On Sat, Jun 20, 2020 at 12:00:43PM -0500, Eric Sandeen wrote: > On 6/19/20 8:56 PM, J. Bruce Fields wrote: > > On Sat, Jun 20, 2020 at 11:49:57AM +1000, Dave Chinner wrote: > > ... > > >> However, other people have different opinions on this matter (and we > >> know that from the people who considered XFS v4 -> v5 going slower > >> because iversion a major regression), and so we must acknowledge > >> those opinions even if we don't agree with them. > > > > Do you have any of those reports handy? Were there numbers? > > I can't answer that but did a little digging. MS_I_VERSION as an option > appeared here: > ... > so the optional enablement was there on day one, without any real explanation > of why. My memory is that they didn't have measurements at first, but worried that there might be a performance issue. Which later mesurements confirmed. But that Jeff Layton's work eliminated most of that. I think ext4 was the focuse of the concern, but xfs might also have had a (less serious) regression, and btrfs might have actually had it worst? But I don't have references and my memory may be wrong. --b.
