Posts made by bendraper

@Alex-Kushleyev So I can see a gadget functionality called gsi.rndis (I'd expect something called rndis.0 or something but maybe that is irrelevant) but when I change the line in /sbin/usb/compositions/901D to use gsi.rndis instead of ncm.0, my network interface on windows disappears and I can no longer ADB. I'd imagine there's something at the kernal level that doesn't like that but I'm out of my wheelhouse there. Setting it back to ncm.0 returns to the previous state but still cannot ping. Wireshark doesn't even see anything happening over that interface

@Alex-Kushleyev It looks like someone on the forums maybe had success setting up the interface as an RNDIS gadget. I'll try to mess with it some more but let me know if you find a solution. Thanks!

@Alex-Kushleyev Yea I can get a connection on a linux machine with no issue. I do need to get it working on a Windows machine for my purposes though. I definitely have the static IP setup correctly. I tried populating the gateway and also not populating it. I confirmed in powershell that there is indeed a route present for that traffic, but they do not want to communicate with each other. Some quick research seems to suggest Windows is very finicky with the CDC NCM protocol and it is more suited to work with an RNDIS interface. Were you able to ping/ssh to a voxl via an ethernet interface on the usb-c port to a windows machine using just the commands you listed above in this post?

@Alex-Kushleyev I saw a post from @tom from a few years ago discussing how there should be the necessary drivers to support an RNDIS gadget so I assume that is my issue. Any insight on how to enable that?

[Apr 1 19:15] configfs-gadget gadget: Wrong NDP SIGN
[ +12.232553] usbpd usbpd0: Type-C Source (default) connected
[  +0.019411] usbpd usbpd0: USB Type-C disconnect
[  +0.004363] android_work: sent uevent USB_STATE=DISCONNECTED
[  +0.001100] msm-dwc3 a600000.ssusb: DWC3 in low power mode
[  +3.492478] usbpd usbpd0: Type-C Source (medium - 1.5A) connected
[  +0.113740] msm-usb-ssphy-qmp 88e8000.ssphy: USB DP QMP PHY: Update TYPEC CTRL(3)
[  +0.057403] msm-dwc3 a600000.ssusb: DWC3 exited from low power mode
[  +0.068603] usbpd usbpd0: Type-C Source (high - 3.0A) connected
[  +0.238255] android_work: sent uevent USB_STATE=CONNECTED
[  +0.041534] configfs-gadget gadget: high-speed config #1: c
[  +0.000644] android_work: sent uevent USB_STATE=CONFIGURED
[  +0.195765] usbpd usbpd0: Type-C Source (medium - 1.5A) connected
[  +0.021780] usbpd usbpd0: Type-C Source (high - 3.0A) connected
[  +0.179818] configfs-gadget gadget: Wrong NDP SIGN
[  +0.000172] configfs-gadget gadget: Wrong NDP SIGN
[  +0.069075] configfs-gadget gadget: Wrong NDP SIGN
[  +0.002827] configfs-gadget gadget: Wrong NDP SIGN
[  +0.176097] configfs-gadget gadget: Wrong NDP SIGN
[  +0.001314] configfs-gadget gadget: Wrong NDP SIGN
[  +0.001331] configfs-gadget gadget: Wrong NDP SIGN
[  +0.250721] configfs-gadget gadget: Wrong NDP SIGN
[  +0.491456] configfs-gadget gadget: Wrong NDP SIGN
[  +0.000164] configfs-gadget gadget: Wrong NDP SIGN
[  +0.000446] configfs-gadget gadget: Wrong NDP SIGN
[  +0.067507] configfs-gadget gadget: Wrong NDP SIGN
[  +0.000984] configfs-gadget gadget: Wrong NDP SIGN
[  +0.012649] configfs-gadget gadget: Wrong NDP SIGN
[  +0.183993] configfs-gadget gadget: Wrong NDP SIGN
[  +0.001107] configfs-gadget gadget: Wrong NDP SIGN
[  +0.211788] configfs-gadget gadget: Wrong NDP SIGN
[  +0.030077] configfs-gadget gadget: Wrong NDP SIGN

@Alex-Kushleyev Should this work on a Voxl to Windows computer connection? I ran the sed and udev rule commands and how the voxl does indeed come up as an ethernet device upon connecting (UsbNcm Host Device) and I can set up the static IPs appropriately, but I cannot ping one to the other for some reason. I can still use the port as an ADB interface and I thought maybe that was the issue but doing adb kill-server on the windows side still doesn't do the trick

@Eric-Katzfey I got about a million warnings when actually building but it did appear to build after initializing the submodules. Thanks!

@Alex-Kushleyev any ideas?

Also experiencing this issue. @Eric-Katzfey could you provide any inside here? I've also tried different branches in the open vins server repo and different branches for install_deps such as staging and sdk_1.6

@Eric-Katzfey dev seems to be broken in that case as I was getting the error

/home/root/src/main.c:42:10: fatal error: voxl_vfc_config.h: No such file or directory

@Eric-Katzfey I tried staging and that ended up working. I was trying dev because voxl-vision-hub repo says to

@Eric-Katzfey sorry, I was thinking of the build script. I’m using dev in accordance with the README in vision hub

@Eric-Katzfey I did not provide an argument

Trying to build latest voxl-vision-hub master with voxl-cross 4.4 and get the following error.

voxl-cross(4.4):~(master)(1.9.20)$ ./build.sh qrb5165
Found voxl-cross version: 4.4
-- ---------------------------------------------------------
-- Using voxl-cross 64-bit toolchain for QRB5165 ubun1 18.04
-- C Compiler  : /usr/bin/aarch64-linux-gnu-gcc-7
-- C++ Compiler: /usr/bin/aarch64-linux-gnu-g++-7
-- Sysroot     : /opt/sysroots/qrb5165_1
-- C flags     : -isystem=/usr/lib/gcc/aarch64-linux-gnu/7/include -isystem=/usr/local/include -isystem=/usr/lib/gcc/aarch64-linux-gnu/7/include-fixed -isystem=/usr/include/aarch64-linux-gnu -isystem=/usr/include -idirafter /usr/include -march=armv8.2-a
-- CXX flags   : -isystem=/usr/include/c++/7 -isystem=/usr/include/aarch64-linux-gnu/c++/7 -isystem=/usr/include/c++/7/backward -isystem=/usr/lib/gcc/aarch64-linux-gnu/7/include -isystem=/usr/local/include -isystem=/usr/lib/gcc/aarch64-linux-gnu/7/include-fixed -isystem=/usr/include/aarch64-linux-gnu -isystem=/usr/include -idirafter /usr/include -march=armv8.2-a
-- Link Flags  :  -B/opt/sysroots/qrb5165_1/usr/lib/ -L/opt/sysroots/qrb5165_1/usr/lib/gcc/aarch64-linux-gnu/7 -L/opt/sysroots/qrb5165_1/usr/lib/aarch64-linux-gnu -L/opt/sysroots/qrb5165_1/usr/lib -L/opt/sysroots/qrb5165_1/lib/aarch64-linux-gnu -L/opt/sysroots/qrb5165_1/usr/lib/aarch64-linux-gnu -L/opt/sysroots/qrb5165_1/lib -L/usr/lib64 -L/usr/lib
-- Configuring done (0.0s)
-- Generating done (0.0s)
-- Build files have been written to: /home/root/build
[ 12%] Built target voxl-calibrate-px4-horizon
[  9%] Built target voxl-inspect-vfc
[ 18%] Building C object src/CMakeFiles/voxl-vision-hub.dir/main.c.o
[ 18%] Building C object src/CMakeFiles/voxl-vision-hub.dir/offboard_backtrack.c.o
[ 21%] Building C object src/CMakeFiles/voxl-vision-hub.dir/offboard_vfc.c.o
[ 25%] Building C object src/CMakeFiles/voxl-vision-hub.dir/offboard_wps.c.o
/home/root/src/offboard_vfc.c:43:10: fatal error: voxl_vfc_config.h: No such file or directory
 #include <voxl_vfc_config.h>
          ^~~~~~~~~~~~~~~~~~~
compilation terminated.
/home/root/src/offboard_backtrack.c:41:10: fatal error: voxl_vfc_config.h: No such file or directory
 #include <voxl_vfc_config.h>
          ^~~~~~~~~~~~~~~~~~~
compilation terminated.
make[2]: *** [src/CMakeFiles/voxl-vision-hub.dir/build.make:331: src/CMakeFiles/voxl-vision-hub.dir/offboard_vfc.c.o] Error 1
make[2]: *** Waiting for unfinished jobs....
make[2]: *** [src/CMakeFiles/voxl-vision-hub.dir/build.make:261: src/CMakeFiles/voxl-vision-hub.dir/offboard_backtrack.c.o] Error 1
/home/root/src/main.c:42:10: fatal error: voxl_vfc_config.h: No such file or directory
 #include "voxl_vfc_config.h"
          ^~~~~~~~~~~~~~~~~~~
compilation terminated.
make[2]: *** [src/CMakeFiles/voxl-vision-hub.dir/build.make:191: src/CMakeFiles/voxl-vision-hub.dir/main.c.o] Error 1
/home/root/src/offboard_wps.c:44:10: fatal error: voxl_vfc_config.h: No such file or directory
 #include "voxl_vfc_config.h"
          ^~~~~~~~~~~~~~~~~~~
compilation terminated.
make[2]: *** [src/CMakeFiles/voxl-vision-hub.dir/build.make:345: src/CMakeFiles/voxl-vision-hub.dir/offboard_wps.c.o] Error 1
make[1]: *** [CMakeFiles/Makefile2:155: src/CMakeFiles/voxl-vision-hub.dir/all] Error 2
make: *** [Makefile:136: all] Error 2
voxl-cross(4.4):~(master)(1.9.20)$

@Eric-Katzfey This process is trying to send SET_POSITION_TARGET_LOCAL_NED by the way. Not sure if the act of setting offboard mode to OFF just blocks that message or something

@Eric-Katzfey So I have a niche issue with that. I'm running a process on the Voxl that is utilizing the local port within vision hub because it is said in the voxl-mavlink-server conf file that you should expose a local port there as opposed to voxl-mavlink server. But now when i set offboard mode to off, my process is blocked off from that local port that is enabled within vision hub... any way you can see to resolve this?

@bendraper Actually I think the message is SET_POSITION_TARGET_LOCAL_NED not SET_ATTITUDE

I have an external ArduPilot flight controller and when I enter Guided mode, my drone starts to descend and move as if it were given a target even though intended behavior is that it should just hold its position until given a target. Upon some digging, I think I've determined that somewhere in vision hub (version 1.8.17), the moment I enter guided mode I get a burst of SET_ATTITUDE_TARGET messages. I have open vins server turned off as well as voxl-px4 for what its worth. Any idea where and why this message may be getting sent?

It is my understanding you guys use Google Fi for your testing. How did you set up a plan. I am battling with Google support who doesn't seem to understand what I'm doing. I just got a regular SIM card on the Fi website and the ping is fine but throughput via an iperf3 test seems to be capped at 30kbps. Support says its because the sim card is only for phones and needs to be activated by a phone. What is the secret to getting a Fi sim to work on a Voxl with the SRC7611 or where did you guys get your sim(URL if you have it). Thanks!

@groupo Thanks for this! I am pretty familiar with the AT Commands I just haven't tried disabling bands yet. Will add it to me to-do list

@tom what debug info would you suggest looking for? I can see that the preferred access tech is 5g although it’s using LTE. And I can see the signal quality is ~40% which doesn’t seem great but it’s been much worse before and I can’t seem to get it above 40 anyways. I have the antennas stuck to some plastic in a diamond array