On Thu, Apr 23, 2009 at 8:29 AM, Korey Sewell <[email protected]> wrote:

> I guess on a 1st cut, that right thing to do is discuss what is the
> accepted way we want to trace instructions across the board and then just
> make sure all CPUs conform to that standard.
>
> Issues in SE mode seem to be with:
> - faulting instructions
> - system calls
>
> They seem to be counted 2x in Simple but once in Atomic. Which one is
> preferable?


Are we sure that this is the case?  All I've seen so far is that they appear
to be traced twice in O3 and once in Simple.  The Simple count appears to be
right (since it matches the number of trace records printed, and it only
prints faulting instructions once).

I think each instruction should be counted exactly once, when it commits
successfully.


> The secondary issue is what's causing the large disparity for things like
> gzip and other regressions? I guess we wont know for sure until someone
> inspects those 1-by-1.


Actually I think this is the primary issue... minor explainable differences
are irritating but OK, but large unexplainable discrepancies are scary.

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

Reply via email to