-
Notifications
You must be signed in to change notification settings - Fork 393
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
AWUS036AXML: ERROR Transfer event TRB DMA ptr not part of current TD (kernel regression) #334
Comments
The ALFA seems to be ok and the driver seems to be ok, but you're running into a kernel regression regarding the XHCI subsystem as reported here: There is nothing I can do and you have to wait for a kernel fix. |
Looks like there are several XHCI related reports: I think we can close this report because it is related to XHCI. |
ZerBea, thanks for the info. I updated the kernel to latest debian unstable: Linux debian 6.4.0-1-amd64 Still no go. If you have any advice for things i can try it would be appreciated. |
A workaround for me is to connect the WiFi device (in may case an ALFA AWUS036ACM) to an USB 2 port. $ uname -r $ hcxdumptool -v If connected to an USB 3 port, the ALFA receive nothing. |
Looks like connecting the ALFA via an external USB hub is working, too: |
After some research (kernel 6.4.7): USB 3 WiFi devices are not working as expected if connected to an USB 3 port BTW: On kernel 6.4.7 there is no dmesg warning like this "ERROR Transfer event TRB DMA ptr not part of current TD".
and hcxdumptool got nothing from the driver:
Looks like the XHCI bug is still not fixed |
Also it looks like there are some more problems regarding USB 3: |
To confirm that the xhci system is not working as expected, you can run either tshark or Wireshark in capture mode in parallel: terminal 2:
tshark will receive nothing, too. |
Oh man thanks for that info! Verified: it's not receiving anything either with "--disable_beacon". Though I thought I observed it working in kernel 5.10 or so. On VirtualBox i disabled USB3 (XHCI) but when I plugged in this USB3 adapter, it issued an error "no available usb ports". When i plugged it in via an old USB 2.0 extension cable, the system recognized the device (lsusb) but didn't recognize it as a wifi adapter (no output in iwconfig). Also, that produced errors in dmesg (unrecognized vendor id..). Yeah, I feel this wraps it up for me, i dont have USB2.0 ports on this SurfacePro so will have to wait for proper XHCI support or swap out for the AWUS036ACHM. |
We got a response on bugzilla: |
Hey ZerBea, I went ahead and applied that patch. I haven't compiled a kernel in god knows how long but felt I needed to give this a good try. Unfortunately, still not getting a positive feedback, seeing same error Double checked that XHCI is enabled in virtualbox, and kernel config has TPM enabled. Running this command Thanks for this wonderful tool and your advice. If you want me to test something, i'll be available gladly.
|
Thanks for this information. |
BTW: I don't think it is related to ALFA devices or the mt76 driver, because other USB 3.0 devices rt2800usb driver) failed, too. |
There is a message for you (regarding the VM): |
Still no go, I'll post the results in the bugzilla. Dusted off my old sony vaio, it had a couple of usb3.0 ports. |
Great, thanks. I don't think the problem is related to the driver. Tested using an USB 2 device on the same port:
iw, ip link and tshark are working as expected as well as hcxdumptool. while all USB 3 devices failed. Next step for me is to order this adapter (ALLNET WA1200AC ) to run some more tests: |
By this commit:
|
Running the same configuration (kernel, driver, ALFA WiFi device) on my Raspberry Pi, everything is working as expected. |
RPI is kernel 5.x? i recall it working on a debian i had with kernel 5.x with an older hcxdumptool and --disable_active_monitor flag By any chance, anything you think i should try? |
is working fine. Information is here: 32 bit: |
I just pulled the latest hcxdumptool, tried it on my pi3 and it says it was able to capture packets but not showing any output. I'm using AWUS036AXML with linux firmware updated as per https://github.com/morrownr/USB-WiFi/blob/main/home/How_to_Install_Firmware_for_Mediatek_based_USB_WiFi_adapters.md . $ uname -a $ lsb_release -a Also tried it again on a box with the previously patched debian kernel 6.5: [ 14.914156] mt7921u 3-4:1.3: firmware: direct-loading firmware mediatek/WIFI_MT7961_patch_mcu_1_2_hdr.bin $ hcxdumptool -v phy idx hw-mac virtual-mac m ifname driver (protocol)0 3 00c0cab3c769 00c0cab3c769 * wlx00c0cab3c769 mt7921u (NETLINK) sudo systemctl stop wpa_supplicant Not really sure what to try next, but open for any advice |
Just checked the WN722n wifi card on the patched kernel 6.5 and that works well: available wlan devices: phy idx hw-mac virtual-mac m ifname driver (protocol)1 4 98ded00948e5 98ded00948e5 + wlx98ded00948e5 ath9k_htc (NETLINK)
|
Thanks for that information. The screen shots confirm that hcxdumptool is working as expected on TL722WN (USB 2) but not on AWUS036AXML (USB 3.2 Gen 1). |
No prob, thank you! |
Hello,
I'm getting this error when running hcxdumptool scanning with the new AWUS036AXML usb adapter. Firmware was updated as per https://store.rokland.com/pages/alfa-awus036axml-driver-support-page. The hcxdumptool returns no output, and this error scrolls constantly in dmesg/messages.
Wondering if anyone has any advice. I'm open to returning this adapter and getting AWUS036ACHM instead if that's best.
Here's the error:
$ dmesg [ 143.833329] xhci_hcd 0000:00:0c.0: ERROR Transfer event TRB DMA ptr not part of current TD ep_index 8 comp_code 13 [ 143.833865] xhci_hcd 0000:00:0c.0: Looking for event-dma 00000000010a0070 trb-start 00000000010a0080 trb-end 00000000010a00b0 seg-start 00000000010a0000 seg-end 00000000010a0ff0
The text was updated successfully, but these errors were encountered: