You do, however it's not clear that you care about the initialization for whatever change you're attempting to evaluate. If you run an atomic cpu for a 1/4 of the benchmark and then create a checkpoint, you can then restore from the checkpoint and get new statistics on a detailed cpu running from that point. In reality you probably want to go ahead and warm-up for a little while and then reset your statistics again, but either way you can evaluate a variety of changes from the checkpoint instead of from the start of the program. If you're running a lot of variations this can significantly reduce the time required to get results (1/4 of the run-time is only executed once not for every variation). Additionally, the statistics you get won't be obfuscated by the warmup period which normally involves creating data structures and not the steady-state behavior found in the middle of the benchmark.
The run-time for splash 2 depends a great deal on the parameters you specify. It can be very short or very long depending on the problem size. Ali On Jun 26, 2008, at 2:38 PM, Sujay Phadke wrote: > Hi Ali, > Could you explain why simulating the initializing code in > Splash-2 > can be avioded? I thought we would have to simulate the entire > thread to get > the correct results. The sim time for splash-2 is about 45mins for the > longest benchmark. > > - Sujay > > ----- Original Message ----- > From: "Ali Saidi" <[EMAIL PROTECTED]> > To: "M5 users mailing list" <[email protected]> > Sent: Tuesday, June 24, 2008 3:22 PM > Subject: Re: [m5-users] error running O3 directly with more than 1 cpu > > >> >> On Jun 23, 2008, at 5:47 PM, Sujay Phadke wrote: >> >>> Well I wanted to run the O3 (n>1) for testing wattch. Is it wrong to >>> run it >>> in detailed mode directly w/o using any switching? >> No, however you'll be spending a very long time simulating, including >> simulating all initialization code in the splash 2 benchmarks that >> isn't incredibly useful. >> >>> >>> If so, can you give me an example file and command line which does >>> switching >>> for say Splash2? >> I don't have one but you can look at how switching is done in >> Simulation.py to get an idea. >> >> Ali >> >> >>> >>> >>> thanks, >>> Sujay >>> >>> ----- Original Message ----- >>> From: "Ali Saidi" <[EMAIL PROTECTED]> >>> To: "M5 users mailing list" <[email protected]> >>> Sent: Monday, June 23, 2008 5:39 PM >>> Subject: Re: [m5-users] error running O3 directly with more than 1 >>> cpu >>> >>> >>>> >>>> On Jun 23, 2008, at 1:23 PM, Sujay Phadke wrote: >>>> >>>>> Hello, >>>>> If I try to run O3 in SE mode directly, without using any >>>>> fastforwarding, it gives me the error: >>>>> >>>>> build/ALPHA_SE/m5.opt configs/splash2/runsplash.py -d -n4 -b >>>>> OceanContig >>>>> >>>>> warn: Increasing stack size by one page. >>>>> warn: ignoring syscall sigprocmask(3, 18446744073709551615, ...) >>>>> warn: ignoring syscall sigprocmask(3, 0, ...) >>>>> warn: ignoring syscall sigaction(8, 4831387552, ...) >>>>> warn: ignoring syscall sigaction(11, 4831387552, ...) >>>>> warn: ignoring syscall sigaction(10, 4831387552, ...) >>>>> warn: ignoring syscall sigaction(4, 4831387552, ...) >>>>> warn: ignoring syscall sigaction(7, 4831387552, ...) >>>>> warn: ignoring syscall sigaction(6, 4831387552, ...) >>>>> warn: ignoring syscall sigaction(12, 4831387552, ...) >>>>> warn: ignoring syscall sigaction(5, 4831387552, ...) >>>>> warn: ignoring syscall sigaction(13, 4831387552, ...) >>>>> nxm_thread_create: no idle contexts available. >>>>> Program aborted at cycle 378945000 >>>>> It works fine of numcpus (-n) = 1, or if I use atomic or timing >>>>> mode. The statement >>>>> >>>>> thread->setStatus(ThreadContext::Unallocated); >>>>> >>>>> appears in src/cpu/simple/base.cc and src/cpu/o3/ >>>>> thread_context_impl.hh. >>>>> >>>> The statement and the context the statement is in are two different >>>> things. From a cursory look of the O3CPU, it appears as though the >>>> state of the thread context is never initialized when the thread >>>> context is created, and the places where it is set to unallocated >>>> are >>>> for switching CPUs. Initializing it to Unallocated should solve the >>>> problem. >>>> >>>> Are you planning on running a detailed CPU for the entirety of the >>>> benchmark? >>>> >>>> Ali >>>> >>>> _______________________________________________ >>>> 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 >>> >> >> _______________________________________________ >> 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 > _______________________________________________ m5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
