commit f7caf682c6b4769b2a3dd2f2241532b98709c1a3
author xueliang.zhong <xueliang.zhong@linaro.org> Wed Mar 01 16:07:02 2017 +0000
committer Nicolas Geoffray <ngeoffray@google.com> Mon May 08 08:33:07 2017 +0000
tree 88ecbd1fa756e722fd79e2b696620cd61ac3305f
parent 91f956c925f015b8cd355e8ec2a697a4ba69f7ab

Instruction scheduling for ARM.

Performance improvements on various benchmarks with this CL:

benchmarks     improvements
---------------------------
algorithm                1%
benchmarksgame           2%
caffeinemark             2%
math                     3%
stanford                 4%

Tested on ARM Cortex-A53 CPU.

The code size impact is negligible.

Test: m test-art-host
Test: m test-art-target
Change-Id: I314c90c09ce27e3d224fc686ef73c7d94a6b5a2c

compiler/Android.bp

diff --git a/compiler/Android.bp b/compiler/Android.bp
index a2b07af..df896dc 100644
--- a/compiler/Android.bp
+++ b/compiler/Android.bp
@@ -115,6 +115,7 @@
                 "optimizing/intrinsics_arm.cc",
                 "optimizing/intrinsics_arm_vixl.cc",
                 "optimizing/nodes_shared.cc",
+                "optimizing/scheduler_arm.cc",
                 "utils/arm/assembler_arm.cc",
                 "utils/arm/assembler_arm_vixl.cc",
                 "utils/arm/assembler_thumb2.cc",