Skip to contentSkip to conversation

x86-mm-fix-pmd_modify-dropping-the-dirty-bit.patch removed from -mm tree

merged mm-hotfixes-stable Andrew Morton 2 participants 3 messages Sep 18 Stable branch

lore
Share & tools
lore

Andrew Morton

author tree notice Applied
#0 lore raw

The quilt patch titled
Subject: x86/mm: fix pmd_modify() dropping the dirty bit
has been removed from the -mm tree. Its filename was
x86-mm-fix-pmd_modify-dropping-the-dirty-bit.patch

This patch was dropped because it was merged into the mm-hotfixes-stable branch of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

------------------------------------------------------
From: Vernon Yang <[email protected]>
Subject: x86/mm: fix pmd_modify() dropping the dirty bit
Date: Thu, 3 Sep 2026 11:16:08 +0800

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.

(Pedro [1]: "FYI we just had a customer case for our downstream kernel for this exact same issue. This fixed it beautifully".)

---

 arch/x86/include/asm/pgtable.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/include/asm/pgtable.h renamed from arch/x86/include/asm/pgtable.h~x86-mm-fix-pmd_modify-dropping-the-dirty-bit +1 -1
@@ -806,7 +806,7 @@ static inline pmd_t pmd_modify(pmd_t pmd
 	pmdval_t val = pmd_val(pmd), oldval = val;
 	pmd_t pmd_result;
 
-	val &= (_HPAGE_CHG_MASK & ~_PAGE_DIRTY);
+	val &= _HPAGE_CHG_MASK;
 	val |= check_pgprot(newprot) & ~_HPAGE_CHG_MASK;
 	val = flip_protnone_guard(oldval, val, PHYSICAL_PMD_PAGE_MASK);
 

_

Patches currently in -mm which might be from [email protected] are

mm-khugepaged-fix-swap-entry-value-to-folio_pfn.patch mm-khugepaged-fix-folio-is-used-after-pte_unmap_unlock.patch mm-khugepaged-fix-folio-is-used-after-folio_put-unlock.patch

#1 lore raw

Hi Andrew,

Quoted 8 lines · Andrew Morton

The quilt patch titled
Subject: x86/mm: fix pmd_modify() dropping the dirty bit
has been removed from the -mm tree. Its filename was
x86-mm-fix-pmd_modify-dropping-the-dirty-bit.patch

This patch was dropped because it was merged into the mm-hotfixes-stable branch of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

You might want to drop this commit. It was already merged in 704340f1cd0dcef829eb62f5b48ae95a2ce17bdf from the tip tree.

(Linus tree f7491d7c81db0e7c304a7bd757a76d2fbeaff80e)

Signature · 1 line
Pedro
#2 lore raw
Quoted 9 lines · On Fri, 18 Sep 2026 12:40:32 +0100 Pedro Falcato <[email protected]> wrote:

Hi Andrew,

Quoted 6 lines · Andrew Morton

The quilt patch titled
Subject: x86/mm: fix pmd_modify() dropping the dirty bit
has been removed from the -mm tree. Its filename was
x86-mm-fix-pmd_modify-dropping-the-dirty-bit.patch

Quoted 7 lines
Quoted 2 lines

This patch was dropped because it was merged into the mm-hotfixes-stable branch of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

You might want to drop this commit. It was already merged in 704340f1cd0dcef829eb62f5b48ae95a2ce17bdf from the tip tree.

(Linus tree f7491d7c81db0e7c304a7bd757a76d2fbeaff80e)

Thanks. Done.