gpgpu-simGPU仿真工具_GPGPU-Sim资源-CSDN文库

gpgpu仿真工具

5星 · 超过95%的资源需积分: 19 152 浏览量 2011-02-07 18:51:56 上传评论收藏 2.53MB GZ 举报

共179个文件

cpp：38个

hpp：33个

h：27个

资源推荐

资源详情

资源评论

收起资源包目录

gpgpu-sim GPU仿真工具（179个子文件）

shader.c 144KB

gpu-sim.c 86KB

dram.c 19KB

addrdec.c 16KB

gpu-cache.c 15KB

delayqueue.c 13KB

gpgpusim_entrypoint.c 6KB

stack.c 4KB

icnt_wrapper.c 4KB

gpu-misc.c 4KB

instructions.cc 89KB

dwf.cc 76KB

cuda-sim.cc 72KB

cuda_runtime_api.cc 48KB

opencl_runtime_api.cc 43KB

ptx_ir.cc 43KB

stat-tool.cc 34KB

visualizer.cc 22KB

option_parser.cc 16KB

warp_tracker.cc 15KB

ptx_sim.cc 14KB

ptx-stats.cc 12KB

dram_sched.cc 9KB

memory.cc 6KB

nvopencl_wrapper.cc 5KB

CHANGES 2KB

gpgpusim.config 1KB

gpgpusim.config 795B

trafficmanager.cpp 34KB

routefunc.cpp 30KB

event_router.cpp 24KB

iq_router.cpp 17KB

interconnect_interface.cpp 17KB

allocator.cpp 11KB

traffic.cpp 7KB

kncube.cpp 6KB

selalloc.cpp 5KB

booksim_config.cpp 5KB

islip.cpp 5KB

rng_double.cpp 4KB

BlackScholes_gold.cpp 4KB

rng.cpp 4KB

maxsize.cpp 4KB

config_utils.cpp 4KB

buffer_state.cpp 3KB

fly.cpp 3KB

arbiter.cpp 3KB

network.cpp 3KB

outputset.cpp 3KB

vc.cpp 3KB

router.cpp 3KB

template_gold.cpp 3KB

injection.cpp 2KB

pim.cpp 2KB

loa.cpp 2KB

stats.cpp 2KB

wavefront.cpp 1KB

module.cpp 1KB

statwraper.cpp 1KB

singlenet.cpp 924B

random_utils.cpp 462B

flit.cpp 352B

misc_utils.cpp 318B

credit.cpp 211B

rng_double_wrapper.cpp 137B

rng_wrapper.cpp 119B

BlackScholes.cu 10KB

template.cu 5KB

template_kernel.cu 3KB

BlackScholes_kernel.cuh 4KB

opcodes.def 5KB

fly26_age 595B

.gdbinit 2KB

gen_ptxinfo 3KB

ptx_ir.h 36KB

mem_latency_stat.h 21KB

shader.h 16KB

ptx_sim.h 13KB

dram.h 9KB

gpu-cache.h 6KB

cflogger.h 6KB

memory.h 5KB

delayqueue.h 5KB

dwf.h 5KB

mem_fetch.h 5KB

gpu-misc.h 5KB

ptx-stats.h 4KB

gpu-sim.h 4KB

option_parser.h 4KB

dram_sched.h 4KB

histogram.h 4KB

icnt_wrapper.h 4KB

warp_tracker.h 4KB

dram_callback.h 4KB

stack.h 4KB

addrdec.h 4KB

opcodes.h 4KB

util.h 3KB

cuda-math.h 3KB

共 179 条

AerialVision Performance Visualizer

Version 1.0

Aaron Ariel, Wilson W. L. Fung, Tor M. Aamodt

University of British Columbia

October 27, 2009

Table of Contents

1. What is AerialVision.................................................................................................................................2

2. Installation................................................................................................................................................2

3. Usage ........................................................................................................................................................3

3.A

Walkthrough Example ................................................................................................................3

3.B

Detailed Description of AerialVision Features...........................................................................9

Startup Screen ......................................................................................................................................9

Visualizer Tab – Creating New Figure ................................................................................................10

Visualizer Tab – Customizing an Existing Figure................................................................................12

Manage Files.......................................................................................................................................17

Source Code Viewer Tab – Performance Metric Selection ...............................................................18

Source Code Viewer Tab – Navigating Through Source Code with Performance Metrics...............20

4. Description of Available Performance Metrics .....................................................................................21

5. Extending AerialVision – Adding Variables ...........................................................................................25

What is AerialVision

AerialVision is a GPU performance analysis tool. Its main features are a Visualizer where various metrics

can be graphed as a function of cycle count, and a Source Code Viewer where various metrics are

displayed in correlation to the corresponding lines of PTX or CUDA.

2. Installation

The set of python scripts that implements AerialVision can be found inside the “aerialvision”

subdirectory of the GPGPU-Sim distribution. AerialVision has been tested to work using Python 2.6.

Earlier distributions of python are not compatible with AerialVision.

To install AerialVision, run the command: “python installscript.py” inside the “aerialvision” subdirectory,

which will download all modules required to run AerialVision using wget (which must be installed on

your system, along with python). These modules may take up to 200MB of disk space in your $HOME

directory. After that, follow the instruction from the install script to modify these environment

variables: PYTHONPATH, CPLUS_INCLUDE_PATH, PKG_CONFIG_PATH. The AerialVision launcher script

“aerialvision.py” can be found inside the “bin” subdirectory of the GPGPU-Sim distribution. Add this

“bin” subdirectory ($GPGPUSIM_ROOT/bin) to your PATH environment variable. You may need to edit

the first line of “aerialvision.py” to point to your local installation of python (if it is not /usr/bin/python).

To begin using AerialVision, run the command: “aerialvision.py”. This should load up the startup screen

as in Figure 1.

Figure 1: Startup Screen in AerialVision

3. Usage

This section is divided into two main parts. The first section gives a brief walkthrough example, so that

the user may get started using AerialVision features as quickly as possible. The second section covers all

AerialVision features in detail.

3.A

Walkthrough Example

We first present a walkthrough example of various AerialVision features. In this manual, we have used

GPGPU-Sim trace files corresponding to the MUMmerGPU CUDA application however users may follow

along using whichever application they choose.

We begin by loading the startup screen by typing ‘python AerialVision.py’ into the command line. Once

the startup screen loads up, the next step is to submit all of the relevant files.

By default, you should now be looking at the fields inside the ‘File Inputs for Visualizer’ tab. In this tab,

we submit files that will enable us to use the time lapse visualizing capabilities of the Visualizer. These

files are by default in the form ‘gpgpusim_visualizer__*.log.gz’. We submit files by clicking the ‘Browse’

button (if you’ve submitted the file before you can click on the ‘Recent Files’ button), and then clicking

‘Add File’ once the file’s path is in the ‘Add Input File’ text field. Finally, choose the appropriate

resolution and parsing options (described in another section of this manual). Figure 2 depicts what your

screen should look like at this point. You should note that you can submit numerous files for visualizing

into this tab; however, for the purposes of this walkthrough we have limited it to one.

Figure 2 File Inputs for Visualizer

Now click on the ‘File Inputs for Source Code Viewer’ tab. In this tab we submit files that present

statistics corresponding to each line of PTX or CUDA source. Before clicking the ‘Add Files’ button, it is

necessary to insert the file paths to three distinct files required by this part of AerialVision. The file that

goes in the ‘Add CUDA Source Code File’ text field is the appropriate CUDA kernel source code file. This

file should end with a ‘*.cu’ extension. In the case of MUMmerGPU, this file is named

‘mummergpu_kernel.cu’ (Not to be confused with ‘mummergpu.cu’). The file that goes in the ‘Add

Corresponding PTX File’ text field is the appropriate PTX file generated by the CUDA compiler. In the

case of MUMmerGPU, this file is named ‘mummergpu.ptx’. Finally, the file that goes in the ‘Add

Corresponding Stat File’ is generated by the GPGPU-Sim and is by default named ‘gpgpu_inst_stats.txt’.

This file should be located in the same folder as the one you launched GPGPU-Sim from. Once you have

filled the three text fields, click the green ‘Add Files’ button. Figure 3 depicts what your screen should

look like at this point. You can now launch AerialVision by clicking the ‘Submit’ button at the bottom. It

should be noted that for your own purposes, it is not necessary to fill both the ‘File Inputs for Visualizer’

and ‘File Inputs for Source Code Viewer tabs as both parts of AerialVision can be used independently of

the other.

Figure 3 File Inputs for Source Code Viewer

Once the ‘Submit’ button has been clicked, AerialVision will parse the input files for the data inside. This

process may take a minute to several minutes depending on the length of the trace files submitted to

AerialVision. Your screen should now look something like Figure 7.

We will now create our first plot using AerialVision. It should be noted that the plots that we produce in

this tutorial will obviously differ from the ones that the user produces as we are simulating a different

application (unless of course the user is simulating MUMmerGPU as well). For starters, let’s investigate

whether off chip memory latency effects performance for our application.

We first need to ‘Choose a File’ by double clicking on the trace file that we want to extra data from.

These files can be found in box ‘A’ in Figure 7. Double clicking on one of the files should turn the

appropriate section of the ‘Options Chosen’ list green.

From this file the first thing we’d like to plot is Instruction Per Cycle (IPC). We do this by selecting

‘globalInsn’ from the list of ‘Y Vars’ (Table 3 and Table 4 in this manual contains a list of all variables and

their descriptions), and then clicking the ‘dy/dx’ checkbox. We click the ‘dy/dx’ checkbox because the

‘globalInsn’ variable contains a running count of instructions executed, and in order to plot IPC we must

take the derivative. For this particular plot, we will select ‘Line’ from the options available for ‘Type of

Graph’ (Table 3 and Table 4 depicts the recommended type of plots for all of the available performance

metrics).

We now have the option of selecting the number of subplots that we’d like to add. For this particular

example, we will choose ‘2’. In order to do this, slide the ‘Add Subplot’ slider to the ‘2’ position and click

‘Submit’.

In the ‘Subplot0’ tab we will once again be plotting IPC, only this time on a per shader basis. We do this

by first selecting a file as we did earlier. The ‘Y Vars’ that we need to select this time is named

‘shaderInsn’ so double click on it and once again check the ‘dy/dx’ checkbox. This time, the ‘Type of

Graph’ that we are going to select is the ‘Parallel Intensity Plot’. You have now chosen all of the

necessary options for ‘Subplot0’.

In the ‘Subplot1’ tab we will be plotting the memory latency. Once again select the appropriate file as

we did earlier. The ‘Y Vars’ that we need to select this time is named ‘averagemflatency’. This time do

not click on the ‘dy/dx’ button and choose the ‘Line’ plot from the ‘Type of Graph’ options. You may

now press the green ‘GraphMe!’ button. If you have followed this walkthrough correctly, all of the fields

in the ‘Option Chosen’ list should be green. After clicking the green ‘GraphMe!’ button, your screen

should now look something like Figure 4.

评论收藏

内容反馈

koscielny_

2015-01-25

这个版本太早了，不如官网去下
jinjizhilu

2013-05-22

可以用，但是有新版了，还是去官网下吧
cuijiyue

2013-03-31

挺好的，就是没有使用教程，英文的看着晕啊

chenjz70

粉丝: 1
资源: 1

gpgpu-sim GPU仿真工具

GPGPU-Sim仿真器

gpgpu-sim_distribution:GPGPU-Sim为运行CUDA和_或OpenCL工作负载的当代NVIDIA GPU提供了详细的仿真模型。 它包括对TensorCores和CUDA Dynamic Parallelism等功能的支持，以及性能可视化工具AeroVisoin和集成的能源模型GPUWattch

gpgpu-sim-docker:用于gpgpu-sim模拟环境的Dockerfile

GPGPU-MiniBench：加速GPGPU微体系结构仿真

提供自动回归脚本、模拟启动实用程序和代码+用于在GPGPU-Sim上在合理时间内完成的模拟的.zip

FusionSim AMD APU模拟仿真器

GPGPU_Sim-Enabled-Turing-WMMA-API：当前（2020.12）的GPGPU-Sim支持多达第一代NVIDIA张量内核。 此发行版包括最多支持第二代张量核心（Turing arch）的模拟器

官网gpgpu-sim安装须知

helloGPGPU-1.0.2.zip GPU入门

GPGPU Programming for Games and Science 无水印pdf

matlab代码检查工具-sim_vehicle_dynamics:TUMRoboraceTeam软件堆栈-车辆仿真

gazebo-auv-sim:远东联邦大学自主水下航行器仿真工具

matlab方位角计算代码-lidar-sim:多通道3D激光雷达的仿真工具

GPGPU-Sim V3.0 中文版手册

sim-dl-runner:管理GPGPU-Sim上的深度学习程序的执行

利用GPGPU方法将大量数据通过纹理传输至GPU进行计算

CUDA by Example. An Introduction to General-Purpose GPU Programming

webgl-gpgpu-demo

gpgpu.rar_GPGPU_GPU_ati gpu_gpgpu测试

Supporting CUDA for an extended RISC-V GPU architecture.pdf

Gpgpu-CloudSim-Cloud-Gaming:具有通用GPU扩展的CloudSim

CPU-GPU-Coherence

gpgpu-rs：一个基于OpenCL的简单GPU计算库

verilog-format的配置文件

完整版 ISO 26262 最新版（2018）1-12部分.rar

中航光电_连接器手册（全）.pdf

STM32通过298N控制电机.zip

英飞凌AURIX TC27x D-Step芯片用户手册(Infineon-TC27x_D-step-UM-v02_02-EN.pdf)

【源码】基于贝叶斯优化的LSTM时间序列预测

最新资源

gpgpu-sim_distribution:GPGPU-Sim为运行CUDA和_或OpenCL工作负载的当代NVIDIA GPU提供了详细的仿真模型。它包括对TensorCores和CUDA Dynamic Parallelism等功能的支持，以及性能可视化工具AeroVisoin和集成的能源模型GPUWattch

GPGPU_Sim-Enabled-Turing-WMMA-API：当前（2020.12）的GPGPU-Sim支持多达第一代NVIDIA张量内核。此发行版包括最多支持第二代张量核心（Turing arch）的模拟器