Re: [PATCH] anobjrmap 1/6 objrmap

[Andrea Arcangeli <andrea@suse.de>]

On Fri, Mar 19, 2004 at 09:11:23AM -0800, Martin J. Bligh wrote:
> Well, I'm pleased to say not only is your code stable in my tests, it's
> also faster than partial objrmap (not by that much, but definitely 
> measurable). And of course, the code's cleaner. Kernbench & SDET are both 
> heavy on fork/exec/exit, so this should give these paths a heavy workout.
> (this was on 16-way NUMA-Q).
> 
> Andrea, are you still working on your code at the moment, or is it ready
> for others to play with? I'll make a run at that as well if you say it's
> ready, though I think I might have lost track of the latest version ;-)

I'm working on my code yes, I think my code is finished, I prefer my
design for the various reasons explained in the other emails (you don't
swap so you can't appreciate the benefits, you only have to check that
performs as well as Hugh's code).

Hugh's and your code is unstable in objrmap, you can find the details in
the email I sent to Hugh, mine is stable (running such simulation for a
few days just fine on 4-way xeon, without my objrmap fixes it live locks
as soon as it hits swap).

You find my anon_vma in 2.6.5-rc1aa2, it's rock solid, just apply the
whole patch and compare it with your other below results. thanks.