From: Christoph Lameter It is better to redo the complete fault if do_swap_page() finds that the page is not in PageSwapCache() because the page migration code may have replaced the swap pte already with a pte pointing to valid memory. do_swap_page() may interpret an invalid swap entry without this patch because we do not reload the pte if we are looping back. The page migration code may already have reused the swap entry referenced by our local swp_entry. Signed-off-by: Christoph Lameter Cc: Hugh Dickins Signed-off-by: Andrew Morton --- mm/memory.c | 7 ------- 1 files changed, 7 deletions(-) diff -puN mm/memory.c~page-migration-make-do_swap_page-redo-the-fault mm/memory.c --- devel/mm/memory.c~page-migration-make-do_swap_page-redo-the-fault 2006-04-10 21:52:07.000000000 -0700 +++ devel-akpm/mm/memory.c 2006-04-10 21:52:10.000000000 -0700 @@ -1879,7 +1879,6 @@ static int do_swap_page(struct mm_struct goto out; entry = pte_to_swp_entry(orig_pte); -again: page = lookup_swap_cache(entry); if (!page) { swapin_readahead(entry, address, vma); @@ -1903,12 +1902,6 @@ again: mark_page_accessed(page); lock_page(page); - if (!PageSwapCache(page)) { - /* Page migration has occured */ - unlock_page(page); - page_cache_release(page); - goto again; - } /* * Back out if somebody else already faulted in this pte. _