[PATCH] D63203: [AMDGPU] gfx1010 dpp16 and dpp8

Wed Jun 12 10:52:26 PDT 2019

kzhuravl added inline comments.

================
Comment at: include/llvm/IR/IntrinsicsAMDGPU.td:1457
+  Intrinsic<[llvm_anyint_ty],
+            [LLVMMatchType<0>, llvm_i32_ty],
+            [IntrNoMem, IntrConvergent]>;
----------------
rampitec wrote:
> kzhuravl wrote:
> > rampitec wrote:
> > > rampitec wrote:
> > > > arsenm wrote:
> > > > > arsenm wrote:
> > > > > > Needs immarg, and to be marked as a source of divergence
> > > > > I think it's a source of divergence? It's missing from the existing DPP intrinsics, but that might be a bug?
> > > > I agree. But in a separate change.
> > > D63216
> > Is D63216 expecting to cause any merge conflicts downstream?
> Yes. Therefor I am going to wait the merge of this change, and only then submit D63216. That way it will avoid conflicts.
Thanks!

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D63203/new/

https://reviews.llvm.org/D63203