Sorry, I set the log path to another file. Both the log file and log.backup
file are found.

It seems that I have to change to use SPEC2006 instead. I'll try it.

Thank you a bunch.
Hu Hong





On 5 August 2013 22:20, Brendan Fitzgerald <[email protected]> wrote:

> By default the output file is called ptlsim.log and is located in the root
> folder of MARSS. The log gets rotated after it hits a size limit so there
> might be two files, ptlsim.log and ptlsim.log.backup.
>
> This isn't a bug within SPEC2000 or any other benchmark. I've spent some
> time looking into it but it doesn't consistently break. I've had SPEC2006
> simulations run to completion with no problems and I've had extremely short
> benchmarks have output stop very quickly.
>
>
> On Mon, Aug 5, 2013 at 10:16 AM, Hong Hu <[email protected]> wrote:
>
>> Hi Brendan,
>>
>> Thanks for you reply.
>>
>> So is there a bug in SPEC2000 or in others? Is there any patches to fix
>> this bug? I searched in marss directory and cannot find any file named
>> ptlsim.log. Do you mean the sim.log?
>>
>> Thanks,
>> Hu Hong
>>
>>
>> On 5 August 2013 22:07, Brendan Fitzgerald <[email protected]>wrote:
>>
>>> There is a known bug where the output will freeze up. To make sure that
>>> your simulation is still running you can tail ptlsim.log
>>>
>>> Brendan
>>>
>>>
>>>  On Mon, Aug 5, 2013 at 9:25 AM, Hong Hu <[email protected]> wrote:
>>>
>>>>  Hi all,
>>>>
>>>> Currently I'm trying to run SPEC CPU2000 benchmarks on marss and meet
>>>> some questions.
>>>>
>>>> I copied the benchmark folder into the qemu image and compiled it
>>>> inside this image. Now I can run the benchmark in the qemu emulator
>>>> environment using "runspec" command. However, when I try to run it in
>>>> simulation environment using command "~/start_sim; runspec ... ;
>>>> ~/stop_sim", the system inside qemu just hanged there. qemu outputs
>>>> "Completed      XXX cycles,       XXX commits:     XXX Hz,      XXX
>>>> insns/sec: rip XXX" at first and then the numbers stop changing.
>>>>
>>>> I searched on the mailing-list and found that SPEC2006 can be run
>>>> correctly. Does anyone have the experience with SPEC2000? or the command I
>>>> used to run the benchmark is not correct?
>>>>
>>>> Any help would be appreciated.
>>>>
>>>> Thanks,
>>>> Hu Hong
>>>>
>>>> _______________________________________________
>>>> 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