Subj : Re: What is the real costs of LOCK on x86 multiprocesor machine?
To   : comp.programming.threads
From : Gianni Mariani
Date : Sat Jul 30 2005 02:50 pm

Mirek Fidler wrote:
> I am unable to find anwswer to above question. Only tidbit of 
> information was that it is 100 cycles on P4, but perhaps that was just 
> the worst case.
> 
> Measuring the same on my uniprocesor AMD64 machine, LOCKed instructin 
> seem to be 3 times slower than regular one, but I guess that has only a 
> little relevance on real MP machine.


If you have some code you would like me to try, I'll run it on an MP 
machine for you (AMD or AMD64).

The other thing to node it that some more modern P4's have the 
hyperthreading thing that looks like an SMP machine.

.