On Sat 20-08-22 23:14:37, Christoph Hellwig wrote: > On Tue, Aug 16, 2022 at 01:21:24PM +0200, Jan Kara wrote: > > 2) I_DIRTY_TIME flag passed to ->dirty_inode() callback. This is admittedly > > bit of a hack. Currently XFS relies on the fact that the only time its > > ->dirty_inode() callback needs to do anything is when VFS decides it is > > time to writeback timestamps and XFS detects this situation by checking for > > I_DIRTY_TIME in inode->i_state. Now to fix the race, we need to first clear > > I_DIRTY_TIME in inode->i_state and only then call the ->dirty_inode() > > callback (otherwise timestamp update can get lost). So the solution I've > > suggested was to propagate the information "timestamp update needed" to XFS > > through I_DIRTY_TIME in flags passed to ->dirty_inode(). > > Maybe we should just add a separate update_lazy_time method to make this > a little more clear? Yes, we could do that if people prefer this. Although I'd say that good documentation at the place in __mark_inode_dirty() where this gets used and in documentation of .dirty_inode might clear the confusion as well. Honza -- Jan Kara <jack@xxxxxxxx> SUSE Labs, CR
