cuda-gdb.pdfcuda-gdb的cuda9.0官方使用文档，对于Linux下调试基于cud

文件名称: cuda-gdb.pdf

所属分类: 深度学习

开发工具:

文件大小: 1mb

下载次数: 0

上传时间: 2019-10-08

提供者: bai****

下载 (1mb)

不能下载？报告错误

详细说明：cuda-gdb的cuda9.0官方使用文档，对于Linux下调试基于cuda的程序，例如实现卷积神经网络深度学习框架，不可或缺的调试帮助文档。7.5. Conditional Breakpoints ∴25 7.6. Watchpoints..... 26 Chapter8. Inspecting Program State…………………………,…………,……………………….27 8 .1. Memory and variables 8.2. Variable Storage and Accessibility. .27 8.3. Inspecting Textures 28 8. 4. Info cuda Commands 28 8. 4.1. info cuda devices ∴29 8. 4.2. info cuda sms 8.4.3. info cuda warps……… 8. 4.4. info cuda lanes 30 8. 4.5. info cuda kernels 8. 4.6. info cuda blocks 。番·。·。番·垂。···。。。···非。。·垂·看看。。··。鲁。鲁 8. 4.7. info cuda threads 8.4.8. info cuda launch trace∴……… 8.49 info cuda launch children ···:···:·· “······ 33 8. 4.10. info cuda contexts 33 8.4.11. info cuda managed. 8.5. Disassembly. 34 8.6. Registers..,,…,…,,,,……,……34 Chapter9. Event notifications.…………………………………………………………….35 9.1. Context Events .35 9.2. Kernel events .35 Chapter 10. Automatic Error Checking..,...... 37 10.1. Checking API Errors 10.2. GPU Error Reporting...... 4···。·····。··。自··《···曲·音···辛。p·······。。·。· 37 10.3, set cuda memcheck 10.4. Autostep.. Chapter 11. Walk-Through Examples. 42 11.1. Example: bitreverse........................ 。·垂。。。P香非。垂。·看;,。。音看 42 11.1.1. Walking through the Code. 43 11.2. Example: autostep 1.2.1. Debugging with Autosteps............ 47 11.3. Example: MPI CUDA Application.. 9 Chapter 12. Advanced settingS. 51 12.1.-cuda-use-lockfile.51 12.2. set cuda break on launch 12.3. set cuda gpu_busy_ check……52 12.4. set cuda launch_blocking 12.5. set cuda notify ∴53 12.6. set cuda ptx_cache 53 12. 7. set cuda single_ stepping_optimizations .. ..............................,..........................53 12.8. set cuda thread selection 54 nvidia.co CUDA Debugger DU-05227042v9.0|ii 12.9. set cuda value_extrapolation...,.. Appendix A. Supported Platforms. 55 Appendix B. Known Issues.………………………………………………………….56 www.nvidia.com CUDA Debugger DU-05227-042V9.0|i LIST OF TABLES Table 1 CUDA EXception Codes ,38 www.nvidia.com CUDA Debugger DU-05227-0429.0|v www.nvidia.com CUDA Debugger DU-05227-042_V9.0|vi Chapter 1 INTRODUCTION This document introduces CudA-cDb, the nvidia CUDa debugger for linux and Mac os 1. 1. What iS CUDA-GDB? CUDA-GDB is the NViDIa tool for debugging CUDa applications running on Linux and Mac. CUDA-GDB is an extension to the x86-64 port of gDB, the gnu Project debugger. The tool provides developers with a mechanism for debugging CUda applications running on actual hardware. This enables developers to debug applications without the potential variations introduced by simulation and emulation environments CUDA-GDB runs on Linux and mac os X, 32-bit and 64-bit. CUDA-GDB is based on GDB 7.6 on both linux and mac OS x 1.2. Supported Features CUDA-GDB is designed to present the user with a seamless debugging environment that allows simultaneous debugging of both GPU and CPU code within the same application. Just as programming in CUDa C is an extension to C programming, debugging with CUDA-GDB is a natural extension to debugging with GDB. The existing gDB debugging features are inherently present for debugging the host code, and additional features have been provided to support debugging Cuda device code CUDA-GDB supports debugging C/C++ and Fortran CUDA applications. ( Fortran debugging support is limited to 64-bit Linux operating system) All the C++ features supported by the nvcc compiler can be debugged by CUDA-GDB CUDA-GDB allows the user to set breakpoints, to single-step CUDA applications, and also to inspect and modify the memory and variables of any given thread running on the nardware CUDA-GDB supports debugging all CUDa applications, whether they use the CUDA driver api, the Cuda runtime apl, or both www.nvidia.com CUDA Debugger DU-05227-042v9.0|1 Introduction CUDA-GDB supports debugging kernels that have been compiled for specific CUDA architectures, such as sm 20 or sm 30, but also supports debugging kernels compiled at runtime, referred to as just-in-time compilation, or JIT compilation for short 1.3. About this document This document is the main documentation for CUDA-GDB and is organized more as a user manual than a reference manual The rest of the document will describe how to install and use CUDA-GDb to debug Cuda kernels and how to use the new CUDA commands that have been added to gdb some walk-through examples are also provided. It is assumcd that the user already knows the basic gdb commands used to debug host applications www.nvidia.com CUDA Debugger DU-05227-042V9.0|2 Chapter 2. RELEASE NOTES 7.0 Release GPU core dump suppot CUDA-GDB supports reading GPU and GPU+CPU core dump files New environment variables: CUDA ENABLE COREDUMP ON EXCEPTION CUDA ENABlE CPU COREDUMP ON EXCEPTIoN and CUDA COREDUMP FilE can be used to enable and configure this feature 6.5 Release CUDA Fortran Support CUDA-GDB supports CUDA Fortran debugging on 64-bit Linux operating systems GDB 7.6.2 Code base The code base for CUDA-GDB was upgraded to GDB 7.6.2 6.0 Release Unified Memory Support Managed variables can be read and written from either a host thread or a device thread. The debugger also annotates memory addresses that reside in managed memory with managed. The list of statically allocated managed variables can be accessed through a new info cuda managed command GDB 7.6 Code base The code base for CUDA-GDB was upgraded from GDB 7.2 to GDB 7.6 Android Support CUDA-GDB can now be used to debug android native applications either locally or remotely www.nvidia.com CUDA Debugger DU-05227-042V9.0|3 Release notes Single-Stepping Optimizations CUDA-GDB can now use optimized methods to single-step the program, which accelerate single-stepping most of the time. This feature can be disabled by issuing set cuda single stepping optimizations off Faster Remote Debugging A lot of effort has gone into making remote debugging considerably faster, up to 2 orders of magnitude. The effort also made local debugging faster Kernel entry breakpoints The set cuda break on launch option will now break on kernels launched from the gpu. also enabling this option does not affect kernel launch notifications Precise Error Attribution On Maxwell architecture(SM5.0), the instruction that triggers an exception will be reported accurately. The application keeps making forward progress and the Pc at which the debugger stops may not match that address but an extra output message dentifies the origin of the exception Live range optimizations To mitigate the issue of variables not being accessible at some code addresses, the debugger offers two new options. With set cuda value extrapolation, the latest known value is displayed with (possibly) prefix With set cuda ptx cache, the latest known value of the PTX register associated with a source variable is displayed with the (cached)prefix Event notifications Kernel event notifications are not displayed by default any more New kernel events verbosity options have been added: set cuda kernel events, set cuda kernel events depth Also set cuda defer kernel launch notifications has been deprecated and has no effect any more 5.5 Release Kernel launch Trace Two new commands, info cuda launch trace and info cuda launch children, are introduced to display the kernel launch trace and the children kernel of a given kernel when Dynamic Parallelism is used Single-GPU Debugging(BETA) is rendering the desktop GUI. This feature also enables debugging of long-runniy CUDA-GDB can now be used to debug a CUda application on the same GPu th ng or indefinite cuda kernels that would otherwise encounter a launch timeout. In addition, multiple CUDA-GDB sessions can debug CUDa applications context- www.nvidia.com CUDA Debugger DU-05227-042V9.0|4

(系统自动生成,下载前可以参看下载内容)