Re: [Gems-users] instruction counts in makeRequest()

[Dan Gibson <degibson@xxxxxxxx>]

I would expect to see repeats, but the instruction count shouldn't decrease or reset. The call to getInstructionCount() eventually calls SIM_step_count(), and I do not know if THAT function returns a monotonically-increasing value or not -- try calling both SIMICS_get_insn_count() and SIM_step_count() directly and compare the results. Their relationship to getInstructionCount() should be made clear in SimicsDriver.C, SimicsProcessor.C, and interface.C.

Regards,
Dan

Carole-Jean Wu wrote:

Hello,

I am finding some weird numbers with instruction counts. Can someone explain whether this makes sense?

in system/Sequencer.C's makeRequest()

Time icnt = g_system_ptr->getDriver()->getInstructionCount(0);
Time mcycle = get_eventQueue_ptr->getTime();


printf("%lld %lld\n", icnt, mcycle);


mcycle is always monotonically increasing, but icnt isn't.

I suppose that icnt means the number of instructions executed so far by CPU0 and icnt should also increase monotonically, right? If that is the case, can someone explain why it is not right now?


Thanks!
Carole

_______________________________________________
Gems-users mailing list
Gems-users@xxxxxxxxxxx
https://lists.cs.wisc.edu/mailman/listinfo/gems-users
Use Google to search the GEMS Users mailing list by adding "site:https://lists.cs.wisc.edu/archive/gems-users/" to your search.

  

-- 
http://www.cs.wisc.edu/~gibson [esc]:wq!