Thanks for this bug report.
I have updated core-models branch to fix this issue.

- Avadh

On Fri, Sep 9, 2011 at 6:44 AM, sparsh1 mittal1 <[email protected]>wrote:

> Hello
> See this (a fraction of the  fer_freq.yaml file for 2-core simulations)Full
> output is attached.
> Here kernel+user is not equal to total. Is it what it is supposed to be? I
> was under impression that total stats should be equal to user+kernel.
>
> My understanding about these stats is that total counts  reflects the whole
> system behavior. If one is particularly interested in application
> performance, user mode is the best choice.
> ---
> sim_stats:
>   tags: [server_name.(none), 2011-09-08, kernel]
>
>    L2_0:
>     annul: 0
>     queueFull: 0
>     cpurequest:
>       redirects: 0
>       count:
>         hit:
>           read:
>             hit: 10189
>             forward: 0
>           write:
>             hit: 3990
>             forward: 0
>         miss:
>           read: 101580
>           write: 118494
>
>  ---
> sim_stats:
>   tags: [server_name.(none), 2011-09-08, user]
>   L2_0:
>     annul: 0
>     queueFull: 0
>     cpurequest:
>       redirects: 0
>       count:
>         hit:
>           read:
>             hit: 34698
>             forward: 0
>           write:
>             hit: 3450
>             forward: 0
>         miss:
>           read: 15761
>           write: 219
>
>
> ---
> sim_stats:
>   tags: [server_name.(none), 2011-09-08, total]
>
>
>   L2_0:
>     annul: 0
>     queueFull: 0
>     cpurequest:
>       redirects: 0
>       count:
>         hit:
>           read:
>             hit: 89774
>             forward: 0
>           write:
>             hit: 14880
>             forward: 0
>         miss:
>           read: 234682
>           write: 237426
>
>
> Thanks and Regards
> Sparsh Mittal
>
>
>
>
> _______________________________________________
> 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