ath9k driver not working properly with AR9285

[Franck]

The ath9k wifi driver provided by linux-backports-modules-2.6.31-19-generic does not work properly: frequent desassociation and packet loss make the connection hardly usable.

The connection is frequently desassociated, as seen in the extract of syslog attached as syslog.txt.
Lot of packets are lost (~30%) as seen in attached ping.txt.
iwconfig.txt attached also.

The version ath9k is seen in attached modinfo.txt. (srcversion of the module is 918698AEA3CBE61D0E996E6)
I am using network-manager.
---
AlsaVersion: Advanced Linux Sound Architecture Driver Version 1.0.21.
AplayDevices:
 **** List of PLAYBACK Hardware Devices ****
 card 0: Intel [HDA Intel], device 0: ALC262 Analog [ALC262 Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
Architecture: amd64
ArecordDevices:
 **** List of CAPTURE Hardware Devices ****
 card 0: Intel [HDA Intel], device 0: ALC262 Analog [ALC262 Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: franck 2515 F.... pulseaudio
CRDA: Error: [Errno 2] Aucun fichier ou dossier de ce type
Card0.Amixer.info:
 Card hw:0 'Intel'/'HDA Intel at 0xe8e00000 irq 22'
   Mixer name : 'Realtek ALC262'
   Components : 'HDA:10ec0262,104d4700,00100302'
   Controls : 18
   Simple ctrls : 11
Card1.Amixer.info:
 Card hw:1 'NVidia'/'HDA NVidia at 0xe3080000 irq 16'
   Mixer name : 'Nvidia ID a'
   Components : 'HDA:10de000a,104d4700,00100100'
   Controls : 0
   Simple ctrls : 0
Card1.Amixer.values:

DistroRelease: Ubuntu 10.04
EcryptfsInUse: Yes
HibernationDevice: RESUME=UUID=33636b1a-e821-4c6a-9d67-39617d0b9ef8
InstallationMedia: Ubuntu 9.10 "Karmic Koala" - Release amd64 (20091027)
IwConfig:
 lo no wireless extensions.

 eth0 no wireless extensions.

 vboxnet0 no wireless extensions.
MachineType: Sony Corporation VPCCW2S1E
NonfreeKernelModules: nvidia
Package: linux (not installed)
ProcCmdLine: BOOT_IMAGE=/vmlinuz-2.6.32-22-generic root=UUID=3b400915-2757-43b4-8047-2e1ac0f50751 ro quiet splash
ProcEnviron:
 LANG=fr_FR.utf8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.32-22.36-generic 2.6.32.11+drm33.2
Regression: No
RelatedPackageVersions: linux-firmware 1.34
Reproducible: Yes
RfKill:
 0: sony-wwan: Wireless WAN
  Soft blocked: no
  Hard blocked: yes
Tags: lucid networking needs-upstream-testing
Uname: Linux 2.6.32-22-generic x86_64
UserGroups: adm admin cdrom dialout lpadmin netdev plugdev sambashare
dmi.bios.date: 11/30/2009
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: R0110Y7
dmi.board.asset.tag: N/A
dmi.board.name: VAIO
dmi.board.vendor: Sony Corporation
dmi.board.version: N/A
dmi.chassis.asset.tag: N/A
dmi.chassis.type: 10
dmi.chassis.vendor: Sony Corporation
dmi.chassis.version: N/A
dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvrR0110Y7:bd11/30/2009:svnSonyCorporation:pnVPCCW2S1E:pvrC603K089:rvnSonyCorporation:rnVAIO:rvrN/A:cvnSonyCorporation:ct10:cvrN/A:
dmi.product.name: VPCCW2S1E
dmi.product.version: C603K089
dmi.sys.vendor: Sony Corporation

---
Architecture: amd64
DistroRelease: Ubuntu 10.04
EcryptfsInUse: Yes
InstallationMedia: Ubuntu 9.10 "Karmic Koala" - Release amd64 (20091027)
Package: linux (not installed)
ProcEnviron:
 LANG=fr_FR.utf8
 SHELL=/bin/bash
Tags: lucid
Uname: Linux 2.6.35-999-generic x86_64
UnreportableReason: The running kernel is not an Ubuntu kernel
UserGroups: adm admin cdrom dialout lpadmin netdev plugdev sambashare

[Rolf Leggewie (r0lf)]

Franck, please try one of the mainline kernels (https://wiki.ubuntu.com/KernelTeam/MainlineBuilds) and report the problem to http://bugzilla.kernel.org Please let us know about the ticket number here. That is your fastest way to get a fix for the problem.

[Franck Routier (franck-routier)]

I tried latest kernel in MailineBuils, and the bug is still here...

So I opened a bug report on bugzilla.kernel.org, with number 15313
(http://bugzilla.kernel.org/show_bug.cgi?id=15313)

[Rolf Leggewie (r0lf)]

great, thank you. Adding upstream ticket so we can properly follow the status from here.

[Archish (archish)]

Hi Franck try the 2.6.32 kernel. It should work without disconnection and also the signal strength should improve.

[Franck Routier (franck-routier)]

@Archish : I have tried the MainlineKernel, which was 2.6.33-rc7 at the moment, and my connection did not improve. I guess 2.6.32 won't be better

[Franck (alci)]

I think my bug might well be a duplicate of #373680 and #460886

I didn't get the point, whether this is a NetworkManager bug, or an ath9k bug, or both combined. There were several assertions saying 'it should be fixed with next kernel release', but as to now (2010-04-07), on lucid beta, using linux-backports-modules-wireless-2.6.32-19.9, the problem persists for me.

[Franck (alci)]

I tried to connect via a manual definition in /etc/network/interfaces and via wicd instead of network-manager, but the problem was still there...

[DarkStang (darkstang)]

I just bought a new Sony VPCCW21FX laptop with this wireless card in it. Running the 2.6.32 kernel I was experience the same issues. I have also compiled and installed the latest ath9k driver from linuxwireless.org, and I'm still having issues. Everything will be working for for a few seconds, then I'll have several seconds of lag, then everything will work for another minute, then 20 seconds of lag, disconnections, etc.

However the odd thing is this. If I have an application open that is constantly sending/receiving data (torrent application, for example) everything works fine. Even if I throttle my down/up speeds down to 10KB/s, that constant connection keeps everything else alive. I hope this helps. If anybody wants me to test anything let me know. I'm a software engineering student so I'm at least slightly competent.

[Franck (alci)]

@DarkStang : maybe if you could comment on upstream bug (http://bugzilla.kernel.org/show_bug.cgi?id=15313) this would be helpful.
Until now upstream has been quite unhelpful (the only comment I got was because I gave a wrong kernel version...), but your input could make the things move on.

[DarkStang (darkstang)]

Posted there... I've never done any work with driver development before. However after graduation (3 weeks), if this still isn't fixed, I might start looking into it. Until then I'm just going to leave qBittorrent open and keep doing my school work. If you find anything new post it. Maybe somebody who knows about network card drivers will get it figured out... hell, I'll buy somebody one of these cards and a case of beer if they make drivers for it that work 100%.

[DarkStang (darkstang)]

I have uploaded these two videos to help people understand what's happening... They are videos of me running speed tests, with Gnome's system monitor opened up so you can view the network activity history.

This first test is without a bittorrent application open. Notice how the connection comes and goes in random spurts. Anywhere from absolutely no data being received, to over 1MB/s.
http://darkstang.com/vids/no-qb-test.ogv

Now this next video is with qbittorrent open, and I'm downloading/seeding some random Ubuntu .iso's. The connection stays solid the entire time.
http://darkstang.com/vids/with-qb-test.ogv

I feel as though I should also mention, there never appears to be an issue with my laptop sending data, only receiving data. I've noticed this several times during file sharing with other computers on the network and gaming with other computers online or on the network.

[[e-mail] (cfraz89-gmail)]

This bug hits me too, AR9285 rev01.
I've had Kubuntu 10.04 RC on here, experienced packet loss.
Arch with 2.6.33 kernel, there was no packet loss, however if traffic got heavy experienced a kernel panic.
Arch 2.6.34rc5, back to packet loss.
Tried installing compat-wireless, no help.
However have noticed that, as I have an xp driver loaded into ndiswrapper (but wont connect to any ap's) if i modprobe ndiswrapper then unload it, connection will stabilise. That's how I work with my card now. Not ideal but saves me from buying an external wifi card :/

[DarkStang (darkstang)]

Okay, so I think I have found the issue. It has nothing to do with the ath9k drivers, it appear to have everything to do with the stock kernel that came with my Linux distro. How that works, I don't know. After finals today I had some time to kill so I downloaded and compiled the 2.6.33 kernel. After some initial tests everything seems to be working fine. I will continue to use this over the next week to see how it goes.

[Franck (alci)]

Do you mean that using the stock kernel compiled at https://wiki.ubuntu.com/KernelTeam/MainlineBuilds, the bug is still here, but downloading a stock kernel and compiling it by yourself, the bug gone ?

[DarkStang (darkstang)]

I meant the generic kernels that Ubuntu provides (2.6.31-32) are the issue. I did not try to download and compile a kernel provided by Ubuntu. Instead I downloaded the 2.6.33.3 Stable kernel from http://kernel.org/ . I used the same .config file from the previous kernel but turned on some logging features for the ath9k driver. I imagine the 2.6.34 RC would do just as well. It's day 2 and the issue still hasn't shown up. Several speed tests show consistent tests. Today I've been playing around streaming music from other machines to this one, and from this one to my PS3. Ping tests have shown 0 packets lost out of hundreds sent. Try downloading and compiling the 2.6.33 kernel and see how it goes for you Franck. I am quite pleased with my new machine now.

[Franck (alci)]

I have tried the 2.6.33.3 and 2.6.34-rc7 mainline kernel packaged at https://wiki.ubuntu.com/KernelTeam/MainlineBuilds and still have the problem... (45% packet loss on ping).

It's been years since I have not compiled a kernel by myself, but I will try it to see if it is better (but I doubt it, as MainLine kernel package should be just that : mainline unpatched kernel compiled and packaged as deb).

I'll post my results ASAP.

[DarkStang (darkstang)]

Last day of finals for me so I'll have way more free time. If that doesn't work for you let me know and I'll post the .deb packages and the ramdisk image that I made.

[DarkStang (darkstang)]

Here is the 64bit kernel and headers that I have compiled for myself, as .deb packages. The ramdisk image is also in there. It was made on a Sony vpccw21fx laptop.

[Franck (alci)]

@DarkStang : where is "here" ? :-)

[DarkStang (darkstang)]

Wow, I spaced out I guess... here.
http://darkstang.com/vpccw21fx/ds-kernel.tar.bz2

[Steven Barrett (damentz)]

I can confirm major packet loss while running 2.6.34, with and without wireless-testing from linville's repository on git.kernel.org. It's even worse on 2.6.34; loading web pages works momentarily after associating with an access point, then it stops entirely until you reassociate using network manager.

[Jeremy Foshee (jeremyfoshee)]

Hi Franck,

This bug was reported a while ago and there hasn't been any activity in it recently. We were wondering if this is still an issue? Can you try with the latest development release of Ubuntu? ISO CD images are available from http://cdimage.ubuntu.com/releases/ .

If it remains an issue, could you run the following command from a Terminal (Applications->Accessories->Terminal). It will automatically gather and attach updated debug information to this report.

apport-collect -p linux 518818

Also, if you could test the latest upstream kernel available that would be great. It will allow additional upstream developers to examine the issue. Refer to https://wiki.ubuntu.com/KernelMainlineBuilds . Once you've tested the upstream kernel, please remove the 'needs-upstream-testing' tag. This can be done by clicking on the yellow pencil icon next to the tag located at the bottom of the bug description and deleting the 'needs-upstream-testing' text. Please let us know your results.

Thanks in advance.

    [This is an automated message. Apologies if it has reached you inappropriately; please just reply to this message indicating so.]

[Franck (alci)]

This bug is still an issue. Testing with latest KernelMainlineBuilds (which appears to be 2.6.35-rc3), the problem is still there (around 40% packet loss when pinging a machine next to me on the network).
The bug is reported upstream, but does not get much love... and I don't know hos to help on this.

Testing with Maverick alpha1 fails on my machine (probably due to another bug where nv driver is unable to get the EDID information out of my laptop screen... but this is another subject.).

[Franck (alci)]

apport information

[Franck (alci)]

apport information

[Franck (alci)]

apport information

[Franck (alci)]

apport information

[Franck (alci)]

apport information

[Franck (alci)]

apport information

[Franck (alci)]

apport information

[Franck (alci)]

apport information

[Franck (alci)]

apport information

[Franck (alci)]

apport information

[Franck (alci)]

apport information

[Franck (alci)]

apport information

[Franck (alci)]

apport information

[Franck (alci)]

apport information

[Franck (alci)]

apport information

[Franck (alci)]

apport information

[Franck (alci)]

apport information

[Franck (alci)]

apport information

[Franck (alci)]

I ran apport twice, once with current Lucid Kernel, once with KernelMainlineBuild kernel (2.6.35.999).

Hope this helps.

[Andres Moreira (elkpichico)]

I can confirm this bug on Ubuntu 10.04. I've tried:

 * ath9k driver shiped with the kernel
 * compat-wireless from linux-backports
 * compat-wireless from linuxwireless.org (stable and bleeding edge snapshot)

and I've the same problem with each one, poor strength signal (more than 5 o 6 meters, it starts disconnecting frequently, terrible annoying), and also, disconnects on high network load (streaming HD over youtube for example).

Also, with the new kernel
 * kernel 2.6.35 (from maverick - Ubuntu kernel team PPA)
 * compat-wireless (from backports and linuxwireless.org)
the sitatuion were WORST!

Actually I'm using Ndiswrapper, it works much better and much reliable, but sucks to use ndiswrapper instead of native ones.

Information of my machine:

$ uname -a
Linux XXXXXX 2.6.32-24-generic #42-Ubuntu SMP Fri Aug 20 14:21:58 UTC 2010 x86_64 GNU/Linux

$ lspci -nn | grep Atheros
02:00.0 Network controller [0280]: Atheros Communications Inc. AR9285 Wireless Network Adapter (PCI-Express) [168c:002b] (rev 01)

$ lspci -vv
....
02:00.0 Network controller: Atheros Communications Inc. AR9285 Wireless Network Adapter (PCI-Express) (rev 01)
 Subsystem: Hewlett-Packard Company Device 303f
 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr+ Stepping- SERR+ FastB2B- DisINTx-
 Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
 Latency: 0, Cache Line Size: 64 bytes
 Interrupt: pin A routed to IRQ 16
 Region 0: Memory at d3400000 (64-bit, non-prefetchable) [size=64K]
 Capabilities: <access denied>
 Kernel driver in use: ndiswrapper
 Kernel modules: ath9k
...

$ lshw

.......
      *-pci:0
             description: PCI bridge
             product: 5 Series/3400 Series Chipset PCI Express Root Port 1
             vendor: Intel Corporation
             physical id: 1c
             bus info: pci@0000:00:1c.0
             version: 05
             width: 32 bits
             clock: 33MHz
             capabilities: pci bus_master cap_list
             configuration: driver=pcieport
             resources: irq:29 ioport:3000(size=4096) memory:d3400000-d43fffff ioport:d0400000(size=16777216)
           *-network
                description: Wireless interface
                product: AR9285 Wireless Network Adapter (PCI-Express)
                vendor: Atheros Communications Inc.
                physical id: 0
                bus info: pci@0000:02:00.0
                logical name: wlan0
                version: 01
                serial: 78:e4:00:05:3c:69
                width: 64 bits
                clock: 33MHz
                capabilities: bus_master cap_list ethernet physical wireless
                configuration: broadcast=yes driver=ndiswrapper+netathwx driverversion=1.55+,12/14/2009,7.7.0.456 ip=192.168.2.157 latency=0 multicast=yes wireless=IEEE 802.11g
                resources: irq:16 memory:d3400000-d340ffff

....

[Gavin Graham (gavingraham)]

Also affected by this bug. uploading apport info now

[Gavin Graham (gavingraham)]

AlsaVersion: Advanced Linux Sound Architecture Driver Version 1.0.23.
Architecture: amd64
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: gavin 1752 F.... pulseaudio
 /dev/snd/pcmC0D0p: gavin 1752 F...m pulseaudio
CRDA: Error: [Errno 2] No such file or directory
Card0.Amixer.info:
 Card hw:0 'Intel'/'HDA Intel at 0xfcb00000 irq 46'
   Mixer name : 'Realtek ALC269'
   Components : 'HDA:10ec0269,144dc06a,00100004'
   Controls : 11
   Simple ctrls : 7
Card1.Amixer.info:
 Card hw:1 'NVidia'/'HDA NVidia at 0xddefc000 irq 16'
   Mixer name : 'Nvidia GT220 HDMI'
   Components : 'HDA:10de000a,10de0101,00100100'
   Controls : 16
   Simple ctrls : 4
DistroRelease: Ubuntu 10.10
HibernationDevice: RESUME=UUID=8f03002f-5adb-4ea8-9797-81165d94d5d4
InstallationMedia: Ubuntu 10.04 LTS "Lucid Lynx" - Release amd64 (20100429)
MachineType: SAMSUNG ELECTRONICS CO., LTD. R580/R590
NonfreeKernelModules: nvidia
Package: linux 2.6.35.20.21
PackageArchitecture: amd64
ProcCmdLine: BOOT_IMAGE=/boot/vmlinuz-2.6.35-20-generic root=UUID=47c93ce9-d27e-402d-8e8f-690710447631 ro quiet splash
ProcEnviron:
 PATH=(custom, user)
 LANG=en_AU.utf8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.35-20.29-generic 2.6.35.4
Regression: Yes
RelatedPackageVersions: linux-firmware 1.38
Reproducible: Yes
Tags: maverick kernel-net regression-potential needs-upstream-testing
Uname: Linux 2.6.35-20-generic x86_64
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare
WifiSyslog:
 Sep 11 06:21:28 gavin-laptop kernel: [ 3533.380757] lo: Disabled Privacy Extensions
 Sep 11 06:24:52 gavin-laptop kernel: [ 3737.082207] lo: Disabled Privacy Extensions
dmi.bios.date: 06/22/2010
dmi.bios.vendor: Phoenix Technologies Ltd.
dmi.bios.version: 11JB.M044.20100622.hkk
dmi.board.asset.tag: Tag 12345
dmi.board.name: R580/R590
dmi.board.vendor: SAMSUNG ELECTRONICS CO., LTD.
dmi.board.version: Not Applicable
dmi.chassis.asset.tag: No Asset Tag
dmi.chassis.type: 9
dmi.chassis.vendor: SAMSUNG ELECTRONICS CO., LTD.
dmi.chassis.version: N/A
dmi.modalias: dmi:bvnPhoenixTechnologiesLtd.:bvr11JB.M044.20100622.hkk:bd06/22/2010:svnSAMSUNGELECTRONICSCO.,LTD.:pnR580/R590:pvrNotApplicable:rvnSAMSUNGELECTRONICSCO.,LTD.:rnR580/R590:rvrNotApplicable:cvnSAMSUNGELECTRONICSCO.,LTD.:ct9:cvrN/A:
dmi.product.name: R580/R590
dmi.product.version: Not Applicable
dmi.sys.vendor: SAMSUNG ELECTRONICS CO., LTD.

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

apport information

[Gavin Graham (gavingraham)]

I've just downloaded and installed the ath9k module from 'compat-wireless-2010-09-12' and it seems after 30min of testing that my problem may be resolved. I will report back after more testing.

[Franck (alci)]

I can confirm that 'compat-wireless-2010-09-12' seems to fix the issue here too (I'm the initial bug reporter).

Is there a chance that linux-backports-modules-wireless-maverick-generic shipping with Ubuntu 10.10 final gets updated to this version ?

[Gavin Graham (gavingraham)]

Well nearing 24hours later and 'compat-wireless-2010-09-12' has certainly fixed my wifi dropouts. I agree with Franck, Is there a chance that linux-backports-modules-wireless-maverick-generic shipping with Ubuntu 10.10 final gets updated to this version ?

[claudios (shatten69)]

After reading your last comments I tried the compat-wireless-2010-09-12 but had no change, still affected.
Would you mind to tell me what kernel are you using?
Thank you

Sony vaio VPCEB1J8E
2.6.35-22-generic-pae #33~lucid1-Ubuntu SMP Sat Sep 18 09:04:16 UTC 2010 i686 GNU/Linux
02:00.0 Network controller [0280]: Atheros Communications Inc. AR9285 Wireless Network Adapter (PCI-Express) [168c:002b] (rev 01)

[Franck (alci)]

I am using latest Maverick kernel, ie 2.6.35-22-generic.

I removed linux-backports-modules-wireless package (ie compat-wireless as packaged in Ubuntu). Then I have downloaded compat-wireless-2010-09-12 from http://www.orbit-lab.org/kernel/compat-wireless-2.6/.

Then I did :

./scripts/driver-select ath9k

(as some parts of compat wireless were not compiling, but ath9k works just fine).

then make, sudo make install.

modinfo ath9k gives me :

filename: /lib/modules/2.6.35-22-generic/updates/drivers/net/wireless/ath/ath9k/ath9k.ko
license: Dual BSD/GPL
description: Support for Atheros 802.11n wireless LAN cards.
author: Atheros Communications
srcversion: 8767105B9B4AB361D41BF33
alias: pci:v0000168Cd00000030sv*sd*bc*sc*i*
alias: pci:v0000168Cd0000002Esv*sd*bc*sc*i*
alias: pci:v0000168Cd0000002Dsv*sd*bc*sc*i*
alias: pci:v0000168Cd0000002Csv*sd*bc*sc*i*
alias: pci:v0000168Cd0000002Bsv*sd*bc*sc*i*
alias: pci:v0000168Cd0000002Asv*sd*bc*sc*i*
alias: pci:v0000168Cd00000029sv*sd*bc*sc*i*
alias: pci:v0000168Cd00000027sv*sd*bc*sc*i*
alias: pci:v0000168Cd00000024sv*sd*bc*sc*i*
alias: pci:v0000168Cd00000023sv*sd*bc*sc*i*
depends: ath9k_hw,mac80211,led-class,ath9k_common,ath,cfg80211
vermagic: 2.6.35-22-generic SMP mod_unload modversions
parm: debug:Debugging mask (uint)
parm: nohwcrypt:Disable hardware encryption (int)
parm: blink:Enable LED blink on activity (int)

And I confirm that the issue is solved here...

[Ahmed Aboul Ela (abol3ela)]

I installed the compat-wireless-2.6.tar.bz2. Though my internet speed returned to normal I frequently get crashes with blank screen ( cannot type anything on it).

This happened after i installed the driver.

[Bryan Moore (moore-bryan)]

This is not fixed... I'm still having both connection drops and inconsistent connections.

[Pablo Quirós (polmac1985)]

It's isn't fixed for me either.

[chkneater (chkneater)]

I have been experiencing a similar but, am using an athlon 64M 3800+ and using a belkin usb card. Previous versions I never had a problem with until I did a kernel upgrade. My last ditch to fix the problem was installing WicD. It experienced the same problems until I altered one of the WicDs' conf. files so it wouldn't load the Dbus. This seems to be working fine, I've had a solid connection for over an hour. Don't know why that worked, but wudevvea.

[Gavin Graham (gavingraham)]

I have previously downloaded (starting at comment #52) compat-wireless-2010-09-12 from http://www.orbit-lab.org/kernel/compat-wireless-2.6/ and reported success with my ath9k NIC and several others have done also with the same result.

What I have noticed since is that while this is working fine, my Samsung R580 laptop has been on A/C power. As soon as I run from battery (and therefore some power-saving comes into action), I get the same result loss of connectivity as I had before I thought it was "fixed".

I am now trying compat-wireless-2010-11-13.tar.bz2 and I will run from battery instead of A/C to see what happens. Stay tuned.

[malclocke (malc-wholemeal)]

I'm also getting this, random connection drops and generally intermittent wireless connectivity. I'm suspecting some interaction with power management is the issue, as I get random connection drops when running with power, but not when running from battery. I.e. the opposite of Gavindi in #86.

Also potentially of interest, sometimes my connection gets hung indefinitely, but if I run mtr to a known IP outside my network the connection starts working again and continues to do so until I kill mtr, at which point the connection hangs again.

Maverick on an Asus U43JC
Linux rudy 2.6.35-22-generic #35-Ubuntu SMP Sat Oct 16 20:45:36 UTC 2010 x86_64 GNU/Linux
03:00.0 Network controller: Atheros Communications Inc. AR9285 Wireless Network Adapter (PCI-Express) (rev 01)

[Gavin Graham (gavingraham)]

I don't agree with the justification but the upstream linux kernel bug (#15313) has been closed.

[Pablo Quirós (polmac1985)]

Latest kernel update made the situation worse for me. Now I can't even use my wireless with the trick in #81.

[Bryan Moore (moore-bryan)]

Compiling compat-wireless bleeding edge makes things more endurable, but does not appear to be a perfect solution.

[Denis Fateyev (dfateyev)]

On Ubuntu 10.10 x64 (2.6.35-24-generic #42-Ubuntu), works fine with compat-wireless-2010-12-26 from http://www.orbit-lab.org/kernel/compat-wireless-2.6/ But, certainly, this is not a solution.

[Bryan Moore (moore-bryan)]

It is unfortunate to see this is still happening. I'm forced to compile the bleeding-edge compat-wireless regularly to see if it's fixed and each time--to my chagrin--it is not. How, exactly, is this not a priority for developers?

[Vladislav Kononenko (vladislavkononenko)]

Disconnection occurs regularly. Connect to access point without having to reboot the laptop fails. It seems that disconnection provoked a lot of traffic (torrent?)
From time to time kernel panic happens:

Jan 8 22:19:05 vk-laptop kernel: [ 107.526002] ------------[ cut here ]------------
Jan 8 22:19:05 vk-laptop kernel: [ 107.526026] WARNING: at /build/buildd/linux-backports-modules-2.6.35-2.6.35/debian/build/build-generic/compat-wireless-2.6.36/drive
rs/net/wireless/ath/ath9k/hw.c:2542 ath9k_hw_gettsf64+0x78/0x90 [ath9k_hw]()
Jan 8 22:19:05 vk-laptop kernel: [ 107.526031] Hardware name: K61IC
Jan 8 22:19:05 vk-laptop kernel: [ 107.526033] Modules linked in: cryptd aes_x86_64 aes_generic binfmt_misc parport_pc ppdev snd_hda_codec_nvhdmi snd_hda_codec_realte
k nvidia(P) snd_hda_intel snd_hda_codec arc4 snd_hwdep joydev snd_pcm ath9k mac80211 snd_seq_midi ath9k_common ath9k_hw ath psmouse snd_rawmidi snd_seq_midi_event asus_
laptop cfg80211 video sparse_keymap i2c_nforce2 snd_seq snd_timer snd_seq_device uvcvideo videodev v4l1_compat v4l2_compat_ioctl32 led_class serio_raw snd output soundc
ore snd_page_alloc lp parport r8169 mii ahci libahci
Jan 8 22:19:05 vk-laptop kernel: [ 107.526081] Pid: 809, comm: phy0 Tainted: P 2.6.35-24-generic #42-Ubuntu
Jan 8 22:19:05 vk-laptop kernel: [ 107.526084] Call Trace:
Jan 8 22:19:05 vk-laptop kernel: [ 107.526094] [<ffffffff8106089f>] warn_slowpath_common+0x7f/0xc0
Jan 8 22:19:05 vk-laptop kernel: [ 107.526099] [<ffffffff810608fa>] warn_slowpath_null+0x1a/0x20
Jan 8 22:19:05 vk-laptop kernel: [ 107.526108] [<ffffffffa01dade8>] ath9k_hw_gettsf64+0x78/0x90 [ath9k_hw]
Jan 8 22:19:05 vk-laptop kernel: [ 107.526115] [<ffffffffa0296403>] ath_beacon_config+0x2c3/0x430 [ath9k]
Jan 8 22:19:05 vk-laptop kernel: [ 107.526122] [<ffffffffa029a454>] ath_set_channel+0x194/0x1c0 [ath9k]
Jan 8 22:19:05 vk-laptop kernel: [ 107.526129] [<ffffffffa029af69>] ath9k_config+0x1a9/0x3c0 [ath9k]
Jan 8 22:19:05 vk-laptop kernel: [ 107.526141] [<ffffffffa0241ab2>] ieee80211_hw_config+0xd2/0x140 [mac80211]
Jan 8 22:19:05 vk-laptop kernel: [ 107.526152] [<ffffffffa0246695>] ieee80211_scan_work+0x205/0x560 [mac80211]
Jan 8 22:19:05 vk-laptop kernel: [ 107.526162] [<ffffffffa0246490>] ? ieee80211_scan_work+0x0/0x560 [mac80211]
Jan 8 22:19:05 vk-laptop kernel: [ 107.526168] [<ffffffff8107a885>] run_workqueue+0xc5/0x1a0
Jan 8 22:19:05 vk-laptop kernel: [ 107.526172] [<ffffffff8107aa03>] worker_thread+0xa3/0x110
Jan 8 22:19:05 vk-laptop kernel: [ 107.526177] [<ffffffff8107f730>] ? autoremove_wake_function+0x0/0x40
Jan 8 22:19:05 vk-laptop kernel: [ 107.526181] [<ffffffff8107a960>] ? worker_thread+0x0/0x110
Jan 8 22:19:05 vk-laptop kernel: [ 107.526185] [<ffffffff8107f1d6>] kthread+0x96/0xa0
Jan 8 22:19:05 vk-laptop kernel: [ 107.526190] [<ffffffff8100aee4>] kernel_thread_helper+0x4/0x10
Jan 8 22:19:05 vk-laptop kernel: [ 107.526194] [<ffffffff8107f140>] ? kthread+0x0/0xa0
Jan 8 22:19:05 vk-laptop kernel: [ 107.526198] [<ffffffff8100aee0>] ? kernel_thread_helper+0x0/0x10
Jan 8 22:19:05 vk-laptop kernel: [ 107.526201] ---[ end trace 7882d3bbb33e1568 ]--...

[Slayer6 (slayer6)]

i tried everything to get this working and finally it does.

working on a sony VPCCW21FX with ar9285 wifi
rebuilt ubuntu kernel 2.6.35
then installed latest compat-wireless from 1/24
http://wireless.kernel.org/download/compat-wireless-2.6/

i am experiencing no packet loss. consistent connection.
works great. as far as i'm concerned this is fixed

[Franck (alci)]

I think latest compat-wireless fix the problem (since at least 1 month). So the last question to mark this bug fixed would probably be : will it be packaged for 11.04 ?

[Peter Magnusson (kmpm)]

compat-wireless >=2011-01-24 works for me as well using maverick on a Zotac MAG.
Before installing this it was close to impossible to get a connection at all and if a connection was made it was close to impossible to use or dropped completely.

After installing I have had no issues what so ever... works like a dream.

[Maccer Davidson (colourless-stick)]

While I don't experience packet loss, my AR928X rev. 01 wireless adapter from the Fujitsu Lifebook A1110 Laptop does not work properly either as of the time I'm posting it. Not in the 3.0 kernel from 11.10, and not in the 3.2 kernel from 12.04 either. While I don't think I experience major packet loss from my networks yet, I cannot connect to any encrypted networks after authentication which are WPA1/WPA2 personal using TKIP/and or AES encryption. Unsecured networks are confirmed to work for some reason. I've also tried probing the module with hardware encryption disabled, but to no effect.

[Gabriel Coimbra (synamics)]

My kernel is: Linux RedBot 3.2.0-26-generic #41-Ubuntu SMP
And its affects me, the bug lets the signal very poor!

[chkneater (chkneater)]

I have a suggestion for those with this problem since I've had the same issue and reported it on ubuntuforums as well as here. The issue probly WON'T be fixed on Non-LTS versions

Having said that nm-manager or any of the others you MAY or May not have tried need to be removed and purged (that means COMPLETE removal in Synaptic) BUT NOT before you D/L and install Wicd.

When I had this issue wicd worked great as long as the other wireless were not conflicting.

Look in your /lib64 or /lib32 libraries (under filesystem, check "show hidden files" under "View" tab) and look for modprobe.conf and see if anything is blacklisted. You may need to blacklist the ath9k driver or totally remove (COMPLETE removal) after you have Wicd installed and running. You may not get good strength at first, but I promise you wicd is more reliable in my opinion, and once you're sure there are no conflicting network programs running or installed it should work good.

One other suggestion, if you have ethernet connection, it may be a good idea to remove and purge the other network programs BEForE installing wicd, ONLY if you have a wired connection tho!! Hope that helps anyone cuz this bug has been around 3+ years and they are not fixing it.

[tuxxxy (0xf777)]

I confirm bug in Kubuntu(Ubuntu), Linux kernel 3.2.0-27-generic, bug is still here. Sometimes WiFi is working, but after reboot or enable/disable wireless network - it stop working and no chances to get it work again.

It alway can scan networks and it see all of them, I can connect and get DHCP address, but real data transfer or pings - not working.

Please, consult me, which information I should provide to fix this bug? dmesg is clean, only "no ipv6 routers is present and nothing else"!