x86/mm: Fix user-space data loss with MADV_FREE and THP
In the Linux kernel, the following vulnerability has been resolved: x86/mm: Fix user-space data loss with MADV_FREE and THP Some of users of Polars (a data analytics library) have lost production data from this bug. They seem to have just the right combination of huge pages, MADV_FREE and heavy reclaim pressure. pmd_modify() masks the old value with (_HPAGE_CHG_MASK & ~_PAGE_DIRTY), silently discarding the hardware dirty bit. The subsequent pmd_mksaveddirty() call is supposed to transfer _PAGE_DIRTY into _PAGE_SAVED_DIRTY when write-protecting, but the dirty bit was already stripped from the value, so there is nothing left to transfer. Contrast with pte_modify(), which keeps _PAGE_DIRTY_BITS in its mask, and pud_modify(), which keeps _HPAGE_CHG_MASK untouched: pmd_modify() is the odd one out. Any pmd_modify() on a writable, dirty PMD loses the dirty state. One visible consequence is data loss with MADV_FREE on PMD-mapped THP: memset(buf, 0x5A, size); // PMD-mapped THP, PMD dirty madvise(buf, size, MADV_FREE); // PMD cleaned but left writable, // folio marked lazyfree memset(buf, 0x5A, size); // hardware sets _PAGE_DIRTY again mprotect(buf, size, PROT_READ); // pmd_modify() drops the dirty bit mprotect(buf, size, PROT_READ|PROT_WRITE); // ... memory pressure ... Reclaim (e.g. under memcg pressure) then finds the lazyfree folio with no dirty bit set anywhere and frees it in __discard_anon_folio_pmd_locked(), even though the data was rewritten after MADV_FREE; subsequent reads fault in fresh zero pages. NUMA hinting alone can trigger the same loss, as do_huge_pmd_numa_page() restores the PMD through pmd_modify() as well. PMD-mapped file THPs are affected too: mprotect()/NUMA hinting dropping the dirty bit means rewritten data is never written back. Fix it by keeping _PAGE_DIRTY in the preserved mask, exactly like pte_modify() and pud_modify() do. The existing pmd_mksaveddirty()/pmd_clear_saveddirty() pair then performs the hardware-dirty <-> saved-dirty transition based on the write bit, preserving the shadow-stack encoding rules.
- Published
- Sep 25, 2026
- Updated
- Sep 25, 2026
- EPSS
- 0.2% · 9th percentile
- CWE
- —
Sign in to follow · You’ll be notified if a followed record enters KEV, gets an exploit or is updated.
Report tools
Action score
0
Monitor
Low priority for now.
- CVSS
- 0 / 40 · —
- CISA KEV
- 0 / 30 · Not listed
- EPSS
- 0 / 30 · 0.2%
Affected systems
—
Versions reported by the vendor
Affected version ranges reported by the assigning authority (Linux). Independent of NVD's CPE analysis and usually ahead of it.
Linux Linux
- 6.6affected
- bb3aadf7d446aaf22c725b274e2c194ac5cb2111 and later · before e4d569fe2f436a9b58c048fefabf56eab68c1b7eaffected · git
- bb3aadf7d446aaf22c725b274e2c194ac5cb2111 and later · before 55ddbc2ca6d5db64d44c3fb8f8705b4d39eb30b7affected · git
- bb3aadf7d446aaf22c725b274e2c194ac5cb2111 and later · before a5ad97cbcfcec31430268441c10eedde73200177affected · git
- bb3aadf7d446aaf22c725b274e2c194ac5cb2111 and later · before f7491d7c81db0e7c304a7bd757a76d2fbeaff80eaffected · git
- before 6.6not affected · semver
- 6.12.111 and later · up to and including 6.12.*not affected · semver
- 6.18.53 and later · up to and including 6.18.*not affected · semver
- 7.2.7 and later · up to and including 7.2.*not affected · semver
- 7.3-rc4 and laternot affected · original_commit_for_fix
Package-level exposure
OSV and GitHub Advisory data: ecosystem, package and range. SBOM matching uses this table.
| Ecosystem | Package | Affected range | Fix |
|---|---|---|---|
| Debian:13 | linux | before 6.12.111-1 | 6.12.111-1 |
| Debian:14 | linux | all versions | — |
Same product
linux: all recordsOther highest-scoring records for the same primary product.
- CVE-2026-74705udp: fix potential use-after-free in tunnel segmentation40Plan
- CVE-2026-74612veth: fix skb length accounting after XDP frag adjustment40Plan
- CVE-2026-74475vxlan: use neigh_ha_snapshot() in route_shortcircuit()40Plan
- CVE-2026-74309vdpa/octeon_ep: fix IRQ-to-ring mapping in interrupt handler40Plan
- CVE-2026-74280crypto: marvell/octeontx - fix DMA cleanup using wrong loop index40Plan
- CVE-2026-74279crypto: cavium/cpt - fix DMA cleanup using wrong loop index40Plan
Remediation
Which version to upgrade to
Fix versions compiled from the vendor, package registries and Microsoft. Verify the vendor's note before upgrading.
| Product / package | Fixed version | Source |
|---|---|---|
| Linux Linux | 55ddbc2ca6d5db64d44c3fb8f8705b4d39eb30b7 | Vendor (CNA) |
| Linux Linux | a5ad97cbcfcec31430268441c10eedde73200177 | Vendor (CNA) |
| Linux Linux | e4d569fe2f436a9b58c048fefabf56eab68c1b7e | Vendor (CNA) |
| Linux Linux | f7491d7c81db0e7c304a7bd757a76d2fbeaff80e | Vendor (CNA) |
| azl3 kernel 6.6.157.1-1 on Azure Linux 3.0 | Release Notes | Microsoft (MSRC) |
| debian:linux | 6.12.111-1 · Debian:13 | Package registry (OSV) |
Exploit status
No known public exploit
No public exploit has been observed yet. That doesn't mean you're safe, only that the bar is a little higher.
Research context
For pentesters and researchers: attack profile, score disagreement, timeline, patch commits, credits, variant and chain candidates, bug bounty scope. All derived from existing data; no exploit code.
Timeline
From publication to today: proof of concept, Metasploit module, CISA KEV and fix record. Dates are as reported by the sources.
No dated events beyond publication.
FIRST EPSS daily score; only changes of 0.01 or more are recorded (step chart).
Patch and commit links
Commit, PR and diff links among the references. A starting point for patch-diffing and variant hunting; fixes, not exploits.
No commit or PR link among the references.
Credits
All researchersFinders, reporters and analysts named in the CNA record. Click a name for that researcher’s other records.
No credits in the CNA record.
Variant candidates
Same product, same weakness class, within 18 months. If the patch missed the root cause, the sibling bug is here.
No nightly-computed relations.
Chain candidates
An authentication bypass and a privilege-requiring bug in the same product, published close together: combined they may become an unauthenticated path.
—
Bug bounty scope
No known public program.
Source: bounty-targets-data (public HackerOne, Bugcrowd, Intigriti, YesWeHack listings).
Technical details
No CVSS vector for this record, so attack conditions can't be derived.
Weakness class (CWE)
—
Attack context
MITRE CAPEC attack patterns and ATT&CK techniques for this weakness class (CWE). A starting point for detection rules and threat hunting.
MITRE has no CAPEC/ATT&CK mapping for this CWE.
Noroxi analysis
No Noroxi analysis for this record yet
We don't hand-write analysis for all 385,000+ vulnerabilities; that wouldn't be honest. For notable, high-impact vulnerabilities our team writes the mechanism, detection and remediation steps.
We use this product, ask for helpChange log
- Fix✗ → ✓
For records you follow, these changes also arrive as notifications. →
References
- git.kernel.org/stable/c/55ddbc2ca6d5db64d44c3fb8f8705b4d39eb30b7
- git.kernel.org/stable/c/a5ad97cbcfcec31430268441c10eedde73200177
- git.kernel.org/stable/c/e4d569fe2f436a9b58c048fefabf56eab68c1b7e
- git.kernel.org/stable/c/f7491d7c81db0e7c304a7bd757a76d2fbeaff80e
Vendor advisories and official records. Exploit/PoC links are deliberately left out.