On Wed 24-09-14 10:13:06, Christoph Hellwig wrote: > On Wed, Sep 24, 2014 at 07:10:20PM +0200, Jan Kara wrote: > > Yeah, I'm aware of that but I decided I won't bother Li Xi with that > > since it's independent issue and 8 bytes aren't that terrible. > > For the inode of which we have so many instances it is. Especially when > only one filesystem can make use of this new field. > > > Personally > > I somewhat prefer what I did in > > http://www.spinics.net/lists/linux-fsdevel/msg74927.html > > where we don't introduce additional method but rather a table with field > > offsets in superblock. If people agree with this, I can cook up a patch for > > quota relatively quickly. > > There's an even better way. We stopped calling the dquot_* functions > from generic code, so now the filesystem can simply pass in a pointer > to the dquot array. I thought that as well before I tried that :) There are situations like when turning quotas on & off where we iterate over all inodes and need to get dquots from the inode... I don't see how to handle those situations without fs callback / table of offsets. Honza -- Jan Kara <jack@xxxxxxx> SUSE Labs, CR -- To unsubscribe from this list: send the line "unsubscribe linux-api" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
