In your splash benchmarks, you will need to insert a resetstats M5 instruction at the beginning and end of the parallel section, recompile the binaries, and then run them on M5.
I'm hoping there is documentation on the gem5.org website on how to do this, but if you look in the util/m5/ directory you'll see some basic setup for compiling in m5 instructions. On Fri, Aug 19, 2011 at 12:20 PM, Jacob Murray <[email protected]> wrote: > Hi, > > I have been trying to isolate the ruby stats for parallel portions of > Splash-2 benchmarks. The m5 hooks seem to only affect m5 stats, so I > haven't really been sucessful in doing a stats dump/reset of ruby at the > beginning of the parallel section. I thought a better way would be to > create a checkpoint at the beginning of the parallel section as is shown > with the Parsec example in the documentation. The documentation mentions, > "Take warm up cache trace for Ruby before reaching most interesting portion > of the program and take the final checkpoint." however it is not very clear > how to do this. Any help would be appreciated. > > Thanks, > > Jacob Murray > ECE PhD Candidate > Washington State University > > _______________________________________________ > gem5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users > -- - Korey
_______________________________________________ gem5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
