ocfs2: prepare some interfaces used in append direct io
authorJoseph Qi <joseph.qi@huawei.com>
Mon, 16 Feb 2015 23:59:50 +0000 (15:59 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Tue, 17 Feb 2015 01:56:04 +0000 (17:56 -0800)
commit026749a86ebff68cb2accdcd29872d36ac148920
tree8fb74b121646906d19041ce09a627b85cdd7d427
parent6f9e2456c9f8904346958b6d9602a755372865b0
ocfs2: prepare some interfaces used in append direct io

Currently in case of append O_DIRECT write (block not allocated yet),
ocfs2 will fall back to buffered I/O.  This has some disadvantages.
Firstly, it is not the behavior as expected.  Secondly, it will consume
huge page cache, e.g.  in mass backup scenario.  Thirdly, modern
filesystems such as ext4 support this feature.

In this patch set, the direct I/O write doesn't fallback to buffer I/O
write any more because the allocate blocks are enabled in direct I/O now.

This patch (of 9):

Prepare some interfaces which will be used in append O_DIRECT write.

Signed-off-by: Joseph Qi <joseph.qi@huawei.com>
Cc: Weiwei Wang <wangww631@huawei.com>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Xuejiufei <xuejiufei@huawei.com>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Cc: alex chen <alex.chen@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
fs/ocfs2/file.c
fs/ocfs2/file.h