[clang] [llvm] [AArch64] Implements FP8 SVE intrinsics for dot-product (PR #118125)

[Momchil Velikov via llvm-commits]

================
@@ -10719,7 +10719,16 @@ Value *CodeGenFunction::EmitSVEDupX(Value *Scalar, llvm::Type *Ty) {
       cast<llvm::VectorType>(Ty)->getElementCount(), Scalar);
 }
 
-Value *CodeGenFunction::EmitSVEDupX(Value* Scalar) {
+Value *CodeGenFunction::EmitSVEDupX(Value *Scalar) {
+  if (auto *Ty = Scalar->getType(); Ty->isVectorTy()) {
----------------
momchil-velikov wrote:

No, that would be inappropriate. It's an invariant, not an option.


https://github.com/llvm/llvm-project/pull/118125