Skip to content

xiaoye77/tx-utils

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

tx-utils

This is a utils pack for nvidia tegra x series.

install

You can just download released zip file and unzip it.

usage

tegrastats2

It's a enhanced pack of Nvidia tegrastats tool. The origin tegrastats locates in ~/tegrastats and it's a executable binary file which shows the current status of tx's cpus and gpu. However it not open sourced and you can not change anything about what it outputs. And I warp it in this script and add time mark.

sudo python3 tegrastats2.py --bin=/home/nvidia/tegrastats --output=./a.log --params "--interval 500"

It should be under sudo privilege if you want to get the gpu status. --bin is the path of original tegrastats location and --output defines the file path you want to write to.

--params adds the optional arguments to tegrastats which has support more functions from jetpack 3.2. --interval 500 means the interval of every probe.

Tegrastats has more options from jetpack 3.2. And I add the tegrastats_jp3.1 and tegrastats_jp_3.2 to this project. You can just download and run tegrastats_jp_3.2 on tx of jetpack 3.1.

visualize

It's a visualization tool of tx status.

It will read the tegrastats2's output and resort the data into a xls file.

python3 visualize.py --start="2018-06-09 02:42:30" --end="2018-06-09 02:43:15" --input="/home/find/ddown/a.log" --output=./freq.xls

If you don't define the --start and --end, you will get the whole logs' data.

tx-utils

这一个nvidia jetson tegra x系列的工具包。主要用来记录tx板子的状态,并将其写入excel表格,便于生成图表。

安装

直接从release下载zip压缩包解压即可。

使用

tegrastats2

Nvidia自带了一个tegrastats工具,默认位于home目录下。可以用来查看cpu和gpu的一些状态信息,我写了这个脚本,在其输出结果中加入了时间。

sudo python3 tegrastats2.py --bin=/home/nvidia/tegrastats --output=./a.log --params "--interval 500"

你应该用sudo来执行这个脚本,因为sudo权限才能让tegrastats获得到gpu的状态。 --bin是tegrastats的路径,--output是log日志输出的路径。

--params是附加给tegrastats的参数,自从jetpack3.2以后,nvidia提供的tegrastats工具越来越强大,给了更多了运行参数。同时我也将新旧版本的tegrastats加到了这个项目中。

visualize

将原始日志内容格式化到excel文件里,并自动生成cpu占用率和gpu占用率的折线图。

python3 visualize.py --start="2018-06-09 02:42:30" --end="2018-06-09 02:43:15" --input="/home/find/ddown/a.log" --output=./freq.xls

如果你不指定--start--end,那么将针对整个日志文件进行格式化。

About

nvidia jetson tegra x series utils

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 100.0%