail page writable instead of restoring the RCU. Every struct page on it is a tail with a zero refcount, so a late cmpxchg fails and leaves the page unchanged. RCU would put rcu_read_lock() back around every get and synchronize_rcu() back into hugetlb allocation and freeing. The cost is the protection against stray writes to tail struct pages. The shared page is shared by all hugetlb folios of the same order in the zone, so such a write now corrupts all of them instead of oopsing. Bootmem folios optimized when the memmap is set up keep their tails read-only. Nothing does a lockless get on them before hugetlb takes them over.
Patching and mitigation
- Patch status
- Fix available (source: CVE record (Linux))
- Fix documented (source date)
- Mitigation
- None documented
- Vendor guidance
- No vendor link on record
Affected products
| Vendor | Product | Affected versions | Fixed in | Per |
|---|---|---|---|---|
| Linux | Linux | from 32c440d67e6cd96a715007d0e62eb970b0c49abc before e570f2c29fc07a1040c3736bf1a1dfa8065d5099 | e570f2c29fc07a1040c3736bf1a1dfa8065d5099 | CVE Program |
| Linux | Linux | from 32c440d67e6cd96a715007d0e62eb970b0c49abc before 0120a603afadba8ab4954bb1ce60f8195f80efc8 | 0120a603afadba8ab4954bb1ce60f8195f80efc8 | CVE Program |
| Linux | Linux | from 6.18.51 before 6.18.64 | 6.18.64 | CVE Program |
| Linux | Linux | from dba93b8cceb0048071389352fc29dcb0a7fe666f before f96a68bab26183433ea542dfb29ec41125f9fe39 | f96a68bab26183433ea542dfb29ec41125f9fe39 | CVE Program |
| Linux | Linux | 7.1 | Not documented | CVE Program |
| Linux | Linux (unaffected) | from 6.18.64 up to and including 6.18.* | Not documented | CVE Program |
| Linux | Linux (unaffected) | from 7.3.5 up to and including 7.3.* | Not documented | CVE Program |
| Linux | Linux (unaffected) | 7.4-rc2 | Not documented | CVE Program |
| Linux | Linux (unaffected) | all versions before 7.1 | Not documented | CVE Program |
Timeline
Disclosure
CVE record published by Linux
Source: CVE Program (opens in a new tab)
Patch
Fix documented by CVE record (Linux)