[PATCH] D126389: [AMDGPU] Improve codegen of extractelement/insertelement in some cases

[Julien Pagès via Phabricator via llvm-commits]

jpages added a comment.

In D126389#3549129 <https://reviews.llvm.org/D126389#3549129>, @arsenm wrote:

> I wouldn’t really expect there to be a performance difference between gpr indexing and movrel requences. Gfx8 has both so you could directly compare there

I'm using a game for my benchmarks and unfortunately this is a ray-tracing title, I can't really test it on GFX8 :(
These tests were performed on GFX1030 for this reason.

If you still have access to the benchmarks done on GFX8, I could try to test again with this patch.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D126389/new/

https://reviews.llvm.org/D126389