esp_hosted_ng - esp32c3 - RSSI/level/signal strengh in WiFi scans is always 0 #495
Open
3 tasks done
Comments
|
Fixed in c13c46a ? |
I don't know, we are using releases versions only. If it's fixed, then are you going to provide a new release version soon ? |
|
Please test with master, at both sides. When the new release will happen, it will be included. |
|
Currently I am not able to run the built firmware (impossible to connect to an AP), but I didn't change the driver yet! This is what I do: Create the firmware using the master branch
Previously I was using the RELEASE firmwares from I am going to update the kernel module now |
|
During the build I get into this issue: I fixed it by commenting At this point I have the ESP32C3 + driver updated to 9f809ff unfortunately it's not starting, I try to find what happens.. I don't know if the problem is related to the comit that I take, or something in my setup. I'm searching |
|
Nice, same issue as here now: #426 |
|
Can you please attach the log from esp as well? What is your Linux based system, also please attach the photo of your setup |
Thanks for your help,
Currently I am not able to have traces on my esp32, I think this is firmware related because with old FW (pre-built in the release) it was OK. Did we have to configure something in menuconfig ? For example: |
|
Also I see something strange. With the most recent version (9f809ff) I see that the driver is not able to stop:
In our usecase it's a problem, because the "reset" pin of the ESP32 is "taken" by the driver. Note: Currently I don't know if the driver refuse to stop due to a bug or if it's a new feature that comes with recent commits. |
|
I just realized that with the recent changes of the driver, the interface So at this point, i can connect to access-point! There are still And at this point I can scan the WiFi networks and get the correct RSSI values (maybe): New issues:
|
|
Old firmware and new host driver might behave strange , @caipiblack . Please flash the same git commit both sides.. |
|
Hello, I have the host driver + ESP32C3 firmware using the same commit: 9f809ff This was the most recent commit of yesterday. Can someone try to reproduce to check if he see the same issues:
Question:
|
|
@kapilkedawat, PTAL |
|
@caipiblack could you please share your kernel version number? |
Hello, This is About the version, note that in My version is less than 5.15.0, (5.10.201) but this function is defined in my side so I don't need this. |
|
@caipiblack thanks for the detail, we will try to check the behavior on nearest available kernel version with us. Compilation error should be resolved by 382dc37 Also you should be able to unload the driver now using same commit, please let us know if that doesn't work. Could you please recheck the IP issue on your setup again. |
|
Hello, I take your commit for the ESP32C3 firmware + Host driver.
etherdevice.h: esp_kernel_ports.h: Logs from dmesg, grepped with 'esp': But your response give me some ideas, here you can see what I change in the driver to make it working on my board: And i don't really configure the parameters of the driver as the driver starts automatically on the first run. So these parameters keep their defaults values: In my case: But if these parameters are not "set" when the driver is started, can it produce This is just an idea, I don't know if the solution is around this. |
|
Yes, my bad, I'll correct the function later. It seems there are some missing checks in You can skip the event processing using this patch(will be committed later): Could you please share your system's route after you give two IPs, also does it work as expected when you add two IPs on Ethernet interface? |
|
Hello, your patch solved the invalid parameters errors! For the IP address it's also working, i found out that the IP address was also set to the ETH interface. So at this point the main issues are solved! There are still minor issues that I fix by my side:
This is just to know if I wait a little to update the commits (after you fix these issues) or if I fix them by my side with patches like now. For the second issue, I have no ideas how you can know if an inline function is "redefined" to prevent "re-redefining it". It can also stay like it was previously, who need to remove it, just patch it to remove it.. Also, is there a 1.0.3 release coming soon ? We always prefer to use "released versions" in our platform instead of "the commit of a day" |
|
@caipiblack thanks for the confirmation, changes will be pushed shortly. Yes, will plan a release in some time, but it may take a week or two. |
If it just take one week or two, I think I'm going to wait for the release! I keep the issue open if someone want to say something / to test something, You can close if it's all good. |
Hello, What is the status of the future new release ? Next week we need to know if we are going to use "the commit of the day" (last commit of the master branch), or a release version If you are not working on a release 1.0.3 then I'm going to use "the last commit", we don't really like to use "un-released versions" but if we have no choice.. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Checklist
How often does this bug occurs?
always
Expected behavior
When I execute a WiFi scan, I am supposed to see correct signal level
Actual behavior (suspected bug)
The RSSI/signal level/signal strengh is always seen to 0.
Error logs or terminal output
No response
Steps to reproduce the behavior
We are using the release 1.0.2, (ESP-Hosted-NG_release_v1.0.2.bin)
Project release version
1.0.2
System architecture
ARM 64-bit (Apple M1/M2, Raspberry Pi 4/5)
Operating system
Linux
Operating system version
Yocto - Hardknott
Linux hubmz-rzg2ul 5.10.201-cip41-yocto-standard #1 SMP PREEMPT Sat Feb 27 02:21:18 UTC 2021 aarch64 GNU/Linux
Shell
Bash
Additional context
No response
The text was updated successfully, but these errors were encountered: