That is a bug, there should be a function something like this in the
BPredUnit:
template <class Impl>
std::string
BPredUnit<Impl>::name() const
{
return cpu->name() + ".BPredUnit";
}
Although the branch predictor doesn't have a cpu pointer so that would
have to be added as well. I don't know why it doesn't. Kevin?
Also, perhaps we could just output the config.ini as an xml file
rather than doing the conversion. It's probably about as readable and
much easier to parse since many tools exist to do that. Anyone want to
comment about this? Nate? Along a similar line it probably wouldn't be
too difficult to add an xml backend to the statistics package.
Ali
On Aug 29, 2008, at 4:54 PM, Rick Strong wrote:
> Hi all,
>
> I am working on a script that turns the config.ini into xml and then
> adds all the m5 stats to the correct node in the xml. This is to make
> things easier for a more detailed power model. However, I noticed
> something strange. Is there a good reason for BPredUnit statistics
> to be
> apart of global? The stats involved are shown below.
>
> -Rick
>
> <stat name="global.BPredUnit.condIncorrect" value="0"/>
> <stat name="global.BPredUnit.BTBLookups" value="0"/>
> <stat name="global.BPredUnit.BTBHits" value="0"/>
> <stat name="global.BPredUnit.condPredicted" value="0"/>
> <stat name="global.BPredUnit.BTBCorrect" value="0"/>
> <stat name="global.BPredUnit.usedRAS" value="0"/>
> <stat name="global.BPredUnit.lookups" value="0"/>
> <stat name="global.BPredUnit.RASInCorrect" value="0"/>
> _______________________________________________
> 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
