Android and the x86 Platform Sebastian Mauer GDG Aachen December Meetup December 16th, 2013 

Who am I? Sebastian Mauer GDG Aachen Co-Lead CS Student Software Engineer I don’t work for Google…yet 

Part I: x86: The new kid on the block? 

Android Platforms MIPS x86 ARM 

Android on x86 is already here 

Android VM Hardware Linux Kernel Dalvik VM App App App App App App 

From Source to Bytecode Java Sourcecode Java Bytecode (.class) Dalvik Bytecode (.dex) JAR Archive Dalvik VM Java VM 

It just works™ • The DalvikVM is already optimized for the x86 Platform • Apps relying on the Android SDK / Dalvik Bytecode
 will automatically benefit from Platform-specific Optimizations (like SSE & Co.) 

Part II: Going native. The NDK. 

What’s the NDK? • NDK stands for Native Development Kit • Allows to compile C/C++ code to native (read: platform specific) executables/libraries. • Has to be compiled for every platform you want to support 

What’s an NDK app? It’s an Android application that uses native libraries. ! Libraries are .so files, usually found inside libs/CPU_ABI/. ! These libs can be generated from native sources inside jni folder, game engines, or required by other 3rd party libraries. ! There is no 100% native application. Even an application purely written in C/C++, using native_app_glue.h, will be executed in the context of the Dalvik Virtual Machine. ! 

NDK Anatomy Native Platform Dalvik VM Dalvik Bytecode NDK compiled binary Java Native Interface (JNI) 

No content

Compile for all the platforms. If you have the source code of your native libraries, you can compile it for several CPU architectures by setting APP_ABI to all in the Makefile “jni/Application.mk”: ! APP_ABI=all The NDK will generate optimized code for all target ABIs You can also pass APP_ABI variable directly to ndk-build, and specify each ABI: ndk-build APP_ABI=x86 ARM v7a libs are built ARM v5 libs are built x86 libs are built mips libs are built Put APP_ABI=all inside Application.mk Run ndk-build… 

Fat Binaries Use lib/armeabi libraries Use lib/armeabi-v7a libraries Use lib/x86 libraries libs/armeabi-v7a libs/x86 libs/armeabi APK file … 

Part III: We must go deeper. Porting Processor-specific code 

SIMD Instructions • SIMD (Single instruction, multiple data) • NEON Instruction set on certain ARM CPUS • MMX™, Intel® SSE, SSE2, SSE3, SSSE3 on Intel® Atom™ processor based platforms 

Easy Way Out: NEONvsSSE.h //******* definition sample ***************** int8x8_t vadd_s8(int8x8_t a, int8x8_t b); // VADD.I8 d0,d0,d0 #ifdef USE_MMX #define vadd_s8 _mm_add_pi8 //MMX #else #define vadd_s8 _mm_add_epi8 #endif //… • Wraps NEON functions and intrinsics to SSE3 – 100% covered • Available at http://intel.ly/10JjuY4 

Memory Alignment By default Easy fix struct TestStruct { int mVar1; long long mVar2; int mVar3; }; struct TestStruct { int mVar1; long long mVar2 __attribute__ ((aligned(8))); int mVar3; }; 

Part IV: Ecosystem & Tools 

HAXM (Hardware Accelerated Execution Manager) 

Intel TBB (Threading Building Blocks) Specify tasks instead of manipulating threads ▪ Intel® Threading Building Blocks (Intel® TBB) maps your logical tasks onto threads with full support for nested parallelism Targets threading for scalable performance ▪ Uses proven efficient parallel patterns ▪ Uses work-stealing to support the load balance of unknown execution time for tasks Open source and licensed versions available on Linux*, Windows*, Mac OS X*, Android*… Open Source version available on: threadingbuildingblocks.org Licensed version available on: software.intel.com/en-us/intel-tbb 

Graphics Performance Analyzer • Profiles performance and Power • Real-time charts of CPU, GPU and power metrics • Conduct real-time experiments with OpenGL-ES* (with state overrides) to help narrow down problems • Triage system-level performance with CPU, GPU and Power metrics Available at intel.com/software/gpa 

HTML5 Development Environment (XDK) • Great tools for free • Convenient, cloud-based build tool lets you target all popular platforms & app stores • Write once, run anywhere; code once, debug once, publish everywhere more on: html5dev-software.intel.com 

Part V: The Future 

The Future: Silvermont New 22nm tri-gate microarchitecture
 ~3X more peak performance or ~5X lower power than previous Atom microarchitecture Intel® Atom™ Processor Z3000 Series (Bay Trail) 
 Next Generation Tablets Merrifield Next Generation Smartphones 

Q&A 

Thanks for your attention.