Description of problem: I'm seeing two problems with my rt61pci wireless right now, often it will repeatedly timeout connecting to the AP when it does correctly connect I see this spammed in dmesg over and over again. Please file bug report to http://rt2x00.serialmonkey.com. phy0 -> rt2x00pci_write_tx_data: Error - Arrived at non-free entry in the non-full queue 2. Version-Release number of selected component (if applicable): kernel-2.6.23.1-41.fc8 How reproducible: Steps to Reproduce: 1. attempt to have NM connect to AP 2. If connection isn't timed out for no good reason Actual results: Notice convinient spamming of logs with request for a bugreport Expected results: Connection to my AP on every boot up, no fail and keep my logs quiet unless something is actually wrong. Additional info:
These kernels have some wireless updates not yet in the -41 kernel you are testing: http://koji.fedoraproject.org/koji/taskinfo?taskID=218502 Would you mind trying to reproduce the issue with those kernels? Does it reappear?
I could successfully reconnect to the wireless after several warm boots, however after a cold boot I could not connect to the AP again without restarting Network Manager a few times. All in all this is a fantastic improvement so far, the success rate has gone up dramatically and so far I have not seen the error message.
Please try these kernels: http://koji.fedoraproject.org/koji/buildinfo?buildID=23734 Are they working satisfactorily for you?
I've been running that for a good 18 hours now, so far it's the best kernel for my wifi. It will still occasionally go into a state where it will claim that the progress of connecting to and getting an IP from the AP times out, this happens most commonly when using WPA. Also this: phy0 -> rt2x00pci_write_tx_data: Error - Arrived at non-free entry in the non-full queue 2. Popped back up in the logs. So we aren't perfect yet but it's far closer to being usable now.
A patch that was reported to fix this problem has been included in kernel-2.6.23.8-63.fc8. Could you test that kernel and confirm whether or not it fixes the problem for you?
John, I'm getting similar error with kernel-2.6.23.8-63.fc8 and rt73usb (Linksys WUSB54GC): phy0 -> rt2x00usb_write_tx_data: Error - Arrived at non-free entry in the non-full queue 2 Also, I got a few of those in messages log: Dec 6 19:23:41 nufin kernel: phy0 -> rt2x00usb_vendor_request: Error - Vendor Request 0x07 failed for offset 0x3040 with error -19. When I pull out the NIC and then plug it back in, the connection seems to be stable.
I'm running kernel-2.6.23.8-63.fc8 and have a wireless NIC which uses the rt61pci device driver. I get a lot of "phy0 -> rt2x00pci_write_tx_data: Error - Arrived at non-free entry in the non-full queue 2." entries in the messages log file with a very high rate. I counted the number of new messages I got in 10 seconds, and they were roughly 100.
I have RaLink RT2561/RT61 802.11g PCI card. I'm loosing connection since kernel-2.6.23.1-42.fc8 kernel. I tryed kernel-2.6.23.8-63.fc8 kernel with no luck.
Should be fixed in wireless-dev commit: http://git.kernel.org/?p=linux/kernel/git/linville/wireless-2.6.git;a=commit;h=a43a3396d2ef2d6d4b8e3de2e79c0a214c403aff
*** Bug 426874 has been marked as a duplicate of this bug. ***
Please try a later .fc8 kernel, one that includes the patch Ivo cited in comment 9 above: http://koji.fedoraproject.org/koji/buildinfo?buildID=28665 Does that correct the issue for you?
It works for me since kernel-2.6.23.9-85.fc8 - I havn't tested the 99th build from koji yet.
I would be glad to retest this on my system, but I am currently blocked from doing so by another rt2x00 issue (buzilla 411481).
I have now run kernel-2.6.23.12-99.fc8 for two days. I see no huge number of messages any longer. However, I still have a problem I had before. The link speed starts at 54 Mbps, but suddenly drops and goes down to 1 Mbps and stays there even though the link quality is good. I should mention that I only have about 5 meters between my wlan router and client, so I expect to have quite high link speed.
Sounds like bug #382781 try setting your configuration to a fixed rate.
...and report any further link speed problems there, please. :-) I'm going to optimistically close this based on comment 14. Please reopen if the problem can be recreated with -99.fc8 or later.