diff options
Diffstat (limited to 'test/Transforms/LoopVectorize/X86')
8 files changed, 217 insertions, 15 deletions
diff --git a/test/Transforms/LoopVectorize/X86/already-vectorized.ll b/test/Transforms/LoopVectorize/X86/already-vectorized.ll index faed77d..fce3b70 100644 --- a/test/Transforms/LoopVectorize/X86/already-vectorized.ll +++ b/test/Transforms/LoopVectorize/X86/already-vectorized.ll @@ -40,7 +40,7 @@ for.end: ; preds = %for.body ; Now, we check for the Hint metadata ; CHECK: [[vect]] = metadata !{metadata [[vect]], metadata [[width:![0-9]+]], metadata [[unroll:![0-9]+]]} -; CHECK: [[width]] = metadata !{metadata !"llvm.vectorizer.width", i32 1} -; CHECK: [[unroll]] = metadata !{metadata !"llvm.vectorizer.unroll", i32 1} +; CHECK: [[width]] = metadata !{metadata !"llvm.loop.vectorize.width", i32 1} +; CHECK: [[unroll]] = metadata !{metadata !"llvm.loop.vectorize.unroll", i32 1} ; CHECK: [[scalar]] = metadata !{metadata [[scalar]], metadata [[width]], metadata [[unroll]]} diff --git a/test/Transforms/LoopVectorize/X86/avx512.ll b/test/Transforms/LoopVectorize/X86/avx512.ll new file mode 100644 index 0000000..a220866 --- /dev/null +++ b/test/Transforms/LoopVectorize/X86/avx512.ll @@ -0,0 +1,35 @@ +; RUN: opt -mattr=+avx512f --loop-vectorize -S < %s | llc -mattr=+avx512f | FileCheck %s + +target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128" +target triple = "x86_64-apple-macosx10.9.0" + +; Verify that we generate 512-bit wide vectors for a basic integer memset +; loop. + +; CHECK-LABEL: f: +; CHECK: vmovdqu32 %zmm{{.}}, ( +; CHECK-NOT: %ymm + +define void @f(i32* %a, i32 %n) { +entry: + %cmp4 = icmp sgt i32 %n, 0 + br i1 %cmp4, label %for.body.preheader, label %for.end + +for.body.preheader: ; preds = %entry + br label %for.body + +for.body: ; preds = %for.body.preheader, %for.body + %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %for.body.preheader ] + %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv + store i32 %n, i32* %arrayidx, align 4 + %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1 + %lftr.wideiv = trunc i64 %indvars.iv.next to i32 + %exitcond = icmp eq i32 %lftr.wideiv, %n + br i1 %exitcond, label %for.end.loopexit, label %for.body + +for.end.loopexit: ; preds = %for.body + br label %for.end + +for.end: ; preds = %for.end.loopexit, %entry + ret void +} diff --git a/test/Transforms/LoopVectorize/X86/lit.local.cfg b/test/Transforms/LoopVectorize/X86/lit.local.cfg index ba763cf..e71f3cc 100644 --- a/test/Transforms/LoopVectorize/X86/lit.local.cfg +++ b/test/Transforms/LoopVectorize/X86/lit.local.cfg @@ -1,4 +1,3 @@ -targets = set(config.root.targets_to_build.split()) -if not 'X86' in targets: +if not 'X86' in config.root.targets: config.unsupported = True diff --git a/test/Transforms/LoopVectorize/X86/metadata-enable.ll b/test/Transforms/LoopVectorize/X86/metadata-enable.ll index 9e4e989..8e0ca41 100644 --- a/test/Transforms/LoopVectorize/X86/metadata-enable.ll +++ b/test/Transforms/LoopVectorize/X86/metadata-enable.ll @@ -9,8 +9,9 @@ ; RUN: opt < %s -mcpu=corei7 -Oz -loop-vectorize -S -unroll-allow-partial=0 | FileCheck %s --check-prefix=OzVEC2 ; RUN: opt < %s -mcpu=corei7 -O3 -disable-loop-vectorization -S -unroll-allow-partial=0 | FileCheck %s --check-prefix=O3DIS -; This file tests the llvm.vectorizer.pragma forcing vectorization even when -; optimization levels are too low, or when vectorization is disabled. +; This file tests the llvm.loop.vectorize.enable metadata forcing +; vectorization even when optimization levels are too low, or when +; vectorization is disabled. 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-unknown-linux-gnu" @@ -170,6 +171,6 @@ for.end: ; preds = %for.body } !0 = metadata !{metadata !0, metadata !1} -!1 = metadata !{metadata !"llvm.vectorizer.enable", i1 1} +!1 = metadata !{metadata !"llvm.loop.vectorize.enable", i1 1} !2 = metadata !{metadata !2, metadata !3} -!3 = metadata !{metadata !"llvm.vectorizer.enable", i1 0} +!3 = metadata !{metadata !"llvm.loop.vectorize.enable", i1 0} diff --git a/test/Transforms/LoopVectorize/X86/vect.omp.force.ll b/test/Transforms/LoopVectorize/X86/vect.omp.force.ll index 84ffb27..074313b 100644 --- a/test/Transforms/LoopVectorize/X86/vect.omp.force.ll +++ b/test/Transforms/LoopVectorize/X86/vect.omp.force.ll @@ -53,7 +53,7 @@ for.end: } !1 = metadata !{metadata !1, metadata !2} -!2 = metadata !{metadata !"llvm.vectorizer.enable", i1 true} +!2 = metadata !{metadata !"llvm.loop.vectorize.enable", i1 true} ; ; This method will not be vectorized, as scalar cost is lower than any of vector costs. diff --git a/test/Transforms/LoopVectorize/X86/vect.omp.force.small-tc.ll b/test/Transforms/LoopVectorize/X86/vect.omp.force.small-tc.ll index 1b979e5..97c31a1 100644 --- a/test/Transforms/LoopVectorize/X86/vect.omp.force.small-tc.ll +++ b/test/Transforms/LoopVectorize/X86/vect.omp.force.small-tc.ll @@ -44,7 +44,7 @@ for.end: } !1 = metadata !{metadata !1, metadata !2} -!2 = metadata !{metadata !"llvm.vectorizer.enable", i1 true} +!2 = metadata !{metadata !"llvm.loop.vectorize.enable", i1 true} ; ; This loop will not be vectorized as the trip count is below the threshold. diff --git a/test/Transforms/LoopVectorize/X86/vectorization-remarks-missed.ll b/test/Transforms/LoopVectorize/X86/vectorization-remarks-missed.ll new file mode 100644 index 0000000..6cdd29b --- /dev/null +++ b/test/Transforms/LoopVectorize/X86/vectorization-remarks-missed.ll @@ -0,0 +1,160 @@ +; RUN: opt < %s -loop-vectorize -S -pass-remarks-missed='loop-vectorize' -pass-remarks-analysis='loop-vectorize' 2>&1 | FileCheck %s + +; C/C++ code for tests +; void test(int *A, int Length) { +; #pragma clang loop vectorize(enable) interleave(enable) +; for (int i = 0; i < Length; i++) { +; A[i] = i; +; if (A[i] > Length) +; break; +; } +; } + +; void test_disabled(int *A, int Length) { +; #pragma clang loop vectorize(disable) interleave(disable) +; for (int i = 0; i < Length; i++) +; A[i] = i; +; } + +; void test_array_bounds(int *A, int *B, int Length) { +; #pragma clang loop vectorize(enable) +; for (int i = 0; i < Length; i++) +; A[i] = A[B[i]]; +; } + +; File, line, and column should match those specified in the metadata +; CHECK: remark: source.cpp:4:5: loop not vectorized: could not determine number of loop iterations +; CHECK: remark: source.cpp:4:5: loop not vectorized: vectorization was not specified +; CHECK: remark: source.cpp:13:5: loop not vectorized: vector width and interleave count are explicitly set to 1 +; CHECK: remark: source.cpp:19:5: loop not vectorized: cannot identify array bounds +; CHECK: remark: source.cpp:19:5: loop not vectorized: vectorization is explicitly enabled + +; CHECK: _Z4testPii +; CHECK-NOT: x i32> +; CHECK: ret + +; CHECK: _Z13test_disabledPii +; CHECK-NOT: x i32> +; CHECK: ret + +; CHECK: _Z17test_array_boundsPiS_i +; CHECK-NOT: x i32> +; CHECK: ret + +target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128" + +; Function Attrs: nounwind optsize ssp uwtable +define void @_Z4testPii(i32* nocapture %A, i32 %Length) #0 { +entry: + %cmp10 = icmp sgt i32 %Length, 0, !dbg !12 + br i1 %cmp10, label %for.body, label %for.end, !dbg !12, !llvm.loop !14 + +for.body: ; preds = %entry, %for.body + %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ] + %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv, !dbg !16 + %0 = trunc i64 %indvars.iv to i32, !dbg !16 + store i32 %0, i32* %arrayidx, align 4, !dbg !16, !tbaa !18 + %cmp3 = icmp sle i32 %0, %Length, !dbg !22 + %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1, !dbg !12 + %1 = trunc i64 %indvars.iv.next to i32 + %cmp = icmp slt i32 %1, %Length, !dbg !12 + %or.cond = and i1 %cmp3, %cmp, !dbg !22 + br i1 %or.cond, label %for.body, label %for.end, !dbg !22 + +for.end: ; preds = %for.body, %entry + ret void, !dbg !24 +} + +; Function Attrs: nounwind optsize ssp uwtable +define void @_Z13test_disabledPii(i32* nocapture %A, i32 %Length) #0 { +entry: + %cmp4 = icmp sgt i32 %Length, 0, !dbg !25 + br i1 %cmp4, label %for.body, label %for.end, !dbg !25, !llvm.loop !27 + +for.body: ; preds = %entry, %for.body + %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ] + %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv, !dbg !30 + %0 = trunc i64 %indvars.iv to i32, !dbg !30 + store i32 %0, i32* %arrayidx, align 4, !dbg !30, !tbaa !18 + %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1, !dbg !25 + %lftr.wideiv = trunc i64 %indvars.iv.next to i32, !dbg !25 + %exitcond = icmp eq i32 %lftr.wideiv, %Length, !dbg !25 + br i1 %exitcond, label %for.end, label %for.body, !dbg !25, !llvm.loop !27 + +for.end: ; preds = %for.body, %entry + ret void, !dbg !31 +} + +; Function Attrs: nounwind optsize ssp uwtable +define void @_Z17test_array_boundsPiS_i(i32* nocapture %A, i32* nocapture readonly %B, i32 %Length) #0 { +entry: + %cmp9 = icmp sgt i32 %Length, 0, !dbg !32 + br i1 %cmp9, label %for.body.preheader, label %for.end, !dbg !32, !llvm.loop !34 + +for.body.preheader: ; preds = %entry + br label %for.body, !dbg !35 + +for.body: ; preds = %for.body.preheader, %for.body + %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %for.body.preheader ] + %arrayidx = getelementptr inbounds i32* %B, i64 %indvars.iv, !dbg !35 + %0 = load i32* %arrayidx, align 4, !dbg !35, !tbaa !18 + %idxprom1 = sext i32 %0 to i64, !dbg !35 + %arrayidx2 = getelementptr inbounds i32* %A, i64 %idxprom1, !dbg !35 + %1 = load i32* %arrayidx2, align 4, !dbg !35, !tbaa !18 + %arrayidx4 = getelementptr inbounds i32* %A, i64 %indvars.iv, !dbg !35 + store i32 %1, i32* %arrayidx4, align 4, !dbg !35, !tbaa !18 + %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1, !dbg !32 + %lftr.wideiv = trunc i64 %indvars.iv.next to i32, !dbg !32 + %exitcond = icmp eq i32 %lftr.wideiv, %Length, !dbg !32 + br i1 %exitcond, label %for.end.loopexit, label %for.body, !dbg !32, !llvm.loop !34 + +for.end.loopexit: ; preds = %for.body + br label %for.end + +for.end: ; preds = %for.end.loopexit, %entry + ret void, !dbg !36 +} + +attributes #0 = { nounwind } + +!llvm.dbg.cu = !{!0} +!llvm.module.flags = !{!9, !10} +!llvm.ident = !{!11} + +!0 = metadata !{i32 786449, metadata !1, i32 4, metadata !"clang version 3.5.0", i1 true, metadata !"", i32 0, metadata !2, metadata !2, metadata !3, metadata !2, metadata !2, metadata !"", i32 2} +!1 = metadata !{metadata !"source.cpp", metadata !"."} +!2 = metadata !{} +!3 = metadata !{metadata !4, metadata !7, metadata !8} +!4 = metadata !{i32 786478, metadata !1, metadata !5, metadata !"test", metadata !"test", metadata !"", i32 1, metadata !6, i1 false, i1 true, i32 0, i32 0, null, i32 256, i1 true, void (i32*, i32)* @_Z4testPii, null, null, metadata !2, i32 1} +!5 = metadata !{i32 786473, metadata !1} +!6 = metadata !{i32 786453, i32 0, null, metadata !"", i32 0, i64 0, i64 0, i64 0, i32 0, null, metadata !2, i32 0, null, null, null} +!7 = metadata !{i32 786478, metadata !1, metadata !5, metadata !"test_disabled", metadata !"test_disabled", metadata !"", i32 10, metadata !6, i1 false, i1 true, i32 0, i32 0, null, i32 256, i1 true, void (i32*, i32)* @_Z13test_disabledPii, null, null, metadata !2, i32 10} +!8 = metadata !{i32 786478, metadata !1, metadata !5, metadata !"test_array_bounds", metadata !"test_array_bounds", metadata !"", i32 16, metadata !6, i1 false, i1 true, i32 0, i32 0, null, i32 256, i1 true, void (i32*, i32*, i32)* @_Z17test_array_boundsPiS_i, null, null, metadata !2, i32 16} +!9 = metadata !{i32 2, metadata !"Dwarf Version", i32 2} +!10 = metadata !{i32 2, metadata !"Debug Info Version", i32 1} +!11 = metadata !{metadata !"clang version 3.5.0"} +!12 = metadata !{i32 3, i32 8, metadata !13, null} +!13 = metadata !{i32 786443, metadata !1, metadata !4, i32 3, i32 3, i32 0, i32 0} +!14 = metadata !{metadata !14, metadata !15, metadata !15} +!15 = metadata !{metadata !"llvm.loop.vectorize.enable", i1 true} +!16 = metadata !{i32 4, i32 5, metadata !17, null} +!17 = metadata !{i32 786443, metadata !1, metadata !13, i32 3, i32 36, i32 0, i32 1} +!18 = metadata !{metadata !19, metadata !19, i64 0} +!19 = metadata !{metadata !"int", metadata !20, i64 0} +!20 = metadata !{metadata !"omnipotent char", metadata !21, i64 0} +!21 = metadata !{metadata !"Simple C/C++ TBAA"} +!22 = metadata !{i32 5, i32 9, metadata !23, null} +!23 = metadata !{i32 786443, metadata !1, metadata !17, i32 5, i32 9, i32 0, i32 2} +!24 = metadata !{i32 8, i32 1, metadata !4, null} +!25 = metadata !{i32 12, i32 8, metadata !26, null} +!26 = metadata !{i32 786443, metadata !1, metadata !7, i32 12, i32 3, i32 0, i32 3} +!27 = metadata !{metadata !27, metadata !28, metadata !29} +!28 = metadata !{metadata !"llvm.loop.vectorize.unroll", i32 1} +!29 = metadata !{metadata !"llvm.loop.vectorize.width", i32 1} +!30 = metadata !{i32 13, i32 5, metadata !26, null} +!31 = metadata !{i32 14, i32 1, metadata !7, null} +!32 = metadata !{i32 18, i32 8, metadata !33, null} +!33 = metadata !{i32 786443, metadata !1, metadata !8, i32 18, i32 3, i32 0, i32 4} +!34 = metadata !{metadata !34, metadata !15} +!35 = metadata !{i32 19, i32 5, metadata !33, null} +!36 = metadata !{i32 20, i32 1, metadata !8, null} diff --git a/test/Transforms/LoopVectorize/X86/vectorization-remarks.ll b/test/Transforms/LoopVectorize/X86/vectorization-remarks.ll index 685d034..f683447 100644 --- a/test/Transforms/LoopVectorize/X86/vectorization-remarks.ll +++ b/test/Transforms/LoopVectorize/X86/vectorization-remarks.ll @@ -1,8 +1,17 @@ ; RUN: opt < %s -loop-vectorize -mtriple=x86_64-unknown-linux -S -pass-remarks='loop-vectorize' 2>&1 | FileCheck -check-prefix=VECTORIZED %s -; RUN: opt < %s -loop-vectorize -force-vector-width=1 -force-vector-unroll=4 -mtriple=x86_64-unknown-linux -S -pass-remarks='.*vectorize.*' 2>&1 | FileCheck -check-prefix=UNROLLED %s +; RUN: opt < %s -loop-vectorize -force-vector-width=1 -force-vector-unroll=4 -mtriple=x86_64-unknown-linux -S -pass-remarks='loop-vectorize' 2>&1 | FileCheck -check-prefix=UNROLLED %s +; RUN: opt < %s -loop-vectorize -force-vector-width=1 -force-vector-unroll=1 -mtriple=x86_64-unknown-linux -S -pass-remarks-analysis='loop-vectorize' 2>&1 | FileCheck -check-prefix=NONE %s -; VECTORIZED: remark: {{.*}}.c:17:8: vectorized loop (vectorization factor: 4, unrolling interleave factor: 1) -; UNROLLED: remark: {{.*}}.c:17:8: unrolled with interleaving factor 4 (vectorization not beneficial) +; This code has all the !dbg annotations needed to track source line information, +; but is missing the llvm.dbg.cu annotation. This prevents code generation from +; emitting debug info in the final output. +; RUN: llc -mtriple x86_64-pc-linux-gnu %s -o - | FileCheck -check-prefix=DEBUG-OUTPUT %s +; DEBUG-OUTPUT-NOT: .loc +; DEBUG-OUTPUT-NOT: {{.*}}.debug_info + +; VECTORIZED: remark: vectorization-remarks.c:17:8: vectorized loop (vectorization factor: 4, unrolling interleave factor: 1) +; UNROLLED: remark: vectorization-remarks.c:17:8: unrolled with interleaving factor 4 (vectorization not beneficial) +; NONE: remark: vectorization-remarks.c:17:8: loop not vectorized: vector width and interleave count are explicitly set to 1 target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128" @@ -37,11 +46,9 @@ for.end: ; preds = %for.body declare void @ibar(i32*) #1 -!llvm.dbg.cu = !{!0} !llvm.module.flags = !{!7, !8} !llvm.ident = !{!9} -!0 = metadata !{i32 786449, metadata !1, i32 12, metadata !"clang version 3.5.0 ", i1 true, metadata !"", i32 0, metadata !2, metadata !2, metadata !3, metadata !2, metadata !2, metadata !"", i32 2} ; [ DW_TAG_compile_unit ] [./vectorization-remarks.c] [DW_LANG_C99] !1 = metadata !{metadata !"vectorization-remarks.c", metadata !"."} !2 = metadata !{} !3 = metadata !{metadata !4} |