On Thu, Apr 23, 2015 at 02:16:04PM -0500, Peter Bergner wrote:
> Ok, I created a separate ttest define_insn that hard codes the operands.
> I switched to using r1 rather than r0 as the second operand, for the
> reason that there could be code that sets r0 directly before this insn
> and I didn't want to create a unneeded read-after-write dependency.
> I thought r1 should be safe to use, since it's only updated in the
> prologue/epilogue and should be constant for the life of the function.
GPR1 is very likely already read before as well :-)
> > Maybe just { powerpc64 } works?
>
> powerpc64 doesn't work. It tells us whether the target will execute
> 64-bit instructions or not.
Ah yes, it is more like a "powerpc64_hw". Oh well.
All looks great to me now, thanks for the changes,
Segher
