Hi Joel, I'm using writescripts.pl from your website. It's very helpful on generating the scripts. Actually all your work with PARSEC on M5 help us a lot too. Also, it's glad to see so many people working on PARSEC in M5's community. :)
Thanks a lot! Best, Jie On Apr 29, 2010, at 12:24 AM, Joel Hestness wrote: > Hi Jie, > That's correct, assuming you're using writescripts.pl from our website to > generate your M5 run scripts. > Joel > > > On Wed, Apr 28, 2010 at 10:36 PM, Jie Meng <[email protected]> wrote: > Hi Joel, > > I'm also using PARSEC on M5 FS mode. And my stats.txt has four parts. > > As far as I understand, they are: > > 1) Statistics for M5 full system booting up > 2) Statistics before reaching ROI > 3) Statistics in the ROI > 4) Statistics dumped when M5 exits > > And we're interest in the multi-thread parallel running part. So I believe > part 3) is what we should focus on. Am I right? > > And hope this could provide some comparable information to Sheng too. > > Thank, > Jie > > > On Apr 27, 2010, at 3:23 PM, Joel Hestness wrote: > >> Hi Sheng, >> If you are using the disk image from our website, the benchmarks are >> compiled to dump M5 statistics before and after the computational region of >> interest of the benchmark, and the simulation dumps statistics when it exits. >> I would encourage you to read through section 6.5 our technical report, in >> particular the part about M5 hooks, for more information on what your >> simulations are doing. >> Good luck, >> Joel >> >> >> On Tue, Apr 27, 2010 at 1:45 PM, sheng qiu <[email protected]> wrote: >> Hi all, >> >> when i running PARSEC on M5 FS mode, the simulation results of each >> application has three parts of the following sections >> ---------- Begin Simulation Statistics ---------- >> ............................... >> ...................... >> ---------- End Simulation Statistics ---------- >> >> my script is: >> #!/bin/sh >> >> # File to run the blackscholes benchmark >> >> cd /parsec/install/bin >> /sbin/m5 resetstats >> ./blackscholes 1 /parsec/install/inputs/blackscholes/in_4K.txt >> /parsec/install/inputs/blackscholes/prices.txt >> /sbin/m5 exit >> So why there are three statistics in one application result? which one is >> the benchmark simulation result? maybe my question is a little trivial. >> >> Thanks, >> Sheng >> >> >> >> >> _______________________________________________ >> m5-users mailing list >> [email protected] >> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users >> >> >> >> -- >> Joel Hestness >> PhD Student, Computer Architecture >> Dept. of Computer Science, University of Texas - Austin >> http://www.cs.utexas.edu/~hestness >> _______________________________________________ >> m5-users mailing list >> [email protected] >> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users > > > _______________________________________________ > m5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/m5-users > > > > -- > Joel Hestness > PhD Student, Computer Architecture > Dept. of Computer Science, University of Texas - Austin > http://www.cs.utexas.edu/~hestness > _______________________________________________ > m5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
_______________________________________________ m5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
