Hi Lukasz, I get exactly the same errors in dmesg as oguzy (I copy his, as my laptop has no network): Dec 30 14:57:03 gulsah-SATELLITE-PRO-C660 kernel: [ 5743.881187] ath: Failed to stop TX DMA! Dec 30 14:57:03 gulsah-SATELLITE-PRO-C660 kernel: [ 5743.895845] ath: DMA failed to stop in 10 ms AR_CR=0x00000024 AR_DIAG_SW=0x02000020 DMADBG_7=0x000062c0 Dec 30 14:57:03 gulsah-SATELLITE-PRO-C660 kernel: [ 5743.895856] ath: Could not stop RX, we could be confusing the DMA engine when we start RX up
I did try the nohwcrypt=1, but that did not work at all (the error was actually getting worse, i.e. almost immediately after having connected to the router). What might be significant is that some of the errors mention 100ms and oguzy and mine mention 10ms. Also when my wireless does work, the speed (8mb/s down) and ping (2ms to router) is normal and the indicator lighton the laptop is on. This changes to off, the errors appear in the logs and the network manager tries to reconnect, claiming that the connection needs a password, which it does already have. With a powerdown and start, the connection comes back up, a simple restart is not enough. After a simple restart, the kernel does not recognize the network card anymore, lspci does not show it to be present on bus 3:0.0 where it previously was. With a simple restart to Windows it does not recognize the card sometimes also. I read somewhere that this type of card needs/gets a firmware reinitialisation every time a reboot occurs, so that might be important. The all or nothing behaviour of the bug I see seems to me more consistent with a race condition than an encryption problem. Maybe this race condition changes when encryption is on/off. I think the above patchby Christian Aravena, which (I think) adds a call to a semaphore in the ath9 driver, is not present in the 3.0 kernel driver and therefore the bug was reintroduced. Is there someone who can verify it not being present in this kernel? That would explain a lot. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/736171 Title: ath: Failed to stop TX DMA in 100 msec after killing last frame To manage notifications about this bug go to: https://bugs.launchpad.net/linux/+bug/736171/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
