rt2x00usb_write_tx_data error

Live forum: http://rt2x00.serialmonkey.com/viewtopic.php?t=4734

ModestUser

15-04-2008 10:33:41

Hi guys!

I am using a DWL-G122, C1 and rt73usb, 2.6.24.4 kernel. Your driver just broke and told me to contact you

phy0 -> rt2x00usb_write_tx_data Error - Arrived at non-free entry in the non-full queue 2.
Please file bug report to http//rt2x00.serialmonkey.com.
phy0 -> rt2x00usb_write_tx_data Error - Arrived at non-free entry in the non-full queue 2.
Please file bug report to http//rt2x00.serialmonkey.com.
phy0 -> rt2x00usb_write_tx_data Error - Arrived at non-free entry in the non-full queue 2.
Please file bug report to http//rt2x00.serialmonkey.com.

These entries are there multiple hundreds of times. No other syslog messages. This has not happened before, so i assume it is not reproducible.

btw, will there by substantial changes in the upcoming 2.6.25 kernel? I have several issues with the rt73usb at the moment.

IvD

15-04-2008 10:48:55


phy0 -> rt2x00usb_write_tx_data Error - Arrived at non-free entry in the non-full queue 2.
Please file bug report to http//rt2x00.serialmonkey.com.
phy0 -> rt2x00usb_write_tx_data Error - Arrived at non-free entry in the non-full queue 2.
Please file bug report to http//rt2x00.serialmonkey.com.
phy0 -> rt2x00usb_write_tx_data Error - Arrived at non-free entry in the non-full queue 2.
Please file bug report to http//rt2x00.serialmonkey.com.

These entries are there multiple hundreds of times. No other syslog messages. This has not happened before, so i assume it is not reproducible.
[/quote1rtbxc7f]

It is indeed a hard to reproduce bug, it is a race condition where it could be possible that 2 events access the queue at exactly the same time after which the queue blows up and starts spitting out these messages.

It has been fixed for the 2.6.26 kernel, but the change was so large and bound to many other patches that it cannot be backported to earlier kernels. The only thing that can be done is reload the module when it happens.


btw, will there by substantial changes in the upcoming 2.6.25 kernel? I have several issues with the rt73usb at the moment.[/quote1rtbxc7f]

A lot of bugfixes and better antenna handling.
Performance improvement currently varies between each -rc release and per person so I can't say anything useful about that at this time.

cciechad

22-05-2008 20:55:03

So Andrew Morton's 2.6.26-rc2-mm1 kernel would include the fix for the queue issue?

IvD

23-05-2008 13:59:32

Yes 2.6.26-rc1 and above should contain the fix.

raqua

23-06-2008 17:46:18



It is indeed a hard to reproduce bug, it is a race condition where it could be possible that 2 events access the queue at exactly the same time after which the queue blows up and starts spitting out these messages.

It has been fixed for the 2.6.26 kernel, but the change was so large and bound to many other patches that it cannot be backported to earlier kernels. The only thing that can be done is reload the module when it happens.
[/quote37jbqjq0]

Not really hard to reproduce on my machine, it happens all the time, if I can give you some logs or other info, just let me know.
Unloading and loading module again does not help, that bug comes back right after that.
I have rt73usb.

Zi7

24-07-2008 18:39:38

Do you run some kind of NetworkManager app to configure network device automatically?
What is your linux distribution?

If you can compile and run a rt2x00.git kernel, this issue should be solved.

olskar

04-03-2009 10:48:35

Ive got the very same error on kernel 2.6.27-9-generic. This is happening every hour, really annoying. Networkmanager 0.7.0, Ubuntu Intrepid and Ralink Technology, Corp. RT2501USB Wireless Adapter

IvD

04-03-2009 17:36:05

Then upgrade your kernel to 2.6.28 or better yet 2.6.29-rcX