[Cluster-devel] [PATCH 1/4] gfs2: fix bogus __mark_inode_dirty(I_DIRTY_SYNC | I_DIRTY_DATASYNC) calls

I_DIRTY_DATASYNC is a strict superset of I_DIRTY_SYNC semantics, as in mark dirty to be written out by fdatasync as well. So dirtying for both flags makes no sense. Signed-off-by: Christoph Hellwig --- fs/gfs2/xattr.c | 8 1 file changed, 4 insertions(+), 4 deletions(-)

[Cluster-devel] [PATCH 3/4] ubifs: fix bogus __mark_inode_dirty(I_DIRTY_SYNC | I_DIRTY_DATASYNC) call

I_DIRTY_DATASYNC is a strict superset of I_DIRTY_SYNC semantics, as in mark dirty to be written out by fdatasync as well. So dirtying for both flags makes no sense. Signed-off-by: Christoph Hellwig --- fs/ubifs/file.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff

[Cluster-devel] [PATCH 4/4] fs: move I_DIRTY_INODE to fs.h

And use it in a few more places rather than opencoding the values. Signed-off-by: Christoph Hellwig --- fs/ext4/inode.c| 4 ++-- fs/fs-writeback.c | 9 +++-- fs/gfs2/super.c| 2 +- include/linux/fs.h | 3 ++- 4 files changed, 8 insertions(+), 10 deletions(-) diff

Re: [Cluster-devel] [PATCH] gfs2: Fsync parent directories

On Wed, Feb 21, 2018 at 08:51:15AM +1100, Dave Chinner wrote: > IOWs, if the filesystem is designed with strictly ordered metadata, > then fsync()ing a new file also implies that all references to the > new file are also on stable storage because they happened before the > fsync on the file was

[Cluster-devel] a few I_DIRTY* related cleanups

Move I_DIRTY_INODE to fs.h, and use the I_DIRTY* defines more consistently.

[Cluster-devel] [PATCH 2/4] ntfs: fix bogus __mark_inode_dirty(I_DIRTY_SYNC | I_DIRTY_DATASYNC) call

I_DIRTY_DATASYNC is a strict superset of I_DIRTY_SYNC semantics, as in mark dirty to be written out by fdatasync as well. So dirtying for both flags makes no sense. Signed-off-by: Christoph Hellwig --- fs/ntfs/mft.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff

Re: [Cluster-devel] a few I_DIRTY* related cleanups

On 21/02/18 15:54, Christoph Hellwig wrote: Move I_DIRTY_INODE to fs.h, and use the I_DIRTY* defines more consistently. Thanks - looks good to me: Acked-by: Steven Whitehouse Steve.