Update for RQX-590 and FPGA version 0x24

Author: QQting

conf.py

@@ -20,13 +20,13 @@
 # -- Project information -----------------------------------------------------
 
 project = 'roscube-doc'
-copyright = '2022, ADLINK Autonomous Business Center'
-author = 'ADLINK Autonomous Business Center'
+copyright = '2023, ADLINK Technology'
+author = 'ADLINK'
 
 # The short X.Y version
 version = ''
 # The full version, including alpha/beta/rc tags
-release = '0.9.0'
+release = '0.9.1'
 
 
 # -- General configuration ---------------------------------------------------
@@ -129,7 +129,7 @@
 #  author, documentclass [howto, manual, or own class]).
 latex_documents = [
     (master_doc, 'roscube-doc.tex', 'roscube-doc Documentation',
-     'ADLINK Autonomous Business Center', 'manual'),
+     'ADLINK', 'manual'),
 ]
 
 

nvidia/advanced/driver_packages/backup_restore.rst

@@ -7,17 +7,19 @@ The tutorial will guide you how to backup and restore system image.
 
     The target machine must be connected to your **Host PC** over USB and entered into **recovery mode**.
 
-+--------------------+------------+
-|Model               |Target Name |
-+====================+============+
-|RQX-580             |rqx_580     |
-+--------------------+------------+
-|RQX-58G             |rqx_58g     |
-+--------------------+------------+
-|NPN-1 (Pico Nano)   |npn1        |
-+--------------------+------------+
-|NPN-2 (Pico AGX NX) |npn2        | 
-+--------------------+------------+
++--------------------+-----------------------+
+|Model               |Target Name            |
++====================+=======================+
+|RQX-590             |jetson-agx-orin-rqx59x |
++--------------------+-----------------------+
+|RQX-580             |rqx_580                |
++--------------------+-----------------------+
+|RQX-58G             |rqx_58g                |
++--------------------+-----------------------+
+|NPN-1 (Pico Nano)   |npn1                   |
++--------------------+-----------------------+
+|NPN-2 (Pico AGX NX) |npn2                   | 
++--------------------+-----------------------+
 
 1. Cloning the Image
 ^^^^^^^^^^^^^^^^^^^^

nvidia/advanced/driver_packages/flash_system.rst

@@ -13,17 +13,19 @@ Once you prepared rootfs in step: *customize rootfs*. You can use flash.sh tool
     # Put machine into recovery mode first.
     sudo ./flash.sh <target name> mmcblk0p1
 
-+--------------------+------------+
-|Model               |Target Name |
-+====================+============+
-|RQX-580             |rqx_580     |
-+--------------------+------------+
-|RQX-58G             |rqx_58g     |
-+--------------------+------------+
-|NPN-1 (Pico Nano)   |npn1        |
-+--------------------+------------+
-|NPN-2 (Pico AGX NX) |npn2        | 
-+--------------------+------------+
++--------------------+-----------------------+
+|Model               |Target Name            |
++====================+=======================+
+|RQX-590             |jetson-agx-orin-rqx59x |
++--------------------+-----------------------+
+|RQX-580             |rqx_580                |
++--------------------+-----------------------+
+|RQX-58G             |rqx_58g                |
++--------------------+-----------------------+
+|NPN-1 (Pico Nano)   |npn1                   |
++--------------------+-----------------------+
+|NPN-2 (Pico AGX NX) |npn2                   | 
++--------------------+-----------------------+
 
 .. note:: 
 

nvidia/packages/index.rst

@@ -1,7 +1,7 @@
 :orphan:
 
-The Packages
-############
+NVIDIA Packages
+###############
 
 Show the usage of application in NVIDIA platform.
 

roscube-x/flash_image/flash_mfi.rst

@@ -8,7 +8,11 @@ How to flash MFI
 
 i) Download mfi image.  
 
-  * Go to the `device web <https://www.adlinktech.com/Products/ROS2_Solution/ROS2_Controller/ROScube-X?lang=en/>`_.
+  * Go to the website:
+  
+    - `RQX-580/58G <https://www.adlinktech.com/Products/ROS2_Solution/ROS2_Controller/RQX-580_58G/>`_.
+    
+    - `RQX-590 series <https://www.adlinktech.com/Products/ROS2_Solution/ROS2_Controller/RQX-59_series/>`_. 
 
   .. image:: images/device-web.png
     :width: 80%

roscube-x/flash_image/recovery_mode.rst

@@ -52,6 +52,8 @@ Check out the mode on **Host PC**.
   * Normal mode will show **L4T**.
   * Recovery mode will show **APX**. 
 
+  If you can't see APX text, please try to change the USB cable or the Host USB port.
+
 .. image:: images/nvidia-usb.png
   :width: 80%
   :align: center

roscube-x/gmsl_camera/camera_driver.rst

@@ -3,88 +3,121 @@
 Camera Driver Installation
 ##########################
 
-You need to install dedicated Camera's driver (e.g. ``leopard-ar0233-gmsl``).
+You need to install dedicated camera driver.
 
-The driver is provided by **ADLINK** and could not be used on other vendor's platform since it is compiled into binary.
+Please ask ADLINK or the camera vendor for the latest camera driver.
 
-And it can only download form **ADLINK ROScube's APT repository**.
-
-1. Install Camera Driver.
+1. Install Camera Driver
 -------------------------
 
-To install **Leopard AR0233 GMSL** camera, please by following terminal commands below:
+To install camera driver, please follow the commands below:
 
 .. code:: bash
     
     sudo apt update
-    # Remove lagecy leopard ar0233 kernel module.
-    sudo apt remove --purge leopard-ar0233-gmsl-kmod
-    sudo apt remove --purge rqx58g-ioboard-gmsl-kmod
-    sudo apt install leopard-ar0233-gmsl
+    sudo apt install -f <camera_driver.deb>
+    
+    # Examples:
+    # 
+    # [Tier IV]
+    # sudo apt install -f tier4-camera-gmsl_1.3.0_arm64.deb
+    # 
+    # [Leopard AR0233]
+    # sudo apt install -f roscube-gmsl-serdes_0.9.3_arm64.deb
+    
+.. warning::
 
-2. Inspect Camera Driver.
+    | You can only install one camera driver in ROScube to avoid conflicts.
+      If there are two or more, please remove them which you don't use.
+    | To remove unused camera driver, please use ``sudo apt remove --purge <camera driver>``
+
+
+2. Apply Device Tree
 -------------------------
 
-To inspect your available camera device tree setting after installing, please use the following terminal command:
+To enable the camera driver, you have to apply the device tree settings.
+Before applying, please use below command to list available modules:
 
 .. code:: bash
 
-    sudo /opt/nvidia/jetson-io/config-by-hardware.py -l
+    sudo /opt/nvidia/jetson-io/config-by-hardware.py --list
 
 .. image:: images/config-hardware-list.png
   :width: 80%
   :align: center
 
-.. warning::
+The camera device tree overlay may be located in different headers depends on the BSP and Jetpack version.
 
-    | It can only install one cammera's driver at the same time.
-      If there are two or more, please remove them which you don't use.
-    | To remove unused camera driver, please use ``sudo apt remove --purge <camera driver>``
+.. image:: images/config-hardware-list-tier4.png
+  :width: 80%
+  :align: center
 
-3. Apply Device Tree Settings.
-------------------------------
+To apply the device tree setings, please follow the commands below:
 
-To apply **Leopard AR0233 device tree** settings to all camera channels, please use the following terminal command:
+For Jetpack 4.6, 5.0, and later:
+================================
 
-**For Jetpack 4.5**, use the following
+If the device tree is located in "Header 1", please enter below command:
 
-Download the patched :download:`libgstnvarguscamerasrc.so <file/libgstnvarguscamerasrc.so>` and replace this file with the original one.
+.. code:: bash
+
+    sudo /opt/nvidia/jetson-io/config-by-hardware.py -n 1='<YOUR_CAMERA_DEVICE_TREE_OVERLAY>'
+    # Examples:
+    #
+    # [Tier IV C1]
+    # sudo /opt/nvidia/jetson-io/config-by-hardware.py -n 1='TIERIV ISX021 GMSL2 Camera Device Tree Overlay'
+    #
+    # [Leopard AR0233]
+    # sudo /opt/nvidia/jetson-io/config-by-hardware.py -n 1='Leopard AR0233 RAW GMSL2 Camera Device Tree Overlay'
+
+
+Otherwise, the device tree is located in "Header 2", please enter below command:
 
 .. code:: bash
 
-    sudo rm /usr/lib/aarch64-linux-gnu/gstreamer-1.0/libgstnvarguscamerasrc.so
-    sudo mv libgstnvarguscamerasrc.so /usr/lib/aarch64-linux-gnu/gstreamer-1.0/libgstnvarguscamerasrc.so
+    sudo /opt/nvidia/jetson-io/config-by-hardware.py -n 2='<YOUR_CAMERA_DEVICE_TREE_OVERLAY>'
 
-Then, apply device tree into your current system.
+.. image:: images/config-hardware-name.png
+  :width: 80%
+  :align: center
+
+
+For Jetpack 4.5:
+================
+
+
+Download the patched :download:`libgstnvarguscamerasrc.so <file/libgstnvarguscamerasrc.so>` and replace this file with the original one.
 
 .. code:: bash
 
-    sudo /opt/nvidia/jetson-io/config-by-hardware.py -n 'Leopard AR0233 GMSL2 Camera Device Tree Overlay'
+    sudo rm /usr/lib/aarch64-linux-gnu/gstreamer-1.0/libgstnvarguscamerasrc.so
+    sudo mv libgstnvarguscamerasrc.so /usr/lib/aarch64-linux-gnu/gstreamer-1.0/libgstnvarguscamerasrc.so
 
-**For Jetpack 4.6**, use the following
+Then, apply device tree into your system.
 
 .. code:: bash
 
-    sudo /opt/nvidia/jetson-io/config-by-hardware.py -n 2='Leopard AR0233 GMSL2 Camera Device Tree Overlay'
+    sudo /opt/nvidia/jetson-io/config-by-hardware.py -n '<YOUR_CAMERA_DEVICE_TREE_OVERLAY>'
 
-.. image:: images/config-hardware-name.png
-  :width: 80%
-  :align: center
+3. Reboot
+----------
 
-.. note::
+Finally, please **reboot** system for the new settings to take effect.
+    
+.. code:: bash
 
-    To reconfigure, please **reboot** system.
+    sudo reboot
 
-4. Check Camera.
+4. Check Camera
 ----------------
 
-After rebooting system, you can check if the cameras are available.
+After rebooting, you can check if the cameras are available.
 
-Please use the following terminal command:
+Please use the following command to check:
 
 .. code:: bash
 
-    ls /dev/gmsl/*
+    ls /dev/video*
 
 .. image:: images/dev-video.png
   :width: 80%
@@ -93,5 +126,5 @@ Please use the following terminal command:
 
 .. warning::
 
-    | If it can't show any video devices, make sure cameras are connected.
-    | And check the error message by ``dmesg | egrep ar0233`` in terminal command.
+    | If there are no video devices, make sure cameras are connected.
+    | And check the error logs by typing the command ``dmesg | egrep -i 'max9295|max9296'`` in terminal.

roscube-x/gmsl_camera/camera_usage.rst

@@ -3,16 +3,38 @@
 Camera Usage
 ############
 
+GStreamer
+---------
+
 Different video frameworks have different methods to run cameras.
 You should check which video framework your camera is and select the right method.
 
-Here we show how to run cameras with gstreamer plugin based on NVIDIA Argus or V4L2.
+Here we show how to run cameras with gstreamer plugin based on V4L2 or NVIDIA Argus.
+
+If your camera has ISP, then you sholud use v4l2src; otherwise, use nvarguscamerasrc for the cameras without ISP.
+
+v4l2src plugin
+==============
+
+To use V4L2, you can read video device under ``/dev/`` directly.
+The camera will be listed like ``/dev/videoX``, while X is from 0-7.
+
+To read camera 1 by gstreamer:
+
+.. code:: bash
+
+    # Example for Tier IV C1 video0
+    gst-launch-1.0 -e v4l2src device=/dev/video0 ! 'video/x-raw,format=UYVY,width=1920,height=1280' ! videoconvert ! fpsdisplaysink video-sink=xvimagesink sync=false
 
-NVIDIA Argus
-------------
+.. image:: images/tier4_video0.png
+  :width: 100%
+  :align: center
+
+nvarguscamerasrc plugin
+=======================
 
 Cameras without ISP can use **Argus API** to preview the camera's video streaming, 
-or you can use **GStreamer NVArgusCameraSrc plugin** to preview the video streaming.
+or you can use **GStreamer nvarguscamerasrc plugin** to preview the video streaming.
 
 Make sure you have modified **sensor-id=0** on the following command.
 
@@ -40,13 +62,15 @@ i. Open a terminal and type command to open ``camera 1's video`` streaming.
 
 .. code:: bash
 
-    gst-launch-1.0 nvarguscamerasrc sensor-id=0 ! 'video/x-raw(memory:NVMM), width=2048, height=1280, framerate=30/1' ! nvvidconv flip-method=0 ! 'video/x-raw, format=(string)I420' ! xvimagesink -e
+    # Example for AR0233 w/o ISP
+    gst-launch-1.0 -e nvarguscamerasrc sensor-id=0 ! 'video/x-raw(memory:NVMM), width=2048, height=1280, framerate=30/1' ! nvvidconv ! 'video/x-raw, format=(string)RGBA' ! videoconvert ! fpsdisplaysink video-sink=xvimagesink sync=false
+
 
 ii. Open **another** terminal and type command to ``open camera 2's video`` streaming.
 
 .. code:: bash
 
-    gst-launch-1.0 nvarguscamerasrc sensor-id=1 ! 'video/x-raw(memory:NVMM), width=2048, height=1280, framerate=30/1' ! nvvidconv flip-method=0 ! 'video/x-raw, format=(string)I420' ! xvimagesink -e
+    gst-launch-1.0 -e nvarguscamerasrc sensor-id=1 ! 'video/x-raw(memory:NVMM), width=2048, height=1280, framerate=30/1' ! nvvidconv ! 'video/x-raw, format=(string)RGBA' ! videoconvert ! fpsdisplaysink video-sink=xvimagesink sync=false
 
 .. image:: images/gst-test.png
   :width: 80%
@@ -63,22 +87,15 @@ If successful, you will see two windows from different cameras, like below.
     If you find that Argus plugin can't operate well, you can restart nvargus-daemon.
     ``sudo systemctl restart nvargus-daemon.service``
 
-V4L2
-----
-
-To use V4L2, you can read video device under ``/dev/`` directly.
-The camera will be listed like ``/dev/videoX``, while X is from 0-7.
-You can also read the camera under ``/dev/gmsl/*``.
-Both two ways are available.
-
-To read camera 1 by gstreamer:
-
-.. code:: bash
-
-    gst-launch-1.0 v4l2src device=/dev/video0 ! videoconvert ! videoscale ! video/x-raw,format=UYVY ! queue ! videoconvert ! ximagesink sync=no
+v4l-utils
+---------
 
-To read camera 1 by v4l2 tools
+v4l2-ctl offers a way to start camera stream and evaluate the FPS without showing the real images.
 
 .. code:: bash
 
-    v4l2-ctl -d /dev/video0 --set-fmt-video=width=1920,height=1080,pixelformat=UYVY --stream-mmap=3 --stream-count=600
+    v4l2-ctl -d /dev/video0  --set-ctrl bypass_mode=0 --stream-mmap
+    
+.. image:: images/v4l2-ctl.png
+  :width: 100%
+  :align: center