On Thu, Jan 04, 2024 at 07:24:28AM +0100, Christoph Hellwig wrote: > On Wed, Jan 03, 2024 at 04:03:24PM -0800, Darrick J. Wong wrote: > > > + /* > > > + * We don't want to bother with kmapping data during repair, so don't > > > + * allow highmem pages to back this mapping. > > > + */ > > > + mapping_set_gfp_mask(inode->i_mapping, GFP_HIGHUSER); > > > > Gonna be fun to see what happens on 32-bit. ;) > > 32-bit with highmem, yes. I suspect we should just not allow online > repair and scrub on that. I've in fact been tempted to see who would > scream if we'd disallow XFS on 32-bit entirel, as that would simplify > a lot of things. You and Dave both. :) I guess we could propose deprecating it like V4 and see if people come out of the woodwork. --D
