On Tue, Mar 18, 2008 at 12:12 PM, Korey Sewell <[EMAIL PROTECTED]> wrote:
> We do? Hmmmm.... Can someone quickly refresh my memory to how EIO
> works with M5???
>
> For example, the function countInst() in src/cpu/simple/base.hh looks like:
> void countInst()
> {
> numInst++;
> numInsts++;
>
> thread->funcExeInst++;
> }
>
> Which of those counts matter for EIO (my guess: thread->funcExeInst)?
yea, if you look at where funcExeInst is declared at the end of
cpu/thread_state.hh the comment says that explicitly.
> Also, where is the code that EIO checks
> to see these instruction counts?
somewhere in encumbered/eio/eio.cc, I'm sure.
>
> Thanks....
>
>
>
> On Tue, Mar 18, 2008 at 2:57 PM, Steve Reinhardt <[EMAIL PROTECTED]> wrote:
> > On Tue, Mar 18, 2008 at 11:52 AM, Korey Sewell <[EMAIL PROTECTED]> wrote:
> > > I thought we didnt care aboue EIO traces anymore?
> >
> > Not a lot, but they're still handy occasionally, and there's no real
> > justification for breaking them at this point.
> >
> >
> > >
> > > Either way, we probably should be keeping a different variable/count
> > > for the EIO compatibility because it seems
> > > to me that correct and EIO arent synonymous...
> >
> > I believe we already do.
> >
> > Steve
> >
> >
>
>
> > _______________________________________________
> > m5-dev mailing list
> > [email protected]
> > http://m5sim.org/mailman/listinfo/m5-dev
> >
>
>
>
> --
> ----------
> Korey L Sewell
> Graduate Student - PhD Candidate
> Computer Science & Engineering
> University of Michigan
> _______________________________________________
> m5-dev mailing list
> [email protected]
> http://m5sim.org/mailman/listinfo/m5-dev
>
_______________________________________________
m5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/m5-dev
