[llvm] [LLVM][CodeGen] Add lowering for scalable vector bfloat operations. (PR #109803)
Matt Arsenault via llvm-commits
llvm-commits at lists.llvm.org
Wed Sep 25 09:12:22 PDT 2024
================
@@ -1070,13 +1071,21 @@ void VectorLegalizer::Expand(SDNode *Node, SmallVectorImpl<SDValue> &Results) {
break;
case ISD::FMINNUM:
case ISD::FMAXNUM:
+ if (SDValue Expanded = ExpandBF16Arith(Node)) {
----------------
arsenm wrote:
Probably shouldn't just bypass the dedicated function to lower something
https://github.com/llvm/llvm-project/pull/109803
More information about the llvm-commits
mailing list