It's possible that the statistics in the O3 cpu model are broken; I know we had problems with them at one point, though I thought they had been fixed.
Steve On Thu, Aug 20, 2009 at 3:01 AM, Arun Rangasamy<[email protected]> wrote: > Hi all, > I want to know the fraction of time spent in kernel/user modes for my > application when doing my timing simulation. > For my simulations, I fast forward a few hundred millions of instructions > and then switch to detailed simulation. > I find in m5stats.txt that only "system.cpu<i>.kern.mode_ticks_kernel" > entries have non-zero values, and > "system.switch_cpus<i>.kern.mode_ticks_kernel" all have zero values. > Could anyone please explain why this is so? > Thank you very much in advance! > Best Regards, > arun > > _______________________________________________ > 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
