Hi Pritha,
The m5 binary should say the period is in ns, not ticks. Thanks, Ali On 01.07.2012 13:10, Pritha Ghoshal wrote: > Hi, > > I am using dumpresetstats in my .rcS script, but I have a doubt in that. The definitions are as follows: > > * resetstats [_delay_ [_period_]]: Reset simulation statistics in _delay_ ticks; repeat this every _period_ ticks. > * dumpstats [_delay_ [_period_]]: Save simulation statistics to a file in _delay_ ticks; repeat this every _period_ ticks. > * dumpresetstats [_delay_ [_period_]]: same as _resetstats; dumpstats_ > > So when I want to have 2s delay and 2s period, I assumed I would have to use 2*10^12 in both the places as 2s corresponds to 2*10^12 ticks. But I actually need to run the following to get that behaviour: > /sbin/m5 dumpresetstats 2000000000 2000000000 > > which is delay = 2*10^9 and period = 2*10^9 > > This is because the calculation in sim/pseudo_inst.cc is as follows: > > Tick when = curTick() + delay * SimClock::Int::ns; > Tick repeat = period * SimClock::Int::ns; > Is this correct? Or should ns be modified to ps as the tick frequency is 10^12? > > Thanks, > Pritha
_______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
