> > Hmm. Recreated in 2.1.55, df sounds like sync... And looks like sync,
> > too (on blinkenlights)... And guess what:
> >
> > And it is because df _calls_ sync!
>
> Honestly, I'm not surprised, but the argument is: When bdflush is
> running, shouldn't all the dirty inodes be written back to disk from
> time to time. If so, sync() shouldn't have to do much work.
by running 'df' you touch the inode, atime gets changed, so sync() has
(some) nonzero work to do. Other inodes might get touched too.
-- mingo