powerpc/64s: __find_linux_pte() synchronization vs pmdp_invalidate()

The change to pmdp_invalidate() to mark the pmd with _PAGE_INVALID
broke the synchronisation against lock free lookups,
__find_linux_pte()'s pmd_none() check no longer returns true for such
cases.

Fix this by adding a check for this condition as well.

Fixes: da7ad366b4 ("powerpc/mm/book3s: Update pmd_present to look at _PAGE_PRESENT bit")
Cc: stable@vger.kernel.org # v4.20+
Suggested-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Reviewed-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This commit is contained in:
Nicholas Piggin 2019-06-07 13:56:36 +10:00 committed by Michael Ellerman
parent 33258a1db1
commit a00196a272

View File

@ -372,13 +372,25 @@ pte_t *__find_linux_pte(pgd_t *pgdir, unsigned long ea,
pdshift = PMD_SHIFT;
pmdp = pmd_offset(&pud, ea);
pmd = READ_ONCE(*pmdp);
/*
* A hugepage collapse is captured by pmd_none, because
* it mark the pmd none and do a hpte invalidate.
* A hugepage collapse is captured by this condition, see
* pmdp_collapse_flush.
*/
if (pmd_none(pmd))
return NULL;
#ifdef CONFIG_PPC_BOOK3S_64
/*
* A hugepage split is captured by this condition, see
* pmdp_invalidate.
*
* Huge page modification can be caught here too.
*/
if (pmd_is_serializing(pmd))
return NULL;
#endif
if (pmd_trans_huge(pmd) || pmd_devmap(pmd)) {
if (is_thp)
*is_thp = true;