Re: [Gems-users] Simulation Time of 16 processors

["Siddharth Rajiv Inamdar" <sri3@xxxxxxxxxxx>]

thanks for your reply... do i remove all the -g settings for the
OPT_FLAGS? I tried doing this but the simulation crashed.

 OPT_FLAGS += -g -ggdb -g3

for example, do i comment the entire line out or only the -g. if i'm
write, g3 is the maximum level of verbosity in debug.

thanks


> I did not run with 16 or more processors, but I run with 4 processors and
> I can
> guess that there is nothing wrong with 20 hours. To make sure everything
> is
> alright, you can do a run for smaller number of instructions.
> A couple of things to check: In Makefile.common, make sure that OPT_FLAGS
> are
> set to the -O2 etc. And -g is not set. Compiling with debugging symbols
> makes
> the memory image of the simulator very big and can cause disk paging on
> 1GB ram
> machine.
>
> --
> Nauman
>
>
> Quoting Siddharth Rajiv Inamdar <sri3@xxxxxxxxxxx>:
>
>> Hi, I was wondering whether anyone has simulated a system with 16 or
>> greater processors and knows how long it takes. We are current running
>> the
>> splash FFT benchmark and it has been running for over 20 hours now. The
>> CPU seems to be busy, so i'm not sure whether we've made a mistake or
>> whether it actually takes this long.
>>
>> The machine we're running it on is an AMD64 machine (booted with solaris
>> image) ,2Ghz 1GB of ram.
>>
>> Thanks
>> Sid
>>
>>
>> _______________________________________________
>> Gems-users mailing list
>> Gems-users@xxxxxxxxxxx
>> https://lists.cs.wisc.edu/mailman/listinfo/gems-users
>>
>
> _______________________________________________
> Gems-users mailing list
> Gems-users@xxxxxxxxxxx
> https://lists.cs.wisc.edu/mailman/listinfo/gems-users
>