tmpfs: undo fallocation on failure
commit1635f6a74152f1dcd1b888231609d64875f0a81a
authorHugh Dickins <hughd@google.com>
Tue, 29 May 2012 22:06:42 +0000 (29 15:06 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Tue, 29 May 2012 23:22:23 +0000 (29 16:22 -0700)
treea2919520e0a8fead3930d23136ca30829f0fc64a
parente2d12e22c59ce714008aa5266d769f8568d74eac
tmpfs: undo fallocation on failure

In the previous episode, we left the already-fallocated pages attached to
the file when shmem_fallocate() fails part way through.

Now try to do better, by extending the earlier optimization of !Uptodate
pages (then always under page lock) to !Uptodate pages (outside of page
lock), representing fallocated pages.  And don't waste time clearing them
at the time of fallocate(), leave that until later if necessary.

Adapt shmem_truncate_range() to shmem_undo_range(), so that a failing
fallocate can recognize and remove precisely those !Uptodate allocations
which it added (and were not independently allocated by racing tasks).

But unless we start playing with swapfile.c and memcontrol.c too, once one
of our fallocated pages reaches shmem_writepage(), we do then have to
instantiate it as an ordinarily allocated page, before swapping out.  This
is unsatisfactory, but improved in the next episode.

Signed-off-by: Hugh Dickins <hughd@google.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Cong Wang <amwang@redhat.com>
Cc: Kay Sievers <kay@vrfy.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/shmem.c