When memory_failure() is called for a already hwpoisoned pfn, kill_accessing_process() will be called to kill current task. However, if the vma of the accessing vaddr is VM_PFNMAP, walk_page_range() will skip the vma in walk_page_test() and return 0. Before commit aaf99ac2ceb7 ("mm/hwpoison: do not send SIGBUS to processes with recovered clean pages"), kill_accessing_process() will return EFAULT. For x86, the current task will be killed in kill_me_maybe(). However, after this commit, kill_accessing_process() simplies return 0, that means UCE is handled properly, but it doesn't actually. In such case, the user task will trigger UCE infinitely. To fix it, add .test_walk callback for hwpoison_walk_ops to scan all vmas. Fixes: aaf99ac2ceb7 ("mm/hwpoison: do not send SIGBUS to processes with recovered clean pages") Acked-by: David Hildenbrand Acked-by: Miaohe Lin Reviewed-by: Jane Chu Cc: Signed-off-by: Jinjiang Tu --- Changelog since v2: * add a comment and cc stable Changelog since v1: * update patch description, suggested by David Hildenbrand mm/memory-failure.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/mm/memory-failure.c b/mm/memory-failure.c index e2e685b971bb..fc30ca4804bf 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -853,9 +853,17 @@ static int hwpoison_hugetlb_range(pte_t *ptep, unsigned long hmask, #define hwpoison_hugetlb_range NULL #endif +static int hwpoison_test_walk(unsigned long start, unsigned long end, + struct mm_walk *walk) +{ + /* We also want to consider pages mapped into VM_PFNMAP. */ + return 0; +} + static const struct mm_walk_ops hwpoison_walk_ops = { .pmd_entry = hwpoison_pte_range, .hugetlb_entry = hwpoison_hugetlb_range, + .test_walk = hwpoison_test_walk, .walk_lock = PGWALK_RDLOCK, }; -- 2.43.0