[PATCH] D71956: [X86] Improve v2i64->v2f32 and v4i64->v4f32 uint_to_fp on avx and avx2 targets.

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon accepted this revision.
RKSimon added a comment.
This revision is now accepted and ready to land.

LGTM - are the TTI costs still accurate for the new codegen?



================
Comment at: llvm/lib/Target/X86/X86ISelLowering.cpp:29111
+      // TODO Any SSE41+ subtarget should work here but BLENDV codegen ends up
+      // a lot worse than it should be.
+      SDValue Zero = DAG.getConstant(0, dl, SrcVT);
----------------
Can you raise a bug+testcase for this please?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D71956/new/

https://reviews.llvm.org/D71956