mm: vm_page_prot: update with WRITE_ONCE/READ_ONCE

vma->vm_page_prot is read lockless from the rmap_walk, it may be updated
concurrently and this prevents the risk of reading intermediate values.

Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Andrea Arcangeli <[email protected]>
Cc: Rik van Riel <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Jan Vorlicek <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
This commit is contained in:
Andrea Arcangeli 2016-10-07 17:01:22 -07:00 • committed by Linus Torvalds
commit 6d2329f887
5 changed files with 13 additions and 11 deletions

View file

@ -234,7 +234,7 @@ static int remove_migration_pte(struct page *new, struct vm_area_struct *vma,
goto unlock;
get_page(new);
pte = pte_mkold(mk_pte(new, vma->vm_page_prot));
pte = pte_mkold(mk_pte(new, READ_ONCE(vma->vm_page_prot)));
if (pte_swp_soft_dirty(*ptep))
pte = pte_mksoft_dirty(pte);