diff options
| author | Alexey Bataev <a.bataev@hotmail.com> | 2019-01-11 20:21:14 +0000 |
|---|---|---|
| committer | Alexey Bataev <a.bataev@hotmail.com> | 2019-01-11 20:21:14 +0000 |
| commit | ce2c8b3360c2270fb95a2a98bef2bedeef476415 (patch) | |
| tree | 2c19359ac4ce3ce868395b08b1666399c946bc4a /llvm/test/Transforms/SLPVectorizer/X86/align.ll | |
| parent | 6b7f5aac7270472115ad98b56919f8dcbc48a419 (diff) | |
| download | bcm5719-llvm-ce2c8b3360c2270fb95a2a98bef2bedeef476415.tar.gz bcm5719-llvm-ce2c8b3360c2270fb95a2a98bef2bedeef476415.zip | |
[SLP]Update test checks for the SPL vectorizer, NFC.
llvm-svn: 350967
Diffstat (limited to 'llvm/test/Transforms/SLPVectorizer/X86/align.ll')
| -rw-r--r-- | llvm/test/Transforms/SLPVectorizer/X86/align.ll | 45 |
1 files changed, 33 insertions, 12 deletions
diff --git a/llvm/test/Transforms/SLPVectorizer/X86/align.ll b/llvm/test/Transforms/SLPVectorizer/X86/align.ll index b74b70900ee..5c7c4ceae81 100644 --- a/llvm/test/Transforms/SLPVectorizer/X86/align.ll +++ b/llvm/test/Transforms/SLPVectorizer/X86/align.ll @@ -1,16 +1,31 @@ +; NOTE: Assertions have been autogenerated by utils/update_test_checks.py ; RUN: opt < %s -basicaa -slp-vectorizer -S -mtriple=x86_64-apple-macosx10.8.0 -mcpu=corei7-avx | FileCheck %s target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128" target triple = "x86_64-apple-macosx10.8.0" ; Simple 3-pair chain with loads and stores -; CHECK-LABEL: @test1 define void @test1(double* %a, double* %b, double* %c) { +; CHECK-LABEL: @test1( +; CHECK-NEXT: entry: +; CHECK-NEXT: [[AGG_TMP_I_I_SROA_0:%.*]] = alloca [3 x double], align 16 +; CHECK-NEXT: [[STORE1:%.*]] = getelementptr inbounds [3 x double], [3 x double]* [[AGG_TMP_I_I_SROA_0]], i64 0, i64 1 +; CHECK-NEXT: [[STORE2:%.*]] = getelementptr inbounds [3 x double], [3 x double]* [[AGG_TMP_I_I_SROA_0]], i64 0, i64 2 +; CHECK-NEXT: [[ARRAYIDX3:%.*]] = getelementptr inbounds double, double* [[A:%.*]], i64 1 +; CHECK-NEXT: [[TMP0:%.*]] = bitcast double* [[A]] to <2 x double>* +; CHECK-NEXT: [[TMP1:%.*]] = load <2 x double>, <2 x double>* [[TMP0]], align 8 +; CHECK-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds double, double* [[B:%.*]], i64 1 +; CHECK-NEXT: [[TMP2:%.*]] = bitcast double* [[B]] to <2 x double>* +; CHECK-NEXT: [[TMP3:%.*]] = load <2 x double>, <2 x double>* [[TMP2]], align 8 +; CHECK-NEXT: [[TMP4:%.*]] = fmul <2 x double> [[TMP1]], [[TMP3]] +; CHECK-NEXT: [[TMP5:%.*]] = bitcast double* [[STORE1]] to <2 x double>* +; CHECK-NEXT: store <2 x double> [[TMP4]], <2 x double>* [[TMP5]], align 8 +; CHECK-NEXT: ret void +; entry: %agg.tmp.i.i.sroa.0 = alloca [3 x double], align 16 -; CHECK: %[[V0:[0-9]+]] = load <2 x double>, <2 x double>* %[[V2:[0-9]+]], align 8 - %i0 = load double, double* %a - %i1 = load double, double* %b + %i0 = load double, double* %a + %i1 = load double, double* %b %mul = fmul double %i0, %i1 %store1 = getelementptr inbounds [3 x double], [3 x double]* %agg.tmp.i.i.sroa.0, i64 0, i64 1 %store2 = getelementptr inbounds [3 x double], [3 x double]* %agg.tmp.i.i.sroa.0, i64 0, i64 2 @@ -19,23 +34,29 @@ entry: %arrayidx4 = getelementptr inbounds double, double* %b, i64 1 %i4 = load double, double* %arrayidx4, align 8 %mul5 = fmul double %i3, %i4 -; CHECK: store <2 x double> %[[V1:[0-9]+]], <2 x double>* %[[V2:[0-9]+]], align 8 store double %mul, double* %store1 store double %mul5, double* %store2, align 16 -; CHECK: ret ret void } ; Float has 4 byte abi alignment on x86_64. We must use the alignmnet of the ; value being loaded/stored not the alignment of the pointer type. -; CHECK-LABEL: @test2 -; CHECK-NOT: align 8 -; CHECK: load <4 x float>{{.*}}, align 4 -; CHECK: store <4 x float>{{.*}}, align 4 -; CHECK: ret - define void @test2(float * %a, float * %b) { +; CHECK-LABEL: @test2( +; CHECK-NEXT: entry: +; CHECK-NEXT: [[A1:%.*]] = getelementptr inbounds float, float* [[A:%.*]], i64 1 +; CHECK-NEXT: [[A2:%.*]] = getelementptr inbounds float, float* [[A]], i64 2 +; CHECK-NEXT: [[A3:%.*]] = getelementptr inbounds float, float* [[A]], i64 3 +; CHECK-NEXT: [[TMP0:%.*]] = bitcast float* [[A]] to <4 x float>* +; CHECK-NEXT: [[TMP1:%.*]] = load <4 x float>, <4 x float>* [[TMP0]], align 4 +; CHECK-NEXT: [[B1:%.*]] = getelementptr inbounds float, float* [[B:%.*]], i64 1 +; CHECK-NEXT: [[B2:%.*]] = getelementptr inbounds float, float* [[B]], i64 2 +; CHECK-NEXT: [[B3:%.*]] = getelementptr inbounds float, float* [[B]], i64 3 +; CHECK-NEXT: [[TMP2:%.*]] = bitcast float* [[B]] to <4 x float>* +; CHECK-NEXT: store <4 x float> [[TMP1]], <4 x float>* [[TMP2]], align 4 +; CHECK-NEXT: ret void +; entry: %l0 = load float, float* %a %a1 = getelementptr inbounds float, float* %a, i64 1 |

