Hello, Does this patch make sense? On Mon, Aug 7, 2017 at 11:46 AM, <nate@xxxxxxxxxx> wrote: > From: Vitaly Zolotusky <vitaly@xxxxxxxxx> > > If the IOCB_DSYNC flag is set a sync is not being performed by > fuse_file_write_iter. Honor IOCB_DSYNC by invoking generic_write_sync > when a write has been successful. generic_write_sync appropriatly syncs > the data which was written when IOCB_DSYNC flag is set. > > Signed-off-by: Vitaly Zolotusky <vitaly@xxxxxxxxx> > Signed-off-by: Nate Clark <nate@xxxxxxxxxx> > --- > fs/fuse/file.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/fs/fuse/file.c b/fs/fuse/file.c > index 3ee4fdc..5e0a39e 100644 > --- a/fs/fuse/file.c > +++ b/fs/fuse/file.c > @@ -1227,6 +1227,8 @@ static ssize_t fuse_file_write_iter(struct kiocb *iocb, struct iov_iter *from) > out: > current->backing_dev_info = NULL; > inode_unlock(inode); > + if (written > 0) > + written = generic_write_sync(iocb, written); > > return written ? written : err; > } > -- > 2.9.4 > -nate
