Re: revivesa status 2008/07/09

[Andrew Doran <ad%netbsd.org@localhost>]

On Fri, Jul 11, 2008 at 04:28:08PM -0400, Thor Lancelot Simon wrote:

> On Fri, Jul 11, 2008 at 09:30:13PM +0200, Manuel Bouyer wrote:
> > On Fri, Jul 11, 2008 at 12:47:45PM +0000, Andrew Doran wrote:
> > > 
> > > In its current form SA threading is a regressive proposition. Even if all
> > > the remaining issues are addressed, what benefits would it bring over and
> > > above 1:1 threading?
> > 
> > Be able to run a netbsd-4 or netbsd-3 userland on a netbsd-5 kernel.
> 
> I think there's another significant benefit: it should give a significant,
> possibly even huge, performance benefit for multithreaded applications on
> platforms which are uniprocessor and have a significant penalty for
> context-switching into the kernel.

I keep hearing this and in my experience it's not true. One test case I can
think of is MySQL and the supersmack benchmark. It makes heavy use of thread
synchronization and issues lots of trival queries. Whatever gains SA makes
in the thread synch area are blown away by the overhead it imposes on kernel
I/O, and so 1:1 beats it on both uni and multiprocessor systems.

> The obvious example is ARM -- and there are a _lot_ of ARM CPUs out
> there in the world in embedded devices, often running software that makes
> heavy use of threading for its UI.

If context switch is slow on ARM then why not make it faster? The entire
system would benefit, not just threading. A lot of time was invested in the
same area on x86 over the last couple of years.

> If I saw a benchmark -- and not an I/O bound one -- from such a platform
> where the 1:1 threading did anywhere near as well as the SA threading I'd
> certainly change my mind, but...

I don't have such a platform, but I can provide benchmarks for x86. Stay
tuned.

Andrew