diff options
| author | Nadav Rotem <nrotem@apple.com> | 2012-09-29 06:33:25 +0000 |
|---|---|---|
| committer | Nadav Rotem <nrotem@apple.com> | 2012-09-29 06:33:25 +0000 |
| commit | a2e7ea2f183d328bc9960c34015ba975daf78a12 (patch) | |
| tree | 924826975e196417d607c3e0b0e3011e7aef0e99 /llvm/test/CodeGen/X86/loop-strength-reduce-3.ll | |
| parent | 4bd4e7e3ba8315c4180da954f0af40bbcb07f624 (diff) | |
| download | bcm5719-llvm-a2e7ea2f183d328bc9960c34015ba975daf78a12.tar.gz bcm5719-llvm-a2e7ea2f183d328bc9960c34015ba975daf78a12.zip | |
A DAGCombine optimization for merging consecutive stores. This optimization is not profitable in many cases
because moden processos can store multiple values in parallel, and preparing the consecutive store requires
some work. We only handle these cases:
1. Consecutive stores where the values and consecutive loads. For example:
int a = p->a;
int b = p->b;
q->a = a;
q->b = b;
2. Consecutive stores where the values are constants. Foe example:
q->a = 4;
q->b = 5;
llvm-svn: 164885
Diffstat (limited to 'llvm/test/CodeGen/X86/loop-strength-reduce-3.ll')
| -rw-r--r-- | llvm/test/CodeGen/X86/loop-strength-reduce-3.ll | 5 |
1 files changed, 2 insertions, 3 deletions
diff --git a/llvm/test/CodeGen/X86/loop-strength-reduce-3.ll b/llvm/test/CodeGen/X86/loop-strength-reduce-3.ll index b1c9fb9c077..e2d77f2a6fd 100644 --- a/llvm/test/CodeGen/X86/loop-strength-reduce-3.ll +++ b/llvm/test/CodeGen/X86/loop-strength-reduce-3.ll @@ -1,8 +1,7 @@ -; RUN: llc < %s -mtriple=i386-apple-darwin -relocation-model=dynamic-no-pic | FileCheck %s +; RUN: llc < %s -mtriple=i386-apple-darwin -mcpu=corei7 -relocation-model=dynamic-no-pic | FileCheck %s ; CHECK: align -; CHECK: movl $4, -4(%ecx) -; CHECK: movl $5, (%ecx) +; CHECK: movlpd %xmm0, -4(%ecx) ; CHECK: addl $4, %ecx ; CHECK: decl %eax ; CHECK: jne |
