[VPlan] Simplify add x, (sub 0, y) -> sub x, y - #187228
Merged
Merged
Conversation
Member
|
@llvm/pr-subscribers-backend-risc-v @llvm/pr-subscribers-vectorizers Author: Ramkumar Ramachandra (artagnon) ChangesPatch is 33.09 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/187228.diff 9 Files Affected:
diff --git a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
index d389db07885c8..3ead9a0a5bff9 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
@@ -1451,6 +1451,10 @@ static void simplifyRecipe(VPSingleDefRecipe *Def, VPTypeAnalysis &TypeInfo) {
return Def->replaceAllUsesWith(
Plan->getZero(TypeInfo.inferScalarType(Def)));
+ if (CanCreateNewRecipe &&
+ match(Def, m_c_Add(m_VPValue(X), m_c_Mul(m_VPValue(Y), m_AllOnes()))))
+ return Def->replaceAllUsesWith(Builder.createSub(X, Y));
+
const APInt *APC;
if (CanCreateNewRecipe && match(Def, m_c_Mul(m_VPValue(A), m_APInt(APC))) &&
APC->isPowerOf2())
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/vector-reverse.ll b/llvm/test/Transforms/LoopVectorize/AArch64/vector-reverse.ll
index 1e417a5588833..857eb578080ee 100644
--- a/llvm/test/Transforms/LoopVectorize/AArch64/vector-reverse.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/vector-reverse.ll
@@ -170,14 +170,13 @@ define i32 @reverse_store_with_partial_reduction(ptr noalias %dst, ptr noalias %
; CHECK-NEXT: [[TMP10:%.*]] = sub nuw nsw i64 [[TMP4]], 1
; CHECK-NEXT: [[TMP20:%.*]] = mul i64 [[TMP10]], -1
; CHECK-NEXT: [[TMP23:%.*]] = getelementptr i16, ptr [[TMP9]], i64 [[TMP20]]
-; CHECK-NEXT: [[TMP14:%.*]] = mul i64 -1, [[TMP4]]
-; CHECK-NEXT: [[TMP13:%.*]] = add i64 [[TMP20]], [[TMP14]]
+; CHECK-NEXT: [[TMP13:%.*]] = sub i64 [[TMP20]], [[TMP4]]
; CHECK-NEXT: [[TMP19:%.*]] = getelementptr i16, ptr [[TMP9]], i64 [[TMP13]]
; CHECK-NEXT: [[TMP16:%.*]] = mul i64 -2, [[TMP4]]
-; CHECK-NEXT: [[TMP25:%.*]] = add i64 [[TMP20]], [[TMP16]]
+; CHECK-NEXT: [[TMP25:%.*]] = sub i64 [[TMP16]], [[TMP10]]
; CHECK-NEXT: [[TMP28:%.*]] = getelementptr i16, ptr [[TMP9]], i64 [[TMP25]]
; CHECK-NEXT: [[TMP21:%.*]] = mul i64 -3, [[TMP4]]
-; CHECK-NEXT: [[TMP15:%.*]] = add i64 [[TMP20]], [[TMP21]]
+; CHECK-NEXT: [[TMP15:%.*]] = sub i64 [[TMP21]], [[TMP10]]
; CHECK-NEXT: [[TMP29:%.*]] = getelementptr i16, ptr [[TMP9]], i64 [[TMP15]]
; CHECK-NEXT: [[REVERSE:%.*]] = call <vscale x 8 x i16> @llvm.vector.reverse.nxv8i16(<vscale x 8 x i16> [[BROADCAST_SPLAT]])
; CHECK-NEXT: store <vscale x 8 x i16> [[REVERSE]], ptr [[TMP23]], align 2
diff --git a/llvm/test/Transforms/LoopVectorize/ARM/mve-gather-scatter-tailpred.ll b/llvm/test/Transforms/LoopVectorize/ARM/mve-gather-scatter-tailpred.ll
index a2e5edd3127c5..612580c399bcc 100644
--- a/llvm/test/Transforms/LoopVectorize/ARM/mve-gather-scatter-tailpred.ll
+++ b/llvm/test/Transforms/LoopVectorize/ARM/mve-gather-scatter-tailpred.ll
@@ -58,8 +58,7 @@ define void @test_stride-1_4i32(ptr readonly %data, ptr noalias nocapture %dst,
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT: [[TMP1:%.*]] = mul nuw nsw i32 [[INDEX]], -1
-; CHECK-NEXT: [[TMP2:%.*]] = add nuw nsw i32 [[TMP1]], 2
+; CHECK-NEXT: [[TMP2:%.*]] = sub i32 2, [[INDEX]]
; CHECK-NEXT: [[TMP3:%.*]] = getelementptr inbounds i32, ptr [[DATA:%.*]], i32 [[TMP2]]
; CHECK-NEXT: [[TMP5:%.*]] = getelementptr inbounds i32, ptr [[TMP3]], i32 -3
; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <4 x i32>, ptr [[TMP5]], align 4
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/dbg-tail-folding-by-evl.ll b/llvm/test/Transforms/LoopVectorize/RISCV/dbg-tail-folding-by-evl.ll
index 3870aa83d5346..051cbf2c7a70e 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/dbg-tail-folding-by-evl.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/dbg-tail-folding-by-evl.ll
@@ -15,11 +15,10 @@ define void @reverse_store(ptr %a, i64 %n) !dbg !3 {
; CHECK-NEXT: [[TMP2:%.*]] = call <vscale x 2 x i64> @llvm.stepvector.nxv2i64()
; CHECK-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[N]], i64 0
; CHECK-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[BROADCAST_SPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
-; CHECK-NEXT: [[TMP3:%.*]] = mul nsw <vscale x 2 x i64> [[TMP2]], splat (i64 -1)
-; CHECK-NEXT: [[INDUCTION:%.*]] = add nsw <vscale x 2 x i64> [[BROADCAST_SPLAT]], [[TMP3]]
+; CHECK-NEXT: [[TMP3:%.*]] = sub <vscale x 2 x i64> [[BROADCAST_SPLAT]], [[TMP2]]
; CHECK-NEXT: br label %[[VECTOR_BODY:.*]]
; CHECK: [[VECTOR_BODY]]:
-; CHECK-NEXT: [[VEC_IND:%.*]] = phi <vscale x 2 x i64> [ [[INDUCTION]], %[[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], %[[VECTOR_BODY]] ]
+; CHECK-NEXT: [[VEC_IND:%.*]] = phi <vscale x 2 x i64> [ [[TMP3]], %[[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[AVL:%.*]] = phi i64 [ [[TMP1]], %[[VECTOR_PH]] ], [ [[AVL_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP4:%.*]] = call i32 @llvm.experimental.get.vector.length.i64(i64 [[AVL]], i32 2, i1 true)
; CHECK-NEXT: [[TMP5:%.*]] = zext i32 [[TMP4]] to i64
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/riscv-vector-reverse.ll b/llvm/test/Transforms/LoopVectorize/RISCV/riscv-vector-reverse.ll
index b5825d9af216c..69578c3d8f455 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/riscv-vector-reverse.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/riscv-vector-reverse.ll
@@ -105,8 +105,7 @@ define void @vector_reverse_i32(ptr noalias %A, ptr noalias %B) {
; RV64-UF2-NEXT: [[TMP10:%.*]] = sub nuw nsw i64 [[TMP5]], 1
; RV64-UF2-NEXT: [[TMP15:%.*]] = mul i64 [[TMP10]], -1
; RV64-UF2-NEXT: [[TMP18:%.*]] = getelementptr inbounds i32, ptr [[TMP9]], i64 [[TMP15]]
-; RV64-UF2-NEXT: [[TMP12:%.*]] = mul i64 -1, [[TMP5]]
-; RV64-UF2-NEXT: [[TMP11:%.*]] = add i64 [[TMP15]], [[TMP12]]
+; RV64-UF2-NEXT: [[TMP11:%.*]] = sub i64 [[TMP15]], [[TMP5]]
; RV64-UF2-NEXT: [[TMP19:%.*]] = getelementptr inbounds i32, ptr [[TMP9]], i64 [[TMP11]]
; RV64-UF2-NEXT: [[WIDE_LOAD:%.*]] = load <vscale x 4 x i32>, ptr [[TMP18]], align 4
; RV64-UF2-NEXT: [[WIDE_LOAD1:%.*]] = load <vscale x 4 x i32>, ptr [[TMP19]], align 4
@@ -317,8 +316,7 @@ define void @vector_reverse_i64(ptr nocapture noundef writeonly %A, ptr nocaptur
; RV64-UF2-NEXT: [[TMP26:%.*]] = sub nuw nsw i64 [[TMP18]], 1
; RV64-UF2-NEXT: [[TMP30:%.*]] = mul i64 [[TMP26]], -1
; RV64-UF2-NEXT: [[TMP33:%.*]] = getelementptr inbounds i32, ptr [[TMP24]], i64 [[TMP30]]
-; RV64-UF2-NEXT: [[TMP27:%.*]] = mul i64 -1, [[TMP18]]
-; RV64-UF2-NEXT: [[TMP25:%.*]] = add i64 [[TMP30]], [[TMP27]]
+; RV64-UF2-NEXT: [[TMP25:%.*]] = sub i64 [[TMP30]], [[TMP18]]
; RV64-UF2-NEXT: [[TMP34:%.*]] = getelementptr inbounds i32, ptr [[TMP24]], i64 [[TMP25]]
; RV64-UF2-NEXT: [[WIDE_LOAD:%.*]] = load <vscale x 4 x i32>, ptr [[TMP33]], align 4
; RV64-UF2-NEXT: [[WIDE_LOAD4:%.*]] = load <vscale x 4 x i32>, ptr [[TMP34]], align 4
@@ -542,8 +540,7 @@ define void @vector_reverse_f32(ptr nocapture noundef writeonly %A, ptr nocaptur
; RV64-UF2-NEXT: [[TMP26:%.*]] = sub nuw nsw i64 [[TMP18]], 1
; RV64-UF2-NEXT: [[TMP30:%.*]] = mul i64 [[TMP26]], -1
; RV64-UF2-NEXT: [[TMP33:%.*]] = getelementptr inbounds float, ptr [[TMP24]], i64 [[TMP30]]
-; RV64-UF2-NEXT: [[TMP27:%.*]] = mul i64 -1, [[TMP18]]
-; RV64-UF2-NEXT: [[TMP25:%.*]] = add i64 [[TMP30]], [[TMP27]]
+; RV64-UF2-NEXT: [[TMP25:%.*]] = sub i64 [[TMP30]], [[TMP18]]
; RV64-UF2-NEXT: [[TMP34:%.*]] = getelementptr inbounds float, ptr [[TMP24]], i64 [[TMP25]]
; RV64-UF2-NEXT: [[WIDE_LOAD:%.*]] = load <vscale x 4 x float>, ptr [[TMP33]], align 4
; RV64-UF2-NEXT: [[WIDE_LOAD4:%.*]] = load <vscale x 4 x float>, ptr [[TMP34]], align 4
@@ -690,8 +687,7 @@ define void @vector_reverse_f32_simplify(ptr noalias %A, ptr noalias %B) {
; RV64-UF2-NEXT: [[TMP10:%.*]] = sub nuw nsw i64 [[TMP5]], 1
; RV64-UF2-NEXT: [[TMP15:%.*]] = mul i64 [[TMP10]], -1
; RV64-UF2-NEXT: [[TMP18:%.*]] = getelementptr inbounds float, ptr [[TMP9]], i64 [[TMP15]]
-; RV64-UF2-NEXT: [[TMP12:%.*]] = mul i64 -1, [[TMP5]]
-; RV64-UF2-NEXT: [[TMP11:%.*]] = add i64 [[TMP15]], [[TMP12]]
+; RV64-UF2-NEXT: [[TMP11:%.*]] = sub i64 [[TMP15]], [[TMP5]]
; RV64-UF2-NEXT: [[TMP19:%.*]] = getelementptr inbounds float, ptr [[TMP9]], i64 [[TMP11]]
; RV64-UF2-NEXT: [[WIDE_LOAD:%.*]] = load <vscale x 4 x float>, ptr [[TMP18]], align 4
; RV64-UF2-NEXT: [[WIDE_LOAD1:%.*]] = load <vscale x 4 x float>, ptr [[TMP19]], align 4
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-interleave.ll b/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-interleave.ll
index 4a12dacbacfc9..5cb9dc17b57d6 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-interleave.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/tail-folding-interleave.ll
@@ -578,8 +578,7 @@ define i32 @load_factor_4_reverse(i64 %n, ptr noalias %a) {
; IF-EVL-NEXT: [[TMP4:%.*]] = call <vscale x 4 x i64> @llvm.stepvector.nxv4i64()
; IF-EVL-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[N]], i64 0
; IF-EVL-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 4 x i64> [[BROADCAST_SPLATINSERT]], <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
-; IF-EVL-NEXT: [[TMP3:%.*]] = mul nsw <vscale x 4 x i64> [[TMP4]], splat (i64 -1)
-; IF-EVL-NEXT: [[INDUCTION:%.*]] = add nsw <vscale x 4 x i64> [[BROADCAST_SPLAT]], [[TMP3]]
+; IF-EVL-NEXT: [[INDUCTION:%.*]] = sub <vscale x 4 x i64> [[BROADCAST_SPLAT]], [[TMP4]]
; IF-EVL-NEXT: br label [[VECTOR_BODY:%.*]]
; IF-EVL: vector.body:
; IF-EVL-NEXT: [[VEC_IND:%.*]] = phi <vscale x 4 x i64> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
@@ -630,8 +629,7 @@ define i32 @load_factor_4_reverse(i64 %n, ptr noalias %a) {
; NO-VP-NEXT: [[TMP7:%.*]] = call <vscale x 4 x i64> @llvm.stepvector.nxv4i64()
; NO-VP-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[N]], i64 0
; NO-VP-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 4 x i64> [[BROADCAST_SPLATINSERT]], <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
-; NO-VP-NEXT: [[TMP8:%.*]] = mul nsw <vscale x 4 x i64> [[TMP7]], splat (i64 -1)
-; NO-VP-NEXT: [[INDUCTION:%.*]] = add nsw <vscale x 4 x i64> [[BROADCAST_SPLAT]], [[TMP8]]
+; NO-VP-NEXT: [[INDUCTION:%.*]] = sub <vscale x 4 x i64> [[BROADCAST_SPLAT]], [[TMP7]]
; NO-VP-NEXT: [[TMP9:%.*]] = mul nsw i64 -1, [[TMP5]]
; NO-VP-NEXT: [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <vscale x 4 x i64> poison, i64 [[TMP9]], i64 0
; NO-VP-NEXT: [[BROADCAST_SPLAT2:%.*]] = shufflevector <vscale x 4 x i64> [[BROADCAST_SPLATINSERT1]], <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer
diff --git a/llvm/test/Transforms/LoopVectorize/first-order-recurrence.ll b/llvm/test/Transforms/LoopVectorize/first-order-recurrence.ll
index 6b4727a1379f8..064a4d6e2fd1f 100644
--- a/llvm/test/Transforms/LoopVectorize/first-order-recurrence.ll
+++ b/llvm/test/Transforms/LoopVectorize/first-order-recurrence.ll
@@ -875,14 +875,14 @@ define i32 @PR27246() {
; UNROLL-NO-IC-NEXT: [[IND_END:%.*]] = sub i32 [[I_016]], [[N_VEC]]
; UNROLL-NO-IC-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[I_016]], i64 0
; UNROLL-NO-IC-NEXT: [[DOTSPLAT:%.*]] = shufflevector <4 x i32> [[DOTSPLATINSERT]], <4 x i32> poison, <4 x i32> zeroinitializer
-; UNROLL-NO-IC-NEXT: [[INDUCTION:%.*]] = add nsw <4 x i32> [[DOTSPLAT]], <i32 0, i32 -1, i32 -2, i32 -3>
+; UNROLL-NO-IC-NEXT: [[INDUCTION:%.*]] = sub <4 x i32> [[DOTSPLAT]], <i32 0, i32 1, i32 2, i32 3>
; UNROLL-NO-IC-NEXT: br label [[VECTOR_BODY:%.*]]
; UNROLL-NO-IC: vector.body:
; UNROLL-NO-IC-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; UNROLL-NO-IC-NEXT: [[VEC_IND:%.*]] = phi <4 x i32> [ [[INDUCTION]], [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
-; UNROLL-NO-IC-NEXT: [[STEP_ADD:%.*]] = add <4 x i32> [[VEC_IND]], splat (i32 -4)
+; UNROLL-NO-IC-NEXT: [[STEP_ADD:%.*]] = sub <4 x i32> [[VEC_IND]], splat (i32 4)
; UNROLL-NO-IC-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 8
-; UNROLL-NO-IC-NEXT: [[VEC_IND_NEXT]] = add nsw <4 x i32> [[STEP_ADD]], splat (i32 -4)
+; UNROLL-NO-IC-NEXT: [[VEC_IND_NEXT]] = sub <4 x i32> [[STEP_ADD]], splat (i32 4)
; UNROLL-NO-IC-NEXT: [[TMP0:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; UNROLL-NO-IC-NEXT: br i1 [[TMP0]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
; UNROLL-NO-IC: middle.block:
@@ -965,7 +965,7 @@ define i32 @PR27246() {
; SINK-AFTER-NEXT: [[IND_END:%.*]] = sub i32 [[I_016]], [[N_VEC]]
; SINK-AFTER-NEXT: [[DOTSPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[I_016]], i64 0
; SINK-AFTER-NEXT: [[DOTSPLAT:%.*]] = shufflevector <4 x i32> [[DOTSPLATINSERT]], <4 x i32> poison, <4 x i32> zeroinitializer
-; SINK-AFTER-NEXT: [[INDUCTION:%.*]] = add nsw <4 x i32> [[DOTSPLAT]], <i32 0, i32 -1, i32 -2, i32 -3>
+; SINK-AFTER-NEXT: [[INDUCTION:%.*]] = sub <4 x i32> [[DOTSPLAT]], <i32 0, i32 1, i32 2, i32 3>
; SINK-AFTER-NEXT: br label [[VECTOR_BODY:%.*]]
; SINK-AFTER: vector.body:
; SINK-AFTER-NEXT: [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
diff --git a/llvm/test/Transforms/LoopVectorize/interleave-with-i65-induction.ll b/llvm/test/Transforms/LoopVectorize/interleave-with-i65-induction.ll
index 1c987c84aa40b..2ad531c5f40cf 100644
--- a/llvm/test/Transforms/LoopVectorize/interleave-with-i65-induction.ll
+++ b/llvm/test/Transforms/LoopVectorize/interleave-with-i65-induction.ll
@@ -12,8 +12,8 @@ define void @i65_induction_with_negative_step(ptr %dst) {
; CHECK: [[VECTOR_BODY]]:
; CHECK-NEXT: [[INDEX:%.*]] = phi i65 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[VECTOR_RECUR:%.*]] = phi <4 x i64> [ <i64 poison, i64 poison, i64 poison, i64 0>, %[[VECTOR_PH]] ], [ [[STEP_ADD:%.*]], %[[VECTOR_BODY]] ]
-; CHECK-NEXT: [[VEC_IND:%.*]] = phi <4 x i64> [ <i64 0, i64 -1, i64 -2, i64 -3>, %[[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], %[[VECTOR_BODY]] ]
-; CHECK-NEXT: [[STEP_ADD]] = add <4 x i64> [[VEC_IND]], splat (i64 -4)
+; CHECK-NEXT: [[VEC_IND:%.*]] = phi <4 x i64> [ <i64 0, i64 -1, i64 -2, i64 -3>, %[[VECTOR_PH]] ], [ [[TMP7:%.*]], %[[VECTOR_BODY]] ]
+; CHECK-NEXT: [[STEP_ADD]] = sub <4 x i64> [[VEC_IND]], splat (i64 4)
; CHECK-NEXT: [[OFFSET_IDX:%.*]] = sub i65 0, [[INDEX]]
; CHECK-NEXT: [[TMP0:%.*]] = trunc i65 [[OFFSET_IDX]] to i64
; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <4 x i64> [[VECTOR_RECUR]], <4 x i64> [[VEC_IND]], <4 x i32> <i32 3, i32 4, i32 5, i32 6>
@@ -26,7 +26,7 @@ define void @i65_induction_with_negative_step(ptr %dst) {
; CHECK-NEXT: store <4 x i64> [[REVERSE]], ptr [[TMP6]], align 8
; CHECK-NEXT: store <4 x i64> [[REVERSE1]], ptr [[TMP8]], align 8
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i65 [[INDEX]], 8
-; CHECK-NEXT: [[VEC_IND_NEXT]] = add <4 x i64> [[STEP_ADD]], splat (i64 -4)
+; CHECK-NEXT: [[TMP7]] = sub <4 x i64> [[STEP_ADD]], splat (i64 4)
; CHECK-NEXT: [[TMP9:%.*]] = icmp eq i65 [[INDEX_NEXT]], 1000
; CHECK-NEXT: br i1 [[TMP9]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK: [[MIDDLE_BLOCK]]:
diff --git a/llvm/test/Transforms/LoopVectorize/iv-select-cmp-decreasing.ll b/llvm/test/Transforms/LoopVectorize/iv-select-cmp-decreasing.ll
index dd3ad4d01b465..4ac2bb36c4967 100644
--- a/llvm/test/Transforms/LoopVectorize/iv-select-cmp-decreasing.ll
+++ b/llvm/test/Transforms/LoopVectorize/iv-select-cmp-decreasing.ll
@@ -41,14 +41,14 @@ define i64 @select_decreasing_induction_icmp_const_start(ptr %a) {
; IC4VF4-NEXT: br label %[[VECTOR_BODY:.*]]
; IC4VF4: [[VECTOR_BODY]]:
; IC4VF4-NEXT: [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
-; IC4VF4-NEXT: [[VEC_IND:%.*]] = phi <4 x i64> [ <i64 19999, i64 19998, i64 19997, i64 19996>, %[[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], %[[VECTOR_BODY]] ]
+; IC4VF4-NEXT: [[VEC_IND:%.*]] = phi <4 x i64> [ <i64 19999, i64 19998, i64 19997, i64 19996>, %[[VECTOR_PH]] ], [ [[TMP19:%.*]], %[[VECTOR_BODY]] ]
; IC4VF4-NEXT: [[VEC_PHI:%.*]] = phi <4 x i64> [ splat (i64 9223372036854775807), %[[VECTOR_PH]] ], [ [[TMP13:%.*]], %[[VECTOR_BODY]] ]
; IC4VF4-NEXT: [[VEC_PHI1:%.*]] = phi <4 x i64> [ splat (i64 9223372036854775807), %[[VECTOR_PH]] ], [ [[TMP14:%.*]], %[[VECTOR_BODY]] ]
; IC4VF4-NEXT: [[VEC_PHI2:%.*]] = phi <4 x i64> [ splat (i64 9223372036854775807), %[[VECTOR_PH]] ], [ [[TMP15:%.*]], %[[VECTOR_BODY]] ]
; IC4VF4-NEXT: [[VEC_PHI3:%.*]] = phi <4 x i64> [ splat (i64 9223372036854775807), %[[VECTOR_PH]] ], [ [[TMP16:%.*]], %[[VECTOR_BODY]] ]
-; IC4VF4-NEXT: [[STEP_ADD:%.*]] = add <4 x i64> [[VEC_IND]], splat (i64 -4)
-; IC4VF4-NEXT: [[STEP_ADD_2:%.*]] = add <4 x i64> [[STEP_ADD]], splat (i64 -4)
-; IC4VF4-NEXT: [[STEP_ADD_3:%.*]] = add <4 x i64> [[STEP_ADD_2]], splat (i64 -4)
+; IC4VF4-NEXT: [[STEP_ADD:%.*]] = sub <4 x i64> [[VEC_IND]], splat (i64 4)
+; IC4VF4-NEXT: [[STEP_ADD_2:%.*]] = sub <4 x i64> [[STEP_ADD]], splat (i64 4)
+; IC4VF4-NEXT: [[STEP_ADD_3:%.*]] = sub <4 x i64> [[STEP_ADD_2]], splat (i64 4)
; IC4VF4-NEXT: [[OFFSET_IDX:%.*]] = sub i64 19999, [[INDEX]]
; IC4VF4-NEXT: [[TMP0:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[OFFSET_IDX]]
; IC4VF4-NEXT: [[TMP2:%.*]] = getelementptr inbounds i64, ptr [[TMP0]], i64 -3
@@ -72,7 +72,7 @@ define i64 @select_decreasing_induction_icmp_const_start(ptr %a) {
; IC4VF4-NEXT: [[TMP15]] = select <4 x i1> [[TMP11]], <4 x i64> [[STEP_ADD_2]], <4 x i64> [[VEC_PHI2]]
; IC4VF4-NEXT: [[TMP16]] = select <4 x i1> [[TMP12]], <4 x i64> [[STEP_ADD_3]], <4 x i64> [[VEC_PHI3]]
; IC4VF4-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 16
-; IC4VF4-NEXT: [[VEC_IND_NEXT]] = add nsw <4 x i64> [[STEP_ADD_3]], splat (i64 -4)
+; IC4VF4-NEXT: [[TMP19]] = sub <4 x i64> [[STEP_ADD_3]], splat (i64 4)
; IC4VF4-NEXT: [[TMP17:%.*]] = icmp eq i64 [[INDEX_NEXT]], 20000
; IC4VF4-NEXT: br i1 [[TMP17]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; IC4VF4: [[MIDDLE_BLOCK]]:
@@ -195,14 +195,14 @@ define i16 @select_decreasing_induction_icmp_table_i16(i16 noundef %val) {
; IC4VF4-NEXT: br label %[[VECTOR_BODY:.*]]
; IC4VF4: [[VECTOR_BODY]]:
; IC4VF4-NEXT: [[INDEX:%.*]] = phi i32 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[PRED_LOAD_CONTINUE44:.*]] ]
-; IC4VF4-NEXT: [[VEC_IND:%.*]] = phi <4 x i16> [ <i16 12, i16 11, i16 10, i16 9>, %[[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], %[[PRED_LOAD_CONTINUE44]] ]
+; IC4VF4-NEXT: [[VEC_IND:%.*]] = phi <4 x i16> [ <i16 12, i16 11, i16 10, i16 9>, %[[VECTOR_PH]] ], [ [[TMP117:%.*]], %[[PRED_LOAD_CONTINUE44]] ]
; IC4VF4-NEXT: [[VEC_PHI:%.*]] = phi <4 x i16> [ splat (i16 32767), %[[VECTOR_PH]] ], [ [[TMP108:%.*]], %[[PRED_LOAD_CONTINUE44]] ]
; IC4VF4-NEXT: [[VEC_PHI1:%.*]] = phi <4 x i16> [ splat (i16 32767), %[[VECTOR_PH]] ], [ [[TMP109:%.*]], %[[PRED_LOAD_CONTINUE44]] ]
; IC4VF4-NEXT: [[VEC_PHI2:%.*]] = phi <4 x i16> [ splat (i16 32767), %[[VECTOR_PH]] ], [ [[TMP110:%.*]], %[[PRED_LOAD_CONTINUE44]] ]
; IC4VF4-NEXT: [[VEC_PHI3:%.*]] = phi <4 x i16> [ splat (i16 32767), %[[VECTOR_PH]] ], [ [[TMP111:%.*]], %[[PRED_LOAD_CONTINUE44]] ]
-; IC4VF4-NEXT: [[STEP_ADD:%.*]] = add <4 x i16> [[VEC_IND]], splat (i16 -4)
-; IC4VF4-NEXT: [[STEP_ADD_2:%.*]] = add <4 x i16> [[STEP_ADD]], splat (i16 -4)
-; IC4VF4-NEXT: [[STEP_ADD_3:%.*]] = add <4 x i16> [[STEP_ADD_2]], splat (i16 -4)
+; IC4VF4-NEXT: [[STEP_ADD:%.*]] = sub <4 x i16> [[VEC_IND]], splat (i16 4)
+; IC4VF4-NEXT: [[STEP_ADD_2:%.*]] = sub <4 x i16> [[STEP_ADD]], splat (i16 4)
+; IC4VF4-NEXT: [[STEP_ADD_3:%.*]] = sub <4 x i16> [[STEP_ADD_2]], splat (i16 4)
; IC4VF4-NEXT: [[DOTCAST:%.*]] = trunc i32 [[INDEX]] to i16
; IC4VF4-NEXT: [[OFFSET_IDX:%.*]] = sub i16 12, [[DOTCAST]]
; IC4VF4-NEXT: [[BROADCAST_SPLATINSERT4:%.*]] = insertelement <4 x i32> poison, i32 [[INDEX]], i64 0
@@ -388,7 +388,7 @@ define i16 @select_decreasing_induction_icmp_table_i16(i16 noundef %val) {
; IC4VF4-NEXT: [[TMP110]] = select...
[truncated]
|
eas
reviewed
Mar 18, 2026
artagnon
changed the title
artagnon
changed the title
lukel97
reviewed
Mar 19, 2026
fhahn
reviewed
Mar 19, 2026
artagnon
changed the title
artagnon
force-pushed
the
vplan-simpl-add-mul-allones-sub
branch
from
March 19, 2026 21:00
5800f05 to
8f52956
Compare
🐧 Linux x64 Test Results
Failed Tests(click on a test name to see its output) lldb-apilldb-api.python_api/run_locker/TestRunLocker.pyIf these failures are unrelated to your changes (for example tests are broken or flaky at HEAD), please open an issue at https://github.com/llvm/llvm-project/issues and add the |
fhahn
reviewed
Mar 26, 2026
artagnon
force-pushed
the
vplan-simpl-add-mul-allones-sub
branch
5 times, most recently
from
March 31, 2026 10:09
9b78b03 to
b1bf08d
Compare
Contributor
Author
|
Gentle ping. |
eas
reviewed
Mar 31, 2026
artagnon
force-pushed
the
vplan-simpl-add-mul-allones-sub
branch
from
March 31, 2026 20:33
b1bf08d to
cadf4cd
Compare
artagnon
force-pushed
the
vplan-simpl-add-mul-allones-sub
branch
from
April 8, 2026 20:36
cadf4cd to
175709c
Compare
Contributor
Author
|
This should be ready now, we can either merge the test change separately, or as part of this PR. |
fhahn
reviewed
Apr 9, 2026
artagnon
force-pushed
the
vplan-simpl-add-mul-allones-sub
branch
from
April 9, 2026 16:54
175709c to
dd54588
Compare
artagnon
force-pushed
the
vplan-simpl-add-mul-allones-sub
branch
2 times, most recently
from
April 20, 2026 16:02
fa0681b to
c068700
Compare
Follow what's done in InstCombine.
artagnon
force-pushed
the
vplan-simpl-add-mul-allones-sub
branch
from
April 21, 2026 06:51
c068700 to
a84f665
Compare
artagnon
enabled auto-merge (squash)
April 21, 2026 06:52
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/197/builds/15420 Here is the relevant piece of the build log for the reference |
YonahGoldberg
pushed a commit
to YonahGoldberg/llvm-project
that referenced
this pull request
Apr 21, 2026
The input IR shown in the partial-reduce-sub dotp test will never reach the vectorizer, due to prior simplification by InstCombine. Change it with this simplification. This is necessary for the simplification in llvm#187228, whose goal is to make llvm#187589 possible.
linuxlonelyeagle
pushed a commit
to linuxlonelyeagle/llvm-project
that referenced
this pull request
Apr 23, 2026
The patch follows what InstCombine does. Alive2 proof: https://alive2.llvm.org/ce/z/0tG6s6
KHicketts
pushed a commit
to KHicketts/llvm-project
that referenced
this pull request
Apr 30, 2026
The patch follows what InstCombine does. Alive2 proof: https://alive2.llvm.org/ce/z/0tG6s6
MacDue
added a commit
that referenced
this pull request
May 17, 2026
pedroMVicente
pushed a commit
to pedroMVicente/llvm-project
that referenced
this pull request
May 19, 2026
…198118) The `neg` form was removed in llvm#187228 (this case now uses the out-of-loop sub, which is preferable, see llvm#189739).
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The patch follows what InstCombine does.
Alive2 proof: https://alive2.llvm.org/ce/z/0tG6s6