Subject: Bug#628444: iwlagn - "MAC is in deep sleep", cannot restore wifi operation
Hi Jonathan!

Sorry I took a while to respond and apologies in advance for them being quite
useless...

> - steps to reproduce, assuming I had the same hardware

Use the computer. My maximum time without hitting the bug has been less than
48 hours before I added the following module options: 11n_disable=1
power_save=0 wd_disable=1 which some googling suggested might solve it. They
do not, but they make it less frequent. With them, I can even get a WEEK
without seeing this.

But notice I have different hw than the original reporter: mine is X200s with
"Intel Corporation Ultimate N WiFi Link 5300" (pci-id: 8086:4236).

> - expected result, actual result, and how the difference indicates a
> bug (should be simple enough in this case)

=) Expected result: wifi keeps working unless I switch it off using rf_kill,
physical switch, unload the module, or turn off the computer.

Actual result: suddenly, out of the blue, in the middle of typing an email in
kmail, see attachment.

Bug: the wifi card has certainly not changed into "Unknown hardware type"
suddenly.

> - how reproducible it is (100% of the time? 50%?)

100% when waiting long enough between reboots.

> - which kernel versions you have tested, and results with each

There were no problems in the 2.6-series. The bug occurs at least in the
Debian kernel versions 3.2.0-1-amd64, 3.0.0-2-amd64, and 3.0.0-1-amd64.

> - full "dmesg" output from booting and reproducing the bug, as an
> attachment

Do not have it now, if really necessary, will get it next time it occurs
(which may be a while: I am back to 2.6.39 because I need to get work done).

> - any other weird observations or workarounds

No workaround. Above module parameters alleviate the issue. This is a
regression, so I suggest someone with time (if anyone has it) bisects 2.6 and
3.2... horrible task, I do not envy anyone doing that.

> upstream to a public mailing list. So the purpose of these questions
> is to collect data on what's known so far as a starting point.

Please CC me if you do.

Cheers,
Juha

--
-----------------------------------------------
| Juha JÃykkÃ, juhaj@xxxxxx |
| http://www.maths.leeds.ac.uk/~juhaj |
-----------------------------------------------

syslog.txt
Description: Text document

(C)2011 mailinglist-archive.com