Re: [Gems-users] instructions executed in ruby

[Dan Gibson <degibson@xxxxxxxx>]

Mishali,
Are you running Ocean and Barnes long enough? Could it be that you're 
only catching the beginning of the parallel section?
Also, is processor 0 spinning on a lock more frequently than the other 
threads? Bear in mind there is not a 1-to-1 correlation between 
instructions executed and "work done" in many parallel benchmarks.
Regards,
Dan Gibson

Mishali Naik wrote:
> Hi All,
>            I simulated 8 processors with MOESI_CMP_NUCA model (GEMS
> with ruby) and ran splash benchmarks OCEAN and BARNES on it. What I
> observed is one of the processors (processor 0) executes about 75-80%
> of the total instructions and the rest of the processors aren't
> executing as many instructions. Has anyone seen this kind of behavior?
> What could the problem be in this scenario?
>           I appreciate all your help.
>
> thanks,
> Mishali
> _______________________________________________
> Gems-users mailing list
> Gems-users@xxxxxxxxxxx
> https://lists.cs.wisc.edu/mailman/listinfo/gems-users
> Use Google to search the GEMS Users mailing list by adding "site:https://lists.cs.wisc.edu/archive/gems-users/"; to your search.
>
>
>