On Fri, Jun 24, 2011 at 11:13 AM, Yingying Tian <[email protected]> wrote:

> To whom it may concern,
>
> I made some modifications on Marss and ran it as a 2-core CMP with 500
> million instructions. Everything went well. Then I ran it with 1 billion
> instructions. most of the benchmarks ran well but one failed with the log
> output as below:
> -------------------------------------------
>  Initializing core 'ooo'
>  Core created: 0
>  Core created: 1
>  Start logging at level 0 in cycle 0
>  Completed             0 cycles,             0 commits:         0 Hz,
>   0 insns/sec: rip 00000000000f1eaa 00000000fffffff0
>  RIP dont match at commit time:
>  SMT per-thread state for t0:
>  Fetchrip: 0000000000000048
> ....
> --------------------------------------------
>
> By just looking at the 'rip' values, it seems that your simulation is
starting from boot sequence (look at 32 bit rip addresses). Currently
simulation of boot sequence is not supported in Marss. Check your
benchmark/checkpoint if its created correctly or not.

- Avadh

> Does anyone know what the problem is? Could you please give me any
> suggestion?
>
>
> Thanks,
> Yingying
>
>
> _______________________________________________
> http://www.marss86.org
> Marss86-Devel mailing list
> [email protected]
> https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
>
>
_______________________________________________
http://www.marss86.org
Marss86-Devel mailing list
[email protected]
https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel

Reply via email to