Hello,
as find_lock_page() already checks with TestSetPageLocked() that page is
locked, there is no need to call lock_page() that will try-lock page
again (chances of page being unlocked in between are small). Call
__lock_page() directly, this saves one atomic operation.
Also, mark truncate-while-slept path as unlikely while we are here.
Andrew, please apply.
Nikita.
Signed-off-by: Nikita Danilov <[email protected]>
mm/filemap.c | 13 +++++++------
1 files changed, 7 insertions(+), 6 deletions(-)
diff -puN mm/filemap.c~find_lock_page-use-__lock_page mm/filemap.c
--- git-linux/mm/filemap.c~find_lock_page-use-__lock_page 2005-11-20 16:33:58.000000000 +0300
+++ git-linux-nikita/mm/filemap.c 2005-11-20 16:41:46.000000000 +0300
@@ -555,11 +555,12 @@ repeat:
page_cache_get(page);
if (TestSetPageLocked(page)) {
read_unlock_irq(&mapping->tree_lock);
- lock_page(page);
+ __lock_page(page);
read_lock_irq(&mapping->tree_lock);
/* Has the page been truncated while we slept? */
- if (page->mapping != mapping || page->index != offset) {
+ if (unlikely(page->mapping != mapping ||
+ page->index != offset)) {
unlock_page(page);
page_cache_release(page);
goto repeat;
_