Don Wiki

使用者工具

網站工具

足跡:
jetson_agx_orin_development

差異處

這裏顯示兩個版本的差異處。

連向這個比對檢視

Both sides previous revision 前次修改
下次修改 		前次修改
jetson_agx_orin_development [2024/04/09 13:18]
don [Check HDMI Status] 		jetson_agx_orin_development [2024/04/16 15:34] (目前版本)
don [AGX Stable Diffusion]
行 13: 行 13:
 Before JetPack 5.1.1\\ Before JetPack 5.1.1\\
 Jetson Orin module to be emulated Flashing command\\ Jetson Orin module to be emulated Flashing command\\
-Jetson AGX Orin 64GB            sudo ./flash.sh jetson-agx-orin-devkit mmcblk0p1\\ +Jetson AGX Orin 64GB\\            ''sudo ./flash.sh jetson-agx-orin-devkit mmcblk0p1''\\ 
-Jetson AGX Orin 32GB            sudo ./flash.sh jetson-agx-orin-devkit-as-jao-32gb mmcblk0p\\ +Jetson AGX Orin 32GB\\            ''sudo ./flash.sh jetson-agx-orin-devkit-as-jao-32gb mmcblk0p''\\ 
-Jetson Orin NX 16GB            sudo ./flash.sh jetson-agx-orin-devkit-as-nx16gb mmcblk0p1\\ +Jetson Orin NX 16GB\\            ''sudo ./flash.sh jetson-agx-orin-devkit-as-nx16gb mmcblk0p1''\\ 
-Jetson Orin NX 8GB            sudo ./flash.sh jetson-agx-orin-devkit-as-nx8gb mmcblk0p1\\ +Jetson Orin NX 8GB\\            ''sudo ./flash.sh jetson-agx-orin-devkit-as-nx8gb mmcblk0p1''\\ 
-Jetson Orin Nano 8GB*            sudo ./flash.sh jetson-agx-orin-devkit-as-nano8gb mmcblk0p1\\ +Jetson Orin Nano 8GB*\\            ''sudo ./flash.sh jetson-agx-orin-devkit-as-nano8gb mmcblk0p1''\\ 
-Jetson Orin Nano 4GB            sudo ./flash.sh jetson-agx-orin-devkit-as-nano4gb mmcblk0p1\\+Jetson Orin Nano 4GB\\            ''sudo ./flash.sh jetson-agx-orin-devkit-as-nano4gb mmcblk0p1''\\
  
  
行 133: 行 133:
 </code> </code>
  
-=== 3. https://catalog.ngc.nvidia.com/orgs/nvidia/resources/deepstream ===+=== 3. Get deepstream-6.3_6.3.0-1_arm64.deb ===
  
-Get deepstream-6.3_6.3.0-1_arm64.deb for Jetson not to download for x86\\ +[[https://catalog.ngc.nvidia.com/orgs/nvidia/resources/deepstream/files?version=6.3|Download deepstream-6.3_6.3.0-1_arm64.deb ]] 
-<code>sudo apt-get install ./deepstream-6.3_6.3.0-1_arm64.deb</code>+ 
 +<color #ed1c24>Get deepstream-6.3_6.3.0-1_arm64.deb for Jetson not to download for x86</color>\\ 
 +<code>sudo apt install ./deepstream-6.3_6.3.0-1_arm64.deb</code>
  
 === 4. Copy the generated libraries to the deepstream directory: === === 4. Copy the generated libraries to the deepstream directory: ===
行 198: 行 200:
 <code>./gpu_burn 1000</code> <code>./gpu_burn 1000</code>
  
-=== Fan Control ===+==== Fan Control ====
 <code>/etc/nvfancontrol.conf </code> <code>/etc/nvfancontrol.conf </code>
 FAN_DEFAULT_PROFILE cool \\ FAN_DEFAULT_PROFILE cool \\
行 204: 行 206:
 <code>sudo systemctl restart nvfancontrol</code> \\ <code>sudo systemctl restart nvfancontrol</code> \\
  
-=== Power Mode Setting ===+==== Power Mode Setting ====
 Nvidia Power Model Tool to config ID=0 “MAXN” \\ Nvidia Power Model Tool to config ID=0 “MAXN” \\
  
行 211: 行 213:
 <code>sudo jetson_clocks</code> \\ <code>sudo jetson_clocks</code> \\
  
-=== AGX Stable Diffusion ===+==== AGX Stable Diffusion ==== 
 + 
 +[[https://huggingface.co/CompVis/stable-diffusion-v-1-4-original|Download sd-v1-4.ckpt]] 
 <code> <code>
-https://github.com/chitoku/stable-diffusion+git clone https://github.com/chitoku/stable-diffusion
 cd stable-diffusion cd stable-diffusion
 git checkout jetson git checkout jetson
 +cp ~/Downloads/sd-v1-4.ckpt /home/aopen/stable-diffusion/models/ldm/stable-diffusion-v1/
 ./docker/run.sh ./docker/run.sh
 python3 scripts/txt2img.py --prompt "Robots marching down a street in Japanese city" --plms python3 scripts/txt2img.py --prompt "Robots marching down a street in Japanese city" --plms
 </code> </code>
  
-=== JETSON AGX Orin performance test ===+==== JETSON AGX Orin performance test ====
 https://github.com/NVIDIA-AI-IOT/jetson_benchmarks https://github.com/NVIDIA-AI-IOT/jetson_benchmarks
- 
- 
  
 <code> <code>
行 318: 行 322:
 </code> </code>
  
 +
 +===== Pinmux Changes =====
 +
 +If your carrier board schematic differs from that of the reference carrier board in Jetson Xavier NX Developer Kit, you must change the pinmux configuration applied by the software.
 +
 +To define your board’s pinmux configuration, download the Jetson Xavier NX pinmux table from the Jetson Download Center. Be sure to get the right version of the table for your SOM.
 +
 +The pinmux table is a spreadsheet that:
 +
 +  * Shows the locations and default pinmux settings
 +
 +  * Acts as a data source for software that defines the pinmux settings in the source code or device tree
 +
 +You must customize the spreadsheet for the configuration of your board, then convert the .dtsi file generated by Excel to a .cfg file. For instructions, see the README file at:
 +
 +Linux_for_Tegra/kernel/pinmux/t19x/
 +You must perform the same conversion for gpio.dtsi and padvoltage.dtsi.
 +
 +
 +
 +[[https://developer.nvidia.com/embedded/downloads#?search=Pin%20and%20Function%20Names%20Guide|pinmux table ]]
  
 ===== GPIO Changes ===== ===== GPIO Changes =====
行 328: 行 353:
   - Search the table for SPI0_CS1.   - Search the table for SPI0_CS1.
   - Confirm that the Customer Usage field contains the GPIO name GPIO3_PZ.07.   - Confirm that the Customer Usage field contains the GPIO name GPIO3_PZ.07.
- 
   - Search PZ.07 using the following command:   - Search PZ.07 using the following command:
  
 cat /sys/kernel/debug/gpio | grep PZ.07 cat /sys/kernel/debug/gpio | grep PZ.07
jetson_agx_orin_development.1712639894.txt.gz · 上一次變更: 2024/04/09 13:18 由 don

頁面工具

若無特別註明，本 wiki 上的內容都是採用以下授權方式： CC Attribution-Share Alike 4.0 International
CC Attribution-Share Alike 4.0 International Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki