[llvm] r292783 - [SLP] Additional test for SLP vectorizer with 31 reduction elements.
Alexey Bataev via llvm-commits
llvm-commits at lists.llvm.org
Mon Jan 23 03:53:18 PST 2017
Author: abataev
Date: Mon Jan 23 05:53:16 2017
New Revision: 292783
URL: http://llvm.org/viewvc/llvm-project?rev=292783&view=rev
Log:
[SLP] Additional test for SLP vectorizer with 31 reduction elements.
Modified:
llvm/trunk/test/Transforms/SLPVectorizer/X86/horizontal-list.ll
Modified: llvm/trunk/test/Transforms/SLPVectorizer/X86/horizontal-list.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Transforms/SLPVectorizer/X86/horizontal-list.ll?rev=292783&r1=292782&r2=292783&view=diff
==============================================================================
--- llvm/trunk/test/Transforms/SLPVectorizer/X86/horizontal-list.ll (original)
+++ llvm/trunk/test/Transforms/SLPVectorizer/X86/horizontal-list.ll Mon Jan 23 05:53:16 2017
@@ -773,3 +773,199 @@ define float @f1(float* nocapture readon
ret float %add.31
}
+define float @loadadd31(float* nocapture readonly %x) {
+; CHECK-LABEL: @loadadd31(
+; CHECK-NEXT: entry:
+; CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds float, float* [[X:%.*]], i64 1
+; CHECK-NEXT: [[TMP0:%.*]] = load float, float* [[ARRAYIDX]], align 4
+; CHECK-NEXT: [[ARRAYIDX_1:%.*]] = getelementptr inbounds float, float* [[X]], i64 2
+; CHECK-NEXT: [[TMP1:%.*]] = load float, float* [[ARRAYIDX_1]], align 4
+; CHECK-NEXT: [[ADD_1:%.*]] = fadd fast float [[TMP1]], [[TMP0]]
+; CHECK-NEXT: [[ARRAYIDX_2:%.*]] = getelementptr inbounds float, float* [[X]], i64 3
+; CHECK-NEXT: [[TMP2:%.*]] = load float, float* [[ARRAYIDX_2]], align 4
+; CHECK-NEXT: [[ADD_2:%.*]] = fadd fast float [[TMP2]], [[ADD_1]]
+; CHECK-NEXT: [[ARRAYIDX_3:%.*]] = getelementptr inbounds float, float* [[X]], i64 4
+; CHECK-NEXT: [[TMP3:%.*]] = load float, float* [[ARRAYIDX_3]], align 4
+; CHECK-NEXT: [[ADD_3:%.*]] = fadd fast float [[TMP3]], [[ADD_2]]
+; CHECK-NEXT: [[ARRAYIDX_4:%.*]] = getelementptr inbounds float, float* [[X]], i64 5
+; CHECK-NEXT: [[TMP4:%.*]] = load float, float* [[ARRAYIDX_4]], align 4
+; CHECK-NEXT: [[ADD_4:%.*]] = fadd fast float [[TMP4]], [[ADD_3]]
+; CHECK-NEXT: [[ARRAYIDX_5:%.*]] = getelementptr inbounds float, float* [[X]], i64 6
+; CHECK-NEXT: [[TMP5:%.*]] = load float, float* [[ARRAYIDX_5]], align 4
+; CHECK-NEXT: [[ADD_5:%.*]] = fadd fast float [[TMP5]], [[ADD_4]]
+; CHECK-NEXT: [[ARRAYIDX_6:%.*]] = getelementptr inbounds float, float* [[X]], i64 7
+; CHECK-NEXT: [[TMP6:%.*]] = load float, float* [[ARRAYIDX_6]], align 4
+; CHECK-NEXT: [[ADD_6:%.*]] = fadd fast float [[TMP6]], [[ADD_5]]
+; CHECK-NEXT: [[ARRAYIDX_7:%.*]] = getelementptr inbounds float, float* [[X]], i64 8
+; CHECK-NEXT: [[TMP7:%.*]] = load float, float* [[ARRAYIDX_7]], align 4
+; CHECK-NEXT: [[ADD_7:%.*]] = fadd fast float [[TMP7]], [[ADD_6]]
+; CHECK-NEXT: [[ARRAYIDX_8:%.*]] = getelementptr inbounds float, float* [[X]], i64 9
+; CHECK-NEXT: [[TMP8:%.*]] = load float, float* [[ARRAYIDX_8]], align 4
+; CHECK-NEXT: [[ADD_8:%.*]] = fadd fast float [[TMP8]], [[ADD_7]]
+; CHECK-NEXT: [[ARRAYIDX_9:%.*]] = getelementptr inbounds float, float* [[X]], i64 10
+; CHECK-NEXT: [[TMP9:%.*]] = load float, float* [[ARRAYIDX_9]], align 4
+; CHECK-NEXT: [[ADD_9:%.*]] = fadd fast float [[TMP9]], [[ADD_8]]
+; CHECK-NEXT: [[ARRAYIDX_10:%.*]] = getelementptr inbounds float, float* [[X]], i64 11
+; CHECK-NEXT: [[TMP10:%.*]] = load float, float* [[ARRAYIDX_10]], align 4
+; CHECK-NEXT: [[ADD_10:%.*]] = fadd fast float [[TMP10]], [[ADD_9]]
+; CHECK-NEXT: [[ARRAYIDX_11:%.*]] = getelementptr inbounds float, float* [[X]], i64 12
+; CHECK-NEXT: [[TMP11:%.*]] = load float, float* [[ARRAYIDX_11]], align 4
+; CHECK-NEXT: [[ADD_11:%.*]] = fadd fast float [[TMP11]], [[ADD_10]]
+; CHECK-NEXT: [[ARRAYIDX_12:%.*]] = getelementptr inbounds float, float* [[X]], i64 13
+; CHECK-NEXT: [[TMP12:%.*]] = load float, float* [[ARRAYIDX_12]], align 4
+; CHECK-NEXT: [[ADD_12:%.*]] = fadd fast float [[TMP12]], [[ADD_11]]
+; CHECK-NEXT: [[ARRAYIDX_13:%.*]] = getelementptr inbounds float, float* [[X]], i64 14
+; CHECK-NEXT: [[TMP13:%.*]] = load float, float* [[ARRAYIDX_13]], align 4
+; CHECK-NEXT: [[ADD_13:%.*]] = fadd fast float [[TMP13]], [[ADD_12]]
+; CHECK-NEXT: [[ARRAYIDX_14:%.*]] = getelementptr inbounds float, float* [[X]], i64 15
+; CHECK-NEXT: [[ARRAYIDX_15:%.*]] = getelementptr inbounds float, float* [[X]], i64 16
+; CHECK-NEXT: [[ARRAYIDX_16:%.*]] = getelementptr inbounds float, float* [[X]], i64 17
+; CHECK-NEXT: [[ARRAYIDX_17:%.*]] = getelementptr inbounds float, float* [[X]], i64 18
+; CHECK-NEXT: [[ARRAYIDX_18:%.*]] = getelementptr inbounds float, float* [[X]], i64 19
+; CHECK-NEXT: [[ARRAYIDX_19:%.*]] = getelementptr inbounds float, float* [[X]], i64 20
+; CHECK-NEXT: [[ARRAYIDX_20:%.*]] = getelementptr inbounds float, float* [[X]], i64 21
+; CHECK-NEXT: [[ARRAYIDX_21:%.*]] = getelementptr inbounds float, float* [[X]], i64 22
+; CHECK-NEXT: [[ARRAYIDX_22:%.*]] = getelementptr inbounds float, float* [[X]], i64 23
+; CHECK-NEXT: [[ARRAYIDX_23:%.*]] = getelementptr inbounds float, float* [[X]], i64 24
+; CHECK-NEXT: [[ARRAYIDX_24:%.*]] = getelementptr inbounds float, float* [[X]], i64 25
+; CHECK-NEXT: [[ARRAYIDX_25:%.*]] = getelementptr inbounds float, float* [[X]], i64 26
+; CHECK-NEXT: [[ARRAYIDX_26:%.*]] = getelementptr inbounds float, float* [[X]], i64 27
+; CHECK-NEXT: [[ARRAYIDX_27:%.*]] = getelementptr inbounds float, float* [[X]], i64 28
+; CHECK-NEXT: [[ARRAYIDX_28:%.*]] = getelementptr inbounds float, float* [[X]], i64 29
+; CHECK-NEXT: [[ARRAYIDX_29:%.*]] = getelementptr inbounds float, float* [[X]], i64 30
+; CHECK-NEXT: [[TMP14:%.*]] = bitcast float* [[ARRAYIDX_14]] to <16 x float>*
+; CHECK-NEXT: [[TMP15:%.*]] = load <16 x float>, <16 x float>* [[TMP14]], align 4
+; CHECK-NEXT: [[ADD_14:%.*]] = fadd fast float undef, [[ADD_13]]
+; CHECK-NEXT: [[ADD_15:%.*]] = fadd fast float undef, [[ADD_14]]
+; CHECK-NEXT: [[ADD_16:%.*]] = fadd fast float undef, [[ADD_15]]
+; CHECK-NEXT: [[ADD_17:%.*]] = fadd fast float undef, [[ADD_16]]
+; CHECK-NEXT: [[ADD_18:%.*]] = fadd fast float undef, [[ADD_17]]
+; CHECK-NEXT: [[ADD_19:%.*]] = fadd fast float undef, [[ADD_18]]
+; CHECK-NEXT: [[ADD_20:%.*]] = fadd fast float undef, [[ADD_19]]
+; CHECK-NEXT: [[ADD_21:%.*]] = fadd fast float undef, [[ADD_20]]
+; CHECK-NEXT: [[ADD_22:%.*]] = fadd fast float undef, [[ADD_21]]
+; CHECK-NEXT: [[ADD_23:%.*]] = fadd fast float undef, [[ADD_22]]
+; CHECK-NEXT: [[ADD_24:%.*]] = fadd fast float undef, [[ADD_23]]
+; CHECK-NEXT: [[ADD_25:%.*]] = fadd fast float undef, [[ADD_24]]
+; CHECK-NEXT: [[ADD_26:%.*]] = fadd fast float undef, [[ADD_25]]
+; CHECK-NEXT: [[ADD_27:%.*]] = fadd fast float undef, [[ADD_26]]
+; CHECK-NEXT: [[ADD_28:%.*]] = fadd fast float undef, [[ADD_27]]
+; CHECK-NEXT: [[RDX_SHUF:%.*]] = shufflevector <16 x float> [[TMP15]], <16 x float> undef, <16 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT: [[BIN_RDX:%.*]] = fadd fast <16 x float> [[TMP15]], [[RDX_SHUF]]
+; CHECK-NEXT: [[RDX_SHUF1:%.*]] = shufflevector <16 x float> [[BIN_RDX]], <16 x float> undef, <16 x i32> <i32 4, i32 5, i32 6, i32 7, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT: [[BIN_RDX2:%.*]] = fadd fast <16 x float> [[BIN_RDX]], [[RDX_SHUF1]]
+; CHECK-NEXT: [[RDX_SHUF3:%.*]] = shufflevector <16 x float> [[BIN_RDX2]], <16 x float> undef, <16 x i32> <i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT: [[BIN_RDX4:%.*]] = fadd fast <16 x float> [[BIN_RDX2]], [[RDX_SHUF3]]
+; CHECK-NEXT: [[RDX_SHUF5:%.*]] = shufflevector <16 x float> [[BIN_RDX4]], <16 x float> undef, <16 x i32> <i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT: [[BIN_RDX6:%.*]] = fadd fast <16 x float> [[BIN_RDX4]], [[RDX_SHUF5]]
+; CHECK-NEXT: [[TMP16:%.*]] = extractelement <16 x float> [[BIN_RDX6]], i32 0
+; CHECK-NEXT: [[TMP17:%.*]] = fadd fast float [[TMP16]], [[TMP13]]
+; CHECK-NEXT: [[TMP18:%.*]] = fadd fast float [[TMP17]], [[TMP12]]
+; CHECK-NEXT: [[TMP19:%.*]] = fadd fast float [[TMP18]], [[TMP11]]
+; CHECK-NEXT: [[TMP20:%.*]] = fadd fast float [[TMP19]], [[TMP10]]
+; CHECK-NEXT: [[TMP21:%.*]] = fadd fast float [[TMP20]], [[TMP9]]
+; CHECK-NEXT: [[TMP22:%.*]] = fadd fast float [[TMP21]], [[TMP8]]
+; CHECK-NEXT: [[TMP23:%.*]] = fadd fast float [[TMP22]], [[TMP7]]
+; CHECK-NEXT: [[TMP24:%.*]] = fadd fast float [[TMP23]], [[TMP6]]
+; CHECK-NEXT: [[TMP25:%.*]] = fadd fast float [[TMP24]], [[TMP5]]
+; CHECK-NEXT: [[TMP26:%.*]] = fadd fast float [[TMP25]], [[TMP4]]
+; CHECK-NEXT: [[TMP27:%.*]] = fadd fast float [[TMP26]], [[TMP3]]
+; CHECK-NEXT: [[TMP28:%.*]] = fadd fast float [[TMP27]], [[TMP2]]
+; CHECK-NEXT: [[TMP29:%.*]] = fadd fast float [[TMP28]], [[TMP1]]
+; CHECK-NEXT: [[TMP30:%.*]] = fadd fast float [[TMP29]], [[TMP0]]
+; CHECK-NEXT: [[ADD_29:%.*]] = fadd fast float undef, [[ADD_28]]
+; CHECK-NEXT: ret float [[TMP30]]
+;
+ entry:
+ %arrayidx = getelementptr inbounds float, float* %x, i64 1
+ %0 = load float, float* %arrayidx, align 4
+ %arrayidx.1 = getelementptr inbounds float, float* %x, i64 2
+ %1 = load float, float* %arrayidx.1, align 4
+ %add.1 = fadd fast float %1, %0
+ %arrayidx.2 = getelementptr inbounds float, float* %x, i64 3
+ %2 = load float, float* %arrayidx.2, align 4
+ %add.2 = fadd fast float %2, %add.1
+ %arrayidx.3 = getelementptr inbounds float, float* %x, i64 4
+ %3 = load float, float* %arrayidx.3, align 4
+ %add.3 = fadd fast float %3, %add.2
+ %arrayidx.4 = getelementptr inbounds float, float* %x, i64 5
+ %4 = load float, float* %arrayidx.4, align 4
+ %add.4 = fadd fast float %4, %add.3
+ %arrayidx.5 = getelementptr inbounds float, float* %x, i64 6
+ %5 = load float, float* %arrayidx.5, align 4
+ %add.5 = fadd fast float %5, %add.4
+ %arrayidx.6 = getelementptr inbounds float, float* %x, i64 7
+ %6 = load float, float* %arrayidx.6, align 4
+ %add.6 = fadd fast float %6, %add.5
+ %arrayidx.7 = getelementptr inbounds float, float* %x, i64 8
+ %7 = load float, float* %arrayidx.7, align 4
+ %add.7 = fadd fast float %7, %add.6
+ %arrayidx.8 = getelementptr inbounds float, float* %x, i64 9
+ %8 = load float, float* %arrayidx.8, align 4
+ %add.8 = fadd fast float %8, %add.7
+ %arrayidx.9 = getelementptr inbounds float, float* %x, i64 10
+ %9 = load float, float* %arrayidx.9, align 4
+ %add.9 = fadd fast float %9, %add.8
+ %arrayidx.10 = getelementptr inbounds float, float* %x, i64 11
+ %10 = load float, float* %arrayidx.10, align 4
+ %add.10 = fadd fast float %10, %add.9
+ %arrayidx.11 = getelementptr inbounds float, float* %x, i64 12
+ %11 = load float, float* %arrayidx.11, align 4
+ %add.11 = fadd fast float %11, %add.10
+ %arrayidx.12 = getelementptr inbounds float, float* %x, i64 13
+ %12 = load float, float* %arrayidx.12, align 4
+ %add.12 = fadd fast float %12, %add.11
+ %arrayidx.13 = getelementptr inbounds float, float* %x, i64 14
+ %13 = load float, float* %arrayidx.13, align 4
+ %add.13 = fadd fast float %13, %add.12
+ %arrayidx.14 = getelementptr inbounds float, float* %x, i64 15
+ %14 = load float, float* %arrayidx.14, align 4
+ %add.14 = fadd fast float %14, %add.13
+ %arrayidx.15 = getelementptr inbounds float, float* %x, i64 16
+ %15 = load float, float* %arrayidx.15, align 4
+ %add.15 = fadd fast float %15, %add.14
+ %arrayidx.16 = getelementptr inbounds float, float* %x, i64 17
+ %16 = load float, float* %arrayidx.16, align 4
+ %add.16 = fadd fast float %16, %add.15
+ %arrayidx.17 = getelementptr inbounds float, float* %x, i64 18
+ %17 = load float, float* %arrayidx.17, align 4
+ %add.17 = fadd fast float %17, %add.16
+ %arrayidx.18 = getelementptr inbounds float, float* %x, i64 19
+ %18 = load float, float* %arrayidx.18, align 4
+ %add.18 = fadd fast float %18, %add.17
+ %arrayidx.19 = getelementptr inbounds float, float* %x, i64 20
+ %19 = load float, float* %arrayidx.19, align 4
+ %add.19 = fadd fast float %19, %add.18
+ %arrayidx.20 = getelementptr inbounds float, float* %x, i64 21
+ %20 = load float, float* %arrayidx.20, align 4
+ %add.20 = fadd fast float %20, %add.19
+ %arrayidx.21 = getelementptr inbounds float, float* %x, i64 22
+ %21 = load float, float* %arrayidx.21, align 4
+ %add.21 = fadd fast float %21, %add.20
+ %arrayidx.22 = getelementptr inbounds float, float* %x, i64 23
+ %22 = load float, float* %arrayidx.22, align 4
+ %add.22 = fadd fast float %22, %add.21
+ %arrayidx.23 = getelementptr inbounds float, float* %x, i64 24
+ %23 = load float, float* %arrayidx.23, align 4
+ %add.23 = fadd fast float %23, %add.22
+ %arrayidx.24 = getelementptr inbounds float, float* %x, i64 25
+ %24 = load float, float* %arrayidx.24, align 4
+ %add.24 = fadd fast float %24, %add.23
+ %arrayidx.25 = getelementptr inbounds float, float* %x, i64 26
+ %25 = load float, float* %arrayidx.25, align 4
+ %add.25 = fadd fast float %25, %add.24
+ %arrayidx.26 = getelementptr inbounds float, float* %x, i64 27
+ %26 = load float, float* %arrayidx.26, align 4
+ %add.26 = fadd fast float %26, %add.25
+ %arrayidx.27 = getelementptr inbounds float, float* %x, i64 28
+ %27 = load float, float* %arrayidx.27, align 4
+ %add.27 = fadd fast float %27, %add.26
+ %arrayidx.28 = getelementptr inbounds float, float* %x, i64 29
+ %28 = load float, float* %arrayidx.28, align 4
+ %add.28 = fadd fast float %28, %add.27
+ %arrayidx.29 = getelementptr inbounds float, float* %x, i64 30
+ %29 = load float, float* %arrayidx.29, align 4
+ %add.29 = fadd fast float %29, %add.28
+ ret float %add.29
+}
+
More information about the llvm-commits
mailing list