; RUN: opt --vec-extabi=true -passes='default<O3>' -mcpu=pwr10 \
; RUN: -pgo-kind=pgo-instr-gen-pipeline -mtriple=powerpc-ibm-aix -S < %s | \
; RUN: FileCheck %s
; RUN: opt -passes='default<O3>' -mcpu=pwr10 -pgo-kind=pgo-instr-gen-pipeline \
; RUN: -mtriple=powerpc64le-unknown-linux-gnu -S < %s | FileCheck %s
; When running this test case under opt + PGO, the SLPVectorizer previously had
; an opportunity to produce wide vector types (such as <256 x i1>) within the
; IR as it deemed these wide vector types to be cheap enough to produce.
; Having this test ensures that the optimizer no longer generates wide vectors
; within the IR.
%0 = type <{ double }>
%1 = type <{ ptr, i8, i8, i8, i8, i32, i32, i32, [1 x i32], [1 x i32], [1 x i32], [24 x i8] }>
declare ptr @__malloc()
; CHECK-NOT: <256 x i1>
; CHECK-NOT: <512 x i1>
define dso_local void @test(ptr %arg, ptr %arg1, ptr %arg2, ptr %arg3, ptr %arg4) {
%i = alloca ptr, align 4
store ptr %arg, ptr %i, align 4
%i7 = alloca ptr, align 4
store ptr %arg1, ptr %i7, align 4
%i9 = alloca ptr, align 4
store ptr %arg2, ptr %i9, align 4
%i10 = alloca ptr, align 4
store ptr %arg3, ptr %i10, align 4
%i11 = alloca ptr, align 4
store ptr %arg4, ptr %i11, align 4
%i14 = alloca %1, align 4
%i15 = alloca i32, align 4
%i16 = alloca i32, align 4
%i17 = alloca i32, align 4
%i18 = alloca i32, align 4
%i20 = alloca i32, align 4
%i21 = alloca i32, align 4
%i22 = alloca i32, align 4
%i23 = alloca i32, align 4
%i25 = alloca double, align 8
%i26 = load ptr, ptr %i9, align 4
%i27 = load i32, ptr %i26, align 4
%i28 = select i1 false, i32 0, i32 %i27
store i32 %i28, ptr %i15, align 4
%i29 = load ptr, ptr %i7, align 4
%i30 = load i32, ptr %i29, align 4
%i31 = select i1 false, i32 0, i32 %i30
store i32 %i31, ptr %i16, align 4
%i32 = load i32, ptr %i15, align 4
%i33 = mul i32 8, %i32
store i32 %i33, ptr %i17, align 4
%i34 = load i32, ptr %i17, align 4
%i35 = load i32, ptr %i16, align 4
%i36 = mul i32 %i34, %i35
store i32 %i36, ptr %i18, align 4
%i37 = load ptr, ptr %i9, align 4
%i38 = load i32, ptr %i37, align 4
%i39 = select i1 false, i32 0, i32 %i38
store i32 %i39, ptr %i22, align 4
%i40 = load ptr, ptr %i10, align 4
%i41 = load i32, ptr %i40, align 4
%i42 = select i1 false, i32 0, i32 %i41
store i32 %i42, ptr %i23, align 4
%i43 = getelementptr inbounds %1, ptr %i14, i32 0, i32 10
%i45 = getelementptr i8, ptr %i43, i32 -12
%i46 = getelementptr inbounds i8, ptr %i45, i32 12
%i48 = load i32, ptr %i23, align 4
%i49 = select i1 false, i32 0, i32 %i48
%i50 = load i32, ptr %i22, align 4
%i51 = select i1 false, i32 0, i32 %i50
%i52 = mul i32 %i51, 8
%i53 = mul i32 %i49, %i52
store i32 %i53, ptr %i46, align 4
%i54 = getelementptr inbounds %1, ptr %i14, i32 0, i32 10
%i56 = getelementptr i8, ptr %i54, i32 -12
%i57 = getelementptr inbounds i8, ptr %i56, i32 36
store i32 8, ptr %i57, align 4
%i61 = call ptr @__malloc()
store ptr %i61, ptr %i14, align 4
br label %bb63
bb63: ; preds = %bb66, %bb
%i64 = load ptr, ptr %i11, align 4
%i65 = load i32, ptr %i64, align 4
br label %bb66
bb66: ; preds = %bb165, %bb63
%i67 = load i32, ptr %i21, align 4
%i68 = icmp sle i32 %i67, %i65
br i1 %i68, label %bb69, label %bb63
bb69: ; preds = %bb66
store i32 1, ptr %i20, align 4
br label %bb70
bb70: ; preds = %bb163, %bb69
%i71 = load i32, ptr %i20, align 4
%i72 = icmp sle i32 %i71, 11
br i1 %i72, label %bb73, label %bb165
bb73: ; preds = %bb70
%i74 = load i32, ptr %i21, align 4
%i76 = mul i32 %i74, 8
%i77 = getelementptr inbounds i8, ptr null, i32 %i76
%i79 = load double, ptr %i77, align 8
%i80 = fcmp fast olt double %i79, 0.000000e+00
%i81 = zext i1 %i80 to i32
%i82 = trunc i32 %i81 to i1
br i1 %i82, label %bb83, label %bb102
bb83: ; preds = %bb73
%i85 = load ptr, ptr %i14, align 4
%i88 = load i32, ptr %i20, align 4
%i89 = getelementptr inbounds %1, ptr %i14, i32 0, i32 10
%i91 = load i32, ptr %i89, align 4
%i92 = mul i32 %i88, %i91
%i93 = getelementptr inbounds i8, ptr %i85, i32 %i92
%i95 = load i32, ptr %i21, align 4
%i96 = getelementptr inbounds %1, ptr %i14, i32 0, i32 10
%i97 = getelementptr inbounds [1 x i32], ptr %i96, i32 0, i32 6
%i98 = load i32, ptr %i97, align 4
%i99 = mul i32 %i95, %i98
%i100 = getelementptr inbounds i8, ptr %i93, i32 %i99
store double 0.000000e+00, ptr %i100, align 8
br label %bb163
bb102: ; preds = %bb73
%i103 = getelementptr i8, ptr null, i32 -8
%i104 = getelementptr inbounds i8, ptr %i103, i32 undef
%i106 = load double, ptr %i104, align 8
%i107 = load ptr, ptr %i, align 4
%i109 = getelementptr i8, ptr %i107, i32 -8
%i110 = getelementptr inbounds i8, ptr %i109, i32 undef
%i112 = load double, ptr %i110, align 8
%i113 = fmul fast double %i106, %i112
%i114 = fcmp fast ogt double 0.000000e+00, %i113
%i115 = zext i1 %i114 to i32
%i116 = trunc i32 %i115 to i1
br i1 %i116, label %bb117, label %bb136
bb117: ; preds = %bb102
%i119 = load ptr, ptr %i14, align 4
%i122 = load i32, ptr %i20, align 4
%i123 = getelementptr inbounds %1, ptr %i14, i32 0, i32 10
%i125 = load i32, ptr %i123, align 4
%i126 = mul i32 %i122, %i125
%i127 = getelementptr inbounds i8, ptr %i119, i32 %i126
%i129 = load i32, ptr %i21, align 4
%i130 = getelementptr inbounds %1, ptr %i14, i32 0, i32 10
%i131 = getelementptr inbounds [1 x i32], ptr %i130, i32 0, i32 6
%i132 = load i32, ptr %i131, align 4
%i133 = mul i32 %i129, %i132
%i134 = getelementptr inbounds i8, ptr %i127, i32 %i133
store double 0.000000e+00, ptr %i134, align 8
br label %bb163
bb136: ; preds = %bb102
%i137 = load double, ptr null, align 8
%i138 = load double, ptr null, align 8
%i139 = fmul fast double %i137, %i138
%i140 = fsub fast double 0.000000e+00, %i139
store double %i140, ptr %i25, align 8
%i141 = load i32, ptr %i21, align 4
%i143 = getelementptr inbounds [1 x i32], ptr null, i32 0, i32 6
%i144 = load i32, ptr %i143, align 4
%i145 = mul i32 %i141, %i144
%i146 = getelementptr inbounds i8, ptr null, i32 %i145
%i148 = load i32, ptr %i20, align 4
%i149 = load i32, ptr %i18, align 4
%i151 = mul i32 %i148, %i149
%i152 = getelementptr i8, ptr null, i32 %i151
%i156 = load double, ptr %i152, align 8
%i157 = load double, ptr %i25, align 8
%i158 = fmul fast double %i156, %i157
%i159 = fadd fast double 0.000000e+00, %i158
%i160 = load double, ptr %i25, align 8
%i161 = fadd fast double 0.000000e+00, %i160
%i162 = fdiv fast double %i159, %i161
store double %i162, ptr %i146, align 8
br label %bb163
bb163: ; preds = %bb136, %bb117, %bb83
%i164 = add nsw i32 %i71, 1
store i32 %i164, ptr %i20, align 4
br label %bb70
bb165: ; preds = %bb70
%i166 = add nsw i32 %i67, 1
store i32 %i166, ptr %i21, align 4
br label %bb66
}
Codebase Browser
llvm