Skip to content

Commit 2d7b55a

Browse files
alexey-bataevalexey-bataev
authored
[SLP]Initial support for copyable elements
Adds initial support for copyable elements, both schedulable and non-schedulable. Adds support only for add for now, other opcodes will added in future. Still some cases are not handled, e.g. stores do not include this, because currently do not check for copyable elements. Reviewers: hiraditya, RKSimon Reviewed By: RKSimon Pull Request: #147366
1 parent ccd35e5 commit 2d7b55a

File tree

9 files changed

+1087
-217
lines changed

9 files changed

+1087
-217
lines changed

‎llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp

Lines changed: 1040 additions & 168 deletions
Large diffs are not rendered by default.

‎llvm/test/Transforms/SLPVectorizer/AArch64/vec3-reorder-reshuffle.ll

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -191,12 +191,12 @@ define i32 @reorder_indices_1(float %0) {
191191
; NON-POW2-NEXT: entry:
192192
; NON-POW2-NEXT: [[NOR1:%.*]] = alloca [0 x [3 x float]], i32 0, align 4
193193
; NON-POW2-NEXT: [[TMP1:%.*]] = load <3 x float>, ptr [[NOR1]], align 4
194-
; NON-POW2-NEXT: [[TMP3:%.*]] = fneg <3 x float> [[TMP1]]
194+
; NON-POW2-NEXT: [[TMP2:%.*]] = shufflevector <3 x float> [[TMP1]], <3 x float> poison, <3 x i32> <i32 1, i32 2, i32 0>
195+
; NON-POW2-NEXT: [[TMP3:%.*]] = fneg <3 x float> [[TMP2]]
195196
; NON-POW2-NEXT: [[TMP4:%.*]] = insertelement <3 x float> poison, float [[TMP0]], i32 0
196197
; NON-POW2-NEXT: [[TMP5:%.*]] = shufflevector <3 x float> [[TMP4]], <3 x float> poison, <3 x i32> zeroinitializer
197198
; NON-POW2-NEXT: [[TMP6:%.*]] = fmul <3 x float> [[TMP3]], [[TMP5]]
198-
; NON-POW2-NEXT: [[TMP10:%.*]] = shufflevector <3 x float> [[TMP6]], <3 x float> poison, <3 x i32> <i32 1, i32 2, i32 0>
199-
; NON-POW2-NEXT: [[TMP7:%.*]] = call <3 x float> @llvm.fmuladd.v3f32(<3 x float> [[TMP1]], <3 x float> zeroinitializer, <3 x float> [[TMP10]])
199+
; NON-POW2-NEXT: [[TMP7:%.*]] = call <3 x float> @llvm.fmuladd.v3f32(<3 x float> [[TMP1]], <3 x float> zeroinitializer, <3 x float> [[TMP6]])
200200
; NON-POW2-NEXT: [[TMP8:%.*]] = call <3 x float> @llvm.fmuladd.v3f32(<3 x float> [[TMP5]], <3 x float> [[TMP7]], <3 x float> zeroinitializer)
201201
; NON-POW2-NEXT: [[TMP9:%.*]] = fmul <3 x float> [[TMP8]], zeroinitializer
202202
; NON-POW2-NEXT: store <3 x float> [[TMP9]], ptr [[NOR1]], align 4

‎llvm/test/Transforms/SLPVectorizer/X86/buildvector-schedule-for-subvector.ll

Lines changed: 1 addition & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -4,17 +4,14 @@
44
define void @test() {
55
; CHECK-LABEL: define void @test() {
66
; CHECK-NEXT: [[BB:.*:]]
7-
; CHECK-NEXT: [[ICMP:%.*]] = icmp samesign ult i32 0, 0
8-
; CHECK-NEXT: [[SELECT:%.*]] = select i1 [[ICMP]], i32 0, i32 0
7+
; CHECK-NEXT: [[SELECT:%.*]] = select i1 false, i32 0, i32 0
98
; CHECK-NEXT: [[ZEXT:%.*]] = zext i32 [[SELECT]] to i64
109
; CHECK-NEXT: [[GETELEMENTPTR:%.*]] = getelementptr ptr addrspace(1), ptr addrspace(1) null, i64 [[ZEXT]]
1110
; CHECK-NEXT: store ptr addrspace(1) null, ptr addrspace(1) [[GETELEMENTPTR]], align 8
1211
; CHECK-NEXT: store volatile i32 0, ptr addrspace(1) null, align 4
1312
; CHECK-NEXT: [[CALL:%.*]] = call i32 null(<2 x double> zeroinitializer)
1413
; CHECK-NEXT: [[TMP2:%.*]] = insertelement <4 x i32> <i32 0, i32 0, i32 0, i32 poison>, i32 [[CALL]], i32 3
1514
; CHECK-NEXT: [[TMP3:%.*]] = icmp eq <4 x i32> [[TMP2]], zeroinitializer
16-
; CHECK-NEXT: [[TMP4:%.*]] = shufflevector <4 x i1> [[TMP3]], <4 x i1> poison, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 poison, i32 poison, i32 poison, i32 poison>
17-
; CHECK-NEXT: [[TMP5:%.*]] = shufflevector <8 x i1> [[TMP4]], <8 x i1> <i1 false, i1 false, i1 false, i1 false, i1 undef, i1 undef, i1 undef, i1 undef>, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 8, i32 9, i32 10, i32 11>
1815
; CHECK-NEXT: ret void
1916
;
2017
bb:

‎llvm/test/Transforms/SLPVectorizer/X86/cast-operand-extracted.ll

Lines changed: 4 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -19,14 +19,14 @@ define void @test(ptr %0, i32 %add651) {
1919
; CHECK-NEXT: [[ARRAYIDX660:%.*]] = getelementptr i8, ptr [[TMP4]], i64 7800
2020
; CHECK-NEXT: [[ARRAYIDX689:%.*]] = getelementptr i8, ptr [[TMP4]], i64 7816
2121
; CHECK-NEXT: [[TMP6:%.*]] = add <2 x i32> [[TMP3]], splat (i32 1)
22-
; CHECK-NEXT: [[TMP8:%.*]] = add <2 x i32> [[TMP6]], [[TMP7]]
23-
; CHECK-NEXT: [[TMP9:%.*]] = insertelement <2 x i32> <i32 1, i32 poison>, i32 [[TMP5]], i32 1
24-
; CHECK-NEXT: [[TMP10:%.*]] = add <2 x i32> [[TMP8]], [[TMP9]]
22+
; CHECK-NEXT: [[TMP10:%.*]] = add <2 x i32> [[TMP6]], [[TMP7]]
2523
; CHECK-NEXT: [[TMP11:%.*]] = insertelement <4 x i32> poison, i32 [[ADD651]], i32 0
2624
; CHECK-NEXT: [[TMP13:%.*]] = insertelement <4 x i32> [[TMP11]], i32 [[TMP2]], i32 1
2725
; CHECK-NEXT: [[TMP19:%.*]] = shufflevector <2 x i32> [[TMP10]], <2 x i32> poison, <4 x i32> <i32 0, i32 1, i32 poison, i32 poison>
2826
; CHECK-NEXT: [[TMP14:%.*]] = shufflevector <4 x i32> [[TMP13]], <4 x i32> [[TMP19]], <4 x i32> <i32 0, i32 1, i32 4, i32 5>
29-
; CHECK-NEXT: [[TMP15:%.*]] = lshr <4 x i32> [[TMP14]], splat (i32 1)
27+
; CHECK-NEXT: [[TMP20:%.*]] = insertelement <4 x i32> <i32 0, i32 0, i32 1, i32 poison>, i32 [[TMP5]], i32 3
28+
; CHECK-NEXT: [[TMP21:%.*]] = add <4 x i32> [[TMP14]], [[TMP20]]
29+
; CHECK-NEXT: [[TMP15:%.*]] = lshr <4 x i32> [[TMP21]], splat (i32 1)
3030
; CHECK-NEXT: [[SHR685:%.*]] = lshr i32 [[TMP2]], 1
3131
; CHECK-NEXT: [[TMP16:%.*]] = trunc <4 x i32> [[TMP15]] to <4 x i16>
3232
; CHECK-NEXT: [[CONV686:%.*]] = trunc i32 [[SHR685]] to i16

‎llvm/test/Transforms/SLPVectorizer/X86/pr35497.ll

Lines changed: 19 additions & 19 deletions
Original file line numberDiff line numberDiff line change
@@ -87,29 +87,29 @@ define void @pr35497(ptr %p, i64 %c) {
8787
; AVX-LABEL: @pr35497(
8888
; AVX-NEXT: entry:
8989
; AVX-NEXT: [[TMP0:%.*]] = load i64, ptr [[P:%.*]], align 1
90-
; AVX-NEXT: [[TMP5:%.*]] = insertelement <2 x i64> poison, i64 [[C:%.*]], i32 0
91-
; AVX-NEXT: [[TMP11:%.*]] = shufflevector <2 x i64> [[TMP5]], <2 x i64> poison, <2 x i32> zeroinitializer
92-
; AVX-NEXT: [[TMP13:%.*]] = lshr <2 x i64> [[TMP11]], splat (i64 6)
90+
; AVX-NEXT: [[TMP1:%.*]] = insertelement <2 x i64> poison, i64 [[C:%.*]], i32 0
91+
; AVX-NEXT: [[TMP2:%.*]] = shufflevector <2 x i64> [[TMP1]], <2 x i64> poison, <2 x i32> zeroinitializer
92+
; AVX-NEXT: [[TMP3:%.*]] = lshr <2 x i64> [[TMP2]], splat (i64 6)
9393
; AVX-NEXT: [[ARRAYIDX2_2:%.*]] = getelementptr inbounds [0 x i64], ptr [[P]], i64 0, i64 4
9494
; AVX-NEXT: [[ARRAYIDX2_5:%.*]] = getelementptr inbounds [0 x i64], ptr [[P]], i64 0, i64 1
95-
; AVX-NEXT: [[TMP1:%.*]] = insertelement <2 x i64> [[TMP11]], i64 [[TMP0]], i32 1
96-
; AVX-NEXT: [[TMP2:%.*]] = shl <2 x i64> [[TMP1]], splat (i64 2)
97-
; AVX-NEXT: [[TMP3:%.*]] = and <2 x i64> [[TMP2]], splat (i64 20)
98-
; AVX-NEXT: [[TMP14:%.*]] = shufflevector <2 x i64> [[TMP3]], <2 x i64> [[TMP1]], <2 x i32> <i32 1, i32 2>
99-
; AVX-NEXT: [[TMP16:%.*]] = shufflevector <2 x i64> [[TMP13]], <2 x i64> [[TMP14]], <2 x i32> <i32 1, i32 3>
100-
; AVX-NEXT: [[TMP6:%.*]] = add <2 x i64> [[TMP14]], [[TMP16]]
101-
; AVX-NEXT: [[TMP17:%.*]] = extractelement <2 x i64> [[TMP6]], i32 1
102-
; AVX-NEXT: store i64 [[TMP17]], ptr [[P]], align 1
103-
; AVX-NEXT: [[TMP4:%.*]] = add nuw nsw <2 x i64> [[TMP3]], [[TMP13]]
104-
; AVX-NEXT: [[TMP12:%.*]] = extractelement <2 x i64> [[TMP6]], i32 0
95+
; AVX-NEXT: [[TMP4:%.*]] = insertelement <2 x i64> [[TMP2]], i64 [[TMP0]], i32 1
96+
; AVX-NEXT: [[TMP5:%.*]] = shl <2 x i64> [[TMP4]], splat (i64 2)
97+
; AVX-NEXT: [[TMP6:%.*]] = and <2 x i64> [[TMP5]], splat (i64 20)
98+
; AVX-NEXT: [[TMP7:%.*]] = shufflevector <2 x i64> [[TMP6]], <2 x i64> [[TMP4]], <2 x i32> <i32 1, i32 2>
99+
; AVX-NEXT: [[TMP8:%.*]] = shufflevector <2 x i64> [[TMP3]], <2 x i64> [[TMP7]], <2 x i32> <i32 1, i32 3>
100+
; AVX-NEXT: [[TMP9:%.*]] = add <2 x i64> [[TMP7]], [[TMP8]]
101+
; AVX-NEXT: [[TMP10:%.*]] = extractelement <2 x i64> [[TMP9]], i32 1
102+
; AVX-NEXT: store i64 [[TMP10]], ptr [[P]], align 1
103+
; AVX-NEXT: [[TMP11:%.*]] = add nuw nsw <2 x i64> [[TMP6]], [[TMP3]]
104+
; AVX-NEXT: [[TMP12:%.*]] = extractelement <2 x i64> [[TMP9]], i32 0
105105
; AVX-NEXT: store i64 [[TMP12]], ptr [[ARRAYIDX2_5]], align 1
106-
; AVX-NEXT: [[TMP7:%.*]] = shl <2 x i64> [[TMP6]], splat (i64 2)
107-
; AVX-NEXT: [[TMP8:%.*]] = and <2 x i64> [[TMP7]], splat (i64 20)
108-
; AVX-NEXT: [[TMP15:%.*]] = extractelement <2 x i64> [[TMP4]], i32 0
106+
; AVX-NEXT: [[TMP13:%.*]] = shl <2 x i64> [[TMP9]], splat (i64 2)
107+
; AVX-NEXT: [[TMP14:%.*]] = and <2 x i64> [[TMP13]], splat (i64 20)
108+
; AVX-NEXT: [[TMP15:%.*]] = extractelement <2 x i64> [[TMP11]], i32 0
109109
; AVX-NEXT: store i64 [[TMP15]], ptr [[P]], align 1
110-
; AVX-NEXT: [[TMP9:%.*]] = lshr <2 x i64> [[TMP4]], splat (i64 6)
111-
; AVX-NEXT: [[TMP10:%.*]] = add nuw nsw <2 x i64> [[TMP8]], [[TMP9]]
112-
; AVX-NEXT: store <2 x i64> [[TMP10]], ptr [[ARRAYIDX2_2]], align 1
110+
; AVX-NEXT: [[TMP16:%.*]] = lshr <2 x i64> [[TMP11]], splat (i64 6)
111+
; AVX-NEXT: [[TMP17:%.*]] = add nuw nsw <2 x i64> [[TMP14]], [[TMP16]]
112+
; AVX-NEXT: store <2 x i64> [[TMP17]], ptr [[ARRAYIDX2_2]], align 1
113113
; AVX-NEXT: ret void
114114
;
115115
entry:

‎llvm/test/Transforms/SLPVectorizer/X86/shuffle-mask-emission.ll

Lines changed: 3 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -5,9 +5,11 @@ define i1 @test() {
55
; CHECK-LABEL: define i1 @test() {
66
; CHECK-NEXT: [[ENTRY:.*:]]
77
; CHECK-NEXT: [[H_PROMOTED118_I_FR:%.*]] = freeze i32 1
8+
; CHECK-NEXT: [[TMP3:%.*]] = insertelement <2 x i32> <i32 poison, i32 0>, i32 [[H_PROMOTED118_I_FR]], i32 0
9+
; CHECK-NEXT: [[TMP4:%.*]] = add <2 x i32> zeroinitializer, [[TMP3]]
10+
; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <2 x i32> [[TMP4]], <2 x i32> poison, <4 x i32> <i32 0, i32 0, i32 1, i32 0>
811
; CHECK-NEXT: [[TMP0:%.*]] = insertelement <4 x i32> <i32 0, i32 0, i32 poison, i32 0>, i32 [[H_PROMOTED118_I_FR]], i32 2
912
; CHECK-NEXT: [[TMP1:%.*]] = add <4 x i32> zeroinitializer, [[TMP0]]
10-
; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <4 x i32> [[TMP0]], <4 x i32> [[TMP1]], <4 x i32> <i32 2, i32 2, i32 7, i32 2>
1113
; CHECK-NEXT: [[TMP5:%.*]] = add <4 x i32> [[TMP1]], [[TMP2]]
1214
; CHECK-NEXT: [[TMP6:%.*]] = and <4 x i32> [[TMP5]], <i32 0, i32 1, i32 1, i32 1>
1315
; CHECK-NEXT: [[TMP7:%.*]] = icmp eq <4 x i32> [[TMP6]], <i32 1, i32 0, i32 0, i32 0>

‎llvm/test/Transforms/SLPVectorizer/X86/vec3-reorder-reshuffle.ll

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -190,12 +190,12 @@ define i32 @reorder_indices_1(float %0) {
190190
; NON-POW2-NEXT: entry:
191191
; NON-POW2-NEXT: [[NOR1:%.*]] = alloca [0 x [3 x float]], i32 0, align 4
192192
; NON-POW2-NEXT: [[TMP1:%.*]] = load <3 x float>, ptr [[NOR1]], align 4
193-
; NON-POW2-NEXT: [[TMP3:%.*]] = fneg <3 x float> [[TMP1]]
193+
; NON-POW2-NEXT: [[TMP2:%.*]] = shufflevector <3 x float> [[TMP1]], <3 x float> poison, <3 x i32> <i32 1, i32 2, i32 0>
194+
; NON-POW2-NEXT: [[TMP3:%.*]] = fneg <3 x float> [[TMP2]]
194195
; NON-POW2-NEXT: [[TMP4:%.*]] = insertelement <3 x float> poison, float [[TMP0]], i32 0
195196
; NON-POW2-NEXT: [[TMP5:%.*]] = shufflevector <3 x float> [[TMP4]], <3 x float> poison, <3 x i32> zeroinitializer
196197
; NON-POW2-NEXT: [[TMP6:%.*]] = fmul <3 x float> [[TMP3]], [[TMP5]]
197-
; NON-POW2-NEXT: [[TMP10:%.*]] = shufflevector <3 x float> [[TMP6]], <3 x float> poison, <3 x i32> <i32 1, i32 2, i32 0>
198-
; NON-POW2-NEXT: [[TMP7:%.*]] = call <3 x float> @llvm.fmuladd.v3f32(<3 x float> [[TMP1]], <3 x float> zeroinitializer, <3 x float> [[TMP10]])
198+
; NON-POW2-NEXT: [[TMP7:%.*]] = call <3 x float> @llvm.fmuladd.v3f32(<3 x float> [[TMP1]], <3 x float> zeroinitializer, <3 x float> [[TMP6]])
199199
; NON-POW2-NEXT: [[TMP8:%.*]] = call <3 x float> @llvm.fmuladd.v3f32(<3 x float> [[TMP5]], <3 x float> [[TMP7]], <3 x float> zeroinitializer)
200200
; NON-POW2-NEXT: [[TMP9:%.*]] = fmul <3 x float> [[TMP8]], zeroinitializer
201201
; NON-POW2-NEXT: store <3 x float> [[TMP9]], ptr [[NOR1]], align 4

‎llvm/test/Transforms/SLPVectorizer/gathered-consecutive-loads-different-types.ll

Lines changed: 9 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -6,20 +6,21 @@ define i32 @test(i8 %0) {
66
; CHECK-SAME: i8 [[TMP0:%.*]]) {
77
; CHECK-NEXT: [[ENTRY:.*:]]
88
; CHECK-NEXT: [[CMP13_NOT_5:%.*]] = icmp eq i64 0, 0
9-
; CHECK-NEXT: [[TMP1:%.*]] = load i8, ptr addrspace(21) getelementptr inbounds (i8, ptr addrspace(21) null, i64 7), align 1
10-
; CHECK-NEXT: [[TMP2:%.*]] = insertelement <2 x i8> <i8 0, i8 poison>, i8 [[TMP1]], i32 1
11-
; CHECK-NEXT: [[TMP3:%.*]] = icmp eq <2 x i8> zeroinitializer, [[TMP2]]
129
; CHECK-NEXT: [[TMP4:%.*]] = load volatile i8, ptr null, align 8
13-
; CHECK-NEXT: [[TMP5:%.*]] = load <2 x i8>, ptr addrspace(21) getelementptr inbounds (i8, ptr addrspace(21) null, i64 8), align 8
10+
; CHECK-NEXT: [[TMP2:%.*]] = load i8, ptr addrspace(21) getelementptr inbounds (i8, ptr addrspace(21) null, i64 9), align 1
11+
; CHECK-NEXT: [[TEST_STRUCTCOPY_14_S14_CM_COERCE_SROA_2_0_COPYLOAD:%.*]] = load i48, ptr addrspace(21) getelementptr inbounds (i8, ptr addrspace(21) null, i64 8), align 8
12+
; CHECK-NEXT: [[TMP12:%.*]] = load i8, ptr addrspace(21) null, align 2
13+
; CHECK-NEXT: [[TMP13:%.*]] = load volatile i8, ptr null, align 2
14+
; CHECK-NEXT: [[TMP5:%.*]] = load <2 x i8>, ptr addrspace(21) getelementptr inbounds (i8, ptr addrspace(21) null, i64 7), align 1
15+
; CHECK-NEXT: [[TMP32:%.*]] = shufflevector <2 x i8> <i8 0, i8 poison>, <2 x i8> [[TMP5]], <2 x i32> <i32 0, i32 2>
16+
; CHECK-NEXT: [[TMP3:%.*]] = icmp eq <2 x i8> zeroinitializer, [[TMP32]]
1417
; CHECK-NEXT: [[TMP6:%.*]] = shufflevector <2 x i8> [[TMP5]], <2 x i8> poison, <8 x i32> <i32 0, i32 1, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
15-
; CHECK-NEXT: [[TMP7:%.*]] = shufflevector <8 x i8> <i8 0, i8 0, i8 poison, i8 0, i8 0, i8 poison, i8 0, i8 0>, <8 x i8> [[TMP6]], <8 x i32> <i32 0, i32 1, i32 8, i32 3, i32 4, i32 9, i32 6, i32 7>
18+
; CHECK-NEXT: [[TMP33:%.*]] = shufflevector <8 x i8> <i8 0, i8 0, i8 poison, i8 0, i8 0, i8 poison, i8 0, i8 0>, <8 x i8> [[TMP6]], <8 x i32> <i32 0, i32 1, i32 9, i32 3, i32 4, i32 poison, i32 6, i32 7>
19+
; CHECK-NEXT: [[TMP7:%.*]] = insertelement <8 x i8> [[TMP33]], i8 [[TMP2]], i32 5
1620
; CHECK-NEXT: [[TMP8:%.*]] = icmp eq <8 x i8> zeroinitializer, [[TMP7]]
17-
; CHECK-NEXT: [[TEST_STRUCTCOPY_14_S14_CM_COERCE_SROA_2_0_COPYLOAD:%.*]] = load i48, ptr addrspace(21) getelementptr inbounds (i8, ptr addrspace(21) null, i64 8), align 8
1821
; CHECK-NEXT: [[TMP9:%.*]] = insertelement <4 x i48> <i48 poison, i48 0, i48 0, i48 0>, i48 [[TEST_STRUCTCOPY_14_S14_CM_COERCE_SROA_2_0_COPYLOAD]], i32 0
1922
; CHECK-NEXT: [[TMP10:%.*]] = trunc <4 x i48> [[TMP9]] to <4 x i8>
2023
; CHECK-NEXT: [[TMP11:%.*]] = icmp eq <4 x i8> zeroinitializer, [[TMP10]]
21-
; CHECK-NEXT: [[TMP12:%.*]] = load i8, ptr addrspace(21) null, align 2
22-
; CHECK-NEXT: [[TMP13:%.*]] = load volatile i8, ptr null, align 2
2324
; CHECK-NEXT: [[TMP14:%.*]] = load <2 x i8>, ptr addrspace(21) getelementptr inbounds (i8, ptr addrspace(21) null, i64 8), align 8
2425
; CHECK-NEXT: [[TMP15:%.*]] = insertelement <8 x i8> <i8 0, i8 poison, i8 0, i8 poison, i8 poison, i8 0, i8 0, i8 0>, i8 [[TMP12]], i32 1
2526
; CHECK-NEXT: [[TMP16:%.*]] = shufflevector <2 x i8> [[TMP14]], <2 x i8> poison, <8 x i32> <i32 0, i32 1, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>

‎llvm/test/Transforms/SLPVectorizer/revec.ll

Lines changed: 5 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -306,13 +306,11 @@ define void @test11(<2 x i64> %0, i64 %1, <2 x i64> %2) {
306306
; CHECK-LABEL: @test11(
307307
; CHECK-NEXT: entry:
308308
; CHECK-NEXT: [[TMP3:%.*]] = insertelement <2 x i64> [[TMP0:%.*]], i64 [[TMP1:%.*]], i32 1
309-
; CHECK-NEXT: [[TMP4:%.*]] = add <2 x i64> <i64 5, i64 0>, [[TMP2:%.*]]
310-
; CHECK-NEXT: [[TMP5:%.*]] = trunc <2 x i64> [[TMP4]] to <2 x i16>
311-
; CHECK-NEXT: [[TMP6:%.*]] = shufflevector <2 x i16> [[TMP5]], <2 x i16> poison, <4 x i32> <i32 0, i32 1, i32 poison, i32 poison>
312-
; CHECK-NEXT: [[TMP7:%.*]] = trunc <2 x i64> [[TMP3]] to <2 x i16>
313-
; CHECK-NEXT: [[TMP10:%.*]] = shufflevector <2 x i16> [[TMP7]], <2 x i16> poison, <4 x i32> <i32 0, i32 1, i32 poison, i32 poison>
314-
; CHECK-NEXT: [[TMP8:%.*]] = shufflevector <4 x i16> [[TMP6]], <4 x i16> [[TMP10]], <4 x i32> <i32 0, i32 1, i32 4, i32 5>
315-
; CHECK-NEXT: [[TMP9:%.*]] = trunc <4 x i16> [[TMP8]] to <4 x i8>
309+
; CHECK-NEXT: [[TMP4:%.*]] = shufflevector <2 x i64> [[TMP2:%.*]], <2 x i64> poison, <4 x i32> <i32 0, i32 1, i32 poison, i32 poison>
310+
; CHECK-NEXT: [[TMP5:%.*]] = shufflevector <2 x i64> [[TMP3]], <2 x i64> poison, <4 x i32> <i32 0, i32 1, i32 poison, i32 poison>
311+
; CHECK-NEXT: [[TMP6:%.*]] = shufflevector <4 x i64> [[TMP4]], <4 x i64> [[TMP5]], <4 x i32> <i32 0, i32 1, i32 4, i32 5>
312+
; CHECK-NEXT: [[TMP7:%.*]] = add <4 x i64> <i64 5, i64 0, i64 0, i64 0>, [[TMP6]]
313+
; CHECK-NEXT: [[TMP9:%.*]] = trunc <4 x i64> [[TMP7]] to <4 x i8>
316314
; CHECK-NEXT: [[TMP11:%.*]] = urem <4 x i8> [[TMP9]], zeroinitializer
317315
; CHECK-NEXT: [[TMP12:%.*]] = icmp ne <4 x i8> [[TMP11]], zeroinitializer
318316
; CHECK-NEXT: ret void

0 commit comments

Comments
 (0)