CVE-2026-97945

x86/mm: Fix user-space data loss with MADV_FREE and THP

N/A
Published: September 25, 2026 at 10:22 AM UTC
Updated: October 19, 2026 at 10:17 AM UTC
1 product affected across 1 vendorLinux Kernel

Using Linux by Linux?

Get instant alerts when new vulnerabilities affect Linux - before attackers find them.

Real-time alertsTrack your software stackFree plan available
Severity
N/A
Exploit prediction
EPSS 0.2%

Description

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.

Affected

Our own answer: ranges merged, affected separated from platform-only, with confidence and review flags.

6.6bb3aadf7d446aaf22c725b274e2c194ac5cb2111
Vendor states this is fixed in 55ddbc2ca6d5db64d44c3fb8f8705b4d39eb30b7 (fixed)
Vendor states this is fixed in a5ad97cbcfcec31430268441c10eedde73200177 (fixed)
Vendor states this is fixed in b5dec6dfb593d81fec15ba43c4fe5d58f10a49a2 (fixed)
Vendor states this is fixed in e4d569fe2f436a9b58c048fefabf56eab68c1b7e (fixed)
Vendor states this is fixed in f7491d7c81db0e7c304a7bd757a76d2fbeaff80e (fixed)
Vendor states this is fixed in 6.12.111 (unaffected)
Vendor states this is fixed in 6.18.53 (unaffected)
Vendor states this is fixed in 6.6 (unaffected)
Vendor states this is fixed in 6.6.158 (unaffected)
Vendor states this is fixed in 7.2.7 (unaffected)
Vendor states this is fixed in 7.3 (unaffected)
Vendor states this is fixed in 7.3-rc4 (unaffected)

Remediation

No guide yet

No remediation guidance has been written for this CVE yet, and no fixed version or vendor advisory is recorded against it.

Quick Info

StatusRECEIVED

Get alerted

Stop reading CVE feeds looking for your own stack

Track the products you run and VulnTracker tells you when a new CVE affects them — with what is known about fixing it, and nothing about the rest.

Free plan, no card