fs: convert a pile of fsync routines to errseq_t based reporting
authorJeff Layton <jlayton@redhat.com>
Fri, 7 Jul 2017 19:20:52 +0000 (15:20 -0400)
committerJeff Layton <jlayton@redhat.com>
Tue, 1 Aug 2017 12:39:29 +0000 (08:39 -0400)
commit3b49c9a1e984b524142afc7536041d8c66877113
tree350dabeb53253f0fabbcd3984fadd8454c1be04a
parentd07a6ac7b6f878c1078b75181cdae060daac5820
fs: convert a pile of fsync routines to errseq_t based reporting

This patch converts most of the in-kernel filesystems that do writeback
out of the pagecache to report errors using the errseq_t-based
infrastructure that was recently added. This allows them to report
errors once for each open file description.

Most filesystems have a fairly straightforward fsync operation. They
call filemap_write_and_wait_range to write back all of the data and
wait on it, and then (sometimes) sync out the metadata.

For those filesystems this is a straightforward conversion from calling
filemap_write_and_wait_range in their fsync operation to calling
file_write_and_wait_range.

Acked-by: Jan Kara <jack@suse.cz>
Acked-by: Dave Kleikamp <dave.kleikamp@oracle.com>
Signed-off-by: Jeff Layton <jlayton@redhat.com>
22 files changed:
arch/powerpc/platforms/cell/spufs/file.c
drivers/staging/lustre/lustre/llite/file.c
drivers/video/fbdev/core/fb_defio.c
fs/9p/vfs_file.c
fs/affs/file.c
fs/afs/write.c
fs/cifs/file.c
fs/exofs/file.c
fs/f2fs/file.c
fs/hfs/inode.c
fs/hfsplus/inode.c
fs/hostfs/hostfs_kern.c
fs/hpfs/file.c
fs/jffs2/file.c
fs/jfs/file.c
fs/ncpfs/file.c
fs/ntfs/dir.c
fs/ntfs/file.c
fs/ocfs2/file.c
fs/reiserfs/dir.c
fs/reiserfs/file.c
fs/ubifs/file.c