[ofa-general] Re: [PATCH 04 of 12] Moves all mmu notifier methods outside the PT lock (first and not last

[Robin Holt]

On Wed, Apr 23, 2008 at 03:44:27PM +0200, Andrea Arcangeli wrote:
> On Tue, Apr 22, 2008 at 04:14:26PM -0700, Christoph Lameter wrote:
> > We want a full solution and this kind of patching makes the patches 
> > difficuilt to review because later patches revert earlier ones.
> 
> I know you rather want to see KVM development stalled for more months
> than to get a partial solution now that already covers KVM and GRU
> with the same API that XPMEM will also use later. It's very unfair on
> your side to pretend to stall other people development if what you
> need has stronger requirements and can't be merged immediately. This
> is especially true given it was publically stated that XPMEM never
> passed all regression tests anyway, so you can't possibly be in such

XPMEM has passed all regression tests using your version 12 notifiers.

I have a bug in xpmem which shows up on our 8x oversubscription tests,
but that is clearly my bug to figure out.  Unfortunately it only shows
up on a 128 processor machine so I have 1024 stack traces to sort
through each time it fails.  Does take a bit of time and a lot of
concentration.

> an hurry like we are, we can't progress without this. Infact we can

SGI is under an equally strict timeline.  We really needed the sleeping
version into 2.6.26.  We may still be able to get this accepted by
vendor distros if we make 2.6.27.

Thanks,
Robin