[llvm-branch-commits] [llvm] [AArch64] SLP can vectorize frem (PR #82488)

[Alexey Bataev via llvm-branch-commits]

================
@@ -869,6 +870,18 @@ TargetTransformInfo::getOperandInfo(const Value *V) {
   return {OpInfo, OpProps};
 }
 
+InstructionCost TargetTransformInfo::getVecLibCallCost(
+    const int OpCode, const TargetLibraryInfo *TLI, VectorType *VecTy,
+    TTI::TargetCostKind CostKind) {
+  Type *ScalarTy = VecTy->getScalarType();
+  LibFunc Func;
+  if (TLI->getLibFunc(OpCode, ScalarTy, Func) &&
+      TLI->isFunctionVectorizable(TLI->getName(Func), VecTy->getElementCount()))
----------------
alexey-bataev wrote:

Yes, and what I'm saying that TTI should handle this since this is target-specific lowering.
If you don't want to propagate TLI, then just copy the legality checks from TLI to TTI and the cost estimation for this specific lowering to the getArithmeticInstructionCost function. This is what we usually do in similar cases for other targets.

https://github.com/llvm/llvm-project/pull/82488