On Tue, Jul 02, 2024 at 11:58:02AM -0400, Jeff Layton wrote: > Yeah, mostly. We shrink struct inode by 8 bytes with that patch, and we > (probably) get a better cache footprint, since i_version ends up in the > same cacheline as the ctime. That's really a separate issue though, so > I'm not too worked up about dropping that patch. > > As a bonus, leaving it split across separate fields means that we can > use unused bits in the nsec field for the flag, so we don't need to > sacrifice any timestamp granularity either. > > I've got a draft rework that does this that I'm testing now. Assuming > it works OK, I'll resend in a few days. So while shrinking the inodes sounds nice, the tradeoff to have to check all timestamps from disk / the server for validity doesn't sound as positive. So I'm glade we're avoiding this at least for.
