diff options
| author | Nadav Rotem <nrotem@apple.com> | 2013-04-14 05:15:53 +0000 |
|---|---|---|
| committer | Nadav Rotem <nrotem@apple.com> | 2013-04-14 05:15:53 +0000 |
| commit | ab105ae95fc473c19d9f0b019fc7c7a16d17b1a5 (patch) | |
| tree | 27247e73ffdef91c52268c935d65afd7c066c1c0 /test/Transforms/SLPVectorizer | |
| parent | 618eda7a60bafff7741a988e27b98bf81d27cb89 (diff) | |
| download | external_llvm-ab105ae95fc473c19d9f0b019fc7c7a16d17b1a5.zip external_llvm-ab105ae95fc473c19d9f0b019fc7c7a16d17b1a5.tar.gz external_llvm-ab105ae95fc473c19d9f0b019fc7c7a16d17b1a5.tar.bz2 | |
SLPVectorizer: Add support for trees that don't start at binary operators, and add the cost of extracting values from the roots of the tree.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179475 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'test/Transforms/SLPVectorizer')
| -rw-r--r-- | test/Transforms/SLPVectorizer/X86/reduction2.ll | 37 |
1 files changed, 37 insertions, 0 deletions
diff --git a/test/Transforms/SLPVectorizer/X86/reduction2.ll b/test/Transforms/SLPVectorizer/X86/reduction2.ll new file mode 100644 index 0000000..9b5d5f7 --- /dev/null +++ b/test/Transforms/SLPVectorizer/X86/reduction2.ll @@ -0,0 +1,37 @@ +; RUN: opt < %s -basicaa -slp-vectorizer -dce -S -mtriple=x86_64-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" +target triple = "i386-apple-macosx10.8.0" + +;CHECK: @foo +;CHECK: load <2 x double> +;CHECK: ret +define double @foo(double* nocapture %D) #0 { + br label %1 + +; <label>:1 ; preds = %1, %0 + %i.02 = phi i32 [ 0, %0 ], [ %10, %1 ] + %sum.01 = phi double [ 0.000000e+00, %0 ], [ %9, %1 ] + %2 = shl nsw i32 %i.02, 1 + %3 = getelementptr inbounds double* %D, i32 %2 + %4 = load double* %3, align 4, !tbaa !0 + %A4 = fmul double %4, %4 + %5 = or i32 %2, 1 + %6 = getelementptr inbounds double* %D, i32 %5 + %7 = load double* %6, align 4, !tbaa !0 + %A7 = fmul double %7, %7 + %8 = fadd double %A4, %A7 + %9 = fadd double %sum.01, %8 + %10 = add nsw i32 %i.02, 1 + %exitcond = icmp eq i32 %10, 100 + br i1 %exitcond, label %11, label %1 + +; <label>:11 ; preds = %1 + ret double %9 +} + +attributes #0 = { nounwind readonly ssp "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf"="true" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "unsafe-fp-math"="false" "use-soft-float"="false" } + +!0 = metadata !{metadata !"double", metadata !1} +!1 = metadata !{metadata !"omnipotent char", metadata !2} +!2 = metadata !{metadata !"Simple C/C++ TBAA"} |