THP split makes non-atomic change of tail page flags. This is almost ok
because tail pages are locked and isolated but this breaks recent changes
in page locking: non-atomic operation could clear bit PG_waiters.
As a result concurrent sequence get_page_unless_zero() -> lock_page()
might block forever. Especially if this page was truncated later.
Fix is trivial: clone flags before unfreezing page reference counter.
This race exists since commit 62906027091f ("mm: add PageWaiters indicating
tasks are waiting for a page bit") while unsave unfreeze itself was added
in commit 8df651c7059e ("thp: cleanup split_huge_page()").
clear_compound_head() also must be called before unfreezing page reference
because after successful get_page_unless_zero() might follow put_page()
which needs correct compound_head().
And replace page_ref_inc()/page_ref_add() with page_ref_unfreeze() which
is made especially for that and has semantic of smp_store_release().
Signed-off-by: Konstantin Khlebnikov
---
mm/huge_memory.c | 36 +++-
1 file changed, 15 insertions(+), 21 deletions(-)
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 87ab9b8f56b5..7a8ead256b2f 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -2355,26 +2355,13 @@ static void __split_huge_page_tail(struct page *head,
int tail,
struct page *page_tail = head + tail;
VM_BUG_ON_PAGE(atomic_read(&page_tail->_mapcount) != -1, page_tail);
- VM_BUG_ON_PAGE(page_ref_count(page_tail) != 0, page_tail);
/*
-* tail_page->_refcount is zero and not changing from under us. But
-* get_page_unless_zero() may be running from under us on the
-* tail_page. If we used atomic_set() below instead of atomic_inc() or
-* atomic_add(), we would then run atomic_set() concurrently with
-* get_page_unless_zero(), and atomic_set() is implemented in C not
-* using locked ops. spin_unlock on x86 sometime uses locked ops
-* because of PPro errata 66, 92, so unless somebody can guarantee
-* atomic_set() here would be safe on all archs (and not only on x86),
-* it's safer to use atomic_inc()/atomic_add().
+* Clone page flags before unfreezing refcount.
+*
+* After successful get_page_unless_zero() might follow flags change,
+* for exmaple lock_page() which set PG_waiters.
*/
- if (PageAnon(head) && !PageSwapCache(head)) {
- page_ref_inc(page_tail);
- } else {
- /* Additional pin to radix tree */
- page_ref_add(page_tail, 2);
- }
-
page_tail->flags &= ~PAGE_FLAGS_CHECK_AT_PREP;
page_tail->flags |= (head->flags &
((1L << PG_referenced) |
@@ -2387,14 +2374,21 @@ static void __split_huge_page_tail(struct page *head,
int tail,
(1L << PG_unevictable) |
(1L << PG_dirty)));
- /*
-* After clearing PageTail the gup refcount can be released.
-* Page flags also must be visible before we make the page non-compound.
-*/
+ /* Page flags must be visible before we make the page non-compound. */
smp_wmb();
+ /*
+* Clear PageTail before unfreezing page refcount.
+*
+* After successful get_page_unless_zero() might follow put_page()
+* which needs correct compound_head().
+*/
clear_compound_head(page_tail);
+ /* Finally unfreeze refcount. Additional reference from page cache. */
+ page_ref_unfreeze(page_tail, 1 + (!PageAnon(head) ||
+ PageSwapCache(head)));
+
if (page_is_young(head))
set_page_young(page_tail);
if (page_is_idle(head))