From: Hugh Dickins <hugh@veritas.com>
Marcelo noticed that the BUG_ON in __set_page_dirty_nobuffers doesn't make
much sense: it lost its way in 2.6.7, amidst so many page_mappings!
It's supposed to be checking that, although page->mapping may suddenly go NULL
from truncation, and although tmpfs swizzles page_mapping(page) between tmpfs
inode address_space and swapper_space, there's sufficient stabilization while
here in __set_page_dirty_nobuffers that the mapping after we locked
mapping->tree_lock is the same as the mapping before we locked
mapping->tree_lock i.e. the lock we hold is the right one.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
25-akpm/mm/page-writeback.c | 7 ++++---
1 files changed, 4 insertions(+), 3 deletions(-)
diff -puN mm/page-writeback.c~__set_page_dirty_nobuffers-mappings mm/page-writeback.c
--- 25/mm/page-writeback.c~__set_page_dirty_nobuffers-mappings 2004-09-05 21:26:02.907820280 -0700
+++ 25-akpm/mm/page-writeback.c 2004-09-05 21:26:02.911819672 -0700
@@ -582,12 +582,13 @@ int __set_page_dirty_nobuffers(struct pa
if (!TestSetPageDirty(page)) {
struct address_space *mapping = page_mapping(page);
+ struct address_space *mapping2;
if (mapping) {
spin_lock_irq(&mapping->tree_lock);
- mapping = page_mapping(page);
- if (page_mapping(page)) { /* Race with truncate? */
- BUG_ON(page_mapping(page) != mapping);
+ mapping2 = page_mapping(page);
+ if (mapping2) { /* Race with truncate? */
+ BUG_ON(mapping2 != mapping);
if (!mapping->backing_dev_info->memory_backed)
inc_page_state(nr_dirty);
radix_tree_tag_set(&mapping->page_tree,
_