Hi Ben and Ali,
I have finally figured out why it wasn't working (thanks to your pointers).
I checked the main functions in both queens benchmark and default gem5
hello program.
In queens;
void main(int argc,char **argv) { ... }
In hello;
*int* main()
{
printf("Hello world!\n");
}
So the problem was the main function of hello. It is not calling an "exit"
type of function. I tried it locally, it worked (or in this case it looks
like it works but there is something going on behind the scenes). I don't
know the underlying mechanics but when I changed the code like this
int main()
{
printf("Hello world!\n");
*return 0;*
}
the problem vanishes. This was my command line by the way;
---------------
./build/X86/gem5.opt --debug-flags=Cache --trace-file=my_trace.out
configs/example/se.py --caches --l2cache --cpu-type=timing -n 2
--cmd="tests/test-progs/queens/bin/x86/linux/queens;tests/test-progs/hello2/bin/x86/linux/hello2"
--options="-c 8;"
---------------
Thanks all.
2013/1/7 Ali Saidi <sa...@umich.edu>
> **
>
> I would guess the exit() call that hello world is performing is exitGroup
> not exit (I don't know why) which causes all threads to exit.
>
>
>
> Ali
>
>
>
>
>
> On 07.01.2013 09:03, Payne, Benjamin wrote:
>
> Hello,
>
>
>
> Could you post the commands you are using to perform the tasks you
> describe?
>
> Also, if it’s not too big, the output status messages from the gem5 run.
>
>
>
> Thanks,
>
>
>
> Ben
>
>
>
> *From:* gem5-users-boun...@gem5.org [mailto:gem5-users-boun...@gem5.org] *On
> Behalf Of *mehmet basaran
> *Sent:* Sunday, January 06, 2013 6:08 PM
> *To:* gem5-users@gem5.org
> *Subject:* [gem5-users] Weird behavior when running different
> [independent] workloads on multicores
>
>
>
> Hi all,
>
> I have configured a CPU with 6 cores. I use queen benchmark and simple
> hello world code.
>
> - When I run queen benchmark (with different parameters) on all the cores,
> each process ends naturally, I get the correct output.
> - But when I run queen benchmark on 5 cores and hello world code on the
> remaining one core, gem5 terminates before 5 cores (queen bench) finishes
> execution.
>
> I checked to see if this is about the time difference required by the each
> workload. I used queen benchmark for all 6 cores, but gave one core much
> smaller workload. Still all the cores finished their jobs before gem5
> terminated.
>
> Should I configure something before running different binaries for each
> processor? Pointers, answers, help or even questions are most welcome.
>
> Thanks
> Mehmet
>
> _______________________________________________
> gem5-users mailing
> listgem5-users@gem5.orghttp://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
>
>
>
>
> _______________________________________________
> gem5-users mailing list
> gem5-users@gem5.org
> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
>
_______________________________________________
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
