I meant to say something vastly smaller than the simpoint, I understand 
that if max-instructions is reached before the checkpoint 'drops' it is 
going to exit.

Sorry,
Will

William George Beazley Jr wrote:


>Lisa,
>
>Ash and I are seeing some benchmarks exiting before the checkpoint is 
>taken with:
>
>Exiting @ cycle 1784286478500 
>because target called exit()
>
>I am not confident on the relationship between cycle and instructions, 
>but I have seen this value at less then half the simpoint.
>
>If I set the simpoint manually to something smaller than 
>max-instructions it drops the checkpoints as expected.
>
>Any ideas?
>
>Thanks,
>Will
>
>
>Lisa Hsu wrote:
>
>
>>Simpoints are an intrinsic feature of workloads - if you want to take a
>>checkpoint at a simpoint, you must do --take-checkpoint 0 --simpoint, as
>>described in our asplos tutorial.  If you look in cpu2000.py, the 
>simpoint
>>parameters are part of each benchmark.
>>Lisa
>>
>>
>>On Fri, Aug 7, 2009 at 3:12 PM, Ashutosh Jain <[email protected]> 
>wrote:
>>
>>> Hi
>>> I want to use the simpoint values of the benchmarks. I do add the 
>command
>>> line parameters "-simpoint --take-checkpoint=100000000". Also I have 
>tried
>>> to run the simulations with taking 0 simpoint value and  benchmark 
>simpoint
>>> value.
>>>
>>> The result in the stats file doesn't get effected, its same every 
>time. So
>>> it seems that simpoint is not working. But when I set the simpoint 
>value, it
>>> reflected changes in config.ini file with the same assigned simpoint 
>value(
>>> whether it is 0 or some value). So it is not clear whether it is 
>working or
>>> not.
>>>
>>> Thanks in advance
>>> Ashutosh Jain
>>>
>>> _______________________________________________
>>> m5-users mailing list
>>> [email protected]
>>> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
>>>
>>
>
>
>------------------------------------------------------------------------
>
>
>_______________________________________________
>>m5-users mailing list
>>[email protected]
>>http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
>
>-----------------------------------
>Will Beazley|Sys. Software Analyst
>409.880.7847|[email protected]
>_______________________________________________
>m5-users mailing list
>[email protected]
>http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
>

-----------------------------------
Will Beazley|Sys. Software Analyst
409.880.7847|[email protected]
_______________________________________________
m5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/m5-users

Reply via email to