[llvm] [CGP]: Optimize mul.overflow. (PR #148343)

[Hassnaa Hamdi via llvm-commits]

================
@@ -6403,6 +6409,229 @@ bool CodeGenPrepare::optimizeGatherScatterInst(Instruction *MemoryInst,
   return true;
 }
 
+// Rewrite the umul_with_overflow intrinsic by checking if both of the
+// operands' value range is within the legal type. If so, we can optimize the
+// multiplication algorithm. This code is supposed to be written during the step
+// of type legalization, but given that we need to reconstruct the IR which is
+// not doable there, we do it here.
+bool CodeGenPrepare::optimizeUMulWithOverflow(Instruction *I) {
+  // Enable this optimization only for aarch64.
+  if (!TLI->getTargetMachine().getTargetTriple().isAArch64())
+    return false;
+  if (TLI->getTypeAction(
+          I->getContext(),
+          TLI->getValueType(*DL, I->getType()->getContainedType(0))) !=
+      TargetLowering::TypeExpandInteger)
+    return false;
+
+  Value *LHS = I->getOperand(0);
+  Value *RHS = I->getOperand(1);
+  auto *Ty = LHS->getType();
+  unsigned VTBitWidth = Ty->getScalarSizeInBits();
+  unsigned VTHalfBitWidth = VTBitWidth / 2;
+  auto *LegalTy = IntegerType::getIntNTy(I->getContext(), VTHalfBitWidth);
+
+  // Skip the optimization if the type with HalfBitWidth is not legal for the
+  // target.
+  if (TLI->getTypeAction(I->getContext(), TLI->getValueType(*DL, LegalTy)) !=
+      TargetLowering::TypeLegal)
+    return false;
+
+  I->getParent()->setName("overflow.res");
+  auto *OverflowResBB = I->getParent();
+  auto *OverflowoEntryBB =
+      I->getParent()->splitBasicBlock(I, "overflow.entry", /*Before*/ true);
+  BasicBlock *NoOverflowBB = BasicBlock::Create(
+      I->getContext(), "overflow.no", I->getFunction(), OverflowResBB);
+  BasicBlock *OverflowBB = BasicBlock::Create(I->getContext(), "overflow",
+                                              I->getFunction(), OverflowResBB);
+  // new blocks should be:
+  //  entry:
+  //    (lhs_lo ne lhs_hi) || (rhs_lo ne rhs_hi) ? overflow, overflow_no
+
+  //  overflow_no:
+  //  overflow:
+  //  overflow.res:
+  //------------------------------------------------------------------------------
----------------
hassnaaHamdi wrote:

done

https://github.com/llvm/llvm-project/pull/148343