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 03:00 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.


OH - and also, the cost of cpu-cpu communication through shared memory 
is somthing else you might want to consider.

I wrote some code many moons ago that I posted on the net...

http://groups-beta.google.com/group/comp.os.linux.development.system/msg/df42233b14e5f962?dmode=source&hl=en

In general, I found that latentcies in absolute time stayed about the 
same until the AMD Opteron came out which shaved a few more cycles out 
of it.  I don't have the data handy.

.