fix uniprocessor lockup in page reclaim
Subsystems (3)
PATCH 1/1 fix uniprocessor lockup in page reclaim
I have a test_and_set_bit(PG_chainlock, page->flags) in page reclaim. Which works fine on SMP. But on uniprocessor, we made pte_chain_unlock() a no-op, so all pages end up with PG_chainlock set. refill_inactive() cannot move any pages onto the inactive list and the machine dies.
The patch removes the test_and_set_bit optimisation in there and just uses pte_chain_lock(). If we want that (dubious) optimisation back then let's do it right and create pte_chain_trylock().
Patch is against recent 2.5.31+BK. Please include for 2.5.32.
mm/vmscan.c renamed from mm/vmscan.c~pte-chain-fix +1 -4
@@ -398,10 +398,7 @@ static /* inline */ void refill_inactive page = list_entry(l_hold.prev, struct page, lru);list_del (&page->lru);if (page->pte.chain) { -if (test_and_set_bit (PG_chainlock, &page->flags)) { -list_add (&page->lru, &l_active); -continue ; - } +pte_chain_lock (page);if (page->pte.chain &&page_referenced (page)) {pte_chain_unlock (page);list_add (&page->lru, &l_active);
.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/