diff options
| author | 2016-07-29 09:51:58 -0700 | |
|---|---|---|
| committer | 2016-08-16 20:19:36 +0000 | |
| commit | 9d4b6da934934c322536ee3309b63ce402740f49 (patch) | |
| tree | 9e7ee5023d6036b98e0560411bb0527efdedca01 /compiler/optimizing/optimizing_compiler.cc | |
| parent | 2af1aa066e3d20edd8fea5d5b6dbbbad73102d52 (diff) | |
jni: Fast path for @FastNative annotated java methods
Adds a faster path for java methods annotated with
dalvik.annotation.optimization.FastNative .
Intended to replace usage of fast JNI (registering with "!(FOO)BAR" descriptors).
Performance Microbenchmark Results (Angler):
* Regular JNI cost in nanoseconds: 115
* Fast JNI cost in nanoseconds: 60
* @FastNative cost in nanoseconds: 36
Summary: Up to 67% faster (vs fast jni) JNI transition cost
Change-Id: Ic23823ae0f232270c068ec999fd89aa993894b0e
Diffstat (limited to 'compiler/optimizing/optimizing_compiler.cc')
| -rw-r--r-- | compiler/optimizing/optimizing_compiler.cc | 9 |
1 files changed, 7 insertions, 2 deletions
diff --git a/compiler/optimizing/optimizing_compiler.cc b/compiler/optimizing/optimizing_compiler.cc index f7c82d1987..6aaa15fa02 100644 --- a/compiler/optimizing/optimizing_compiler.cc +++ b/compiler/optimizing/optimizing_compiler.cc @@ -283,8 +283,13 @@ class OptimizingCompiler FINAL : public Compiler { CompiledMethod* JniCompile(uint32_t access_flags, uint32_t method_idx, - const DexFile& dex_file) const OVERRIDE { - return ArtQuickJniCompileMethod(GetCompilerDriver(), access_flags, method_idx, dex_file); + const DexFile& dex_file, + JniOptimizationFlags optimization_flags) const OVERRIDE { + return ArtQuickJniCompileMethod(GetCompilerDriver(), + access_flags, + method_idx, + dex_file, + optimization_flags); } uintptr_t GetEntryPointOf(ArtMethod* method) const OVERRIDE |