[llvm] [AMDGPU] Fix vector legalization for bf16 valu ops (PR #158439)
Matt Arsenault via llvm-commits
llvm-commits at lists.llvm.org
Sun Sep 14 23:45:29 PDT 2025
================
@@ -607,6 +607,8 @@ InstructionCost GCNTTIImpl::getArithmeticInstrCost(
case ISD::FSUB:
if (ST->hasPackedFP32Ops() && SLT == MVT::f32)
NElts = (NElts + 1) / 2;
+ if (ST->hasBF16PackedInsts() && SLT == MVT::bf16)
----------------
arsenm wrote:
F16 case here missing is probably a bug. We should also probably be expressing this in terms of a legality query on the 2 x element
https://github.com/llvm/llvm-project/pull/158439
More information about the llvm-commits
mailing list