[PATCH] AArch64: add support for llvm.aarch64.hint intrinsic

[Jim Grosbach]

Is hasSideEffects=1 not sufficient?

-j
> On Jul 8, 2014, at 8:17 AM, Tim Northover <t.p.northover at gmail.com> wrote:
> 
> I don't think the "mayLoad = ?" is any better. TableGen is still inferring "mayLoad = 1", we're just concealing that from the reader.
> 
> I think I'd go for the original code, with a comment that really these should be "mayLoad = 0" but that intrinsics' side effects aren't modelled with enough granularity for that to be possible.
> 
> Cheers.
> 
> Tim.
> 
> http://reviews.llvm.org/D4403
> 
> 
> 
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits