On Tue, May 8, 2012 at 2:03 AM, tim yenni <[email protected]> wrote:
> Hi, > > I am a new research assistant and I am using Marss as the simulator for > the research. I had run the simulation with Marss 0.2, QEMU and some of the > parsec benchmark. However, recently I try to use Marss 0.3 to run the > simulation and I get a different statistics. May I know is there any major > changes on the code which could possibly causing the problem? I am sorry if > I asked something which does not make sense here, but I am totally a > beginner and seeking for help from everywhere I can. > Thank you in advance. Hope to get some guide here. > > In 0.3 release we have fixed many issues related to non-determinism in multicore benchmarks. Also there has been many bug fixes in coherent cache and memory hierarchy. Due to these fixes there will be differences between results from 0.2 and 0.3 release and I will rely more on 0.3 release results. All cores pipeline and cache logic is same as 0.2 release. - Avadh Regards, > > Ophelia > > _______________________________________________ > http://www.marss86.org > Marss86-Devel mailing list > [email protected] > https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel > >
_______________________________________________ http://www.marss86.org Marss86-Devel mailing list [email protected] https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
