[llvm] [VectorCombine] Add type shrinking and zext propagation for fixed-width vector types (PR #104606)

Simon Pilgrim via llvm-commits llvm-commits at lists.llvm.org
Fri Aug 16 08:37:16 PDT 2024


================
@@ -2493,6 +2494,106 @@ bool VectorCombine::foldSelectShuffle(Instruction &I, bool FromReduction) {
   return true;
 }
 
+/// Check if instruction depends on ZExt and this ZExt can be moved after the
+/// instruction. Move ZExt if it is profitable
+bool VectorCombine::shrinkType(llvm::Instruction &I) {
+  Value *ZExted, *OtherOperand;
+  if (match(&I, m_c_BinOp(m_ZExt(m_Value(ZExted)), m_Value(OtherOperand)))) {
+    if (I.getOpcode() != Instruction::And && I.getOpcode() != Instruction::Or &&
+        I.getOpcode() != Instruction::Xor && I.getOpcode() != Instruction::LShr)
+      return false;
----------------
RKSimon wrote:

You might be able to clean this up as:
```c
if (!match(&I, m_c_BitwiseLogic(m_ZExt(m_Value(ZExted)), m_Value(OtherOperand))) &&
    !match(&I, m_LShr(m_ZExt(m_Value(ZExted)), m_Value(OtherOperand))))
  return false;
```

https://github.com/llvm/llvm-project/pull/104606


More information about the llvm-commits mailing list