Arch Linux

Please read this before reporting a bug:
https://wiki.archlinux.org/title/Bug_reporting_guidelines

Do NOT report bugs when a package is just outdated, or it is in the AUR. Use the 'flag out of date' link on the package page, or the Mailing List.

REPEAT: Do NOT report bugs for outdated packages!
Tasklist

FS#18771 - [kernel26] Cannot use my atheros wifi chipset with 2.6.33

Attached to Project: Arch Linux
Opened by Frederic Bezies (fredbezies) - Sunday, 21 March 2010, 11:25 GMT
Last edited by Roman Kyrylych (Romashka) - Wednesday, 29 September 2010, 11:36 GMT
Task Type Bug Report
Category Upstream Bugs
Status Closed
Assigned To Tobias Powalowski (tpowa)
Thomas Bächler (brain0)
Architecture All
Severity Medium
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 0
Private No

Details

Description: I installed today an archlinux, and after I switched to testing with kernel 2.6.33, either network manager or wicd cannot find any wifi connection. I got this in /var/logs/errors.log

Mar 21 12:17:59 fredo-arch kernel: ath5k phy0: gain calibration timeout (2412MHz)
Mar 21 12:17:59 fredo-arch kernel: ath5k phy0: gain calibration timeout (2417MHz)
Mar 21 12:18:00 fredo-arch kernel: ath5k phy0: gain calibration timeout (2422MHz)
Mar 21 12:18:00 fredo-arch kernel: ath5k phy0: gain calibration timeout (2427MHz)
Mar 21 12:18:01 fredo-arch kernel: ath5k phy0: gain calibration timeout (2432MHz)
Mar 21 12:18:01 fredo-arch kernel: ath5k phy0: gain calibration timeout (2437MHz)
Mar 21 12:18:01 fredo-arch kernel: ath5k phy0: gain calibration timeout (2442MHz)
Mar 21 12:18:02 fredo-arch kernel: ath5k phy0: gain calibration timeout (2447MHz)
Mar 21 12:18:02 fredo-arch kernel: ath5k phy0: gain calibration timeout (2452MHz)
Mar 21 12:18:03 fredo-arch kernel: ath5k phy0: gain calibration timeout (2457MHz)


Additional info:

[fred@fredo-arch src]$ pacman -Qi kernel26
Nom : kernel26
Version : 2.6.33.1-1

Steps to reproduce: just upgrade to testing ;)
This task depends upon

Closed by  Roman Kyrylych (Romashka)
Wednesday, 29 September 2010, 11:36 GMT
Reason for closing:  No response
Additional comments about closing:  No response in +1 month.
Comment by Frederic Bezies (fredbezies) - Sunday, 21 March 2010, 12:02 GMT
If I use a wifi connection, it resets itself after about 10 minutes :(
Comment by Mauro Santos (R00KIE) - Monday, 22 March 2010, 16:29 GMT
This seems related with the problem I have, check here http://bbs.archlinux.org/viewtopic.php?id=93450

In my case the problems start when I plug or unplug the notebook's charger and only a cold reboot will fix the problem, this happens with ath5k and madwifi.
I have confirmed that in my case downgrading only kernel26{,-headers,-firmware} with make it work as expected.
Comment by Stan Holenda (texoft) - Sunday, 11 April 2010, 10:01 GMT
Confirm with 2.6.33.2-1 I got this right after system startup:

ath5k phy0: gain calibration timeout (2412MHz)
ath5k phy0: gain calibration timeout (2412MHz)
ath5k phy0: gain calibration timeout (2417MHz)

Last kernel working for me is 2.6.32.8-1. Even this one however sometimes fails after resume (from suspend).
Comment by Stan Holenda (texoft) - Sunday, 11 April 2010, 10:11 GMT Comment by Mauro Santos (R00KIE) - Sunday, 11 April 2010, 13:38 GMT
@Stan Holenda (texoft)
Do you have any cpu frequency scaling support modules loaded? Such as powernow_k8 or acpi_cpufreq.
If you do, try to blacklist them and the cpufreq daemon and try to reproduce the problem.

If things work then report it upstream [1] as this doesn't seem to be a problem specific to arch.

I have also reported this specific problem in our bug tracker [2].

[1] https://bugzilla.kernel.org/show_bug.cgi?id=15693
[2] http://bugs.archlinux.org/task/18982
Comment by Thomas Dziedzic (tomd123) - Monday, 17 May 2010, 04:24 GMT
upstream "needs info"
Comment by Frederic Bezies (fredbezies) - Monday, 17 May 2010, 05:00 GMT
Sorry for the delay.

@mauro : I have the same problem. And for now, the only tweak is to use wicd with a wired connection, which is not ideal. But NetworkManager seems to show this problem faster than wicd.
Comment by Mauro Santos (R00KIE) - Monday, 17 May 2010, 11:16 GMT
Sorry I didn't report back with any useful info as of yet, I've been quite busy lately and didn't have the needed free time to test this, as it requires my intervention (requires me to plug/unplug the charger).

I see kernel 2.6.34 has landed in [testing], that I can test fairly quickly and report back if it works or not. Meanwhile the workaround of using phc-k8 has been working for me.

Edit:
Still the same problem with 2.6.34 in testing. I really need to free a day to bisect the kernel and try to find the offending commit.
Comment by Gerardo Exequiel Pozzi (djgera) - Monday, 09 August 2010, 22:51 GMT
  • Field changed: Status (Assigned → Waiting on Response)
  • Field changed: Category (Packages: Core → Upstream Bugs)
status with 2.6.35?

Loading...