[llvm] r324612 - [SLPVectorizer] auto-generate complete checks; NFC
Sanjay Patel via llvm-commits
llvm-commits at lists.llvm.org
Thu Feb 8 07:16:26 PST 2018
Author: spatel
Date: Thu Feb 8 07:16:26 2018
New Revision: 324612
URL: http://llvm.org/viewvc/llvm-project?rev=324612&view=rev
Log:
[SLPVectorizer] auto-generate complete checks; NFC
Modified:
llvm/trunk/test/Transforms/SLPVectorizer/X86/cse.ll
Modified: llvm/trunk/test/Transforms/SLPVectorizer/X86/cse.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Transforms/SLPVectorizer/X86/cse.ll?rev=324612&r1=324611&r2=324612&view=diff
==============================================================================
--- llvm/trunk/test/Transforms/SLPVectorizer/X86/cse.ll (original)
+++ llvm/trunk/test/Transforms/SLPVectorizer/X86/cse.ll Thu Feb 8 07:16:26 2018
@@ -1,3 +1,4 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -basicaa -slp-vectorizer -S -mtriple=i386-apple-macosx10.8.0 -mcpu=corei7-avx | FileCheck %s
target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128-n8:16:32-S128"
@@ -10,13 +11,30 @@ target triple = "i386-apple-macosx10.8.0
; G[3] = 8+G[6]*4;
;}
-;CHECK-LABEL: @test(
-;CHECK: load <2 x double>
-;CHECK: fadd <4 x double>
-;CHECK: store <4 x double>
-;CHECK: ret i32
-
define i32 @test(double* nocapture %G) {
+; CHECK-LABEL: @test(
+; CHECK-NEXT: entry:
+; CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds double, double* [[G:%.*]], i64 5
+; CHECK-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds double, double* [[G]], i64 6
+; CHECK-NEXT: [[TMP0:%.*]] = bitcast double* [[ARRAYIDX]] to <2 x double>*
+; CHECK-NEXT: [[TMP1:%.*]] = load <2 x double>, <2 x double>* [[TMP0]], align 8
+; CHECK-NEXT: [[TMP2:%.*]] = fmul <2 x double> <double 4.000000e+00, double 3.000000e+00>, [[TMP1]]
+; CHECK-NEXT: [[ARRAYIDX5:%.*]] = getelementptr inbounds double, double* [[G]], i64 1
+; CHECK-NEXT: [[ARRAYIDX9:%.*]] = getelementptr inbounds double, double* [[G]], i64 2
+; CHECK-NEXT: [[TMP3:%.*]] = extractelement <2 x double> [[TMP1]], i32 1
+; CHECK-NEXT: [[MUL11:%.*]] = fmul double [[TMP3]], 4.000000e+00
+; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x double> [[TMP2]], i32 0
+; CHECK-NEXT: [[TMP5:%.*]] = insertelement <4 x double> undef, double [[TMP4]], i32 0
+; CHECK-NEXT: [[TMP6:%.*]] = extractelement <2 x double> [[TMP2]], i32 1
+; CHECK-NEXT: [[TMP7:%.*]] = insertelement <4 x double> [[TMP5]], double [[TMP6]], i32 1
+; CHECK-NEXT: [[TMP8:%.*]] = insertelement <4 x double> [[TMP7]], double [[TMP4]], i32 2
+; CHECK-NEXT: [[TMP9:%.*]] = insertelement <4 x double> [[TMP8]], double [[MUL11]], i32 3
+; CHECK-NEXT: [[TMP10:%.*]] = fadd <4 x double> <double 1.000000e+00, double 6.000000e+00, double 7.000000e+00, double 8.000000e+00>, [[TMP9]]
+; CHECK-NEXT: [[ARRAYIDX13:%.*]] = getelementptr inbounds double, double* [[G]], i64 3
+; CHECK-NEXT: [[TMP11:%.*]] = bitcast double* [[G]] to <4 x double>*
+; CHECK-NEXT: store <4 x double> [[TMP10]], <4 x double>* [[TMP11]], align 8
+; CHECK-NEXT: ret i32 undef
+;
entry:
%arrayidx = getelementptr inbounds double, double* %G, i64 5
%0 = load double, double* %arrayidx, align 8
@@ -45,13 +63,27 @@ entry:
; A[2] = A[2] * 7.6 * n + 3.0;
; A[3] = A[3] * 7.4 * n + 4.0;
;}
-; CHECK-LABEL: @foo(
-; CHECK: load <4 x double>
-; CHECK: fmul <4 x double>
-; CHECK: fmul <4 x double>
-; CHECK: fadd <4 x double>
-; CHECK: store <4 x double>
+
define i32 @foo(double* nocapture %A, i32 %n) {
+; CHECK-LABEL: @foo(
+; CHECK-NEXT: entry:
+; CHECK-NEXT: [[CONV:%.*]] = sitofp i32 [[N:%.*]] to double
+; CHECK-NEXT: [[ARRAYIDX3:%.*]] = getelementptr inbounds double, double* [[A:%.*]], i64 1
+; CHECK-NEXT: [[ARRAYIDX9:%.*]] = getelementptr inbounds double, double* [[A]], i64 2
+; CHECK-NEXT: [[ARRAYIDX15:%.*]] = getelementptr inbounds double, double* [[A]], i64 3
+; CHECK-NEXT: [[TMP0:%.*]] = bitcast double* [[A]] to <4 x double>*
+; CHECK-NEXT: [[TMP1:%.*]] = load <4 x double>, <4 x double>* [[TMP0]], align 8
+; CHECK-NEXT: [[TMP2:%.*]] = fmul <4 x double> <double 7.900000e+00, double 7.700000e+00, double 7.600000e+00, double 7.400000e+00>, [[TMP1]]
+; CHECK-NEXT: [[TMP3:%.*]] = insertelement <4 x double> undef, double [[CONV]], i32 0
+; CHECK-NEXT: [[TMP4:%.*]] = insertelement <4 x double> [[TMP3]], double [[CONV]], i32 1
+; CHECK-NEXT: [[TMP5:%.*]] = insertelement <4 x double> [[TMP4]], double [[CONV]], i32 2
+; CHECK-NEXT: [[TMP6:%.*]] = insertelement <4 x double> [[TMP5]], double [[CONV]], i32 3
+; CHECK-NEXT: [[TMP7:%.*]] = fmul <4 x double> [[TMP6]], [[TMP2]]
+; CHECK-NEXT: [[TMP8:%.*]] = fadd <4 x double> <double 6.000000e+00, double 2.000000e+00, double 3.000000e+00, double 4.000000e+00>, [[TMP7]]
+; CHECK-NEXT: [[TMP9:%.*]] = bitcast double* [[A]] to <4 x double>*
+; CHECK-NEXT: store <4 x double> [[TMP8]], <4 x double>* [[TMP9]], align 8
+; CHECK-NEXT: ret i32 undef
+;
entry:
%0 = load double, double* %A, align 8
%mul = fmul double %0, 7.900000e+00
@@ -91,13 +123,37 @@ entry:
; }
; We can't merge the gather sequences because one does not dominate the other.
-; CHECK-LABEL: @test2(
-; CHECK: insertelement
-; CHECK: insertelement
-; CHECK: insertelement
-; CHECK: insertelement
-; CHECK: ret
+
define i32 @test2(double* nocapture %G, i32 %k) {
+; CHECK-LABEL: @test2(
+; CHECK-NEXT: [[TMP1:%.*]] = icmp eq i32 [[K:%.*]], 0
+; CHECK-NEXT: [[TMP2:%.*]] = getelementptr inbounds double, double* [[G:%.*]], i64 5
+; CHECK-NEXT: [[TMP3:%.*]] = load double, double* [[TMP2]], align 8
+; CHECK-NEXT: [[TMP4:%.*]] = fmul double [[TMP3]], 4.000000e+00
+; CHECK-NEXT: br i1 [[TMP1]], label [[TMP14:%.*]], label [[TMP5:%.*]]
+; CHECK: [[TMP6:%.*]] = getelementptr inbounds double, double* [[G]], i64 6
+; CHECK-NEXT: [[TMP7:%.*]] = load double, double* [[TMP6]], align 8
+; CHECK-NEXT: [[TMP8:%.*]] = fmul double [[TMP7]], 3.000000e+00
+; CHECK-NEXT: [[TMP9:%.*]] = insertelement <2 x double> undef, double [[TMP4]], i32 0
+; CHECK-NEXT: [[TMP10:%.*]] = insertelement <2 x double> [[TMP9]], double [[TMP8]], i32 1
+; CHECK-NEXT: [[TMP11:%.*]] = fadd <2 x double> <double 1.000000e+00, double 6.000000e+00>, [[TMP10]]
+; CHECK-NEXT: [[TMP12:%.*]] = getelementptr inbounds double, double* [[G]], i64 1
+; CHECK-NEXT: [[TMP13:%.*]] = bitcast double* [[G]] to <2 x double>*
+; CHECK-NEXT: store <2 x double> [[TMP11]], <2 x double>* [[TMP13]], align 8
+; CHECK-NEXT: br label [[TMP24:%.*]]
+; CHECK: [[TMP15:%.*]] = getelementptr inbounds double, double* [[G]], i64 2
+; CHECK-NEXT: [[TMP16:%.*]] = getelementptr inbounds double, double* [[G]], i64 6
+; CHECK-NEXT: [[TMP17:%.*]] = load double, double* [[TMP16]], align 8
+; CHECK-NEXT: [[TMP18:%.*]] = fmul double [[TMP17]], 3.000000e+00
+; CHECK-NEXT: [[TMP19:%.*]] = insertelement <2 x double> undef, double [[TMP4]], i32 0
+; CHECK-NEXT: [[TMP20:%.*]] = insertelement <2 x double> [[TMP19]], double [[TMP18]], i32 1
+; CHECK-NEXT: [[TMP21:%.*]] = fadd <2 x double> <double 7.000000e+00, double 8.000000e+00>, [[TMP20]]
+; CHECK-NEXT: [[TMP22:%.*]] = getelementptr inbounds double, double* [[G]], i64 3
+; CHECK-NEXT: [[TMP23:%.*]] = bitcast double* [[TMP15]] to <2 x double>*
+; CHECK-NEXT: store <2 x double> [[TMP21]], <2 x double>* [[TMP23]], align 8
+; CHECK-NEXT: br label [[TMP24]]
+; CHECK: ret i32 undef
+;
%1 = icmp eq i32 %k, 0
%2 = getelementptr inbounds double, double* %G, i64 5
%3 = load double, double* %2, align 8
@@ -138,13 +194,27 @@ define i32 @test2(double* nocapture %G,
; A[2] = A[2] * 7.9 * n + 6.0;
; A[3] = A[3] * 7.9 * n + 6.0;
;}
-; CHECK-LABEL: @foo4(
-; CHECK: load <4 x double>
-; CHECK: fmul <4 x double>
-; CHECK: fmul <4 x double>
-; CHECK: fadd <4 x double>
-; CHECK: store <4 x double>
+
define i32 @foo4(double* nocapture %A, i32 %n) {
+; CHECK-LABEL: @foo4(
+; CHECK-NEXT: entry:
+; CHECK-NEXT: [[CONV:%.*]] = sitofp i32 [[N:%.*]] to double
+; CHECK-NEXT: [[ARRAYIDX3:%.*]] = getelementptr inbounds double, double* [[A:%.*]], i64 1
+; CHECK-NEXT: [[ARRAYIDX9:%.*]] = getelementptr inbounds double, double* [[A]], i64 2
+; CHECK-NEXT: [[ARRAYIDX15:%.*]] = getelementptr inbounds double, double* [[A]], i64 3
+; CHECK-NEXT: [[TMP0:%.*]] = bitcast double* [[A]] to <4 x double>*
+; CHECK-NEXT: [[TMP1:%.*]] = load <4 x double>, <4 x double>* [[TMP0]], align 8
+; CHECK-NEXT: [[TMP2:%.*]] = fmul <4 x double> <double 7.900000e+00, double 7.900000e+00, double 7.900000e+00, double 7.900000e+00>, [[TMP1]]
+; CHECK-NEXT: [[TMP3:%.*]] = insertelement <4 x double> undef, double [[CONV]], i32 0
+; CHECK-NEXT: [[TMP4:%.*]] = insertelement <4 x double> [[TMP3]], double [[CONV]], i32 1
+; CHECK-NEXT: [[TMP5:%.*]] = insertelement <4 x double> [[TMP4]], double [[CONV]], i32 2
+; CHECK-NEXT: [[TMP6:%.*]] = insertelement <4 x double> [[TMP5]], double [[CONV]], i32 3
+; CHECK-NEXT: [[TMP7:%.*]] = fmul <4 x double> [[TMP6]], [[TMP2]]
+; CHECK-NEXT: [[TMP8:%.*]] = fadd <4 x double> <double 6.000000e+00, double 6.000000e+00, double 6.000000e+00, double 6.000000e+00>, [[TMP7]]
+; CHECK-NEXT: [[TMP9:%.*]] = bitcast double* [[A]] to <4 x double>*
+; CHECK-NEXT: store <4 x double> [[TMP8]], <4 x double>* [[TMP9]], align 8
+; CHECK-NEXT: ret i32 undef
+;
entry:
%0 = load double, double* %A, align 8
%mul = fmul double %0, 7.900000e+00
@@ -180,13 +250,36 @@ entry:
; A[2] = A[2] * n;
; A[3] = A[3] * (n+4);
;}
-;CHECK-LABEL: @partial_mrg(
-;CHECK: insertelement <2 x double>
-;CHECK: insertelement <2 x double>
-;CHECK: insertelement <2 x double>
-;CHECK-NOT: insertelement <2 x double>
-;CHECK: ret
+
define i32 @partial_mrg(double* nocapture %A, i32 %n) {
+; CHECK-LABEL: @partial_mrg(
+; CHECK-NEXT: entry:
+; CHECK-NEXT: [[CONV:%.*]] = sitofp i32 [[N:%.*]] to double
+; CHECK-NEXT: [[ARRAYIDX2:%.*]] = 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: [[TMP2:%.*]] = insertelement <2 x double> undef, double [[CONV]], i32 0
+; CHECK-NEXT: [[TMP3:%.*]] = insertelement <2 x double> [[TMP2]], double [[CONV]], i32 1
+; CHECK-NEXT: [[TMP4:%.*]] = fmul <2 x double> [[TMP3]], [[TMP1]]
+; CHECK-NEXT: [[TMP5:%.*]] = bitcast double* [[A]] to <2 x double>*
+; CHECK-NEXT: store <2 x double> [[TMP4]], <2 x double>* [[TMP5]], align 8
+; CHECK-NEXT: [[CMP:%.*]] = icmp slt i32 [[N]], 4
+; CHECK-NEXT: br i1 [[CMP]], label [[RETURN:%.*]], label [[IF_END:%.*]]
+; CHECK: if.end:
+; CHECK-NEXT: [[ARRAYIDX7:%.*]] = getelementptr inbounds double, double* [[A]], i64 2
+; CHECK-NEXT: [[ARRAYIDX11:%.*]] = getelementptr inbounds double, double* [[A]], i64 3
+; CHECK-NEXT: [[TMP6:%.*]] = bitcast double* [[ARRAYIDX7]] to <2 x double>*
+; CHECK-NEXT: [[TMP7:%.*]] = load <2 x double>, <2 x double>* [[TMP6]], align 8
+; CHECK-NEXT: [[ADD:%.*]] = add nsw i32 [[N]], 4
+; CHECK-NEXT: [[CONV12:%.*]] = sitofp i32 [[ADD]] to double
+; CHECK-NEXT: [[TMP8:%.*]] = insertelement <2 x double> [[TMP2]], double [[CONV12]], i32 1
+; CHECK-NEXT: [[TMP9:%.*]] = fmul <2 x double> [[TMP8]], [[TMP7]]
+; CHECK-NEXT: [[TMP10:%.*]] = bitcast double* [[ARRAYIDX7]] to <2 x double>*
+; CHECK-NEXT: store <2 x double> [[TMP9]], <2 x double>* [[TMP10]], align 8
+; CHECK-NEXT: br label [[RETURN]]
+; CHECK: return:
+; CHECK-NEXT: ret i32 0
+;
entry:
%0 = load double, double* %A, align 8
%conv = sitofp i32 %n to double
@@ -222,6 +315,28 @@ return:
@a = external global double, align 8
define void @PR19646(%class.B.53.55* %this) {
+; CHECK-LABEL: @PR19646(
+; CHECK-NEXT: entry:
+; CHECK-NEXT: br i1 undef, label [[IF_END13:%.*]], label [[IF_END13]]
+; CHECK: sw.epilog7:
+; CHECK-NEXT: [[DOTIN:%.*]] = getelementptr inbounds [[CLASS_B_53_55:%.*]], %class.B.53.55* [[THIS:%.*]], i64 0, i32 0, i32 1
+; CHECK-NEXT: [[TMP0:%.*]] = load double, double* [[DOTIN]], align 8
+; CHECK-NEXT: [[ADD:%.*]] = fadd double undef, 0.000000e+00
+; CHECK-NEXT: [[ADD6:%.*]] = fadd double [[ADD]], [[TMP0]]
+; CHECK-NEXT: [[TMP1:%.*]] = load double, double* @a, align 8
+; CHECK-NEXT: [[ADD8:%.*]] = fadd double [[TMP1]], 0.000000e+00
+; CHECK-NEXT: [[_DY:%.*]] = getelementptr inbounds [[CLASS_B_53_55]], %class.B.53.55* [[THIS]], i64 0, i32 0, i32 2
+; CHECK-NEXT: [[TMP2:%.*]] = load double, double* [[_DY]], align 8
+; CHECK-NEXT: [[ADD10:%.*]] = fadd double [[ADD8]], [[TMP2]]
+; CHECK-NEXT: br i1 undef, label [[IF_THEN12:%.*]], label [[IF_END13]]
+; CHECK: if.then12:
+; CHECK-NEXT: [[TMP3:%.*]] = load double, double* undef, align 8
+; CHECK-NEXT: br label [[IF_END13]]
+; CHECK: if.end13:
+; CHECK-NEXT: [[X_1:%.*]] = phi double [ 0.000000e+00, [[IF_THEN12]] ], [ [[ADD6]], [[SW_EPILOG7:%.*]] ], [ undef, [[ENTRY:%.*]] ], [ undef, [[ENTRY]] ]
+; CHECK-NEXT: [[B_0:%.*]] = phi double [ [[TMP3]], [[IF_THEN12]] ], [ [[ADD10]], [[SW_EPILOG7]] ], [ undef, [[ENTRY]] ], [ undef, [[ENTRY]] ]
+; CHECK-NEXT: unreachable
+;
entry:
br i1 undef, label %if.end13, label %if.end13
More information about the llvm-commits
mailing list