diff options
| author | Duncan Sands <baldrick@free.fr> | 2012-09-29 10:25:35 +0000 |
|---|---|---|
| committer | Duncan Sands <baldrick@free.fr> | 2012-09-29 10:25:35 +0000 |
| commit | 454627252b1cc43e81949d41eb20e9ea9560da58 (patch) | |
| tree | f17db9a4b50aee70ecbabfe49763d454584dc6f1 /test/CodeGen/X86/loop-strength-reduce-2.ll | |
| parent | 0eb5dadf657d38da9a8c7fe44c660bcfb6933038 (diff) | |
| download | external_llvm-454627252b1cc43e81949d41eb20e9ea9560da58.zip external_llvm-454627252b1cc43e81949d41eb20e9ea9560da58.tar.gz external_llvm-454627252b1cc43e81949d41eb20e9ea9560da58.tar.bz2 | |
Speculatively revert commit 164885 (nadav) in the hope of ressurecting a pile of
buildbots. Original commit message:
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;
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@164890 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'test/CodeGen/X86/loop-strength-reduce-2.ll')
| -rw-r--r-- | test/CodeGen/X86/loop-strength-reduce-2.ll | 10 |
1 files changed, 6 insertions, 4 deletions
diff --git a/test/CodeGen/X86/loop-strength-reduce-2.ll b/test/CodeGen/X86/loop-strength-reduce-2.ll index b094fed..b546462 100644 --- a/test/CodeGen/X86/loop-strength-reduce-2.ll +++ b/test/CodeGen/X86/loop-strength-reduce-2.ll @@ -1,18 +1,20 @@ -; RUN: llc < %s -march=x86 -mcpu=corei7 -relocation-model=pic | FileCheck %s -check-prefix=PIC -; RUN: llc < %s -march=x86 -mcpu=corei7 -relocation-model=static | FileCheck %s -check-prefix=STATIC +; RUN: llc < %s -march=x86 -relocation-model=pic | FileCheck %s -check-prefix=PIC +; RUN: llc < %s -march=x86 -relocation-model=static | FileCheck %s -check-prefix=STATIC ; ; Make sure the common loop invariant A is hoisted up to preheader, ; since too many registers are needed to subsume it into the addressing modes. ; It's safe to sink A in when it's not pic. ; PIC: align -; PIC: movlpd %xmm0, -4([[REG:%e[a-z]+]]) +; PIC: movl $4, -4([[REG:%e[a-z]+]]) +; PIC: movl $5, ([[REG]]) ; PIC: addl $4, [[REG]] ; PIC: decl {{%e[[a-z]+}} ; PIC: jne ; STATIC: align -; STATIC: movlpd %xmm0, -4(%ecx) +; STATIC: movl $4, -4(%ecx) +; STATIC: movl $5, (%ecx) ; STATIC: addl $4, %ecx ; STATIC: decl %eax ; STATIC: jne |