[LoopVectorize] Test in-loop reductions with tail folding for scalable vectors

[kmclaughlin]

Summary

Adds -prefer-inloop-reductions to the RUN line of sve-tail-folding.ll & adds
a new test where in-loop reductions cannot be used (@cond_xor_reduction). NFC.

[david-arm]

LGTM!

llvm/test/Transforms/LoopVectorize/AArch64/sve-tail-folding.ll

658

nit: I think you'll need to rebase the patch before merging because we now use the get.active.lane.mask intrinsic to generate the loop predicate.

680

I think the reason why we don't use in-loop reductions here is because it's a conditional reduction, right?

682

nit: I think you can just delete everything from this line onwards because we'll never reach the scalar tail anyway.

[kmclaughlin]

llvm/test/Transforms/LoopVectorize/AArch64/sve-tail-folding.ll

680

Yes, this test changes to use in-loop reductions in D117580 which adds support for conditional reductions.

llvm/test/Transforms/LoopVectorize/AArch64/sve-tail-folding.ll

-; RUN: opt -S -hints-allow-reordering=false -loop-vectorize -prefer-predicate-over-epilogue=predicate-dont-vectorize < %s | FileCheck %s
-; RUN: opt -S -hints-allow-reordering=false -loop-vectorize -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue < %s | FileCheck %s
+; RUN: opt -S -hints-allow-reordering=false -loop-vectorize -prefer-predicate-over-epilogue=predicate-dont-vectorize -prefer-inloop-reductions < %s | FileCheck %s
+; RUN: opt -S -hints-allow-reordering=false -loop-vectorize -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue -prefer-inloop-reductions < %s | FileCheck %s
 
 ; CHECK-NOT: vector.body:
 
 target triple = "aarch64-unknown-linux-gnu"
 
 
 define void @simple_memset(i32 %val, i32* %ptr, i64 %n) #0 {
 ; CHECK-LABEL: @simple_memset(
while.end.loopexit:                               ; preds = %while.body
   ret void
 }
 
 
 define i32 @add_reduction_i32(i32* %ptr, i64 %n) #0 {
 ; CHECK-LABEL: @add_reduction_i32(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[UMAX:%.*]] = call i64 @llvm.umax.i64(i64 [[N:%.*]], i64 1)
-; CHECK-NEXT:    br i1 false, label %scalar.ph, label %vector.ph
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label %vector.ph
 ; CHECK:       vector.ph:
 ; CHECK-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
 ; CHECK-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 4
 ; CHECK-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
 ; CHECK-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 4
 ; CHECK-NEXT:    [[TMP4:%.*]] = sub i64 [[TMP3]], 1
 ; CHECK-NEXT:    [[N_RND_UP:%.*]] = add i64 [[UMAX]], [[TMP4]]
 ; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP1]]
 ; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
 ; CHECK-NEXT:    br label %vector.body
 ; CHECK:       vector.body:
 ; CHECK-NEXT:    [[INDEX1:%.*]] = phi i64 [ 0, %vector.ph ], [ [[INDEX_NEXT2:%.*]], %vector.body ]
-; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <vscale x 4 x i32> [ insertelement (<vscale x 4 x i32> zeroinitializer, i32 0, i32 0), %vector.ph ], [ [[TMP16:%.*]], %vector.body ]
-; CHECK-NEXT:    [[BROADCAST_SPLATINSERT3:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[INDEX1]], i32 0
-; CHECK-NEXT:    [[BROADCAST_SPLAT4:%.*]] = shufflevector <vscale x 4 x i64> [[BROADCAST_SPLATINSERT3]], <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
+; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi i32 [ 0, %vector.ph ], [ [[TMP14:%.*]], %vector.body ]
+; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[INDEX1]], i32 0
+; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 4 x i64> [[BROADCAST_SPLATINSERT]], <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    [[TMP5:%.*]] = call <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64()
 ; CHECK-NEXT:    [[TMP6:%.*]] = add <vscale x 4 x i64> [[TMP5]], zeroinitializer
 ; CHECK-NEXT:    [[TMP7:%.*]] = mul <vscale x 4 x i64> [[TMP6]], shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 1, i32 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer)
-; CHECK-NEXT:    [[INDUCTION:%.*]] = add <vscale x 4 x i64> [[BROADCAST_SPLAT4]], [[TMP7]]
+; CHECK-NEXT:    [[INDUCTION:%.*]] = add <vscale x 4 x i64> [[BROADCAST_SPLAT]], [[TMP7]]
 ; CHECK-NEXT:    [[TMP8:%.*]] = add i64 [[INDEX1]], 0
 ; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <vscale x 4 x i1> @llvm.get.active.lane.mask.nxv4i1.i64(i64 [[TMP8]], i64 [[UMAX]])
-; CHECK-NEXT:    [[TMP10:%.*]] = getelementptr i32, i32* [[PTR:%.*]], i64 [[TMP8]]
-; CHECK-NEXT:    [[TMP11:%.*]] = getelementptr i32, i32* [[TMP10]], i32 0
-; CHECK-NEXT:    [[TMP12:%.*]] = bitcast i32* [[TMP11]] to <vscale x 4 x i32>*
-; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 4 x i32> @llvm.masked.load.nxv4i32.p0nxv4i32(<vscale x 4 x i32>* [[TMP12]], i32 4, <vscale x 4 x i1> [[ACTIVE_LANE_MASK]], <vscale x 4 x i32> poison)
-; CHECK-NEXT:    [[TMP13]] = add <vscale x 4 x i32> [[VEC_PHI]], [[WIDE_MASKED_LOAD]]
-; CHECK-NEXT:    [[TMP16:%.*]] = select <vscale x 4 x i1> [[ACTIVE_LANE_MASK]], <vscale x 4 x i32> [[TMP13]], <vscale x 4 x i32> [[VEC_PHI]]
-; CHECK-NEXT:    [[TMP14:%.*]] = call i64 @llvm.vscale.i64()
-; CHECK-NEXT:    [[TMP15:%.*]] = mul i64 [[TMP14]], 4
-; CHECK-NEXT:    [[INDEX_NEXT2]] = add i64 [[INDEX1]], [[TMP15]]
+; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr i32, i32* [[PTR:%.*]], i64 [[TMP8]]
+; CHECK-NEXT:    [[TMP10:%.*]] = getelementptr i32, i32* [[TMP9]], i32 0
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i32* [[TMP10]] to <vscale x 4 x i32>*
+; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 4 x i32> @llvm.masked.load.nxv4i32.p0nxv4i32(<vscale x 4 x i32>* [[TMP11]], i32 4, <vscale x 4 x i1> [[ACTIVE_LANE_MASK]], <vscale x 4 x i32> poison)
+; CHECK-NEXT:    [[TMP12:%.*]] = select <vscale x 4 x i1> [[ACTIVE_LANE_MASK]], <vscale x 4 x i32> [[WIDE_MASKED_LOAD]], <vscale x 4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP13:%.*]] = call i32 @llvm.vector.reduce.add.nxv4i32(<vscale x 4 x i32> [[TMP12]])
+; CHECK-NEXT:    [[TMP14]] = add i32 [[TMP13]], [[VEC_PHI]]
+; CHECK-NEXT:    [[TMP15:%.*]] = call i64 @llvm.vscale.i64()
+; CHECK-NEXT:    [[TMP16:%.*]] = mul i64 [[TMP15]], 4
+; CHECK-NEXT:    [[INDEX_NEXT2]] = add i64 [[INDEX1]], [[TMP16]]
 ; CHECK-NEXT:    [[TMP17:%.*]] = icmp eq i64 [[INDEX_NEXT2]], [[N_VEC]]
-; CHECK-NEXT:    br i1 [[TMP17]], label %middle.block, label %vector.body, !llvm.loop [[LOOP20:![0-9]+]]
+; CHECK-NEXT:    br i1 [[TMP17]], label [[MIDDLE_BLOCK:%.*]], label %vector.body, !llvm.loop [[LOOP18:![0-9]+]]
 ; CHECK:       middle.block:
-; CHECK-NEXT:    [[TMP18:%.*]] = call i32 @llvm.vector.reduce.add.nxv4i32(<vscale x 4 x i32> [[TMP16]])
-; CHECK-NEXT:    br i1 true, label %while.end.loopexit, label %scalar.ph
+; CHECK-NEXT:    br i1 true, label [[WHILE_END_LOOPEXIT:%.*]], label %scalar.ph
 ;
 entry:
   br label %while.body
 
 while.body:                                       ; preds = %while.body, %entry
   %index = phi i64 [ %index.next, %while.body ], [ 0, %entry ]
   %red = phi i32 [ %red.next, %while.body ], [ 0, %entry ]
   %gep = getelementptr i32, i32* %ptr, i64 %index
   %val = load i32, i32* %gep
   %red.next = add i32 %red, %val
   %index.next = add nsw i64 %index, 1
   %cmp10 = icmp ult i64 %index.next, %n
   br i1 %cmp10, label %while.body, label %while.end.loopexit, !llvm.loop !0
 
 while.end.loopexit:                               ; preds = %while.body
   ret i32 %red.next
 }
 
-
 define float @add_reduction_f32(float* %ptr, i64 %n) #0 {
 ; CHECK-LABEL: @add_reduction_f32(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[UMAX:%.*]] = call i64 @llvm.umax.i64(i64 [[N:%.*]], i64 1)
 ; CHECK-NEXT:    br i1 false, label %scalar.ph, label %vector.ph
 ; CHECK:       vector.ph:
 ; CHECK-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
 ; CHECK-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 4
while.body:                                       ; preds = %while.body, %entry
   %index.next = add nsw i64 %index, 1
   %cmp10 = icmp ult i64 %index.next, %n
   br i1 %cmp10, label %while.body, label %while.end.loopexit, !llvm.loop !0
 
 while.end.loopexit:                               ; preds = %while.body
   ret float %red.next
 }
 
+define i32 @cond_xor_reduction(i32* noalias %a, i32* noalias %cond, i64 %N) #0 {
+; CHECK-LABEL: @cond_xor_reduction(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label %scalar.ph, label %vector.ph
+; CHECK:       vector.ph:
+; CHECK-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
+; CHECK-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 4
+; CHECK-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
+; CHECK-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 4
+; CHECK-NEXT:    [[TMP4:%.*]] = sub i64 [[TMP3]], 1
+; CHECK-NEXT:    [[N_RND_UP:%.*]] = add i64 [[N:%.*]], [[TMP4]]
+; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP1]]
+; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
+; CHECK-NEXT:    br label %vector.body
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, %vector.ph ], [ [[INDEX_NEXT:%.*]], %vector.body ]
+; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <vscale x 4 x i32> [ insertelement (<vscale x 4 x i32> zeroinitializer, i32 7, i32 0), %vector.ph ], [ [[PREDPHI:%.*]], %vector.body ]
+; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[INDEX]], i32 0
+; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 4 x i64> [[BROADCAST_SPLATINSERT]], <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP5:%.*]] = call <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64()
+; CHECK-NEXT:    [[TMP6:%.*]] = add <vscale x 4 x i64> [[TMP5]], zeroinitializer
+; CHECK-NEXT:    [[TMP7:%.*]] = mul <vscale x 4 x i64> [[TMP6]], shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 1, i32 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer)
+; CHECK-NEXT:    [[INDUCTION:%.*]] = add <vscale x 4 x i64> [[BROADCAST_SPLAT]], [[TMP7]]
+; CHECK-NEXT:    [[TMP8:%.*]] = add i64 [[INDEX]], 0
+; CHECK-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = call <vscale x 4 x i1> @llvm.get.active.lane.mask.nxv4i1.i64(i64 [[TMP8]], i64 [[N]])
+; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i32, i32* [[COND:%.*]], i64 [[TMP8]]
+; CHECK-NEXT:    [[TMP10:%.*]] = getelementptr inbounds i32, i32* [[TMP9]], i32 0
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i32* [[TMP10]] to <vscale x 4 x i32>*
+; CHECK-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 4 x i32> @llvm.masked.load.nxv4i32.p0nxv4i32(<vscale x 4 x i32>* [[TMP11]], i32 4, <vscale x 4 x i1> [[ACTIVE_LANE_MASK]], <vscale x 4 x i32> poison)
+; CHECK-NEXT:    [[TMP12:%.*]] = icmp eq <vscale x 4 x i32> [[WIDE_MASKED_LOAD]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 5, i32 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr i32, i32* [[A:%.*]], i64 [[TMP8]]
+; CHECK-NEXT:    [[TMP14:%.*]] = select <vscale x 4 x i1> [[ACTIVE_LANE_MASK]], <vscale x 4 x i1> [[TMP12]], <vscale x 4 x i1> zeroinitializer
+; CHECK-NEXT:    [[TMP15:%.*]] = getelementptr i32, i32* [[TMP13]], i32 0
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast i32* [[TMP15]] to <vscale x 4 x i32>*
+; CHECK-NEXT:    [[WIDE_MASKED_LOAD1:%.*]] = call <vscale x 4 x i32> @llvm.masked.load.nxv4i32.p0nxv4i32(<vscale x 4 x i32>* [[TMP16]], i32 4, <vscale x 4 x i1> [[TMP14]], <vscale x 4 x i32> poison)
+; CHECK-NEXT:    [[TMP17:%.*]] = xor <vscale x 4 x i32> [[VEC_PHI]], [[WIDE_MASKED_LOAD1]]
+; CHECK-NEXT:    [[TMP18:%.*]] = xor <vscale x 4 x i1> [[TMP12]], shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i32 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer)
+; CHECK-NEXT:    [[TMP19:%.*]] = select <vscale x 4 x i1> [[ACTIVE_LANE_MASK]], <vscale x 4 x i1> [[TMP18]], <vscale x 4 x i1> zeroinitializer
+; CHECK-NEXT:    [[PREDPHI]] = select <vscale x 4 x i1> [[TMP14]], <vscale x 4 x i32> [[TMP17]], <vscale x 4 x i32> [[VEC_PHI]]
+; CHECK-NEXT:    [[TMP20:%.*]] = select <vscale x 4 x i1> [[ACTIVE_LANE_MASK]], <vscale x 4 x i32> [[PREDPHI]], <vscale x 4 x i32> [[VEC_PHI]]
+; CHECK-NEXT:    [[TMP21:%.*]] = call i64 @llvm.vscale.i64()
+; CHECK-NEXT:    [[TMP22:%.*]] = mul i64 [[TMP21]], 4
+; CHECK-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP22]]
+; CHECK-NEXT:    [[TMP23:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; CHECK-NEXT:    br i1 [[TMP23]], label [[MIDDLE_BLOCK:%.*]], label %vector.body, !llvm.loop [[LOOP22:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    [[TMP24:%.*]] = call i32 @llvm.vector.reduce.xor.nxv4i32(<vscale x 4 x i32> [[TMP20]])
+; CHECK-NEXT:    br i1 true, label [[FOR_END:%.*]], label %scalar.ph
+;
+entry:
+  br label %for.body
+
+for.body:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.inc ]
+  %rdx = phi i32 [ 7, %entry ], [ %res, %for.inc ]
+  %arrayidx = getelementptr inbounds i32, i32* %cond, i64 %iv
+  %0 = load i32, i32* %arrayidx
+  %tobool = icmp eq i32 %0, 5
+  br i1 %tobool, label %if.then, label %for.inc
+
+if.then:
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i64 %iv
+  %1 = load i32, i32* %arrayidx2
+  %xor = xor i32 %rdx, %1
+  br label %for.inc
+
+for.inc:
+  %res = phi i32 [ %rdx, %for.body ], [ %xor, %if.then ]
+  %iv.next = add i64 %iv, 1
+  %exitcond.not = icmp eq i64 %iv.next, %N
+  br i1 %exitcond.not, label %for.end, label %for.body, !llvm.loop !0
+
+for.end:
+  ret i32 %res
+}
 
 ; Negative tests where we don't expect tail-folding
 
 ; Integer divides can throw exceptions and since we can't scalarize conditional
 ; divides for scalable vectors we just don't bother vectorizing.
 define void @simple_idiv(i32* noalias %dst, i32* noalias %src, i64 %n) #0 {
 ; CHECK-LABEL: @simple_idiv(
 ; CHECK-NOT:   vector.body