Re: [Gems-users] sparc decode failure -- followup

[Luke Yen <lyen@xxxxxxxxxxx>]

   Regarding the decode errors, Opal does _not_ implement most (if not
all) of the VIS instructions, because the workloads we study do not use those
instructions.

  If you are studying workloads using those instructions you have to
implement the functionality in opal/system/dx.i

   Luke

On Thu, 5 Oct 2006, Thomas Yeh wrote:

> I just checked the stats, and I see 2 stats:
>
> Percent functional:
> Percent Correct:
>
> Does the percent functional correlate to the amount of instructions that
> failed the decode due to lack of support? And does Correct correlate to
> the % of instructions that were correctly handled by opal?
>
> thx,
> Tom
>
>
>
> On Thu, 5 Oct 2006, Thomas Yeh wrote:
>
> > I am hitting lots of instruction decode failure messages, and I am
> > wondering if gems handle the VIS instructions. I turned on -mvis when I
> > compiled the benchmark.
> >
> > I've seen the previous posts, and I am wondering what is the best way to
> > tune the compiler to prevent generating a lot of these unsupported
> > instructions.
> >
> > thx
> > Tom
> >
> >
> _______________________________________________
> Gems-users mailing list
> Gems-users@xxxxxxxxxxx
> https://lists.cs.wisc.edu/mailman/listinfo/gems-users
> Use Google to search the GEMS Users mailing list by adding "site:https://lists.cs.wisc.edu/archive/gems-users/"; to your search.
>