Open venoom27 opened 1 year ago
Hi, thanks for the report! I think the FW_RTW89
option is not needed: this driver can only work with Realtek 8852AE, 8851BE, 8852BE, and 8852CE. Ideally, it should not be loaded but it is worth to check. At the same time, I noticed you had to blacklist the if_rtw88
(FreeBSD) driver from the host -- there is a slight chance that there is interference still and it is worth to check if that is not loaded either. However, based on its man page, it cannot handle this specific model.
What does lsmod
say on wifibox console
(inside the guest)?
Hrm, looks like Arch Linux has a specific rtw88
driver (a sibling of rtw89
) and does not use the one shipped with the vanilla kernel. Realtek's own drivers is known to be unreliable and of low performance. I will port this package and get back with the wifibox-alpine
version that you can test.
I have added the rtw88
driver for wifibox-alpine
that could be enabled by the FW_RTW88
configuration option. This is version 20230820
that can be found in the dev
branch of the pgj/freebsd-wifibox-port
repository.
I installed wifibox-alpine-20230820 with UDS_PASSTHRU, RW_RTW88, KERN_LTS. COMP_XZ, and APP_WPA_SUPPLICANT selected. I am still getting packet lost.
wifibox:~# ping www.google.com
PING www.google.com (172.253.62.103): 56 data bytes
64 bytes from 172.253.62.103: seq=0 ttl=104 time=19.925 ms
64 bytes from 172.253.62.103: seq=1 ttl=104 time=21.201 ms
64 bytes from 172.253.62.103: seq=2 ttl=104 time=182.102 ms
64 bytes from 172.253.62.103: seq=23 ttl=104 time=771.687 ms
64 bytes from 172.253.62.103: seq=24 ttl=104 time=92.161 ms
64 bytes from 172.253.62.103: seq=25 ttl=104 time=16.242 ms
64 bytes from 172.253.62.103: seq=26 ttl=104 time=15.035 ms
64 bytes from 172.253.62.103: seq=27 ttl=104 time=23.029 ms
64 bytes from 172.253.62.103: seq=28 ttl=104 time=53.374 ms
64 bytes from 172.253.62.103: seq=29 ttl=104 time=65.433 ms
64 bytes from 172.253.62.103: seq=30 ttl=104 time=23.593 ms
64 bytes from 172.253.62.103: seq=31 ttl=104 time=32.064 ms
64 bytes from 172.253.62.103: seq=32 ttl=104 time=30.643 ms
64 bytes from 172.253.62.103: seq=33 ttl=104 time=182.652 ms
64 bytes from 172.253.62.103: seq=34 ttl=104 time=50.970 ms
64 bytes from 172.253.62.103: seq=35 ttl=104 time=14.300 ms
64 bytes from 172.253.62.103: seq=36 ttl=104 time=15.080 ms
64 bytes from 172.253.62.103: seq=37 ttl=104 time=14.693 ms
64 bytes from 172.253.62.103: seq=38 ttl=104 time=15.888 ms
64 bytes from 172.253.62.103: seq=39 ttl=104 time=128.183 ms
^C
--- www.google.com ping statistics ---
40 packets transmitted, 20 packets received, 50% packet loss
round-trip min/avg/max = 14.300/88.412/771.687 ms
That is sad. Could you please include the uname -a
output from the guest? But it looks like we will need to get through the "vanilla Arch Linux as a guest" experiment (as it was sketched in #58) to make it sure that the problem is not due to the underlying virtualization and the host OS.
Description
I installed both the stable version then tried the dev port LTS version and I am keep getting packet loss to the point that it almost is impossible to use.
Host operating system
Wireless NIC
Wifibox version
Disk image type and version
The kind of VM image in use, e.g. Wifibox/Alpine, and its version. wifibox-alpine-20230810
config options selected
Wifibox-core-0.11.0
Changes to the default configuration files
Logs
Additional context
Add any other context about the problem here that might help the investigation.
I am able to use the same hardware on Arch Linux when I dual boot with no drop out issues. I have also had the same kind of issues with a Macbook pro that I already been working through with support on. https://github.com/pgj/freebsd-wifibox/issues/38
Have you tried to turn it on and off?