COMFAST CF-953AX - chipset: mt7921au

[foen73]

[Info] Yesterday my new toy arrived and really worked out of the box, at least with the mainline kernel 5.18.12. With standard Ubuntu kernels, I tested 5.15.0.41.43 & 5.17.0.1013.12 (OEM), it does not work without further action.

Speedtest by Ookla

Server: iway AG - Zurich (id = 3188)
ISP: Vodafone Germany
Latency: 24.94 ms (39.31 ms jitter)
Download: 390.00 Mbps (data used: 460.7 MB)
Upload: 25.01 Mbps (data used: 43.2 MB)
Packet Loss: 0.0%

lsusb
Bus 002 Device 007: ID 0e8d:7961 MediaTek Inc. Wireless_Device

uname -a
Linux foen-ThinkPad-T570 5.18.12-051812-generic #202207150942 SMP PREEMPT_DYNAMIC Fri Jul 15 10:10:55 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux

dmesg
[ 1038.765594] usb 2-2: new SuperSpeed USB device number 7 using xhci_hcd
[ 1038.788041] usb 2-2: New USB device found, idVendor=0e8d, idProduct=7961, bcdDevice= 1.00
[ 1038.788045] usb 2-2: New USB device strings: Mfr=6, Product=7, SerialNumber=8
[ 1038.788047] usb 2-2: Product: Wireless_Device
[ 1038.788048] usb 2-2: Manufacturer: MediaTek Inc.
[ 1038.788049] usb 2-2: SerialNumber: 000000000
[ 1041.272871] Bluetooth: hci0: Device setup in 2418728 usecs
[ 1043.297145] Bluetooth: hci0: Opcode 0x c03 failed: -110
[ 1045.313182] Bluetooth: hci0: Failed to read MSFT supported features (-110)
[ 1047.329075] Bluetooth: hci0: AOSP get vendor capabilities (-110)
[ 1047.457597] usb 2-2: reset SuperSpeed USB device number 7 using xhci_hcd
[ 1047.507236] mt7921u 2-2:1.3: HW/SW Version: 0x8a108a10, Build Time: 20220209150832a

[ 1047.602000] Bluetooth: hci0: Device setup in 117882 usecs
[ 1047.666384] Bluetooth: hci0: AOSP extensions version v1.00
[ 1047.666390] Bluetooth: hci0: AOSP quality report is supported
[ 1047.757820] mt7921u 2-2:1.3: WM Firmware Version: ____010000, Build Time: 20220209150915
[ 1049.373255] mt7921u 2-2:1.3 wlxe0e1a934a1bb: renamed from wlan0
[ 1053.039692] wlxe0e1a934a1bb: authenticate with 88:dc:96:8d:1a:fd
[ 1053.576871] wlxe0e1a934a1bb: send auth to 88:dc:96:8d:1a:fd (try 1/3)
[ 1053.585139] wlxe0e1a934a1bb: authenticated
[ 1053.592626] wlxe0e1a934a1bb: associate with 88:dc:96:8d:1a:fd (try 1/3)
[ 1053.608944] wlxe0e1a934a1bb: RX AssocResp from 88:dc:96:8d:1a:fd (capab=0x1511 status=0 aid=7)
[ 1053.636803] wlxe0e1a934a1bb: associated
[ 1053.687419] wlxe0e1a934a1bb: Limiting TX power to 20 (23 - 3) dBm as advertised by 88:dc:96:8d:1a:fd
[ 1053.735668] IPv6: ADDRCONF(NETDEV_CHANGE): wlxe0e1a934a1bb: link becomes ready

Interestingly, with the standard Ubuntu kernel the BT part works, which I am neither interested in nor have I tested for functionality, but at least the dmseg output looks good here.

[morrownr]

Hi @foen73

Thanks for the report.

[ 1047.757820] mt7921u 2-2:1.3: WM Firmware Version: ____010000, Build Time: 20220209150915

There is a later firmware that is available. To install/update, go to the following link and go to the mt7921 section:

https://github.com/morrownr/USB-WiFi/blob/main/home/How_to_Install_Firmware_for_Mediatek_based_USB_WiFi_adapters.md

Would you mind running iw list and post the results for the CF-953AX? Keep in mind that if you have 2 wifi devices, there will be two outputs.

Also, would you mind deactivating usb_modeswitch so we can make sure that this is a single-state adapter? If so, look at the last section at the below link:

https://github.com/morrownr/USB-WiFi/blob/main/home/How_to_Modeswitch.md

Thanks

[foen73]

Hi @morrownr,

First of all, I would like to thank you very much for the great provision of your useful information. I already have the current firmware on the computer, but I was primarily interested in exploring the extent to which a real out of box experience (related to Ubuntu 22.04) is given. That's why I'm especially happy to confirm that it's definitely a single state adapter. I will attach the iw list output as a text file.

iw list of CF-953AX.txt

Kind regards,
Sebastian

[morrownr]

@foen73

Thanks for posting the iw list.

I am puzzled. I have a laptop with a card based on a mt7921e chipset. I upgraded to kernel 5.19 and the latest firmware earlier today. AP and AP/VLAN mode support is now showing as available as was expected. I checked iw list. It showed a dual band card as expected.

Now tell me why iw list is showing your CF-953AX to be a tri-band adapter that is capable of the 6 GHz channels? Do you have the capability to test if the 6 GHz channels are actually working?

Regards

[foen73]

Hi @morrownr,

Unfortunately, I don't have any Wi-Fi 6E access points or compatible hardware. I hardly believe that a corresponding radio module was built in. The packaging explicitly mentions dual band.
Perhaps another manufacturer will take pity on the chipset and fully exploit its possibilities.

[foen73]

Fast iperf3 benchmark under optimal conditions. 4 meters from AP, in line of sight, latest Kernel (5.19.0-051900rc7-generic) and latest firmware. The counterpart is an EnGenius ECW220v2 access point.

[ ID] Interval Transfer Bitrate Retr Cwnd
[ 5] 0.00-1.00 sec 61.7 MBytes 518 Mbits/sec 58 1.32 MBytes
[ 5] 1.00-2.00 sec 62.5 MBytes 524 Mbits/sec 0 1.46 MBytes
[ 5] 2.00-3.00 sec 62.5 MBytes 524 Mbits/sec 0 1.57 MBytes
[ 5] 3.00-4.00 sec 62.5 MBytes 524 Mbits/sec 0 1.65 MBytes
[ 5] 4.00-5.00 sec 62.5 MBytes 524 Mbits/sec 2 1.23 MBytes
[ 5] 5.00-6.00 sec 62.5 MBytes 524 Mbits/sec 0 1.30 MBytes
[ 5] 6.00-7.00 sec 63.8 MBytes 535 Mbits/sec 0 1.35 MBytes
[ 5] 7.00-8.00 sec 61.2 MBytes 514 Mbits/sec 0 1.39 MBytes
[ 5] 8.00-9.00 sec 62.5 MBytes 524 Mbits/sec 0 1.41 MBytes
[ 5] 9.00-10.00 sec 62.5 MBytes 524 Mbits/sec 0 1.42 MBytes

---

[ ID] Interval Transfer Bitrate Retr
[ 5] 0.00-10.00 sec 624 MBytes 524 Mbits/sec 60 sender
[ 5] 0.00-10.00 sec 622 MBytes 521 Mbits/sec receiver

[morrownr]

I don't have any Wi-Fi 6E access points or compatible hardware.

I don't have a WiFi 6E capable access point either.

I hardly believe that a corresponding radio module was built in. The packaging explicitly mentions dual band.

I agree that it would seem unlikely that 6E is supported but I have seen many mistakes in ads and documentation. I will believe what we determine to be true with testing.

The COMFAST ad is the first time I have seen the term MT7921AU and I keep up with this stuff pretty closely. I decided last night to do some more searching. There are very few references on the internet to MT7921AU but the following is a copy and paste from an article I found:

MT7921AU supports Wi-Fi 6E and Bluetooth 5.2 from a single, highly integrated platform. With 1.2 Gbps speeds, consumers can
enjoy smoother streaming and gaming for uninterrupted entertainment. Wi-Fi 6E is designed to make use of uncongested
bandwidth in the 6GHz spectrum and offers numerous advantages over previous Wi-Fi generations, such as significantly faster
multi-gigabit data throughput, lower latencies and the latest security and connection reliability features.

It could be that there is an error in the mt7921u driver where it is reporting 6E support when it should not. I think you mentioned that you are using a Thinkpad laptop. Is it possible you could find a friend that has a 6E capable access point so that you can test?

[morrownr]

There is a good thread going in issue #87 and you are welcome to join.

We are trying to confirm that 6 GHz is supported.

[sjtmbg]

Hi,

I received my CF-953AX (at least that's what the box says and what's printed on the device) but it's detected as vendor: product 3574:6211, ie, completely different from this thread. So do I have a different chipset? Any advice?

Regards, Steve Larson

[bjlockie]

I don't know. https%3A%2F%2Fusb.org%2Fsites%2Fdefault%2Ffiles%2Fvendor_ids051920_0.pdf&usg=AOvVaw2rFgug5g7rSrm7-kqgfchO says 0x3575 is Sitecom Europe B.V. but I also found this: https://the-sz.com/products/usbid/index.php?v=3574&p=6211&n= Shenzhen Four Seas Global Link Network Technology Co., Ltd. which is Comfast. Maybe Comfast is not USB certified. Do you have a link to what you bought? I would ask Comfast what 3574:6211 means. Do you have access to Windows? Someone might see something so please post: $ sudo lsusb -d 3574:6211 -v

[morrownr]

I'm kinda scratching my head. I'd like to the entire output of

$ lsusb

[sjtmbg]

Hi,

Thanks for the responses. Here is the output from "lsusb -d 3574:6211":

lsusb.3574_6211.txt

@bjlockie - I'm surprised by the vendorID info you found. Both dmesg and lsusb call it "MediaTek Inc." "Wireless_Device".

I was able to get connected using this device with the mt7921u driver by putting the VendorID and ProductID in "/sys/bus/usb/drivers/mt7921u/new_id"

I bought the device from an outfit called "Sunsky":

https://www.sunsky-online.com/p/EDA003280301A/COMFAST-CF-953AX-1800Mbps-USB-3.0-WiFi6-Wireless-Network-Card-with-Antenna-Black-.htm

No, I don't have access to Windows. I got it working on Ubuntu 22.10, 5.19.0-31-generic after I noticed that the firmware files in @morrownr's firmware guide were already present on my system. I might try playing with it on fedora and arch, where I have more recent (6.1.12) kernels.

Thank you very very much to @morrownr for all your fine work on these devices, which I rely on every day.

Regards,

Steve

[morrownr]

@sjtmbg

"/sys/bus/usb/drivers/mt7921u/new_id"

I really appreciate this bit of info. I had never heard of this but have now posted it on the site in a couple of places for users to see.

Here is the example. It is for the Netgear A8000:

su
modprobe mt7921u
echo 0846 9060 > /sys/bus/usb/drivers/mt7921u/new_id

"lsusb -d 3574:6211"

This info really should go in the kernel. Do you know how to submit patches to linux-wireless?

[morrownr]

@morrownr:

What are you testing with?

953 on 6.2 kernel and udev 250

@rindeal

I added your guidance to the Netgear A8000 and Comfast CF-953AX adapters as listed in the Plug and Play list, Main Menu item 2. Would you mind reading what I put for both adapters to see if I did it correctly. I don't have any adapters that don't already have the VID/PID in the kernel and it would take time to take a VID/PID out and recompile.

If things are correct, do you mind if I give credit to you?

@morrownr

[sjtmbg]

Thank you to @rindeal for the persistent udev solution, works great for me.

Having recently upgraded some boxes to ubuntu 23.04, I see that the 8822bu driver comes with the 6.2.0 kernel. For probably a year now I've been compiling and using the 88x2bu.ko driver and getting download speeds that max out with what I get from my ISP (~200Mbps). With the driver that comes with the 6.2.0 kernel, my typical download speeds are about 25 Mbps. Any suggestions where I should start a new thread on this?

[morrownr]

Having recently upgraded some boxes to ubuntu 23.04, I see that the 8822bu driver comes with the 6.2.0 kernel. For probably a year now I've been compiling and using the 88x2bu.ko driver and getting download speeds that max out with what I get from my ISP (~200Mbps). With the driver that comes with the 6.2.0 kernel, my typical download speeds are about 25 Mbps. Any suggestions where I should start a new thread on this?

Just start a new thread here Issues. Maybe the title should ask for what performance people are seeing with the rtl8812bu in-kernel driver?

I have been testing with kernel 6.3. Performance and stability is much better. 6.4 should even be better.

Do you use the 88x2bu driver here? The install and remove scripts automatically blacklist and unblacklist the in-kernel driver so you don't have to.

[sjtmbg]

Thanks for that info, I hadn't realized the install script did the blacklisting. I usually run make manually because I think I learn more that way. I've used the the 88x2bu driver here in the past but more recently have used RinCat's (that's where my performance numbers are coming from).

[rindeal]

@morrownr:
I added your guidance to the Netgear A8000 and Comfast CF-953AX adapters as listed in the Plug and Play list, Main Menu item 2. Would you mind reading what I put for both adapters to see if I did it correctly. I don't have any adapters that don't already have the VID/PID in the kernel and it would take time to take a VID/PID out and recompile.

The upstream patch for Comfast 952/953 is included in kernel 6.3, not 6.2. Otherwise LGTM.
Also you might want to consider providing a link to just download the file like this 90-mt7921u-wifi6-usb.rules.zip, as it might be quicker and less prone to errors.

@morrownr:
If things are correct, do you mind if I give credit to you?

Yes, but thank you for asking.

[morrownr]

This is based on a sample size of 1 - is that enough to submit the patch?

Generally I like to see 3 or 4 individuals reporting the same thing before recoomending someone submit a patch so that give you time to learn how to submit a patch. There will be others that come along. If you need help, let me know.

[sjtmbg]

I started working on the patch when I came across this:

https://www.spinics.net/lists/kernel/msg4639080.html

So it looks like someone beat me to it - does that sound right to you?

One point of interest: the comment in that patch, referring to 3574:6211, mentions the CF-952AX (ie, not 953-AX).

[morrownr]

So it looks like someone beat me to it - does that sound right to you?

Yes, it does but I did not remember that the VID/PID was the same as yours.

This is really confusing now.

In the 3rd paragraph after the Recent Changes section at the below link, I talk about the issue of the 952AX and the new VID/PID but I had no idea that it included situations such as yours:

https://github.com/morrownr/USB-WiFi/blob/main/home/USB_WiFi_Adapters_that_are_supported_with_Linux_in-kernel_drivers.md

Does the adapter say 953AX on it somewhere? ... or just the packaging?

https://github.com/morrownr/USB-WiFi/blob/main/home/USB_WiFi_Adapters_that_are_supported_with_Linux_in-kernel_drivers.md

3574:6211

[sjtmbg]

So it looks like someone beat me to it - does that sound right to you?

Yes, it does but I did not remember that the VID/PID was the same as yours.

Yep, and I can confirm that the VID/PID is in the wireless-testing code already (in .../mediatek/mt76/mt7921/usb.c).

Does the adapter say 953AX on it somewhere? ... or just the packaging?

It's on both the device and the packaging. To be sure, it's on a sticker on the device, so would be pretty easy to mess with.

[zachmathew]

Hi there @morrownr ,

Can confirm what @sjtmbg is saying. I have the CF-953AX as well, also got it from SunSky. My device looks identical with the same sticker etc. Also curiously, the blue led indicator on it doesn't seem to work at all. I thought it was a dud until i looked at the lsusb output.

I also have the same vendor and device id i.e 3574:6211

lsusb gives : Bus 002 Device 003: ID 3574:6211 MediaTek Inc. Wireless_Device
Managed to get it working by modprobing the pre-existing mt7921u module on the kernel (I'm on Arch on the 6.1.10 kernel) and then writing 3574 6211 to the new_id file in /sys/bus/usb/drivers/mt7921u
Attaching the output of sudo lsusb -d 3574:6211 -v

953ax_lsusb-v.txt

Hope this info helps. Let me know if you need anything more

[fs30000]

I got a comfast-CF-953AX too.

lsusb shows:
Bus 002 Device 003: ID 0e8d:7961 MediaTek Inc. Wireless_Device

On the pci.ids file it shows up as:
7961 MT7921 802.11ax PCI Express Wireless Network Adapter

But can't get it to work in AP mode, with hostapd. Put it on windows, couldn't activate the mobile hotspot. Using comfast drivers. Hmmm....

[morrownr]

But can't get it to work in AP mode, with hostapd.

Gimme some details. What distro, computer and band are you using?

[fs30000]

Ok, put the adapter on a Lenovo thinkpad laptop. Running proxmox 7.3. Kernel is 5.15.85-1-pve.

It doesn't show up in "ip a". The module mt7921e isn't loaded automatically. After loading it, the adapter still doesn't show up in "ip a". I'm guessing there's many usb ids for this comfast adapter. There isn't /sys/bus/usb/drivers/mt7921u/, so can't echo the usb id there.

ID 0e8d:7961 MediaTek Inc. Wireless_Device

** lsusb -d 0e8d:7961 -v**

Device Descriptor:
  bLength                18
  bDescriptorType         1
  bcdUSB               3.20
  bDeviceClass          239 Miscellaneous Device
  bDeviceSubClass         2
  bDeviceProtocol         1 Interface Association
  bMaxPacketSize0         9
  idVendor           0x0e8d MediaTek Inc.
  idProduct          0x7961
  bcdDevice            1.00
  iManufacturer           6 MediaTek Inc.
  iProduct                7 Wireless_Device
  iSerial                 8 000000000
  bNumConfigurations      1
  Configuration Descriptor:
    bLength                 9
    bDescriptorType         2
    wTotalLength       0x01fa
    bNumInterfaces          4
    bConfigurationValue     1
    iConfiguration          9 Config_01
    bmAttributes         0xa0
      (Bus Powered)
      Remote Wakeup
    MaxPower              160mA
    Interface Association:
      bLength                 8
      bDescriptorType        11
      bFirstInterface         0
      bInterfaceCount         3
      bFunctionClass        224 Wireless
      bFunctionSubClass       1 Radio Frequency
      bFunctionProtocol       1 Bluetooth
      iFunction               5 BT_FUNCTION
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        0
      bAlternateSetting       0
      bNumEndpoints           3
      bInterfaceClass       224 Wireless
      bInterfaceSubClass      1 Radio Frequency
      bInterfaceProtocol      1 Bluetooth
      iInterface              1 BT_ACL_If
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x81  EP 1 IN
        bmAttributes            3
          Transfer Type            Interrupt
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0040  1x 64 bytes
        bInterval               1
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x82  EP 2 IN
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x02  EP 2 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               0
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        1
      bAlternateSetting       0
      bNumEndpoints           2
      bInterfaceClass       224 Wireless
      bInterfaceSubClass      1 Radio Frequency
      bInterfaceProtocol      1 Bluetooth
      iInterface              2 BT_SCO_If
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x83  EP 3 IN
        bmAttributes            1
          Transfer Type            Isochronous
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0000  1x 0 bytes
        bInterval               4
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x03  EP 3 OUT
        bmAttributes            1
          Transfer Type            Isochronous
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0000  1x 0 bytes
        bInterval               4
        bMaxBurst               0
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        1
      bAlternateSetting       1
      bNumEndpoints           2
      bInterfaceClass       224 Wireless
      bInterfaceSubClass      1 Radio Frequency
      bInterfaceProtocol      1 Bluetooth
      iInterface              2 BT_SCO_If
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x83  EP 3 IN
        bmAttributes            1
          Transfer Type            Isochronous
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0009  1x 9 bytes
        bInterval               4
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x03  EP 3 OUT
        bmAttributes            1
          Transfer Type            Isochronous
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0009  1x 9 bytes
        bInterval               4
        bMaxBurst               0
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        1
      bAlternateSetting       2
      bNumEndpoints           2
      bInterfaceClass       224 Wireless
      bInterfaceSubClass      1 Radio Frequency
      bInterfaceProtocol      1 Bluetooth
      iInterface              2 BT_SCO_If
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x83  EP 3 IN
        bmAttributes            1
          Transfer Type            Isochronous
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0011  1x 17 bytes
        bInterval               4
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x03  EP 3 OUT
        bmAttributes            1
          Transfer Type            Isochronous
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0011  1x 17 bytes
        bInterval               4
        bMaxBurst               0
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        1
      bAlternateSetting       3
      bNumEndpoints           2
      bInterfaceClass       224 Wireless
      bInterfaceSubClass      1 Radio Frequency
      bInterfaceProtocol      1 Bluetooth
      iInterface              2 BT_SCO_If
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x83  EP 3 IN
        bmAttributes            1
          Transfer Type            Isochronous
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0019  1x 25 bytes
        bInterval               4
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x03  EP 3 OUT
        bmAttributes            1
          Transfer Type            Isochronous
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0019  1x 25 bytes
        bInterval               4
        bMaxBurst               0
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        1
      bAlternateSetting       4
      bNumEndpoints           2
      bInterfaceClass       224 Wireless
      bInterfaceSubClass      1 Radio Frequency
      bInterfaceProtocol      1 Bluetooth
      iInterface              2 BT_SCO_If
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x83  EP 3 IN
        bmAttributes            1
          Transfer Type            Isochronous
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0021  1x 33 bytes
        bInterval               4
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x03  EP 3 OUT
        bmAttributes            1
          Transfer Type            Isochronous
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0021  1x 33 bytes
        bInterval               4
        bMaxBurst               0
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        1
      bAlternateSetting       5
      bNumEndpoints           2
      bInterfaceClass       224 Wireless
      bInterfaceSubClass      1 Radio Frequency
      bInterfaceProtocol      1 Bluetooth
      iInterface              2 BT_SCO_If
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x83  EP 3 IN
        bmAttributes            1
          Transfer Type            Isochronous
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0031  1x 49 bytes
        bInterval               4
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x03  EP 3 OUT
        bmAttributes            1
          Transfer Type            Isochronous
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0031  1x 49 bytes
        bInterval               4
        bMaxBurst               0
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        1
      bAlternateSetting       6
      bNumEndpoints           2
      bInterfaceClass       224 Wireless
      bInterfaceSubClass      1 Radio Frequency
      bInterfaceProtocol      1 Bluetooth
      iInterface              2 BT_SCO_If
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x83  EP 3 IN
        bmAttributes            1
          Transfer Type            Isochronous
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x003f  1x 63 bytes
        bInterval               4
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x03  EP 3 OUT
        bmAttributes            1
          Transfer Type            Isochronous
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x003f  1x 63 bytes
        bInterval               4
        bMaxBurst               0
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        2
      bAlternateSetting       0
      bNumEndpoints           2
      bInterfaceClass       224 Wireless
      bInterfaceSubClass      1 Radio Frequency
      bInterfaceProtocol      1 Bluetooth
      iInterface              3 BT_ISO_If
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x8a  EP 10 IN
        bmAttributes            3
          Transfer Type            Interrupt
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0040  1x 64 bytes
        bInterval               1
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x0a  EP 10 OUT
        bmAttributes            3
          Transfer Type            Interrupt
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0040  1x 64 bytes
        bInterval               1
        bMaxBurst               0
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        2
      bAlternateSetting       1
      bNumEndpoints           2
      bInterfaceClass       224 Wireless
      bInterfaceSubClass      1 Radio Frequency
      bInterfaceProtocol      1 Bluetooth
      iInterface              3 BT_ISO_If
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x8a  EP 10 IN
        bmAttributes            3
          Transfer Type            Interrupt
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval               1
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x0a  EP 10 OUT
        bmAttributes            3
          Transfer Type            Interrupt
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0200  1x 512 bytes
        bInterval               1
        bMaxBurst               0
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        3
      bAlternateSetting       0
      bNumEndpoints           9
      bInterfaceClass       255 Vendor Specific Class
      bInterfaceSubClass    255 Vendor Specific Subclass
      bInterfaceProtocol    255 Vendor Specific Protocol
      iInterface              4 WiFi_If
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x84  EP 4 IN
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x85  EP 5 IN
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x08  EP 8 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x04  EP 4 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x05  EP 5 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x06  EP 6 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x07  EP 7 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x09  EP 9 OUT
        bmAttributes            2
          Transfer Type            Bulk
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0400  1x 1024 bytes
        bInterval               0
        bMaxBurst               0
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x86  EP 6 IN
        bmAttributes            3
          Transfer Type            Interrupt
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0002  1x 2 bytes
        bInterval               1
        bMaxBurst               0
Binary Object Store Descriptor:
  bLength                 5
  bDescriptorType        15
  wTotalLength       0x0016
  bNumDeviceCaps          2
  USB 2.0 Extension Device Capability:
    bLength                 7
    bDescriptorType        16
    bDevCapabilityType      2
    bmAttributes   0x0000f11e
      BESL Link Power Management (LPM) Supported
    BESL value      256 us
    Deep BESL value    61440 us
  SuperSpeed USB Device Capability:
    bLength                10
    bDescriptorType        16
    bDevCapabilityType      3
    bmAttributes         0x00
    wSpeedsSupported   0x000e
      Device can operate at Full Speed (12Mbps)
      Device can operate at High Speed (480Mbps)
      Device can operate at SuperSpeed (5Gbps)
    bFunctionalitySupport   1
      Lowest fully-functional device speed is Full Speed (12Mbps)
    bU1DevExitLat          10 micro seconds
    bU2DevExitLat         180 micro seconds
can't get debug descriptor: Resource temporarily unavailable
Device Status:     0x0000
  (Bus Powered)

[fs30000]

Again, no idea why the text gets so uneven with the Add Code thing here...

EDIT: Ok, used the quadruple ` thingies. Looks ok now.

[jaearr]

I picked up a Comfast CF-953AX for $15 on Aliexpress about a year ago and I finally addressed some issues with it. Looking at the other discussions on this device I might be the only one. I have the version that has idVendor=0e8d and idProduct=7961. I have a few old computers, so I tested on Ubuntu 24.04 and Fedora 40.

1. It really doesn't like to work right if I'm using it at super speed. When lsusb -t lists it at 5000M, it is slow to find access points and in most cases refuses to connect to them even if the signal strength is high. I attached it to a USB2 extension cable and it resolves this problem even though it reduces the USB link speed to 480M. I didn't see any earlier comments showing link speed, so if anyone can do their own comparison that would help.
2. The CF-953AX resets itself or messes up USB (lsusb takes ~60s to run) if you let it load the bluetooth firmware from /lib/firmware/mediatek/BT_RAM_CODE_MT7961_1_2_hdr.bin. I installed the current firmware as of today (202407161551027) as described and it was strange that it would sometimes work fine and other times make the device chain reset. Bluetooth, when it did function, was unable to find any nearby devices. Adding modules_blacklist=btmtk to the kernel boot options makes all of bluetooth not start, so deleting the bluetooth firmware was required.
3. If you want to also block the appearance of bluetooth from the system since it would create a /sys/class/bluetooth/hci0 entry that would be non functional. The CF-953AX has a property that makes it possible to block out with a udev rule. However if the above firmware for bluetooth is present it was still loading it and chain resetting the device.
   /etc/udev/rules.d/70-mt7921au-nobluetooth.rules:
   DRIVERS=="btusb", ATTRS{interface}=="BT_ACL_If", ATTR{authorized}="0"
   This allowed my USB CSR bluetooth adapter or Intel bluetooth adapter to be the only visible one in the computer.

Some earlier discussion had hardware aliases added so kernel modules would load. Since my adapter has broken bluetooth, is there a way to remove device support with a command or would I have to find the hardware alias file for it and recompile the kernel/module? I would like to minimize the modifications to the system due to badly behaving hardware.

[jaearr]

retracting my earlier statement about the 5000M link speed issue. Was able to fix it by moving the adapter after getting a usb extension base that had a longer cable and used a different design.

[morrownr]

Hi @jaearr

I've seen some other reports of similar behavior. What happens if you simply delete the bluetooth firmware file?

Try that and then tell me if any of the other issues are still there.

I do not what is causing this problem. It does not happen on all adapter because it does not happen on my Edup adapter with the same chip.

[jaearr]

I would not be surprised if the issue I'm having is due to Comfast's implementation.

Deleting /lib/firmware/mediatek/BT_RAM_CODE_MT7961_1_2_hdr.bin does not resolve the USB3 problem and bluetooth still creates /sys/class/bluetooth/hci0 for the device.

a subset of lsusb -t without the firmware and without the udev rule:

/:  Bus 009.Port 001: Dev 001, Class=root_hub, Driver=xhci_hcd/2p, 480M
    |__ Port 002: Dev 010, If 0, Class=Wireless, Driver=btusb, 480M
    |__ Port 002: Dev 010, If 1, Class=Wireless, Driver=btusb, 480M
    |__ Port 002: Dev 010, If 2, Class=Wireless, Driver=[none], 480M
    |__ Port 002: Dev 010, If 3, Class=Vendor Specific Class, Driver=mt7921u, 480M

subset of lsusb -t without the firmware and with the udev rule:

/:  Bus 009.Port 001: Dev 001, Class=root_hub, Driver=xhci_hcd/2p, 480M
    |__ Port 002: Dev 006, If 0, Class=Wireless, Driver=[none], 480M
    |__ Port 002: Dev 006, If 1, Class=Wireless, Driver=[none], 480M
    |__ Port 002: Dev 006, If 2, Class=Wireless, Driver=[none], 480M
    |__ Port 002: Dev 006, If 3, Class=Vendor Specific Class, Driver=mt7921u, 480M

and these were the list of properties from the relevant device from udevadm info -a -p /sys/class/bluetooth/hci0

  looking at parent device '/devices/pci0000:00/0000:00:01.0/0000:03:00.0/usb9/9-2/9-2:1.0':
    KERNELS=="9-2:1.0"
    SUBSYSTEMS=="usb"
    DRIVERS=="btusb"
    ATTRS{authorized}=="1"
    ATTRS{bAlternateSetting}==" 0"
    ATTRS{bInterfaceClass}=="e0"
    ATTRS{bInterfaceNumber}=="00"
    ATTRS{bInterfaceProtocol}=="01"
    ATTRS{bInterfaceSubClass}=="01"
    ATTRS{bNumEndpoints}=="03"
    ATTRS{coredump}=="(not readable)"
    ATTRS{iad_bFirstInterface}=="00"
    ATTRS{iad_bFunctionClass}=="e0"
    ATTRS{iad_bFunctionProtocol}=="01"
    ATTRS{iad_bFunctionSubClass}=="01"
    ATTRS{iad_bInterfaceCount}=="03"
    ATTRS{interface}=="BT_ACL_If"
    ATTRS{supports_autosuspend}=="1"

and dmesg when plugging in using USB3 and the bluetooth firmware deleted:

[ 3851.001186] usb 10-2: new SuperSpeed USB device number 2 using xhci_hcd
[ 3851.016891] usb 10-2: New USB device found, idVendor=0e8d, idProduct=7961, bcdDevice= 1.00
[ 3851.016899] usb 10-2: New USB device strings: Mfr=6, Product=7, SerialNumber=8
[ 3851.016903] usb 10-2: Product: Wireless_Device
[ 3851.016906] usb 10-2: Manufacturer: MediaTek Inc.
[ 3851.016909] usb 10-2: SerialNumber: 000000000
[ 3851.048945] bluetooth hci0: Direct firmware load for mediatek/BT_RAM_CODE_MT7961_1_2_hdr.bin failed with error -2
[ 3851.048958] Bluetooth: hci0: Failed to load firmware file (-2)
[ 3851.048963] Bluetooth: hci0: Failed to set up firmware (-2)
[ 3851.048966] Bluetooth: hci0: HCI Enhanced Setup Synchronous Connection command is advertised, but not supported.
[ 3851.175951] usb 10-2: reset SuperSpeed USB device number 2 using xhci_hcd
[ 3851.199792] bluetooth hci0: Direct firmware load for mediatek/BT_RAM_CODE_MT7961_1_2_hdr.bin failed with error -2
[ 3851.199801] Bluetooth: hci0: Failed to load firmware file (-2)
[ 3851.199804] Bluetooth: hci0: Failed to set up firmware (-2)
[ 3851.199806] Bluetooth: hci0: HCI Enhanced Setup Synchronous Connection command is advertised, but not supported.
[ 3851.248819] mt7921u 10-2:1.3: HW/SW Version: 0x8a108a10, Build Time: 20240716150944a

[ 3851.510924] mt7921u 10-2:1.3: WM Firmware Version: ____010000, Build Time: 20240716151027
[ 3853.115056] mt7921u 10-2:1.3 wls1u2i3: renamed from wlan0
[ 3856.640652] wls1u2i3: authenticate with xx:xx:xx:xx:xx:xx (local address=yy:yy:yy:yy:yy:yy)
[ 3856.788535] wls1u2i3: send auth to xx:xx:xx:xx:xx:xx (try 1/3)
[ 3857.244474] wls1u2i3: send auth to xx:xx:xx:xx:xx:xx (try 2/3)
[ 3857.788404] wls1u2i3: send auth to xx:xx:xx:xx:xx:xx (try 3/3)
[ 3858.340397] wls1u2i3: authentication with xx:xx:xx:xx:xx:xx timed out
[ 3862.359084] wls1u2i3: authenticate with xx:xx:xx:xx:xx:xx (local address=yy:yy:yy:yy:yy:yy)
[ 3862.367862] wls1u2i3: send auth to xx:xx:xx:xx:xx:xx (try 1/3)
[ 3862.949277] wls1u2i3: send auth to xx:xx:xx:xx:xx:xx (try 2/3)
[ 3863.502749] wls1u2i3: send auth to xx:xx:xx:xx:xx:xx (try 3/3)
[ 3864.053248] wls1u2i3: authentication with xx:xx:xx:xx:xx:xx timed out
[ 3868.486343] wls1u2i3: authenticate with xx:xx:xx:xx:xx:xx (local address=yy:yy:yy:yy:yy:yy)
[ 3868.495163] wls1u2i3: send auth to xx:xx:xx:xx:xx:xx (try 1/3)
[ 3868.964329] wls1u2i3: send auth to xx:xx:xx:xx:xx:xx (try 2/3)
[ 3869.518725] wls1u2i3: send auth to xx:xx:xx:xx:xx:xx (try 3/3)
[ 3870.068057] wls1u2i3: authentication with xx:xx:xx:xx:xx:xx timed out
[ 3875.050601] wls1u2i3: authenticate with xx:xx:xx:xx:xx:xx (local address=yy:yy:yy:yy:yy:yy)
[ 3875.059412] wls1u2i3: send auth to xx:xx:xx:xx:xx:xx (try 1/3)
[ 3875.532081] wls1u2i3: send auth to xx:xx:xx:xx:xx:xx (try 2/3)
[ 3876.084025] wls1u2i3: send auth to xx:xx:xx:xx:xx:xx (try 3/3)
[ 3876.627992] wls1u2i3: authentication with xx:xx:xx:xx:xx:xx timed out

iw dev wls1u2i3 station dump lists that AP at -55.00dBm on USB3 but lists signal avg -50 [-50, -58] dBm when connected on USB2. So I would say this device has two strikes against Comfast: including a broken bluetooth interface and not thoroughly testing it for USB3 operation (it can connect to a device that's 1 foot away, so it technically does work with USB3).

[comfastgroup]

I suggest you contact them. [comfast wifi]https://comfastgroup.com

[bjlockie]

I don't use bluetooth and I doubt it works but wifi seems ok.

[   11.936548] Bluetooth: hci0: urb 0000000013a4efbf failed to resubmit (2)
[   13.472150] Bluetooth: hci0: Opcode 0x c03 failed: -110
[   13.554642] usb 2-2.3: reset SuperSpeed USB device number 3 using xhci_hcd
[   13.715819] usbcore: registered new interface driver mt7921u
[   13.720841] mt7921u 2-2.3:1.3: HW/SW Version: 0x8a108a10, Build Time: 20240716150944a

[   13.756562] mt7921u 2-2.3:1.3: WM Firmware Version: ____010000, Build Time: 20240716151027
[   15.584165] Bluetooth: hci0: Opcode 0x c03 failed: -110

$ lsusb -t
/:  Bus 02.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/4p, 5000M
    |__ Port 2: Dev 2, If 0, Class=Hub, Driver=hub/4p, 5000M
        |__ Port 3: Dev 3, If 0, Class=Wireless, Driver=btusb, 5000M
        |__ Port 3: Dev 3, If 1, Class=Wireless, Driver=btusb, 5000M
        |__ Port 3: Dev 3, If 2, Class=Wireless, Driver=, 5000M
        |__ Port 3: Dev 3, If 3, Class=Vendor Specific Class, Driver=mt7921u, 5000M
        |__ Port 4: Dev 4, If 0, Class=Mass Storage, Driver=uas, 5000M
/:  Bus 01.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/1p, 480M
    |__ Port 1: Dev 2, If 0, Class=Hub, Driver=hub/4p, 480M
        |__ Port 2: Dev 3, If 0, Class=Hub, Driver=hub/4p, 480M
        |__ Port 4: Dev 4, If 0, Class=Human Interface Device, Driver=usbfs, 1.5M

Raspbian
6.1.21-v8+

[jaearr]

Thanks. I don't have a Raspberry Pi with USB3 to test if range was affected like my PC, but the Raspberry Pi 3B I tested the CF-953AX on did not have the bluetooth issue I had in Fedora 40. Regardless, the bluetooth did not work in Raspbian (6.6), hcitool -i hci0 scan responded with the interface was down.

[morrownr]

I'm operating slow due to health reasons but I've done a little testing today.

I tested two adapters with the mt7921au chip. Here are the two adapters:

Alfa AXML
Edup EDUP EP-AX1672

Test distro: Ubuntu 24.04
Test CPU: AMD

1. I tested both adapters by booting to see if wifi was working. No wifi interface was adapter with the Alfa adapter but the wifi interface was available on the Edup and it worked as expected. dmesg was clean for the Edup adapter but had dirt in for the Alfa adapter. Most lines indicating a problem seemed to point to bluetooth or usb as the problem.

2. I deleted the bluetooth firmware file for the mt7921au chip in Ubuntu. Now both adapters brought up wifi interface and worked normally. The Alfa adapter, even though wifi worked fine, did show the following with dmesg:

[    3.053925] bluetooth hci1: Direct firmware load for mediatek/BT_RAM_CODE_MT7961_1_2_hdr.bin failed with error -2
[    3.053934] Bluetooth: hci1: Failed to load firmware file (-2)
[    3.053938] Bluetooth: hci1: Failed to set up firmware (-2)
[    3.053942] Bluetooth: hci1: HCI Enhanced Setup Synchronous Connection command is advertised, but not supported.

So, direct firmware for bluetooth failed and then "HCI Enhanced Setup Synchronous Connection command is advertised, but not supported."

Neither adapter advertises that it supports bluetooth, nor should they since USB3 and BT can interfere with each other so who wants USB2 wifi. There is no doubt that the mt8021au chip is capable of BT but it should be turned on off properly for USB adapters. It is appears that BT is indeed turned off properly in the Edup adapter but not with the Alfa adapter. This is my theory anyway...

Other wise, why would the system be detecting and trying to bring BT online, which evidently blocks wifi when it fails. This seems to be the problem with the Comfast adapter in question here.

With that said, This problem only came up earlier this year. I suspect a change with something in BT has broken this. I do not think the problem is with wifi. It could be eeprom settings in the adapters... settings that the maker though were correct but now that BT is working differently, boom, we have a big bug.

Thoughts?

[jaearr]

The interface for Bluetooth is exposed through USB with or without firmware loaded. I suspect fixing bugs in the BT firmware is the source of the problem and because Comfast did a strange implementation that this problem exists. Just like you use the Edup adapter without this problem, the adapter listed from bjlockie's thread doesn't have the problem and is cheaper than the Comfast.

[jaearr]

The kernel version in Ubuntu 24.10 is 6.11.0-8 which works as previously mentioned (deleting the bluetooth firmware in /lib/firmware/mediatek/BT*). Fedora 41 uses 6.11.4-301 and OpenSuSE Tumbleweed uses 6.11.5-1 which don't work: after the bluetooth fails to initialize, Linux does not start WiFi. As much as I hate creating unnecessary e-waste, I'm going to discard the Comfast CF-953AX because of this bluetooth issue creating problems and pick up another Fenvi FU-AX1800 clone since those work trouble-free.

[morrownr]

Further testing:

I added the mt7921au firmware file back into my first test system but I added a little usb BT adapter to provide bluetooth support. Both adapters then showed wifi that worked with no problem. The Alfa adapter still have some junk in the log but it did not affect wifi operation.

I then moved to a RasPi4B that had internal BT on. Same result...wifi worked with both adapters.

It appears that as long as there is a functional BT chip active in the system, wifi is happy. The problem with wifi seems to come about if the are true:

1. If an adapter that seems to be telling the system that BT is available is in use such as the Alfa and Comfast adapters mentioned here.
2. When the mt7921au BT firmware is installed.
3. When no other BT is available in the system.

Testing various kernels to determine when this problem started can be helpful. I know nothing about bluetooth myself as I rarely use it and I know nothing of the bluetooth system in Linux.

[jaearr]

When I was testing on an 64bit x86 laptop with an Intel AC7260 which has a USB Bluetooth interface, I had issues with the adapter crashing/disconnecting after Ubuntu 24.04 attempted to load the BT firmware. Do you know what chipset was in your bluetooth adapter that appeared to not cause trouble on 64 bit x86?

[morrownr]

@jaearr

Do you know what chipset was in your bluetooth adapter that appeared to not cause trouble on 64 bit x86?

Laptop = mt7921 M.2 x86_64
Desktop = mt7922 PCIe x86_64
Mini PC = mt7925 M.2 AMD64

I have a nano bt adapter that I could throw into the mix.

The interface for Bluetooth is exposed through USB with or without firmware loaded. I suspect fixing bugs in the BT firmware is the source of the problem and because Comfast did a strange implementation that this problem exists.

I have seen this reported with owners of the Alfa AXML so it is not just the early Comfast adapters. I suspect that your analysis that the problem has to do with a bug in BT is correct. It would be good to figure out exactly what happened.

What I advise users to do is delete the bluetooth firmware file. It may show a couple of ugly lines in the system log but wifi should then work okay. None of the current mt7921au based adapters in the Plug and Play List show this problem as far as I know.

[EasyNetDev]

Hi,

Today I retested again my CF-953AX adapter on my BananaPi R4 router. It has an USB 3.0 port.

Hardare: BananaPi R4 with MT7988
Kernel : 6.12.0-rc1-rsslro-EasyOS
OS : Debian Trixie ARM64

As soon I'm plug in the adapter this happen:

[  191.741046] usb 2-1.4: new SuperSpeed USB device number 3 using xhci-mtk
[  191.762591] usb 2-1.4: New USB device found, idVendor=0e8d, idProduct=7961, bcdDevice= 1.00
[  191.770935] usb 2-1.4: New USB device strings: Mfr=6, Product=7, SerialNumber=8
[  191.778242] usb 2-1.4: Product: Wireless_Device
[  191.782767] usb 2-1.4: Manufacturer: MediaTek Inc.
[  191.787545] usb 2-1.4: SerialNumber: 000000000
[  191.890735] Bluetooth: Core ver 2.22
[  191.894353] Bluetooth: Starting self testing
[  191.908474] Bluetooth: ECDH test passed in 9629 usecs
[  191.916812] Bluetooth: SMP test passed in 3181 usecs
[  191.921783] Bluetooth: Finished self testing
[  191.926111] NET: Registered PF_BLUETOOTH protocol family
[  191.931430] Bluetooth: HCI device and connection manager initialized
[  191.937783] Bluetooth: HCI socket layer initialized
[  191.942661] Bluetooth: L2CAP socket layer initialized
[  191.947716] Bluetooth: SCO socket layer initialized
[  191.967181] cfg80211: Loading compiled-in X.509 certificates for regulatory database
[  191.986247] Loaded X.509 cert 'sforshee: 00b28ddf47aef9cea7'
[  191.992452] Loaded X.509 cert 'wens: 61c038651aabdcf94bd0ac7ff06c7248db18c600'
[  191.996298] usbcore: registered new interface driver btusb
[  192.056031] cfg80211: loaded regulatory.db is malformed or signature is missing/invalid
[  192.103626] Bluetooth: hci0: HW/SW Version: 0x008a008a, Build Time: 20240826151221
[  192.174571] Bluetooth: hci0: urb 00000000e41d32ac failed to resubmit (2)
[  196.750950] Bluetooth: hci0: command 0xfd98 tx timeout
[  196.756102] Bluetooth: hci0: failed to reset (-19)
[  196.756108] Bluetooth: hci0: Failed to apply iso setting (-110)
[  196.766794] Bluetooth: hci0: Device setup in 4648446 usecs
[  196.772275] Bluetooth: hci0: HCI Enhanced Setup Synchronous Connection command is advertised, but not supported.
[  198.830906] Bluetooth: hci0: Opcode 0x0c03 failed: -110
[  200.910864] Bluetooth: hci0: Failed to read MSFT supported features (-110)
[  202.990836] Bluetooth: hci0: AOSP get vendor capabilities (-110)
[  202.997247] Unable to handle kernel paging request at virtual address dead000000000108
[  203.005170] Mem abort info:
[  203.007951]   ESR = 0x0000000096000044
[  203.011696]   EC = 0x25: DABT (current EL), IL = 32 bits
[  203.016995]   SET = 0, FnV = 0
[  203.020037]   EA = 0, S1PTW = 0
[  203.023171]   FSC = 0x04: level 0 translation fault
[  203.028035] Data abort info:
[  203.030906]   ISV = 0, ISS = 0x00000044, ISS2 = 0x00000000
[  203.036377]   CM = 0, WnR = 1, TnD = 0, TagAccess = 0
[  203.041419]   GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
[  203.046718] [dead000000000108] address between user and kernel address ranges
[  203.053848] Internal error: Oops: 0000000096000044 [#1] SMP
[  203.059410] Modules linked in: mt7921u(+) mt792x_usb mt7921_common mt792x_lib mt76_connac_lib mt76_usb mt76 mac80211 libarc4 btusb btrtl btintel btbcm cfg80211 btmtk bluetooth crct10dif_ce polyval_ce polyval_generic evbug evdev sm4 sha1_ce governor_passive gpio_keys mtk_cci_de
vfreq fuse nfnetlink ip_tables x_tables
[  203.087325] CPU: 2 UID: 0 PID: 3398 Comm: (udev-worker) Not tainted 6.12.0-rc1-rsslro-EasyOS #1
[  203.096011] Hardware name: Banana Pi BPI-R4 (DT)
[  203.100615] pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[  203.107564] pc : hci_unregister_dev+0x60/0x220 [bluetooth]
[  203.113076] lr : hci_unregister_dev+0x3c/0x220 [bluetooth]
[  203.118576] sp : ffff80008268b630
[  203.121878] x29: ffff80008268b630 x28: ffff800079cef1d8 x27: ffff0000c8b6b400
[  203.129004] x26: ffff0000c03c6800 x25: ffff0000c03c68a8 x24: ffff800079bc7000
[  203.136128] x23: 0000000000000000 x22: ffff800079bc7078 x21: ffff0000c756e000
[  203.143251] x20: ffff800079a1d000 x19: ffff0000c756e000 x18: ffffffffffffffff
[  203.150375] x17: 0000000000000901 x16: 0000000000000000 x15: ffff0000c37ae160
[  203.157499] x14: ffff0000c9dde240 x13: ffff0000c14a5200 x12: 0000000000000000
[  203.164623] x11: 00000000000000a8 x10: 0000000000000980 x9 : ffff80008268b440
[  203.171746] x8 : 0000000000000000 x7 : 0000000000000000 x6 : 0000000000000001
[  203.178869] x5 : 0000000000000001 x4 : ffff800079a1d4d8 x3 : 0000000000000000
[  203.185992] x2 : dead000000000100 x1 : dead000000000122 x0 : 0000000000000100
[  203.193117] Call trace:
[  203.195551]  hci_unregister_dev+0x60/0x220 [bluetooth]
[  203.200709]  btusb_disconnect+0x50/0x150 [btusb]
[  203.205318]  usb_unbind_interface+0x78/0x2b4
[  203.209583]  device_remove+0x70/0x80
[  203.213152]  device_release_driver_internal+0x1cc/0x224
[  203.218363]  device_release_driver+0x18/0x24
[  203.222621]  usb_forced_unbind_intf+0xa0/0xc8
[  203.226966]  usb_reset_device+0xd4/0x2a0
[  203.230880]  mt7921u_probe+0xac/0x260 [mt7921u]
[  203.235403]  usb_probe_interface+0xd8/0x318
[  203.239575]  really_probe+0xc0/0x39c                                                                                                                                                             
[  203.243139]  __driver_probe_device+0x7c/0x14c                                                                                                                                                    
[  203.247484]  driver_probe_device+0x3c/0x120                                                                                                                                                      
[  203.251654]  __driver_attach+0xc4/0x200                                                                                                                                                          
[  203.255478]  bus_for_each_dev+0x78/0xd8                                                                                                                                                          
[  203.259302]  driver_attach+0x24/0x30
[  203.262867]  bus_add_driver+0x110/0x240
[  203.266693]  driver_register+0x68/0x124
[  203.270517]  usb_register_driver+0x80/0x150
[  203.274688]  mt7921u_driver_init+0x28/0x1000 [mt7921u]
[  203.279815]  do_one_initcall+0x44/0x25c
[  203.283641]  do_init_module+0x5c/0x220
[  203.287382]  load_module+0x1b14/0x20fc
[  203.291120]  init_module_from_file+0x88/0xcc
[  203.295378]  __arm64_sys_finit_module+0x194/0x3d8
[  203.300071]  invoke_syscall+0x48/0x104
[  203.303811]  el0_svc_common.constprop.0+0xc0/0xe0
[  203.308504]  do_el0_svc+0x1c/0x28
[  203.311808]  el0_svc+0x34/0xe0
[  203.314854]  el0t_64_sync_handler+0x120/0x12c
[  203.319199]  el0t_64_sync+0x194/0x198
[  203.322851] Code: 88e07c41 35000840 a9400662 d2802000 (f9000441) 
[  203.328932] ---[ end trace 0000000000000000 ]---
[  203.333536] Kernel panic - not syncing: Oops: Fatal exception
[  203.339269] SMP: stopping secondary CPUs
[  203.343183] Kernel Offset: disabled
[  203.346658] CPU features: 0x00,00000102,00200000,4200420b
[  203.352043] Memory Limit: none
[  203.355088] Rebooting in 10 seconds..

In this case the device is crashing even earlier than I was expecting.

[morrownr]

Thanks for adding this information. Right now I am aware of 3 adapters that suffer from this bluetooth problem:

Comfast CF-853AX
Alfa AXML
Alfa AXM

All 3 have been removed from The Plug and Play List. Not to say that there is necessarily a problem with the adapters but rather that list if for people to buy new adapters and we don't know when this is going to be fixed.

You can blacklist btusb and that stops the problem but then no bluetooth at all.

[fhteagle]

I agree with taking the CF-953AX off the plug and play list. Of the three mt7921 based adapters/cards I was using to make APs, this has been the least stable (even with bluetooth fully disabled/blacklisted/masked on the host system), most quirky/temperamental, etc. In comparison, the AWUS036AXML has been a very stable and capable AP source (again with BT disabled).

[morrownr]

@fhteagle

Thanks for the note. When mt7921au based adapters first came out, several users reported good results with the CF-953AX so I put the adapter in the Plug and Play List. I then bought a CF-951AX for myself and it became apparent fairly quickly that it was not a good quality adapter so it was added and then removed as the problems showed themselves. Stability was a problem and it appears that quality control on the assembly line was to blame but that could be caused by bad engineering. My opinion of Comfast is not good these days.

I am now in direct contact with Alfa. I am hoping to facilitate work that can lead to fixing this BT problem once and for all. I have an AXML that has the problem but I also have a EDUP EP-AX1672 that does not show the problem. I think all of the adapters in the mt7921au section of the Plug and Play List are good and do not show this problem. I have one pending adapter that is posted in Issues waiting for people to report as I may make an exception to my "no TP_link policy" as it might work out in this case.

It certainly appears to me that a patch to the bluetooth stack is what caused this. I have not identified the patch but I just cannot see it being anything else. Hopefully this BT problem is fixed soon so we can get back to enjoying our AXML's with having to do workarounds.

[fhteagle]

Well my CF953-AX seems to be dying. Even with btusb module blacklisted, it is causing system hang / soft freezes. dmesg says:

mt7921u 2-3:1.3: vendor request req:63 off:4230 failed:-110

2 years in heavy testing service, so I am not mad, but annoyed enough to want it out of my stash of stuff lol.

If anyone wants it for parts / testing / think you can resurrect it, let me know.

[bjlockie]

I've been using mine as a client 24x7 for probably more than 2 years.
No problems with mine. :-)

I have kernel 6.1.21.
RaspianOS. :-)

[morrownr]

@fhteagle

Well my CF953-AX seems to be dying. Even with btusb module blacklisted, it is causing system hang / soft freezes.

I suspect your adapter is not dying. There are 3 mt7921au based adapters that are struggling: (and were removed from The Plug and Play List in 2024)

Comfast CF-953AX
Alfa AXML
Alfa AXM

Those were 3 of the first available adapters with the mt7921au chip. Problems were first noted around the time kernel 6.7 was released, about 1 year ago. The various workarounds, including deleting the bluetooth firmware file, used to work but no longer work with the most recent distros such as Ubuntu 24.10.

I am working the issue. I will probably post an issue for this problem soon so that we have a place to compare notes.

[morrownr]

@bjlockie

No problems with mine. :-)

Which kernel are you using?

[fhteagle]

@morrownr - I did try going back to 6.6.69-lts kernel before writing up my issue, but no change there. But I can certainly wait awhile to see if the driver situation improves. But I still suspect hardware failure of my unit, based on instant and repetitive (each 1-4 seconds) I get usb errors in journalctl. That, combined with the fact that even plugging the 953 in with bt* modules unloaded, bluetooth service stopped, and the bluetooth firmware file banished, even a local terminal soft freezes with up to a minute delay on even simple commands such as ls. Remove the device from the usb port, and the soft freeze is instantly over.

@bjlockie - when you plug in your 953, does the little blue LED next to the "Wifi6" lettering light up? Mine used to, but does not anymore....

[bjlockie]

@bjlockie - when you plug in your 953, does the little blue LED next to the "Wifi6" lettering light up? Mine used to, but does not anymore....

Mine doesn't but I've never checked it before so maybe it never did.

[morrownr]

@fhteagle

I did try going back to 6.6.69-lts kernel before writing up my issue...

You might need to go back with the firmware files as well. I can show you how if need be.

But then again, the adapter could be going bad. It happens. Over the last 2 years, I have observed things that would make me reluctant to add a Comfast adapter to The Plug and Play List again. I question the quality of their manufacturing.