Force to predicate a reduction operation when fold by EVL.

Author: Mel-Chen

llvm/lib/Transforms/Vectorize/LoopVectorize.cpp

@@ -1515,6 +1515,17 @@ class LoopVectorizationCostModel {
     return InLoopReductions.contains(Phi);
   }
 
+  /// Returns true if the predicated reduction select should be used.
+  bool usePredicatedReductionSelect(unsigned Opcode, Type *PhiTy) const {
+    // Force to use predicated reduction select since the EVL of the
+    // second-to-last iteration might not be VF*UF.
+    if (foldTailWithEVL())
+      return true;
+    return PreferPredicatedReductionSelect ||
+           TTI.preferPredicatedReductionSelect(
+               Opcode, PhiTy, TargetTransformInfo::ReductionFlags());
+  }
+
   /// Estimate cost of an intrinsic call instruction CI if it were vectorized
   /// with factor VF.  Return the cost of the instruction, including
   /// scalarization overhead if it's needed.
@@ -9344,10 +9355,8 @@ void LoopVectorizationPlanner::adjustRecipesForReductions(
                cast<VPInstruction>(&U)->getOpcode() ==
                    VPInstruction::ComputeReductionResult;
       });
-      if (PreferPredicatedReductionSelect ||
-          TTI.preferPredicatedReductionSelect(
-              PhiR->getRecurrenceDescriptor().getOpcode(), PhiTy,
-              TargetTransformInfo::ReductionFlags()))
+      if (CM.usePredicatedReductionSelect(
+              PhiR->getRecurrenceDescriptor().getOpcode(), PhiTy))
         PhiR->setOperand(1, NewExitingVPV);
     }
 

llvm/test/Transforms/LoopVectorize/RISCV/inloop-reduction.ll

@@ -136,16 +136,16 @@ define i32 @add_i16_i32(ptr nocapture readonly %x, i32 %n) {
 ; IF-EVL-OUTLOOP:       vector.body:
 ; IF-EVL-OUTLOOP-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; IF-EVL-OUTLOOP-NEXT:    [[EVL_BASED_IV:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_EVL_NEXT:%.*]], [[VECTOR_BODY]] ]
-; IF-EVL-OUTLOOP-NEXT:    [[VEC_PHI:%.*]] = phi <vscale x 4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP11:%.*]], [[VECTOR_BODY]] ]
+; IF-EVL-OUTLOOP-NEXT:    [[VEC_PHI:%.*]] = phi <vscale x 4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP12:%.*]], [[VECTOR_BODY]] ]
 ; IF-EVL-OUTLOOP-NEXT:    [[TMP5:%.*]] = sub i32 [[N]], [[EVL_BASED_IV]]
 ; IF-EVL-OUTLOOP-NEXT:    [[TMP6:%.*]] = call i32 @llvm.experimental.get.vector.length.i32(i32 [[TMP5]], i32 4, i1 true)
 ; IF-EVL-OUTLOOP-NEXT:    [[TMP7:%.*]] = add i32 [[EVL_BASED_IV]], 0
 ; IF-EVL-OUTLOOP-NEXT:    [[TMP8:%.*]] = getelementptr inbounds i16, ptr [[X:%.*]], i32 [[TMP7]]
 ; IF-EVL-OUTLOOP-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i16, ptr [[TMP8]], i32 0
 ; IF-EVL-OUTLOOP-NEXT:    [[VP_OP_LOAD:%.*]] = call <vscale x 4 x i16> @llvm.vp.load.nxv4i16.p0(ptr align 2 [[TMP9]], <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), i32 [[TMP6]])
 ; IF-EVL-OUTLOOP-NEXT:    [[TMP10:%.*]] = sext <vscale x 4 x i16> [[VP_OP_LOAD]] to <vscale x 4 x i32>
-; IF-EVL-OUTLOOP-NEXT:    [[TMP11]] = add <vscale x 4 x i32> [[VEC_PHI]], [[TMP10]]
-; IF-EVL-OUTLOOP-NEXT:    [[TMP12:%.*]] = call <vscale x 4 x i32> @llvm.vp.merge.nxv4i32(<vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> [[TMP11]], <vscale x 4 x i32> [[VEC_PHI]], i32 [[TMP6]])
+; IF-EVL-OUTLOOP-NEXT:    [[TMP11:%.*]] = add <vscale x 4 x i32> [[VEC_PHI]], [[TMP10]]
+; IF-EVL-OUTLOOP-NEXT:    [[TMP12]] = call <vscale x 4 x i32> @llvm.vp.merge.nxv4i32(<vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> [[TMP11]], <vscale x 4 x i32> [[VEC_PHI]], i32 [[TMP6]])
 ; IF-EVL-OUTLOOP-NEXT:    [[INDEX_EVL_NEXT]] = add i32 [[TMP6]], [[EVL_BASED_IV]]
 ; IF-EVL-OUTLOOP-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], [[TMP4]]
 ; IF-EVL-OUTLOOP-NEXT:    [[TMP13:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
@@ -350,19 +350,60 @@ define i32 @smin(ptr %a, i64 %n, i32 %start) {
 ;
 ; IF-EVL-OUTLOOP-LABEL: @smin(
 ; IF-EVL-OUTLOOP-NEXT:  entry:
+; IF-EVL-OUTLOOP-NEXT:    [[TMP0:%.*]] = sub i64 -1, [[N:%.*]]
+; IF-EVL-OUTLOOP-NEXT:    [[TMP1:%.*]] = call i64 @llvm.vscale.i64()
+; IF-EVL-OUTLOOP-NEXT:    [[TMP2:%.*]] = mul i64 [[TMP1]], 4
+; IF-EVL-OUTLOOP-NEXT:    [[TMP3:%.*]] = icmp ult i64 [[TMP0]], [[TMP2]]
+; IF-EVL-OUTLOOP-NEXT:    br i1 [[TMP3]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; IF-EVL-OUTLOOP:       vector.ph:
+; IF-EVL-OUTLOOP-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
+; IF-EVL-OUTLOOP-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 4
+; IF-EVL-OUTLOOP-NEXT:    [[TMP6:%.*]] = sub i64 [[TMP5]], 1
+; IF-EVL-OUTLOOP-NEXT:    [[N_RND_UP:%.*]] = add i64 [[N]], [[TMP6]]
+; IF-EVL-OUTLOOP-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP5]]
+; IF-EVL-OUTLOOP-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
+; IF-EVL-OUTLOOP-NEXT:    [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
+; IF-EVL-OUTLOOP-NEXT:    [[TMP8:%.*]] = mul i64 [[TMP7]], 4
+; IF-EVL-OUTLOOP-NEXT:    [[MINMAX_IDENT_SPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[START:%.*]], i64 0
+; IF-EVL-OUTLOOP-NEXT:    [[MINMAX_IDENT_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[MINMAX_IDENT_SPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
+; IF-EVL-OUTLOOP-NEXT:    br label [[VECTOR_BODY:%.*]]
+; IF-EVL-OUTLOOP:       vector.body:
+; IF-EVL-OUTLOOP-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; IF-EVL-OUTLOOP-NEXT:    [[EVL_BASED_IV:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_EVL_NEXT:%.*]], [[VECTOR_BODY]] ]
+; IF-EVL-OUTLOOP-NEXT:    [[VEC_PHI:%.*]] = phi <vscale x 4 x i32> [ [[MINMAX_IDENT_SPLAT]], [[VECTOR_PH]] ], [ [[TMP16:%.*]], [[VECTOR_BODY]] ]
+; IF-EVL-OUTLOOP-NEXT:    [[TMP9:%.*]] = sub i64 [[N]], [[EVL_BASED_IV]]
+; IF-EVL-OUTLOOP-NEXT:    [[TMP10:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[TMP9]], i32 4, i1 true)
+; IF-EVL-OUTLOOP-NEXT:    [[TMP11:%.*]] = add i64 [[EVL_BASED_IV]], 0
+; IF-EVL-OUTLOOP-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[TMP11]]
+; IF-EVL-OUTLOOP-NEXT:    [[TMP13:%.*]] = getelementptr inbounds i32, ptr [[TMP12]], i32 0
+; IF-EVL-OUTLOOP-NEXT:    [[VP_OP_LOAD:%.*]] = call <vscale x 4 x i32> @llvm.vp.load.nxv4i32.p0(ptr align 4 [[TMP13]], <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), i32 [[TMP10]])
+; IF-EVL-OUTLOOP-NEXT:    [[TMP14:%.*]] = icmp slt <vscale x 4 x i32> [[VP_OP_LOAD]], [[VEC_PHI]]
+; IF-EVL-OUTLOOP-NEXT:    [[TMP15:%.*]] = select <vscale x 4 x i1> [[TMP14]], <vscale x 4 x i32> [[VP_OP_LOAD]], <vscale x 4 x i32> [[VEC_PHI]]
+; IF-EVL-OUTLOOP-NEXT:    [[TMP16]] = call <vscale x 4 x i32> @llvm.vp.merge.nxv4i32(<vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> [[TMP15]], <vscale x 4 x i32> [[VEC_PHI]], i32 [[TMP10]])
+; IF-EVL-OUTLOOP-NEXT:    [[TMP17:%.*]] = zext i32 [[TMP10]] to i64
+; IF-EVL-OUTLOOP-NEXT:    [[INDEX_EVL_NEXT]] = add i64 [[TMP17]], [[EVL_BASED_IV]]
+; IF-EVL-OUTLOOP-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP8]]
+; IF-EVL-OUTLOOP-NEXT:    [[TMP18:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; IF-EVL-OUTLOOP-NEXT:    br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
+; IF-EVL-OUTLOOP:       middle.block:
+; IF-EVL-OUTLOOP-NEXT:    [[TMP19:%.*]] = call i32 @llvm.vector.reduce.smin.nxv4i32(<vscale x 4 x i32> [[TMP16]])
+; IF-EVL-OUTLOOP-NEXT:    br i1 true, label [[FOR_END:%.*]], label [[SCALAR_PH]]
+; IF-EVL-OUTLOOP:       scalar.ph:
+; IF-EVL-OUTLOOP-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; IF-EVL-OUTLOOP-NEXT:    [[BC_MERGE_RDX:%.*]] = phi i32 [ [[TMP19]], [[MIDDLE_BLOCK]] ], [ [[START]], [[ENTRY]] ]
 ; IF-EVL-OUTLOOP-NEXT:    br label [[FOR_BODY:%.*]]
 ; IF-EVL-OUTLOOP:       for.body:
-; IF-EVL-OUTLOOP-NEXT:    [[IV:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
-; IF-EVL-OUTLOOP-NEXT:    [[RDX:%.*]] = phi i32 [ [[START:%.*]], [[ENTRY]] ], [ [[SMIN:%.*]], [[FOR_BODY]] ]
-; IF-EVL-OUTLOOP-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[IV]]
-; IF-EVL-OUTLOOP-NEXT:    [[TMP0:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
-; IF-EVL-OUTLOOP-NEXT:    [[CMP_I:%.*]] = icmp slt i32 [[TMP0]], [[RDX]]
-; IF-EVL-OUTLOOP-NEXT:    [[SMIN]] = select i1 [[CMP_I]], i32 [[TMP0]], i32 [[RDX]]
+; IF-EVL-OUTLOOP-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
+; IF-EVL-OUTLOOP-NEXT:    [[RDX:%.*]] = phi i32 [ [[BC_MERGE_RDX]], [[SCALAR_PH]] ], [ [[SMIN:%.*]], [[FOR_BODY]] ]
+; IF-EVL-OUTLOOP-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[IV]]
+; IF-EVL-OUTLOOP-NEXT:    [[TMP20:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
+; IF-EVL-OUTLOOP-NEXT:    [[CMP_I:%.*]] = icmp slt i32 [[TMP20]], [[RDX]]
+; IF-EVL-OUTLOOP-NEXT:    [[SMIN]] = select i1 [[CMP_I]], i32 [[TMP20]], i32 [[RDX]]
 ; IF-EVL-OUTLOOP-NEXT:    [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
-; IF-EVL-OUTLOOP-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV_NEXT]], [[N:%.*]]
-; IF-EVL-OUTLOOP-NEXT:    br i1 [[EXITCOND_NOT]], label [[FOR_END:%.*]], label [[FOR_BODY]]
+; IF-EVL-OUTLOOP-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV_NEXT]], [[N]]
+; IF-EVL-OUTLOOP-NEXT:    br i1 [[EXITCOND_NOT]], label [[FOR_END]], label [[FOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
 ; IF-EVL-OUTLOOP:       for.end:
-; IF-EVL-OUTLOOP-NEXT:    [[SMIN_LCSSA:%.*]] = phi i32 [ [[SMIN]], [[FOR_BODY]] ]
+; IF-EVL-OUTLOOP-NEXT:    [[SMIN_LCSSA:%.*]] = phi i32 [ [[SMIN]], [[FOR_BODY]] ], [ [[TMP19]], [[MIDDLE_BLOCK]] ]
 ; IF-EVL-OUTLOOP-NEXT:    ret i32 [[SMIN_LCSSA]]
 ;
 ; IF-EVL-INLOOP-LABEL: @smin(

llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-cond-reduction.ll

@@ -53,15 +53,15 @@ define i32 @cond_add(ptr %a, i64 %n, i32 %start) {
 ; IF-EVL-OUTLOOP-NEXT:    [[VP_OP_LOAD:%.*]] = call <vscale x 4 x i32> @llvm.vp.load.nxv4i32.p0(ptr align 4 [[TMP17]], <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), i32 [[TMP11]])
 ; IF-EVL-OUTLOOP-NEXT:    [[TMP18:%.*]] = icmp sgt <vscale x 4 x i32> [[VP_OP_LOAD]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 3, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
 ; IF-EVL-OUTLOOP-NEXT:    [[TMP19:%.*]] = select <vscale x 4 x i1> [[TMP18]], <vscale x 4 x i32> [[VP_OP_LOAD]], <vscale x 4 x i32> zeroinitializer
-; IF-EVL-OUTLOOP-NEXT:    [[TMP20]] = add <vscale x 4 x i32> [[TMP19]], [[VEC_PHI]]
-; IF-EVL-OUTLOOP-NEXT:    [[TMP21:%.*]] = call <vscale x 4 x i32> @llvm.vp.merge.nxv4i32(<vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> [[TMP20]], <vscale x 4 x i32> [[VEC_PHI]], i32 [[TMP11]])
+; IF-EVL-OUTLOOP-NEXT:    [[TMP21:%.*]] = add <vscale x 4 x i32> [[TMP19]], [[VEC_PHI]]
+; IF-EVL-OUTLOOP-NEXT:    [[TMP20]] = call <vscale x 4 x i32> @llvm.vp.merge.nxv4i32(<vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> [[TMP21]], <vscale x 4 x i32> [[VEC_PHI]], i32 [[TMP11]])
 ; IF-EVL-OUTLOOP-NEXT:    [[TMP22:%.*]] = zext i32 [[TMP11]] to i64
 ; IF-EVL-OUTLOOP-NEXT:    [[INDEX_EVL_NEXT1]] = add i64 [[TMP22]], [[EVL_BASED_IV1]]
 ; IF-EVL-OUTLOOP-NEXT:    [[INDEX_EVL_NEXT]] = add i64 [[EVL_BASED_IV]], [[TMP8]]
 ; IF-EVL-OUTLOOP-NEXT:    [[TMP23:%.*]] = icmp eq i64 [[INDEX_EVL_NEXT]], [[N_VEC]]
 ; IF-EVL-OUTLOOP-NEXT:    br i1 [[TMP23]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
 ; IF-EVL-OUTLOOP:       middle.block:
-; IF-EVL-OUTLOOP-NEXT:    [[TMP24:%.*]] = call i32 @llvm.vector.reduce.add.nxv4i32(<vscale x 4 x i32> [[TMP21]])
+; IF-EVL-OUTLOOP-NEXT:    [[TMP24:%.*]] = call i32 @llvm.vector.reduce.add.nxv4i32(<vscale x 4 x i32> [[TMP20]])
 ; IF-EVL-OUTLOOP-NEXT:    br i1 true, label [[FOR_END:%.*]], label [[SCALAR_PH]]
 ; IF-EVL-OUTLOOP:       scalar.ph:
 ; IF-EVL-OUTLOOP-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
@@ -305,15 +305,15 @@ define i32 @cond_add_pred(ptr %a, i64 %n, i32 %start) {
 ; IF-EVL-OUTLOOP-NEXT:    [[TMP19:%.*]] = add <vscale x 4 x i32> [[VEC_PHI]], [[VP_OP_LOAD]]
 ; IF-EVL-OUTLOOP-NEXT:    [[TMP20:%.*]] = xor <vscale x 4 x i1> [[TMP18]], shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer)
 ; IF-EVL-OUTLOOP-NEXT:    [[TMP21:%.*]] = select <vscale x 4 x i1> [[TMP15]], <vscale x 4 x i1> [[TMP20]], <vscale x 4 x i1> zeroinitializer
-; IF-EVL-OUTLOOP-NEXT:    [[PREDPHI]] = select <vscale x 4 x i1> [[TMP21]], <vscale x 4 x i32> [[VEC_PHI]], <vscale x 4 x i32> [[TMP19]]
-; IF-EVL-OUTLOOP-NEXT:    [[TMP22:%.*]] = call <vscale x 4 x i32> @llvm.vp.merge.nxv4i32(<vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> [[PREDPHI]], <vscale x 4 x i32> [[VEC_PHI]], i32 [[TMP11]])
+; IF-EVL-OUTLOOP-NEXT:    [[PREDPHI1:%.*]] = select <vscale x 4 x i1> [[TMP21]], <vscale x 4 x i32> [[VEC_PHI]], <vscale x 4 x i32> [[TMP19]]
+; IF-EVL-OUTLOOP-NEXT:    [[PREDPHI]] = call <vscale x 4 x i32> @llvm.vp.merge.nxv4i32(<vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> [[PREDPHI1]], <vscale x 4 x i32> [[VEC_PHI]], i32 [[TMP11]])
 ; IF-EVL-OUTLOOP-NEXT:    [[TMP23:%.*]] = zext i32 [[TMP11]] to i64
 ; IF-EVL-OUTLOOP-NEXT:    [[INDEX_EVL_NEXT]] = add i64 [[TMP23]], [[EVL_BASED_IV]]
 ; IF-EVL-OUTLOOP-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP8]]
 ; IF-EVL-OUTLOOP-NEXT:    [[TMP24:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
 ; IF-EVL-OUTLOOP-NEXT:    br i1 [[TMP24]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
 ; IF-EVL-OUTLOOP:       middle.block:
-; IF-EVL-OUTLOOP-NEXT:    [[TMP27:%.*]] = call i32 @llvm.vector.reduce.add.nxv4i32(<vscale x 4 x i32> [[TMP22]])
+; IF-EVL-OUTLOOP-NEXT:    [[TMP27:%.*]] = call i32 @llvm.vector.reduce.add.nxv4i32(<vscale x 4 x i32> [[PREDPHI]])
 ; IF-EVL-OUTLOOP-NEXT:    br i1 true, label [[FOR_END:%.*]], label [[SCALAR_PH]]
 ; IF-EVL-OUTLOOP:       scalar.ph:
 ; IF-EVL-OUTLOOP-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]

llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-inloop-reduction.ll

@@ -1804,16 +1804,16 @@ define i32 @anyof_icmp(ptr %a, i64 %n, i32 %start, i32 %inv) {
 ; IF-EVL:       vector.body:
 ; IF-EVL-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; IF-EVL-NEXT:    [[EVL_BASED_IV:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_EVL_NEXT:%.*]], [[VECTOR_BODY]] ]
-; IF-EVL-NEXT:    [[VEC_PHI:%.*]] = phi <vscale x 4 x i1> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP15:%.*]], [[VECTOR_BODY]] ]
+; IF-EVL-NEXT:    [[VEC_PHI:%.*]] = phi <vscale x 4 x i1> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP16:%.*]], [[VECTOR_BODY]] ]
 ; IF-EVL-NEXT:    [[TMP9:%.*]] = sub i64 [[N]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[TMP10:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[TMP9]], i32 4, i1 true)
 ; IF-EVL-NEXT:    [[TMP11:%.*]] = add i64 [[EVL_BASED_IV]], 0
 ; IF-EVL-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[TMP11]]
 ; IF-EVL-NEXT:    [[TMP13:%.*]] = getelementptr inbounds i32, ptr [[TMP12]], i32 0
 ; IF-EVL-NEXT:    [[VP_OP_LOAD:%.*]] = call <vscale x 4 x i32> @llvm.vp.load.nxv4i32.p0(ptr align 4 [[TMP13]], <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), i32 [[TMP10]])
 ; IF-EVL-NEXT:    [[TMP14:%.*]] = icmp slt <vscale x 4 x i32> [[VP_OP_LOAD]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 3, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
-; IF-EVL-NEXT:    [[TMP15]] = or <vscale x 4 x i1> [[VEC_PHI]], [[TMP14]]
-; IF-EVL-NEXT:    [[TMP16:%.*]] = call <vscale x 4 x i1> @llvm.vp.merge.nxv4i1(<vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i1> [[TMP15]], <vscale x 4 x i1> [[VEC_PHI]], i32 [[TMP10]])
+; IF-EVL-NEXT:    [[TMP15:%.*]] = or <vscale x 4 x i1> [[VEC_PHI]], [[TMP14]]
+; IF-EVL-NEXT:    [[TMP16]] = call <vscale x 4 x i1> @llvm.vp.merge.nxv4i1(<vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i1> [[TMP15]], <vscale x 4 x i1> [[VEC_PHI]], i32 [[TMP10]])
 ; IF-EVL-NEXT:    [[TMP17:%.*]] = zext i32 [[TMP10]] to i64
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add i64 [[TMP17]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP8]]
@@ -1931,16 +1931,16 @@ define i32 @anyof_fcmp(ptr %a, i64 %n, i32 %start, i32 %inv) {
 ; IF-EVL:       vector.body:
 ; IF-EVL-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; IF-EVL-NEXT:    [[EVL_BASED_IV:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_EVL_NEXT:%.*]], [[VECTOR_BODY]] ]
-; IF-EVL-NEXT:    [[VEC_PHI:%.*]] = phi <vscale x 4 x i1> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP15:%.*]], [[VECTOR_BODY]] ]
+; IF-EVL-NEXT:    [[VEC_PHI:%.*]] = phi <vscale x 4 x i1> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP16:%.*]], [[VECTOR_BODY]] ]
 ; IF-EVL-NEXT:    [[TMP9:%.*]] = sub i64 [[N]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[TMP10:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[TMP9]], i32 4, i1 true)
 ; IF-EVL-NEXT:    [[TMP11:%.*]] = add i64 [[EVL_BASED_IV]], 0
 ; IF-EVL-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[TMP11]]
 ; IF-EVL-NEXT:    [[TMP13:%.*]] = getelementptr inbounds float, ptr [[TMP12]], i32 0
 ; IF-EVL-NEXT:    [[VP_OP_LOAD:%.*]] = call <vscale x 4 x float> @llvm.vp.load.nxv4f32.p0(ptr align 4 [[TMP13]], <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), i32 [[TMP10]])
 ; IF-EVL-NEXT:    [[TMP14:%.*]] = fcmp fast olt <vscale x 4 x float> [[VP_OP_LOAD]], shufflevector (<vscale x 4 x float> insertelement (<vscale x 4 x float> poison, float 3.000000e+00, i64 0), <vscale x 4 x float> poison, <vscale x 4 x i32> zeroinitializer)
-; IF-EVL-NEXT:    [[TMP15]] = or <vscale x 4 x i1> [[VEC_PHI]], [[TMP14]]
-; IF-EVL-NEXT:    [[TMP16:%.*]] = call <vscale x 4 x i1> @llvm.vp.merge.nxv4i1(<vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i1> [[TMP15]], <vscale x 4 x i1> [[VEC_PHI]], i32 [[TMP10]])
+; IF-EVL-NEXT:    [[TMP15:%.*]] = or <vscale x 4 x i1> [[VEC_PHI]], [[TMP14]]
+; IF-EVL-NEXT:    [[TMP16]] = call <vscale x 4 x i1> @llvm.vp.merge.nxv4i1(<vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i1> [[TMP15]], <vscale x 4 x i1> [[VEC_PHI]], i32 [[TMP10]])
 ; IF-EVL-NEXT:    [[TMP17:%.*]] = zext i32 [[TMP10]] to i64
 ; IF-EVL-NEXT:    [[INDEX_EVL_NEXT]] = add i64 [[TMP17]], [[EVL_BASED_IV]]
 ; IF-EVL-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP8]]