On Wed, 2019-10-16 at 23:51 -0700, Christoph Hellwig wrote: > On Wed, Oct 16, 2019 at 09:53:30PM -0700, Darrick J. Wong wrote: > > > The problem is that this will probably be used in logging later > > > and > > > there's a lot of logging that uses the upper case variant. > > > > > > OTOH if all the log messages were changed to use lower case "xfs" > > > then > > > one of the problems I see with logging (that name inconsistency) > > > would > > > go away. > > > > > > So I'm not sure what I should do here. > > > > I would just leave it 'XFS' for consistency, but I might be in the > > back > > pocket of Big Letter. ;) > > I isn't really used for much, and Al already has a patch from Eric in > on > of his trees that kills the field in favor of the file_system_type > name > field: > > https://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs.git/commit/?h=work.mount-parser-later&id=543fdf1d617edd6d681fcc50e16478e832a7a2ac Yes, I was thinking about that when I replied. > > So we should not spell them differently if we can. I don't think it will make much difference in the long run, I'll just change it to lower case for the time being. I think the implications of the mount api on logging need to be discussed further (see my thoughts in the series cover letter), there's not just a log entry case issue. I plan on making some RFC patches for this later. Since I'm doing the mount api change for xfs I'll base my RFC patches on xfs and they will need to go to a wider audience (fsdevel as well I expect). Ian
