PT-2026-152544
CVE-2026-99664
·
Publicado
2026-11-20
·
Atualizado
2026-11-20
Descrição por
Pontuação
N/A
Não disponívelNão há classificações de severidade ou métricas disponíveis. Quando houver, atualizaremos as informações correspondentes na página.
In the Linux kernel, the following vulnerability has been resolved:
mm/gup: recheck the PUD under PTL in follow pud mask()
follow pud mask() reads the PUD without a lock. If it is a leaf, it takes
the lock and calls follow huge pud(), which reads the entry again, but
only tests whether it is present.
A leaf can be zapped concurrently and a page fault can fill the empty
entry with a page table.
This currently only applies when device-dax makes use of PUD entries,
and the fault after e.g., MADV DONTNEED would find THPs to be disabled.
That requires a sequence like
- Fault in the PUD
- Disable THP
- Have GUP race against concurrent MADV DONTNEED + refault
Which is rather unlikely to happen in practice, which is why we never
received a report about such a problem.
Before commit 1b1676180246 ("mm/gup: handle huge pud for
follow pud mask()"), we performed that check in follow devmap pud() by
testing for pud devmap().
Let's fix it by just structuring follow pud mask() similar to
follow pmd mask(), minus protnone stuff that does not apply and minus
FOLL SPLIT PMD handling. That makes it much easier to spot any such
inconsistencies.
Perform the pud bad() check early in follow pmd mask() and just forward
the PUD value. We can drop the pud present() check in follow huge pud(),
just like we have in follow huge pmd().
Even though we are performing more changes than strictly necessary, the
fix should be easy to backport.
Encontrou algum problema na descrição? Tem algo a acrescentar? Fique à vontade para nos escrever 👾