RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-26 04:32:44 +01:00
Code Issues Projects Releases Wiki Activity

[AArch64] Disable complex GEP optimization by default.

Browse Source 
Enough concerns were raised that this optimization is pessimising some code patterns.

The obvious fix, to add a Reassociate run afterwards, causes even more pessimisation in some cases due to fewer complex addressing modes being matched. As there isn't a trivial fix for this, backing this out by default until someone gets a chance to fix the addressing mode matcher.

llvm-svn: 235491
This commit is contained in:
James Molloy 2015-04-22 09:11:38 +00:00
parent ae34e5b3f0
commit ffea00f649
2 changed files with 4 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
lib/Target/AArch64/AArch64TargetMachine.cpp
View File

@ -85,7 +85,7 @@ EnableA53Fix835769("aarch64-fix-cortex-a53-835769", cl::Hidden,
static cl::opt<bool> static cl::opt<bool>
EnableGEPOpt("aarch64-gep-opt", cl::Hidden, EnableGEPOpt("aarch64-gep-opt", cl::Hidden,
cl::desc("Enable optimizations on complex GEPs"), cl::desc("Enable optimizations on complex GEPs"),
cl::init(true)); cl::init(false));
// FIXME: Unify control over GlobalMerge. // FIXME: Unify control over GlobalMerge.
static cl::opt<cl::boolOrDefault> static cl::opt<cl::boolOrDefault>

6
test/CodeGen/AArch64/aarch64-gep-opt.ll
View File

@ -1,6 +1,6 @@
; RUN: llc -O3 -verify-machineinstrs %s -o - | FileCheck %s ; RUN: llc -O3 -aarch64-gep-opt=true -verify-machineinstrs %s -o - | FileCheck %s
; RUN: llc -O3 -print-after=codegenprepare -mcpu=cyclone < %s >%t 2>&1 && FileCheck --check-prefix=CHECK-NoAA <%t %s ; RUN: llc -O3 -aarch64-gep-opt=true -print-after=codegenprepare -mcpu=cyclone < %s >%t 2>&1 && FileCheck --check-prefix=CHECK-NoAA <%t %s
; RUN: llc -O3 -print-after=codegenprepare -mcpu=cortex-a53 < %s >%t 2>&1 && FileCheck --check-prefix=CHECK-UseAA <%t %s ; RUN: llc -O3 -aarch64-gep-opt=true -print-after=codegenprepare -mcpu=cortex-a53 < %s >%t 2>&1 && FileCheck --check-prefix=CHECK-UseAA <%t %s
target datalayout = "e-m:e-i64:64-i128:128-n32:64-S128" target datalayout = "e-m:e-i64:64-i128:128-n32:64-S128"
target triple = "aarch64-linux-gnueabi" target triple = "aarch64-linux-gnueabi"