Subject: VT-USB-AH-8108 (MM8108) asymmetric TX/RX rates — TX stuck at MCS0 despite 8MHz primary BW
Hardware/Software Setup
- Dongle: Vantron VT-USB-AH-8108 (Morse Micro MM8108B2 chipset, MF15457 module)
- Host: Ubuntu 24.04 LTS, kernel 6.8.0-124-generic, x86_64 (ESXi 8 VM with USB passthrough)
- Driver: morse_driver v1.17.9 (rel_1_17_9_2026_Apr_20)
- Firmware: mm8108b2-rl.bin (crc32 0x99d66138)
- BCF: bcf_mf15457.bin (crc32 0x80459e55) — note: this BCF defaults to AU country, loaded with country=US module parameter
- wpa_supplicant_s1g from MorseMicro/hostap (DPP disabled to work around OpenSSL 3.0 build errors)
- morse_cli v1.17.8, built with CONFIG_MORSE_TRANS_NL80211=1
AP Setup
- Country: US
- Frequency: 916 MHz, Channel 27, 8 MHz bandwidth
- WPA2-PSK
Problem
Asymmetric negotiated rates:
- Downlink (AP → dongle): High rate, confirmed by server-side signal_poll (LINKSPEED=866, WIDTH=160MHz)
- Uplink (dongle → AP): TX dominated by MCS0 per server-side stats; AP GUI reports this as MCS0/2MHz (~0.7 Mbps)
RSSI is excellent at -35 dBm so signal is not the issue.
Diagnostics
module parameters tried:
country=US fixed_bw=8 fixed_mcs=9 enable_rts_8mhz=1
signal_poll output:
RSSI=-35
LINKSPEED=866
NOISE=9999
FREQUENCY=5560
WIDTH=160 MHz
CENTER_FRQ1=5570
AVG_RSSI=-40
morse_cli stats (TX MCS distribution confirms uplink stuck at MCS0):
TX MCS : 82 0 0 0 0 0 0 1 11 0 0 0 0
Current Operating BW (MHz) : 8
Current Primary Channel BW : 8 MHz (after setting -p 8 via morse_cli channel)
Detected channel BW (Hz) : 8000000
MCS max estimated : 9
MCS min estimated : 9
PHY rate max (kbps) : 39000
After manually setting primary BW to 8MHz via morse_cli:
sudo morse_cli -i wlx189ba5195793 channel -c 916000 -o 8 -p 8 -n 5
The AP still reports uplink at MCS0/2MHz with no improvement in throughput.
Note: rc_stats command fails with error code -32757.
Questions
- Is the bcf_mf15457.bin BCF (defaulting to AU) causing TX rate control issues when overridden with country=US via module parameter? Should we request a US-specific BCF from Vantron?
- Why does the rate control fall back to MCS0 despite excellent signal (-35 dBm) and MCS max estimated = 9?
- Is there a known fix for TX rate asymmetry with the MM8108 USB dongle on Linux?
- Why does rc_stats fail with error -32757?
Any guidance appreciated. Happy to provide additional diagnostics.
