This looks pretty obviously correct to me. I'll add it to my patch
queue and make sure that it doesn't break anything.
Nate
On Mon, Sep 14, 2009 at 10:14 PM, Vince Weaver <vi...@csl.cornell.edu> wrote:
> Hello
>
> the SPARC udivcc instruction was attempting to read bits 63:32 of a
> uint32_t, which didn't work very well.
>
> The below patch changes the value to uint64_t which fixes my test case.
>
> Vince
>
> diff -r 3b2d7fdff6b1 src/arch/sparc/isa/decoder.isa
> --- a/src/arch/sparc/isa/decoder.isa Fri Sep 11 16:19:31 2009 -0500
> +++ b/src/arch/sparc/isa/decoder.isa Tue Sep 15 01:14:42 2009 -0400
> @@ -226,7 +226,8 @@
> if(Rs2_or_imm13.udw == 0) fault = new DivisionByZero;
> else Rd = Rs1.udw / Rs2_or_imm13.udw;}});
> 0x1E: IntOpCcRes::udivcc({{
> - uint32_t resTemp, val2 = Rs2_or_imm13.udw;
> + uint64_t resTemp;
> + uint32_t val2 = Rs2_or_imm13.udw;
> int32_t overflow = 0;
> if(val2 == 0) fault = new DivisionByZero;
> else
> _______________________________________________
> m5-dev mailing list
> m5-dev@m5sim.org
> http://m5sim.org/mailman/listinfo/m5-dev
>
>
_______________________________________________
m5-dev mailing list
m5-dev@m5sim.org
http://m5sim.org/mailman/listinfo/m5-dev
