quota: move unmount handling into the filesystem
authorChristoph Hellwig <hch@infradead.org>
Wed, 19 May 2010 11:16:42 +0000 (07:16 -0400)
committerJan Kara <jack@suse.cz>
Mon, 24 May 2010 12:09:12 +0000 (14:09 +0200)
commite0ccfd959cd8907bcb66cc2042e0f4fd7fcbff2b
treea62a84034243e75e07bb063526e10ea0b837062f
parent0f0dd62fddcbd0f6830ed8ef3d3426ccc46b9250
quota: move unmount handling into the filesystem

Currently the VFS calls into the quotactl interface for unmounting
filesystems.  This means filesystems with their own quota handling
can't easily distinguish between user-space originating quotaoff
and an unount.  Instead move the responsibily of the unmount handling
into the filesystem to be consistent with all other dquot handling.

Note that we do call dquot_disable a lot later now, e.g. after
a sync_filesystem.  But this is fine as the quota code does all its
writes via blockdev's mapping and that is synced even later.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jan Kara <jack@suse.cz>
fs/ext2/super.c
fs/ext3/super.c
fs/ext4/super.c
fs/jfs/super.c
fs/reiserfs/super.c
fs/super.c
fs/udf/super.c
fs/ufs/super.c
include/linux/quotaops.h