Re: ZVC: Increase threshold for larger processor configurationss

[Andrew Morton <akpm@osdl.org>]

On Thu, 29 Jun 2006 23:15:55 -0700 (PDT)
Christoph Lameter <clameter@sgi.com> wrote:

> On Thu, 29 Jun 2006, Christoph Lameter wrote:
> 
> > > Did you consider my earlier suggestion about these counters?  That, over the
> > > short-term, they tend to count in only one direction?  So we can do
> > Uhh... We are overcompensating right? Pretty funky idea that is new to me 
> > and that would require some thought.
> > 
> > This would basically increase the stepping by 50% if we are only going in 
> > one direction.
> 
> A patch that does this:
> 
> 
> ZVC: overcompensate while incrementing ZVC counters
> 
> Overcompensate by a balance factor when incrementing or decrementing
> ZVC counters anticipating continual increase in the same direction.

Looks sensible.

Please check that none of this is racy wrt memory hotplug
(process_zones->vm_stat_setup).

> Note that I have not been able to see any effect off this approach on
> an 8p system where I tested this.
> I probably will have a chance to test it on larger systems (160p) tomorrow.

OK.  But let's not rush - it's only fine-tuning.  I'm thinking we should
get what we have in -mm4 into -rc1 - I think it's stable enough, and we
don't want to be carrying all those changes splatered across the VM for the
next two months.  Let's aim to get the well-measured fine-tuning in place
for -rc2, OK?

Are you aware of any to-do items remaining in the -mm4 patches?  The NFS
changes need a review from Trond - hopefully he'll be able to find 5-10
minutes to do that sometime?
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/