> Because I want to model the branch penalty of an in-order core this
> will probably not affect the total number of inpredicted branches, nor
> the state of the branch predictor.
>
I'm a little confused here. Are you are saying that you want to know many
cycles do mispredicted branches contribute to overall execution?

In terms of if you used the inorder model,
Is it not sufficient to estimate this by first counting the # of
mispredicted branches and then multiplying that by the # of stages behind
where that branch is resolved? If it is, you can just add another M5
Statistic that will calculate this for you.

In terms of using the SimpleCPU model,
I'm not sure what the best way to do that is since there is no concept of
stages there and since there is no speculative path in the SimpleCPU, I'm
not sure how you would figure out mispredictions, branch penalty, etc. Maybe
someone else can give some additional thoughts here.

-- 
- Korey
_______________________________________________
m5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/m5-dev

Reply via email to