[llvm] [VPlan] Introduce CSE pass (PR #151872)

Florian Hahn via llvm-commits llvm-commits at lists.llvm.org
Mon Aug 4 04:37:15 PDT 2025


================
@@ -122,6 +122,49 @@ void VPDef::dump() const {
 }
 #endif
 
+bool VPValue::isIdenticalTo(const VPValue *Other) const {
+  if (getVPValueID() != Other->getVPValueID() ||
+      hasDefiningRecipe() != Other->hasDefiningRecipe() ||
+      !getUnderlyingValue() != !Other->getUnderlyingValue())
+    return false;
+
+  auto *I = dyn_cast_or_null<Instruction>(getUnderlyingValue());
+  auto *OtherI = dyn_cast_or_null<Instruction>(Other->getUnderlyingValue());
+  if (I && OtherI)
+    return I->isIdenticalTo(OtherI);
----------------
fhahn wrote:

We should not look at the underlying instruction here. There are multiple cases where this will lead to incorrect results:
 * recipe operands have been modified
 * multiple recipes can have the same underlying instruction after unrolling, recipes for all parts will have the same underlying instruction, but produce values for different parts

https://github.com/llvm/llvm-project/pull/151872


More information about the llvm-commits mailing list