1; This test checks that the !llvm.loop metadata has been updated after inlining 2; so that the start and end locations refer to the inlined DILocations. 3 4; RUN: opt -inline -always-inline %s -S 2>&1 | FileCheck %s 5; CHECK: br i1 %{{.*}}, label %middle.block.i, label %vector.body.i, !dbg !{{[0-9]+}}, !llvm.loop [[VECTOR:![0-9]+]] 6; CHECK: br i1 %{{.*}}, label %for.cond.cleanup.loopexit.i, label %for.body.i, !dbg !{{[0-9]+}}, !llvm.loop [[SCALAR:![0-9]+]] 7; CHECK-DAG: [[VECTOR]] = distinct !{[[VECTOR]], [[START:![0-9]+]], [[END:![0-9]+]], [[IS_VECTORIZED:![0-9]+]]} 8; CHECK-DAG: [[SCALAR]] = distinct !{[[SCALAR]], [[START]], [[END]], [[NO_UNROLL:![0-9]+]], [[IS_VECTORIZED]]} 9; CHECK-DAG: [[IS_VECTORIZED]] = !{!"llvm.loop.isvectorized", i32 1} 10; CHECK-DAG: [[NO_UNROLL]] = !{!"llvm.loop.unroll.runtime.disable"} 11 12; This IR can be generated by running: 13; clang -emit-llvm -S -gmlt -O2 inlined.cpp -o - -mllvm -opt-bisect-limit=53 |\ 14; opt -loop-vectorize 15; 16; Where inlined.cpp contains: 17; extern int *Array; 18; static int bar(unsigned x) 19; { 20; int Ret = 0; 21; for (unsigned i = 0; i < x; ++i) 22; { 23; Ret += Array[i] * i; 24; } 25; return Ret; 26; } 27; 28; int foo(unsigned x) 29; { 30; int Bar = bar(x); 31; return Bar; 32; } 33 34@"?Array@@3PEAHEA" = external dso_local local_unnamed_addr global i32*, align 8 35 36define dso_local i32 @"?foo@@YAHI@Z"(i32 %x) local_unnamed_addr !dbg !8 { 37entry: 38 %call = call fastcc i32 @"?bar@@YAHI@Z"(i32 %x), !dbg !10 39 ret i32 %call, !dbg !11 40} 41 42define internal fastcc i32 @"?bar@@YAHI@Z"(i32 %x) unnamed_addr !dbg !12 { 43entry: 44 %cmp7 = icmp eq i32 %x, 0, !dbg !13 45 br i1 %cmp7, label %for.cond.cleanup, label %for.body.lr.ph, !dbg !13 46 47for.body.lr.ph: ; preds = %entry 48 %0 = load i32*, i32** @"?Array@@3PEAHEA", align 8, !dbg !14, !tbaa !15 49 %wide.trip.count = zext i32 %x to i64, !dbg !14 50 %min.iters.check = icmp ult i64 %wide.trip.count, 8, !dbg !13 51 br i1 %min.iters.check, label %scalar.ph, label %vector.ph, !dbg !13 52 53vector.ph: ; preds = %for.body.lr.ph 54 %n.mod.vf = urem i64 %wide.trip.count, 8, !dbg !13 55 %n.vec = sub i64 %wide.trip.count, %n.mod.vf, !dbg !13 56 br label %vector.body, !dbg !13 57 58vector.body: ; preds = %vector.body, %vector.ph 59 %index = phi i64 [ 0, %vector.ph ], [ %index.next, %vector.body ], !dbg !13 60 %vec.ind = phi <4 x i64> [ <i64 0, i64 1, i64 2, i64 3>, %vector.ph ], [ %vec.ind.next, %vector.body ] 61 %vec.phi = phi <4 x i32> [ zeroinitializer, %vector.ph ], [ %17, %vector.body ] 62 %vec.phi2 = phi <4 x i32> [ zeroinitializer, %vector.ph ], [ %18, %vector.body ] 63 %vec.ind4 = phi <4 x i32> [ <i32 0, i32 1, i32 2, i32 3>, %vector.ph ], [ %vec.ind.next7, %vector.body ], !dbg !19 64 %step.add = add <4 x i64> %vec.ind, <i64 4, i64 4, i64 4, i64 4> 65 %1 = add i64 %index, 0, !dbg !13 66 %2 = add i64 %index, 1, !dbg !13 67 %3 = add i64 %index, 2, !dbg !13 68 %4 = add i64 %index, 3, !dbg !13 69 %5 = add i64 %index, 4, !dbg !13 70 %6 = add i64 %index, 5, !dbg !13 71 %7 = add i64 %index, 6, !dbg !13 72 %8 = add i64 %index, 7, !dbg !13 73 %9 = getelementptr inbounds i32, i32* %0, i64 %1, !dbg !19 74 %10 = getelementptr inbounds i32, i32* %0, i64 %5, !dbg !19 75 %11 = getelementptr inbounds i32, i32* %9, i32 0, !dbg !19 76 %12 = bitcast i32* %11 to <4 x i32>*, !dbg !19 77 %wide.load = load <4 x i32>, <4 x i32>* %12, align 4, !dbg !19, !tbaa !20 78 %13 = getelementptr inbounds i32, i32* %9, i32 4, !dbg !19 79 %14 = bitcast i32* %13 to <4 x i32>*, !dbg !19 80 %wide.load3 = load <4 x i32>, <4 x i32>* %14, align 4, !dbg !19, !tbaa !20 81 %step.add5 = add <4 x i32> %vec.ind4, <i32 4, i32 4, i32 4, i32 4>, !dbg !19 82 %15 = mul <4 x i32> %wide.load, %vec.ind4, !dbg !19 83 %16 = mul <4 x i32> %wide.load3, %step.add5, !dbg !19 84 %17 = add <4 x i32> %15, %vec.phi, !dbg !19 85 %18 = add <4 x i32> %16, %vec.phi2, !dbg !19 86 %index.next = add i64 %index, 8, !dbg !13 87 %vec.ind.next = add <4 x i64> %step.add, <i64 4, i64 4, i64 4, i64 4> 88 %vec.ind.next7 = add <4 x i32> %step.add5, <i32 4, i32 4, i32 4, i32 4>, !dbg !19 89 %19 = icmp eq i64 %index.next, %n.vec, !dbg !13 90 br i1 %19, label %middle.block, label %vector.body, !dbg !13, !llvm.loop !22 91 92middle.block: ; preds = %vector.body 93 %bin.rdx = add <4 x i32> %18, %17, !dbg !19 94 %rdx.shuf = shufflevector <4 x i32> %bin.rdx, <4 x i32> poison, <4 x i32> <i32 2, i32 3, i32 undef, i32 undef>, !dbg !19 95 %bin.rdx8 = add <4 x i32> %bin.rdx, %rdx.shuf, !dbg !19 96 %rdx.shuf9 = shufflevector <4 x i32> %bin.rdx8, <4 x i32> poison, <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>, !dbg !19 97 %bin.rdx10 = add <4 x i32> %bin.rdx8, %rdx.shuf9, !dbg !19 98 %20 = extractelement <4 x i32> %bin.rdx10, i32 0, !dbg !19 99 %cmp.n = icmp eq i64 %wide.trip.count, %n.vec, !dbg !13 100 br i1 %cmp.n, label %for.cond.cleanup.loopexit, label %scalar.ph, !dbg !13 101 102scalar.ph: ; preds = %middle.block, %for.body.lr.ph 103 %bc.resume.val = phi i64 [ %n.vec, %middle.block ], [ 0, %for.body.lr.ph ] 104 %bc.merge.rdx = phi i32 [ 0, %for.body.lr.ph ], [ %20, %middle.block ] 105 br label %for.body, !dbg !13 106 107for.cond.cleanup.loopexit: ; preds = %middle.block, %for.body 108 %add.lcssa = phi i32 [ %add, %for.body ], [ %20, %middle.block ], !dbg !19 109 br label %for.cond.cleanup, !dbg !25 110 111for.cond.cleanup: ; preds = %for.cond.cleanup.loopexit, %entry 112 %Ret.0.lcssa = phi i32 [ 0, %entry ], [ %add.lcssa, %for.cond.cleanup.loopexit ], !dbg !14 113 ret i32 %Ret.0.lcssa, !dbg !25 114 115for.body: ; preds = %for.body, %scalar.ph 116 %indvars.iv = phi i64 [ %bc.resume.val, %scalar.ph ], [ %indvars.iv.next, %for.body ] 117 %Ret.08 = phi i32 [ %bc.merge.rdx, %scalar.ph ], [ %add, %for.body ] 118 %arrayidx = getelementptr inbounds i32, i32* %0, i64 %indvars.iv, !dbg !19 119 %21 = load i32, i32* %arrayidx, align 4, !dbg !19, !tbaa !20 120 %22 = trunc i64 %indvars.iv to i32, !dbg !19 121 %mul = mul i32 %21, %22, !dbg !19 122 %add = add i32 %mul, %Ret.08, !dbg !19 123 %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1, !dbg !13 124 %exitcond = icmp eq i64 %indvars.iv.next, %wide.trip.count, !dbg !13 125 br i1 %exitcond, label %for.cond.cleanup.loopexit, label %for.body, !dbg !13, !llvm.loop !26 126} 127 128!llvm.dbg.cu = !{!0} 129!llvm.module.flags = !{!3, !4, !5, !6} 130!llvm.ident = !{!7} 131 132!0 = distinct !DICompileUnit(language: DW_LANG_C_plus_plus, file: !1, producer: "clang version 9.0.0 (https://github.com/llvm/llvm-project.git b1e28d9b6a16380ccf1456fe0695f639364407a9)", isOptimized: true, runtimeVersion: 0, emissionKind: LineTablesOnly, enums: !2, nameTableKind: None) 133!1 = !DIFile(filename: "inlined.cpp", directory: "") 134!2 = !{} 135!3 = !{i32 2, !"CodeView", i32 1} 136!4 = !{i32 2, !"Debug Info Version", i32 3} 137!5 = !{i32 1, !"wchar_size", i32 2} 138!6 = !{i32 7, !"PIC Level", i32 2} 139!7 = !{!"clang version 9.0.0 (https://github.com/llvm/llvm-project.git b1e28d9b6a16380ccf1456fe0695f639364407a9)"} 140!8 = distinct !DISubprogram(name: "foo", scope: !1, file: !1, line: 13, type: !9, scopeLine: 14, flags: DIFlagPrototyped, spFlags: DISPFlagDefinition | DISPFlagOptimized, unit: !0, retainedNodes: !2) 141!9 = !DISubroutineType(types: !2) 142!10 = !DILocation(line: 15, scope: !8) 143!11 = !DILocation(line: 16, scope: !8) 144!12 = distinct !DISubprogram(name: "bar", scope: !1, file: !1, line: 3, type: !9, scopeLine: 4, flags: DIFlagPrototyped, spFlags: DISPFlagLocalToUnit | DISPFlagDefinition | DISPFlagOptimized, unit: !0, retainedNodes: !2) 145!13 = !DILocation(line: 6, scope: !12) 146!14 = !DILocation(line: 0, scope: !12) 147!15 = !{!16, !16, i64 0} 148!16 = !{!"any pointer", !17, i64 0} 149!17 = !{!"omnipotent char", !18, i64 0} 150!18 = !{!"Simple C++ TBAA"} 151!19 = !DILocation(line: 8, scope: !12) 152!20 = !{!21, !21, i64 0} 153!21 = !{!"int", !17, i64 0} 154!22 = distinct !{!22, !13, !23, !24} 155!23 = !DILocation(line: 9, scope: !12) 156!24 = !{!"llvm.loop.isvectorized", i32 1} 157!25 = !DILocation(line: 10, scope: !12) 158!26 = distinct !{!26, !13, !23, !27, !24} 159!27 = !{!"llvm.loop.unroll.runtime.disable"} 160