On Wed, 26 Jan 2005 14:19:29 -0500
J Chapman Flack <flack@cs.purdue.edu> wrote:
> > > That wasn't the point. What if I have a stable non-busy system?
> > > sync(8) lies currently and doesn't force all data out to the
> > > disk...
> > 
> > It always lied, thus "sync, sync, halt".
> 
> The best explanation I was ever able to get for the "sync, sync, halt"
> rule of thumb was that the extra typing gave you something to do
> during any time it might take for the writes initiated by 'sync' to
> complete.

That's not quite how I was told.  The idea is (Was?  I'm not sure if it
is still true) that the first sync scheduled all dirty buffers to be
written to disk.  The second sync blocked until the effects of the first
one were complete so the idea was not that the second sync did anything
but that it would not return until the effects of the first one were
completed.

Of course, I may very well be full of the brown stuff but that is the
way I heard it.

-- 
D'Arcy J.M. Cain <darcy@NetBSD.org>
http://www.NetBSD.org/