[VPlan] Fold PredPHI constant -> constant

Author: artagnon

llvm/lib/Transforms/Vectorize/VPlan.cpp

@@ -317,13 +317,7 @@ Value *VPTransformState::get(const VPValue *Def, bool NeedsScalar) {
     LastLane = 0;
   }
 
-  auto *LastInst = dyn_cast<Instruction>(get(Def, LastLane));
-  if (!LastInst) {
-    Value *VectorValue = GetBroadcastInstrs(ScalarValue);
-    set(Def, VectorValue);
-    return VectorValue;
-  }
-
+  auto *LastInst = cast<Instruction>(get(Def, LastLane));
   // Set the insert point after the last scalarized instruction or after the
   // last PHI, if LastInst is a PHI. This ensures the insertelement sequence
   // will directly follow the scalar definitions.

llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp

@@ -2804,15 +2804,8 @@ InstructionCost VPBranchOnMaskRecipe::computeCost(ElementCount VF,
 
 void VPPredInstPHIRecipe::execute(VPTransformState &State) {
   assert(State.Lane && "Predicated instruction PHI works per instance.");
-  Value *ScalarPred = State.get(getOperand(0), *State.Lane);
-  Instruction *ScalarPredInst = dyn_cast<Instruction>(ScalarPred);
-  if (!ScalarPredInst) {
-    if (State.hasScalarValue(this, *State.Lane))
-      State.reset(this, ScalarPred, *State.Lane);
-    else
-      State.set(this, ScalarPred, *State.Lane);
-    return;
-  }
+  Instruction *ScalarPredInst =
+      cast<Instruction>(State.get(getOperand(0), *State.Lane));
   BasicBlock *PredicatedBB = ScalarPredInst->getParent();
   BasicBlock *PredicatingBB = PredicatedBB->getSinglePredecessor();
   assert(PredicatingBB && "Predicated block has no single predecessor.");

llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp

@@ -1011,6 +1011,16 @@ static void simplifyRecipe(VPRecipeBase &R, VPTypeAnalysis &TypeInfo) {
           .Default([](auto *) { return false; }))
     return;
 
+  // Fold PredPHI constant -> constant.
+  if (auto *PredPHI = dyn_cast<VPPredInstPHIRecipe>(&R)) {
+    VPlan *Plan = R.getParent()->getPlan();
+    VPValue *Op = PredPHI->getOperand(0);
+    if (!Op->isLiveIn() || !Op->getLiveInIRValue())
+      return;
+    if (auto *C = dyn_cast<Constant>(Op->getLiveInIRValue()))
+      PredPHI->replaceAllUsesWith(Plan->getOrAddLiveIn(C));
+  }
+
   // VPScalarIVSteps can only be simplified after unrolling. VPScalarIVSteps for
   // part 0 can be replaced by their start value, if only the first lane is
   // demanded.

llvm/test/Transforms/LoopVectorize/X86/drop-poison-generating-flags.ll

@@ -690,8 +690,7 @@ define void @pr70590_recipe_without_underlying_instr(i64 %n, ptr noalias %dst) {
 ; CHECK:       [[PRED_SREM_IF5]]:
 ; CHECK-NEXT:    br label %[[PRED_SREM_CONTINUE6]]
 ; CHECK:       [[PRED_SREM_CONTINUE6]]:
-; CHECK-NEXT:    [[TMP11:%.*]] = add i64 poison, -3
-; CHECK-NEXT:    [[TMP12:%.*]] = add i64 [[INDEX]], [[TMP11]]
+; CHECK-NEXT:    [[TMP12:%.*]] = add i64 [[INDEX]], poison
 ; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr [5 x i8], ptr @c, i64 0, i64 [[TMP12]]
 ; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr i8, ptr [[TMP13]], i32 0
 ; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x i8>, ptr [[TMP14]], align 1
@@ -768,8 +767,7 @@ define void @recipe_without_underlying_instr_lanes_used(i64 %n, ptr noalias %dst
 ; CHECK:       [[PRED_SREM_IF5]]:
 ; CHECK-NEXT:    br label %[[PRED_SREM_CONTINUE6]]
 ; CHECK:       [[PRED_SREM_CONTINUE6]]:
-; CHECK-NEXT:    [[TMP6:%.*]] = add i64 poison, -3
-; CHECK-NEXT:    [[TMP7:%.*]] = add i64 [[INDEX]], [[TMP6]]
+; CHECK-NEXT:    [[TMP7:%.*]] = add i64 [[INDEX]], poison
 ; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr [5 x i8], ptr @c, i64 0, i64 [[TMP7]]
 ; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr i8, ptr [[TMP8]], i32 0
 ; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x i8>, ptr [[TMP9]], align 1