[ILUG] Hyperthreading and SMP testing

[John P. Looney]

On Wed, Apr 14, 2004 at 11:27:55AM +0100, Kenn Humborg mentioned:
> 
> It's always a very good idea to test multi-threaded
> code on an SMP machine, since it increases the chances
> of race conditions, deadlocks, etc, occurring.
> 
> However, is it good enough to test on a Pentium with
> hyperthreading, rather than a real SMP machine?  How
> close is hyperthreading to real SMP?

 I don't think so. Not everything will run in parallel. Some stuff gets
blocked and I'm fairly sure most people wouldn't know what would get
tested and what wouldn't.

 If you don't have an SMP system, fair enough, it's a cheap substitute.
But that's all.

John