Bug#598144: rt2500pci: very low transfers and link quality

[David Sanchez Herrero <david.krusty@gmail.com>]

Hi Jonathan,

Thanks for reply about this problem, but the laptop with rt2500 wireless card died 6 or 7 months ago (a graphic card crash), so I can't try the 3.1 kernel version. I think that after a lot of tests (compiling a new module version using compat-wireless project, compiling the drivers from Ralink, updating the kernel version, disabling Network-Manager to manually configure the card, ...), I finally got the wireless card working fine, but now I can't remember which was the final solution to the problem.

I'm sure that there was something bad with the rt2500pci module because I found a lot of information about the problem, associated to different distros, and I could verify with a Live CD that the problem was in Ubuntu too.

I'm thinking about analyze the laptop hard disk this Christmas to recover some files to my new computer, so if I find some information about the solutión that I applied, I promise to write again to tell you.

Meanwhile, I think that you can mark this bug as solved if you want.

Greetings, David.

El 3 de diciembre de 2011 08:19, Jonathan Nieder <jrnieder@gmail.com> escribió:

# forwarded last year:
#  http://thread.gmane.org/gmane.linux.kernel.wireless.general/56714
# no response, so let's try again.
notforwarded 598144
quit

Heya,

Regarding version 2.6.32-20, David Sánchez Herrero wrote:

> I have been using rt2500pci since rt2500 legacy drivers source package
> dissapeared from the repository, and I have had this problems from the first
> time until now. The legacy drivers worked fine, but with the new ones I have
> very low transfers and poor link quality.

That's no good.  So let's see.

[...]
> it's
> impossible to make anything because in all terminals appears one message
> continously:
>
> cfg80211: Calling CRDA to update world regulatory domain.
>
> I can see that I lose the connection, probably because this message. This
> happens with 2.6.36-rc5-686 kernel. Any idea about solving this problem?

Very odd.  Does it still happen?

[...]
> This problem doesn't exists with 2.6.32-5-686 kernel, but with it, I
> have again the error message that appears in my first report:
>
> phy0 -> rt2500pci_set_device_state: Error - Device failed to enter state 1 (-16).

Which kernel version is that?  You can get the currently running
kernel version by running "cat /proc/version" and looking for the
string in parentheses.

I don't see any relevant fixes upstream (amazingly, since the driver
has had a lot of work done).  If you can get a chance to test a 3.1.y
or later kernel from sid, that would be very helpful.  (Then we can
get help from upstream.)  For graphics, while testing you can use the
vesa driver or nouveau.

Thanks for writing and hope that helps,
Jonathan

[...]
> 01:00.0 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller [10ec:8168] (rev 01)
>       Subsystem: Micro-Star International Co., Ltd. Device [1462:3fcc]