Difference between revisions of "Jetson nsight system"
(→Installation) |
(→Installation) |
||
| Line 2: | Line 2: | ||
=Installation= | =Installation= | ||
| − | Jetson Nsight System must be installed via SDKManager | + | Jetson Nsight System must be installed via SDKManager |
| − | NSys User Guide: https://docs.nvidia.com/nsight-systems/UserGuide/index.html | + | NSys User Guide: https://docs.nvidia.com/nsight-systems/UserGuide/index.html |
==Install NS on x86 Linux Host== | ==Install NS on x86 Linux Host== | ||
| − | 1. Install Nsight System via SDKManager <br> | + | '''1. Install Nsight System via SDKManager''' <br> |
<gallery mode=nolines widths=260> | <gallery mode=nolines widths=260> | ||
NS_Install1.PNG|Step#1: Select "Host Machine" | NS_Install1.PNG|Step#1: Select "Host Machine" | ||
| Line 12: | Line 12: | ||
Just click '''Continue''' to install Nsight System on x86 Linux System. | Just click '''Continue''' to install Nsight System on x86 Linux System. | ||
| − | 2. Verify Installation<br> | + | '''2. Verify Installation'''<br> |
After installation is done, you can open it with "nsight-sys" command as below. | After installation is done, you can open it with "nsight-sys" command as below. | ||
<gallery mode=nolines> | <gallery mode=nolines> | ||
| Line 20: | Line 20: | ||
==Install NS on Jetson Device== | ==Install NS on Jetson Device== | ||
'''Note''': on the newly installed Jetson system, there is not Nsight System on it. <br> | '''Note''': on the newly installed Jetson system, there is not Nsight System on it. <br> | ||
| − | 1. Installation Steps <br> | + | '''1. Installation Steps''' <br> |
# On x86, launch the Nsight system installed via SDKManager as described above | # On x86, launch the Nsight system installed via SDKManager as described above | ||
# In Nsight System, create a new project, and connect to Jetson device as below, this step will install the Nsight target binaries ontO Jetson device | # In Nsight System, create a new project, and connect to Jetson device as below, this step will install the Nsight target binaries ontO Jetson device | ||
| Line 26: | Line 26: | ||
Install_NS_libs_onto_Jetson_Device.png | Install_NS_libs_onto_Jetson_Device.png | ||
</gallery> | </gallery> | ||
| − | 2. Verify Installation <br> | + | '''2. Verify Installation''' <br> |
After installation, nsys locates under /opt/nvidia/nsight_systems/ . <br> | After installation, nsys locates under /opt/nvidia/nsight_systems/ . <br> | ||
Revision as of 01:42, 7 January 2021
This page describes how to use Nsight System on Jetson L4T system
Contents
Installation
Jetson Nsight System must be installed via SDKManager NSys User Guide: https://docs.nvidia.com/nsight-systems/UserGuide/index.html
Install NS on x86 Linux Host
1. Install Nsight System via SDKManager
Step#1: Select "Host Machine"
Step#2: Install "NVIDIA Nsight Systems"
Just click Continue to install Nsight System on x86 Linux System.
2. Verify Installation
After installation is done, you can open it with "nsight-sys" command as below.
Install NS on Jetson Device
Note: on the newly installed Jetson system, there is not Nsight System on it.
1. Installation Steps
- On x86, launch the Nsight system installed via SDKManager as described above
- In Nsight System, create a new project, and connect to Jetson device as below, this step will install the Nsight target binaries ontO Jetson device
2. Verify Installation
After installation, nsys locates under /opt/nvidia/nsight_systems/ .
Profile
Remote Profile (UI)
User can run Nsight System on Host and remotely profile the application running on Jetson. User can select several options to enable the corresponding proiling.
Local Profile
User can check the profile option
$ /opt/nvidia/nsight_systems/nsys profile --help
Profile Application
Profile
Run an application and capture its profile into a QDSTRM file, and view it in Nsys GUI profiler
One exmaple to profile TRT inference:
$ sudo /opt/nvidia/nsight_systems/nsys profile -t cuda,nvtx,nvmedia,osrt --accelerator-trace=nvmedia --show-output=true --force-overwrite=true --delay=20 --duration=90 --output=%p /usr/src/tensorrt/bin/trtexec --loadEngine=yolo_dla_0_bs20.engine --useDLACore=0 --batch=20
Options: --accelerator-trace=nvmedia : enable profile DLA --delay : start profiling after 20 seconds --duration : profile time
View Log
Import QDSTRM file into GUI Nsys:
1. TensorRT and DLA Inference Time
2. CUDA Kernel Run Time
3. Enqueue Time
