[PATCH] D131959: [AMDGPU] Fix SDST operand of V_DIV_SCALE to always be VCC

[Stanislav Mekhanoshin via Phabricator via llvm-commits]

rampitec added a comment.

In D131959#3783259 <https://reviews.llvm.org/D131959#3783259>, @Pierre-vh wrote:

>> Did you change AMDGPUdiv_scale to produce glue? So that you have something to glue this copy to?
>
> Isn't `CurDAG->SelectNodeTo(N, Opc, CurDAG->getVTList(VT, MVT::i1, MVT::Glue), Ops);` enough for that? Do I need to change the DAG Op definition itself?

I think so, but I believe we have reached the conclusion we do not need this w/a. We are pretty confident it works as is and the w/a tends to be invasive. So I'd suggest to shelve this just in case, abandon the patch and reject the ticket.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D131959/new/

https://reviews.llvm.org/D131959