ANDROID DEVELOPER TOOLS TRAINING GTC 2014. Sébastien Dominé, NVIDIA



Similar documents
TEGRA X1 DEVELOPER TOOLS SEBASTIEN DOMINE, SR. DIRECTOR SW ENGINEERING

How To Develop For A Powergen 2.2 (Tegra) With Nsight) And Gbd (Gbd) On A Quadriplegic (Powergen) Powergen Powergen 3

CUDA Tools for Debugging and Profiling. Jiri Kraus (NVIDIA)

GPU Tools Sandra Wienke

Optimizing Application Performance with CUDA Profiling Tools

GPU Profiling with AMD CodeXL

Getting Started with Tizen SDK : How to develop a Web app. Hong Gyungpyo 洪 競 杓 Samsung Electronics Co., Ltd

Guided Performance Analysis with the NVIDIA Visual Profiler

NVIDIA CUDA GETTING STARTED GUIDE FOR MAC OS X

STLinux Software development environment

Android Development: Part One

Development With ARM DS-5. Mervyn Liu FAE Aug. 2015

Performance Optimization and Debug Tools for mobile games with PlayCanvas

Reminders. Lab opens from today. Many students want to use the extra I/O pins on

How To Develop Android On Your Computer Or Tablet Or Phone

NVIDIA CUDA GETTING STARTED GUIDE FOR MAC OS X

NVIDIA CUDA GETTING STARTED GUIDE FOR MICROSOFT WINDOWS

Finding Performance and Power Issues on Android Systems. By Eric W Moore

Visualizing gem5 via ARM DS-5 Streamline. Dam Sunwoo ARM R&D December 2012

Developer Tools. Tim Purcell NVIDIA

Android Development: a System Perspective. Javier Orensanz

Eddy Integrated Development Environment, LemonIDE for Embedded Software System Development

Getting Started with CodeXL

Università Degli Studi di Parma. Distributed Systems Group. Android Development. Lecture 1 Android SDK & Development Environment. Marco Picone

Introducing the IBM Software Development Kit for PowerLinux

AMD CodeXL 1.7 GA Release Notes

01. Introduction of Android

Introduction to OpenCV for Tegra. Shalini Gupta, Nvidia

Project SHIELD and Tegra 4: Redefining AFK

NVIDIA Tools For Profiling And Monitoring. David Goodwin

High Performance or Cycle Accuracy?

Cloud Gaming & Application Delivery with NVIDIA GRID Technologies. Franck DIARD, Ph.D. GRID Architect, NVIDIA

Running a Program on an AVD

GPU Performance Analysis and Optimisation

Introduction to Android

Lecture 3 Mobile App Development (Android, ios, BlackBerry, Windows Mobile) <lecturer, date>

Overview of CS 282 & Android

NVIDIA Tegra Android Platform Support Pack Getting Started Guide

l What is Android? l Getting Started l The Emulator l Hello World l ADB l Text to Speech l Other APIs (camera, bitmap, etc)

HPC Wales Skills Academy Course Catalogue 2015

Tizen SDK Annual Report Key Improvements. Changseok Oh (Samsung), Yeongkyoon Lee (S-Core)

OMX, Android, GStreamer How do I decide what to use? 15 July 2011

NVIDIA GRID OVERVIEW SERVER POWERED BY NVIDIA GRID. WHY GPUs FOR VIRTUAL DESKTOPS AND APPLICATIONS? WHAT IS A VIRTUAL DESKTOP?

Embedded Development Tools

Android Development. Lecture AD 0 Android SDK & Development Environment. Università degli Studi di Parma. Mobile Application Development

Easing embedded Linux software development for SBCs

XID ERRORS. vr352 May XID Errors

[PACKTl. Flash Development for Android Cookbook. Flash, Flex, and AIR. Joseph Labrecque. Over 90 recipes to build exciting Android applications with

An Easier Way for Cross-Platform Data Acquisition Application Development

CLOUD GAMING WITH NVIDIA GRID TECHNOLOGIES Franck DIARD, Ph.D., SW Chief Software Architect GDC 2014

Next Generation GPU Architecture Code-named Fermi

Application Framework: Apertis Hands-on

Multi-/Many-core Modeling at Freescale

Real-time Debugging using GDB Tracepoints and other Eclipse features

Applications to Computational Financial and GPU Computing. May 16th. Dr. Daniel Egloff

An Introduction to Android

NVIDIA GeForce GTX 580 GPU Datasheet

E6895 Advanced Big Data Analytics Lecture 14:! NVIDIA GPU Examples and GPU on ios devices

VMware and NVIDIA: Bringing Workstations to the cloud

What else can you do with Android? Inside Android. Chris Simmonds. Embedded Linux Conference Europe Copyright 2010, 2net Limited.

for Java developers Building Mobile Applications Introduction 1 Building Mobile Applications

Basics of VTune Performance Analyzer. Intel Software College. Objectives. VTune Performance Analyzer. Agenda

Development Kit (MCSDK) Training

Instrumentation Software Profiling

Get the Best out of NVIDIA GPUs for 3D Design and Engineering in the Cloud

ODP Application proof point: OpenFastPath. ODP mini-summit

Red Hat Linux Internals

The MaXX Desktop. Workstation Environment. Revised Road Map Version 0.7. for Graphics Professionals

Node.JS Appliances on Embedded Linux Devices. Mehmet Fatih Karagöz & Cevahir Turgut

The "Eclipse Classic" version is recommended. Otherwise, a Java or RCP version of Eclipse is recommended.

Complete Integrated Development Platform Copyright Atmel Corporation

QuickSpecs. NVIDIA Quadro K5200 8GB Graphics INTRODUCTION. NVIDIA Quadro K5200 8GB Graphics. Technical Specifications

NVIDIA CUDA INSTALLATION GUIDE FOR MICROSOFT WINDOWS

ENEA BARE METAL PERFORMANCE TOOLS FOR NETLOGIC XLP AND CAVIUM OCTEON PLUS

Manjrasoft Market Oriented Cloud Computing Platform

The power of root on Android emulators

Enhanced Project Management for Embedded C/C++ Programming using Software Components

Android NDK Native Development Kit

Virtual Desktop VMware View Horizon

Developing with Android Studio

Installation Guide. (Version ) Midland Valley Exploration Ltd 144 West George Street Glasgow G2 2HG United Kingdom

Equalizer. Parallel OpenGL Application Framework. Stefan Eilemann, Eyescale Software GmbH

Q N X S O F T W A R E D E V E L O P M E N T P L A T F O R M v Steps to Developing a QNX Program Quickstart Guide

CS3813 Performance Monitoring Project

Introduction to NaviGenie SDK Client API for Android

1) SETUP ANDROID STUDIO

Intel Integrated Native Developer Experience (INDE): IDE Integration for Android*

Performance Architect Remote Storage (Intern)

System Requirements G E N E R A L S Y S T E M R E C O M M E N D A T I O N S

Workshop on Android and Applications Development

QuickSpecs. NVIDIA Quadro K5200 8GB Graphics INTRODUCTION. NVIDIA Quadro K5200 8GB Graphics. Overview. NVIDIA Quadro K5200 8GB Graphics J3G90AA

Application-Level Debugging and Profiling: Gaps in the Tool Ecosystem. Dr Rosemary Francis, Ellexus

Development_Setting. Step I: Create an Android Project

Hands-on CUDA exercises

Fahim Uddin 1. Java SDK

Vulkan on NVIDIA GPUs. Piers Daniell, Driver Software Engineer, OpenGL and Vulkan

IOTIVITY AND EMBEDDED LINUX SUPPORT. Kishen Maloor Intel Open Source Technology Center

Accelerate your Mobile Apps and Games for Android on ARM. Matthew Du Puy Software Engineer, ARM

Developing applications on Yocto. Lianhao Lu Intel Corporation Feb. 29th, 2012

Virtualization: Hypervisors for Embedded and Safe Systems. Hanspeter Vogel Triadem Solutions AG

Transcription:

ANDROID DEVELOPER TOOLS TRAINING GTC 2014 Sébastien Dominé, NVIDIA

AGENDA NVIDIA Developer Tools Introduction Multi-core CPU tools Graphics Developer Tools Compute Developer Tools

NVIDIA Developer Tools Build, Debug and Profile Deploy IDE integration NVIDIA Nsight Performance Monitoring SDK Hardware Support GPU and CPU Performance and Power Monitoring GPU Shader and Kernel Debugging and Profiling Standalone Tools Local and Remote Debugging Tegra Android Developer Pack

Tegra Android Development Pack For Windows, OSX, Linux GET STARTED in minutes NOT hours or days INSTALLS all tools required for Tegra Android CPU DEBUGGING with Nsight Tegra GPU DEBUGGING with PerfHUD ES/Tegra Graphics Debugger OPTIMIZE applications with Tegra System Profiler REFERENCE docs, samples & tutorials OPTIMIZED for Tegra Android development FLASHES Tegra DevKit with OS Image CONFIGURED for debugging and profiling INCLUDES some Kernel symbols and DS-5 support

TEGRA CPU DEVELOPER TOOLS CPU DEBUGGING Supports Android/Linux standard tools Eclipse/NDK/JDK Android Studio Gdb/gdbserver 7.3.x (Android) Nsight Tegra Visual Studio Edition Gdb/Jdb within Visual Studio

NVIDIA NSIGHT TEGRA Android NDK/JDK application development Project Management Android Debugging Logcat Filtering

NSIGHT TEGRA 1.5 Microsoft Visual Studio 2013 NDK r9c / Android SDK 22.3 Support for external build systems (makefile) Improved MSBUILD compatibility with NDK Improved project/solution loading performance

Android GDB debugging in Visual Studio Set breakpoints in both Java and Native (C/C++) Build Native Android projects in Visual Studio using vs-android, ndk-build or makefiles. Use the familiar Visual Studio Locals, Watches, Memory and Breakpoints windows. NVIDIA CONFIDENTIAL

TEGRA CPU DEVELOPER TOOLS CPU PROFILING Supports Android/Linux standard tools Systrace/atrace/ftrace Perf L1-dcache icache-load store prefetch{-misses} l2cache-dr/dw/instr-misses Oprofile ARM DS-5 Gator Tegra System Profiler

TEGRA SYSTEM PROFILER MULTI-CORE CPU PROFILER FOR ALL TEGRA PLATFORMS Windows, Linux and OSX host application Maximize multi-core A15/A9 CPU utilization Quickly identify CPU hot spots and hot paths Quickly identify L1/L2 cache issues Easily prepare a device for profiling Easily deploy applications for profiling

WHAT S NEW WITH TEGRA SYSTEM PROFILER 2.0 Supports Tegra K1 and Tegra Note 7 CPU usage broken down by module New backtrace options simplify call stack collection Improved performance and filtering options Supports Linux and OSX in addition to Windows

Identify call chain hot spots Capture multi-core CPU utilization, L1/L2 cache counters NVIDIA CONFIDENTIAL

GRAPHICS DEVELOPER TOOLS Nsight Visual Studio Edition Frame Debugging and Profiling Shader Debugging and Pixel History System trace Tegra Graphics Debugger Frame Debugging and Profiling Shader Debugging and Pixel History (v2.0) PerfKit API to sample performance SW and HW counters

NVIDIA NSIGHT HOMOGENEOUS APPLICATION DEVELOPMENT FOR CPU+GPU, GRAPHICS AND COMPUTE GPU Shader Debugger Pixel History Graphics Inspector and Profiler System Analysis

Demo Basic Frame Debugging, Shader Debugging, Pixel History Using FaceWorks Fast frame scrubbing and HUD GPU Shader Debugging and Pixel history

TEGRA GRAPHICS DEBUGGER NEXT-GEN GRAPHICS DEVELOPMENT TOOLS FOR TEGRA K1 Supports OpenGL 4.x, OpenGL ES 2.0/3.0/3.1 + numerous extensions Monitor key software and hardware performance metrics Debug draw calls and related states and resources Live capture of a single rendering frame Edit and recompile shaders live Automatic GPU bottleneck analysis Advanced timings for draw calls and kernel dispatches

Current target view Dynamic shader editing Resource revisions

PERFHUD ES 2.2 GRAPHICS DEVELOPMENT TOOLS FOR TEGRA 4 AND PRIOR Supports OpenGL ES 1.0 and 2.0 Monitor key software and hardware performance metrics Examine frames to reveal rendering problems Debug API calls, parameters, return values and errors Execute directed tests to identify bottlenecks Edit and apply shaders dynamically at runtime Monitor CPU and GPU utilization

PERFKIT 3.1 GPU and Software Performance Counter API Performance Monitoring Automated bottleneck analysis Graphics and Compute Supports Kepler Architecture and higher on Android Application Available on Windows (upcoming Linux and Android/Logan) Nsight VSE / Tegra Graphics Profiler PerfKit Samples NVPMAPI Standard NVIDIA Driver OpenGL ES OpenGL Direct3D NVIDIA GPU http://www.nvidia.com/object/nvperfkit_home.html

COMPUTE DEVELOPER TOOLS Nsight Eclipse Edition Integrated CUDA development environment for Linux and MAC Nsight Visual Studio Edition Integrated CUDA development environment for Windows CUDA 6.0 Toolkit command line tools for Android RenderScript Tegra K1 acceleration / No developer tools

NVIDIA NSIGHT HOMOGENEOUS APPLICATION DEVELOPMENT FOR CPU+GPU COMPUTE PLATFORMS BUILD, DEBUG AND PROFILE ON REMOTE TARGETS WITH CUDA 6.0 CUDA-Aware Editor CUDA Debugger CPU+GPU CUDA Profiler

CUDA 6.0 STANDALONE TOOLS VISUAL PROFILER Trace CUDA activities Kernel Profiler Performance instrumentation with source code correlation Guided Expert Analysis NVPROF Generates execution summary Gather Performance events and metrics CUDA-MEMCHECK Out of bounds memory access detection Detects Race Condition CUDA-GDB Command line CUDA debugging Debug CPU and GPU code

CONCLUSION Full TEGRA K1 support Full OpenGL 4.x and ES 2.0/3.0/3.1 Full support for Renderscript and CUDA 6.0 Advanced Visual Studio integration Advanced CPU profiling Consistent user experience from PC to Tegra Android devices Many years of discreet PC GPU developer tools experience leverage to provide a smooth user experience!

FINDING MORE INFORMATION https://developer.nvidia.com/tegra-development https://devtalk.nvidia.com/ Developer Tools