fs: don't allow to complete sync iocbs through aio_complete
authorChristoph Hellwig <hch@lst.de>
Wed, 11 Feb 2015 18:59:44 +0000 (19:59 +0100)
committerAl Viro <viro@zeniv.linux.org.uk>
Fri, 13 Mar 2015 16:10:22 +0000 (12:10 -0400)
commit599bd19bdc4c6b20fd91d50f2f79dececbaf80c1
tree782dec7d04dd789c841e3798ee4c920c9493fe2c
parent9d5722b7777e64de2d932f46cfee7765fdcc60d6
fs: don't allow to complete sync iocbs through aio_complete

The AIO interface is fairly complex because it tries to allow
filesystems to always work async and then wakeup a synchronous
caller through aio_complete.  It turns out that basically no one
was doing this to avoid the complexity and context switches,
and we've already fixed up the remaining users and can now
get rid of this case.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
fs/aio.c
fs/ecryptfs/file.c
fs/read_write.c
include/linux/aio.h
net/socket.c