summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorHugh Dickins <hughd@google.com>2016-07-10 16:46:32 -0700
committerLinus Torvalds <torvalds@linux-foundation.org>2016-07-10 20:08:44 -0700
commit7f556567036cb7f89aabe2f0954b08566b4efb53 (patch)
tree020fc68e31c9361c49786606a819dc5be9a2a8a6
parent617a8d6bc19edd075e8111c6770f79cae75be51f (diff)
tmpfs: fix regression hang in fallocate undo
The well-spotted fallocate undo fix is good in most cases, but not when fallocate failed on the very first page. index 0 then passes lend -1 to shmem_undo_range(), and that has two bad effects: (a) that it will undo every fallocation throughout the file, unrestricted by the current range; but more importantly (b) it can cause the undo to hang, because lend -1 is treated as truncation, which makes it keep on retrying until every page has gone, but those already fully instantiated will never go away. Big thank you to xfstests generic/269 which demonstrates this. Fixes: b9b4bb26af01 ("tmpfs: don't undo fallocate past its last page") Cc: stable@vger.kernel.org Signed-off-by: Hugh Dickins <hughd@google.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
-rw-r--r--mm/shmem.c8
1 files changed, 5 insertions, 3 deletions
diff --git a/mm/shmem.c b/mm/shmem.c
index 24463b67b6ef..171dee7a131f 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -2225,9 +2225,11 @@ static long shmem_fallocate(struct file *file, int mode, loff_t offset,
error = shmem_getpage(inode, index, &page, SGP_FALLOC);
if (error) {
/* Remove the !PageUptodate pages we added */
- shmem_undo_range(inode,
- (loff_t)start << PAGE_SHIFT,
- ((loff_t)index << PAGE_SHIFT) - 1, true);
+ if (index > start) {
+ shmem_undo_range(inode,
+ (loff_t)start << PAGE_SHIFT,
+ ((loff_t)index << PAGE_SHIFT) - 1, true);
+ }
goto undone;
}