ath9k not working on resume

Bug #441475 reported by David Huggins-Daines
76
This bug affects 12 people
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
Fix Released
Undecided
Unassigned

Bug Description

After suspending and resuming my eeepc 1005ha, the wireless no longer works. When I run "iwlist wlan0 scan" I get:

dhuggins@bluesy:~$ iwlist wlan0 scan
wlan0 Failed to read scan data : Resource temporarily unavailable

Removing and re-inserting the ath9k module makes it work again - scanning works and I can connect to the network.

Is there some way to force this module to be unloaded and reloaded on suspend/resume? Or, I suppose, it would be even better if the driver supported suspend and resume :-)

ProblemType: Bug
AplayDevices:
 **** List of PLAYBACK Hardware Devices ****
 card 0: Intel [HDA Intel], device 0: ALC269 Analog [ALC269 Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
Architecture: i386
ArecordDevices:
 **** List of CAPTURE Hardware Devices ****
 card 0: Intel [HDA Intel], device 0: ALC269 Analog [ALC269 Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: dhuggins 1652 F.... pulseaudio
CRDA: Error: [Errno 2] No such file or directory
Card0.Amixer.info:
 Card hw:0 'Intel'/'HDA Intel at 0xf7db8000 irq 22'
   Mixer name : 'Realtek ALC269'
   Components : 'HDA:10ec0269,10438398,00100004'
   Controls : 14
   Simple ctrls : 9
Date: Sat Oct 3 10:46:55 2009
DistroRelease: Ubuntu 9.10
HibernationDevice: RESUME=UUID=f471126a-a40b-4a27-8bc9-3f595ba87e18
MachineType: ASUSTeK Computer INC. 1005HA
Package: linux-image-2.6.31-11-generic 2.6.31-11.36
ProcCmdLine: BOOT_IMAGE=/boot/vmlinuz-2.6.31-11-generic root=UUID=fff848f3-a294-47cc-8234-fd3aa6de444b ro quiet splash
ProcEnviron:
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-11.36-generic
RelatedPackageVersions:
 linux-backports-modules-2.6.31-11-generic N/A
 linux-firmware 1.19
SourcePackage: linux
Uname: Linux 2.6.31-11-generic i686
dmi.bios.date: 06/18/2009
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: 0601
dmi.board.asset.tag: To Be Filled By O.E.M.
dmi.board.name: 1005HA
dmi.board.vendor: ASUSTeK Computer INC.
dmi.board.version: x.xx
dmi.chassis.asset.tag: 0x00000000
dmi.chassis.type: 10
dmi.chassis.vendor: ASUSTeK Computer INC.
dmi.chassis.version: x.x
dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvr0601:bd06/18/2009:svnASUSTeKComputerINC.:pn1005HA:pvrx.x:rvnASUSTeKComputerINC.:rn1005HA:rvrx.xx:cvnASUSTeKComputerINC.:ct10:cvrx.x:
dmi.product.name: 1005HA
dmi.product.version: x.x
dmi.sys.vendor: ASUSTeK Computer INC.

Revision history for this message
David Huggins-Daines (dhuggins) wrote :
Revision history for this message
David Huggins-Daines (dhuggins) wrote :

sorry for the duplicates of this bug report, folks ... network problem...

Revision history for this message
Venkat Raghavan (venkatraghavan) wrote :

I have the same issue too. I will update to the latest karmic kernel and then report back.

Revision history for this message
Przemek K. (azrael) wrote :

"Is there some way to force this module to be unloaded and reloaded on suspend/resume?"

Yes, there is.
Add "ath9k" to the following line in /etc/default/acpi-support
MODULES=""

so it looks like this:
MODULES="ath9k"

If that doesn't work, you can try adding also the dependencies of this module, namely:
mac80211 cfg80211

or these, if you have linux-backports-modules installed:
lbm_cw_mac80211 lbm_cw_cfg80211

Revision history for this message
David Huggins-Daines (dhuggins) wrote :

Actually, since 2.6.31-13 I'm not seeing this problem anymore. Can others verify this? I haven't looked at the changelogs but ath9k seems to be generally working a lot better...

Revision history for this message
David Huggins-Daines (dhuggins) wrote :

hmm, nope, it's still there (now using backports-modules). anyway the acpi-support workaround seems to do okay.

Revision history for this message
Luarvik (tuvikene) wrote :

The ath9k problem disappears with 2.6.32. With 2.6.31 kernels after suspend the wireless is gone. The module unloading is a workaround on these kernels.

Revision history for this message
Swâmi Petaramesh (swami-petaramesh) wrote :

I confirm this bug (that didn't exist in 9.04 Jaunty) on my Asus EeePC 1002HA since upgrading to 9.10 Karmic.

Wi-Fi hardware is Atheros AR928X

Wi-Fi works, but after resuming (after a suspend to RAM), the Wi-Fi speed drops to a crawl and the connection gets dropped, then reconnects, then drops again...

Manually rmmod'ing then modprobe'ing ath9k fixes the issue everytime, until next suspend/resume...

Adding ath9k to the "MODULES" list in /etc/default/acpi-support has absolutely no positive influence here (although I haven't tried to add the dependancies as well, yet)

Revision history for this message
^_Pepe_^ (jose-angel-fernandez-freire) wrote :

Please, can we mark this bug as Fix Commited?

Thanks

Revision history for this message
Swâmi Petaramesh (swami-petaramesh) wrote :

« Please, can we mark this bug as Fix Commited? »

Definitely not as far as I'm concerned. I haven't observed any visible improvement in ath9k with respect to suspend/resume after having applied all the latest (as of today) released Karmic kernel updates.

Still need to rmmod/modprobe it to get Wi-Fi to work properly at decent speed.

Revision history for this message
^_Pepe_^ (jose-angel-fernandez-freire) wrote :

Hi Swâmi,

Some reporters say that 2.6.32 kernel solved the problem. So, that's why my question is "Commited" but not "Released".

Please, can you try with that kernel? You can find it here: http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.32/

Regards

Revision history for this message
Dionysius Almeida (wilsonalmeida) wrote :

 A small observation since i filed bug #388290, the current driver works fine for me on compaq v2000 series.
The only exception is that the lights on my pci card doen't work like they used to be. Earlier the rx and tx lights used to blink to show activity. Now I see only one steady light on either the tx or rx led ( since they are not labeled, i cannot say which one.)

Revision history for this message
Dionysius Almeida (wilsonalmeida) wrote :

FYI: My pci card is D-Link DWL-G650.

Revision history for this message
Swâmi Petaramesh (swami-petaramesh) wrote :

Hi ^_Pepe_^, I have installed :

# uname -r
2.6.32-020632-generic

At first sight everything seems to work good. I have suspended my laptop to RAM, and after wake up, Wi-Fi seems to work perfectly now.

My machine is: Asus EeePC 1002HA, with Atheros Communications Inc. AR928X Wireless Network Adapter (PCI-Express), using ath9k.

Revision history for this message
Swâmi Petaramesh (swami-petaramesh) wrote :

Important addendum : WARNING : With 2.6.32-020632-generic my screen starts "flashing and flickering" from time to time in X11 (Intel driver), and my machine just happened to lock completely (which never happened before I loaded this kernel).

This screen flickering happened in the past with some broken kernel/X11 combination (in Intrepid if I remember well), so I believe there's a serious issue with Intel X11 and this kernel version, and it should not be released.

Revision history for this message
^_Pepe_^ (jose-angel-fernandez-freire) wrote :

Swâmi,

Thanks for your comments.

I'm going to mark the wifi issue as Commited, and I encourage you to add a new bug regarding this flickering behaviour.

Thanks

Changed in linux (Ubuntu):
status: New → Fix Committed
Revision history for this message
madbiologist (me-again) wrote :

Linux kernel 2.6.33-rc3 has just been released. The release notes at http://lkml.org/lkml/2010/1/5/512 include the following comment:

ath9k: fix suspend by waking device prior to stop

Based on past kernel releases I would expect between 3 and 5 more RC's (at approximately 1-week intervals) until 2.6.33 final. Then we just need someone to make a PPA for Karmic, followed later by a PPA for Lucid when it is released.

Revision history for this message
Artem Litovskikh (artemios-) wrote :

UNR 9.10 wi-fi working fine after resume from standby.
I also have installed this:
•Wireless (Atheros AR9285) works out of the box, but connection is flaky. To fix, open a terminal and type 'sudo apt-get install linux-backports-modules-karmic'
from https://wiki.ubuntu.com/HardwareSupport/Machines/Netbooks#Asus Eee PC 1005HA

Revision history for this message
com_h (patsjunk) wrote :

Since doing a fresh install of 10.04 (Previously using 9.10 and bug drove Me nuts), this bug seems to have gone.

I resume on my ASUS N10J and its Atheros card doesn't miss a beat.

Hope this helps.

Revision history for this message
Carl Beech (carl-beech) wrote : Re: [Bug 441475] Re: ath9k not working on resume

On 12/05/10 22:16, com_h wrote:
> Since doing a fresh install of 10.04 (Previously using 9.10 and bug
> drove Me nuts), this bug seems to have gone.
>
> I resume on my ASUS N10J and its Atheros card doesn't miss a beat.
>
> Hope this helps.
>
>
Hi,

Yes, it's much improved, however, I've still gone back to using
ndiswrapper - the ath9k does fail if you move large(r) files about e.g.
100Mb+.... we have a NAS with films on it and watch them from that
rather than getting DVD's out of the cupboard!

Carl.

Revision history for this message
martinbures (needemesleepe) wrote :

I am having this problem. On resume, I am unable to reconnect.

Linux martin-laptop 2.6.32-22-generic #33-Ubuntu SMP Wed Apr 28 13:28:05 UTC 2010 x86_64 GNU/Linux

Revision history for this message
martinbures (needemesleepe) wrote :

Following suspend or sleep, I am able to reconnect by doing:
rmmod ath9k

and then:
modprobe ath9k

relevant output from dmesg:

[36462.458868] ath9k: Unable to reset channel (5825 Mhz) reset status -5
[36462.458873] ath9k: Unable to set channel
[36462.585143] ath9k: timeout (100000 us) on reg 0x9860: 0x00049d19 & 0x00000001 != 0x00000000
[36462.585147] ath9k: Unable to reset channel (2437 Mhz) reset status -5
[36462.585172] ath9k: Unable to set channel
[36484.023808] ath9k 0000:0b:00.0: PCI INT A disabled
[36484.023883] ath9k: Driver unloaded
[36489.629394] ath9k 0000:0b:00.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
[36489.629431] ath9k 0000:0b:00.0: setting latency timer to 64
[36489.763840] ath: EEPROM regdomain: 0x64
[36489.763843] ath: EEPROM indicates we should expect a direct regpair map
[36489.763846] ath: Country alpha2 being used: 00
[36489.763848] ath: Regpair used: 0x64
[36489.764822] phy1: Selected rate control algorithm 'ath9k_rate_control'
[36489.765679] Registered led device: ath9k-phy1::radio
[36489.765696] Registered led device: ath9k-phy1::assoc
[36489.766322] Registered led device: ath9k-phy1::tx
[36489.766507] Registered led device: ath9k-phy1::rx
[36489.766518] phy1: Atheros AR5418 MAC/BB Rev:2 AR5133 RF Rev:81: mem=0xffffc900050c0000, irq=16
[36489.814925] ADDRCONF(NETDEV_UP): wlan0: link is not ready
[36503.115472] wlan0: deauthenticating from 00:16:b6:dd:26:db by local choice (reason=3)
[36503.115964] wlan0: direct probe to AP 00:16:b6:dd:26:db (try 1)
[36503.119110] wlan0: direct probe responded
[36503.119119] wlan0: authenticate with AP 00:16:b6:dd:26:db (try 1)
[36503.121112] wlan0: authenticated
[36503.121127] wlan0: associate with AP 00:16:b6:dd:26:db (try 1)
[36503.123600] wlan0: RX AssocResp from 00:16:b6:dd:26:db (capab=0x411 status=0 aid=2)
[36503.123603] wlan0: associated
[36503.124146] ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready
[36513.952615] wlan0: no IPv6 routers present

Revision history for this message
martinbures (needemesleepe) wrote :

The above fix in /etc/default/acpi-support works for me for suspend. However when the laptop goes to sleep, I must manually load and unload the module. Is there a way that I can fix this also?

Revision history for this message
Renate (yollywau) wrote :

uname -r
2.6.35-5.slh.4-aptosid-686
The above fix in /etc/devault/acpi-support doesn't work for me.

modprobe -r ath9k

root@siduxbox:/home/user# modprobe ath9k
root@siduxbox:/home/user# ifup wlan0
wpa_supplicant: /sbin/wpa_supplicant daemon failed to start
run-parts: /etc/network/if-pre-up.d/wpasupplicant exited with return code 1
Internet Systems Consortium DHCP Client 4.1.1-P1
Copyright 2004-2010 Internet Systems Consortium.
All rights reserved.
For info, please visit https://www.isc.org/software/dhcp/

SIOCSIFFLAGS: Input/output error
Listening on LPF/wlan0/94:0c:6d:c4:a6:6c
Sending on LPF/wlan0/94:0c:6d:c4:a6:6c
Sending on Socket/fallback
DHCPDISCOVER on wlan0 to 255.255.255.255 port 67 interval 7
DHCPDISCOVER on wlan0 to 255.255.255.255 port 67 interval 13
DHCPDISCOVER on wlan0 to 255.255.255.255 port 67 interval 20
DHCPDISCOVER on wlan0 to 255.255.255.255 port 67 interval 15
DHCPDISCOVER on wlan0 to 255.255.255.255 port 67 interval 6
No DHCPOFFERS received.
No working leases in persistent database - sleeping.

The "route" command doesn't find the fritz.box any more.....

wlan0: no IPv6 routers present

/etc/init.d/networking restart
Running /etc/init.d/networking restart is deprecated because it may not enable again some interfaces ... (warning).
Reconfiguring network interfaces...Internet Systems Consortium DHCP Client 4.1.1-P1
Copyright 2004-2010 Internet Systems Consortium.
All rights reserved.
For info, please visit https://www.isc.org/software/dhcp/

Listening on LPF/wlan0/94:0c:6d:c4:a6:6c
Sending on LPF/wlan0/94:0c:6d:c4:a6:6c
Sending on Socket/fallback
DHCPRELEASE on wlan0 to 192.168.178.1 port 67
send_packet: Network is unreachable
send_packet: please consult README file regarding broadcast address.
done.

Changed in linux (Ubuntu):
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.