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: b9b4bb2 ("tmpfs: don't undo fallocate past its last page")
Cc: [email protected]
Signed-off-by: Hugh Dickins <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

Author: Hugh Dickins

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;
 		}