Atheros AR5212 poor connection speed

Bug #463625 reported by Leann Ogasawara
32
This bug affects 5 people
Affects Status Importance Assigned to Milestone
Linux
Fix Released
Medium
linux (Ubuntu)
Won't Fix
Medium
Unassigned

Bug Description

I'm opening this new bug on behalf of bug reporter Christansen (lpid: happylinux). Will attach the relevant debug files and comments they originally posted to bug 461419:

Christiansen wrote:

"The information collected was using the Karmic RC Live CD and default ath5k module from the linux-image. Connectionspeed was checked several times using the ISPs reliable internet speed test, and the result was less than 256Kbit/ps downspeed and very unstable - link was checked to more than 2 Mbit/ps with an other Kubuntu 8.04 with the default ath_pci module loaded, this was done in between the Karmic RC tests - The boxes was connected to the AP one at a time, and no other access for other client boxes was allowed.

The LBM was tested on an new installed Karmic RC fra the same live CD and on the same box, and was updated with all updates including the the LBM. The test results was even worse than with the default ath5k module.

A test on an enterprise Cisco WiFi network was done too with the same box, and the connection speed was even less than with the default module and wireless stack.

Compiling an ath-pci module from source (Ver. 10.0.5.6), and loading the module on the other hand showed the expected connection speed on the enterprise network, which is approximately 18.000 Kbit/ps (18 Mbps effective on a 54G network).

I then successfully compiled the the wireless-compat stack dated 2009-10-27, and loaded the ath5k module without any errors seen in dmesg. Ran a connection speed test and got same result as with the LBM module, rebooted and reran the test with same bad result. So unfortunately it seems like the bug is still present in the latest modules too.
Modules loaded after "sudo make install" was:
ath5k 122372 0
mac80211 210508 1 ath5k
ath 8828 1 ath5k
cfg80211 131048 3 ath5k,mac80211,ath
led_class 4096 2 ath5k,thinkpad_acpi
"

Revision history for this message
Leann Ogasawara (leannogasawara) wrote :
Revision history for this message
Leann Ogasawara (leannogasawara) wrote :
Revision history for this message
Leann Ogasawara (leannogasawara) wrote :
Revision history for this message
Leann Ogasawara (leannogasawara) wrote :
Revision history for this message
Leann Ogasawara (leannogasawara) wrote :
Changed in linux (Ubuntu):
importance: Undecided → Medium
status: New → Triaged
Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

Hi Christiansen,

Thanks for all the great testing and information you provided. I've since opened a bug upstream for you as well. If you can, please subscribe yourself to that the upstream bug and follow up with any questions/comments posted by any upstream developers in that bug.

http://bugzilla.kernel.org/show_bug.cgi?id=14511

Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

Note, you made need to first create an account with bugzilla.kernel.org if you don't have one already in order to subscribe to the upstream bug.

Revision history for this message
Christiansen (happylinux) wrote :

Hi Leann,

Thanks for all the work you have put into the bug-handeling until now, I will subscribe to the upstream bug as soon as possible, and hopefully later tonight. I hope that all the testing could lead to a solution, as the ath5k never have been working with this type of Lenovo/IBM adapter - checked with other distros for the past 2 or 3 kernel releases too. I'm pretty sure that the adapter is widely used, as the adapter is one of the adapters used in a bunch of newer ThinkPads according to www.thinkwiki.org.

Changed in linux:
status: Unknown → Confirmed
Revision history for this message
Christiansen (happylinux) wrote :

Lucid Alpha 2 (kernel 2.6.32-10.14) is also affected by this bug.

Revision history for this message
Christiansen (happylinux) wrote :

Lucid Alpha 3 (kernel 2.6.32-10.14) still suffering, and now even only connects to an wireless AP (tested Cisco. Linksys, 3COM) in very rear cases.
Meanwhile bugs keeps coming in on launchpad and even more people continues to rapport in on bug #461419 for this, in ThinkPads, widely used adapter.

  Ethernet controller [0200]: Atheros Communications Inc. AR5212 802.11abg NIC [168c:1014] (rev 01)
  Subsystem: IBM Device [1014:058a]

Nothing much seems to happen on the related upstream kernel bug, and compiling latest compat-wireless drivers unfortunately doesn't seem to change things either.

Revision history for this message
ashgtx (ashwinhgtx-yahoo) wrote :

This bug is preventing my friend from moving to Ubuntu. :(

Revision history for this message
Christiansen (happylinux) wrote :

@ashgtx
what adapter (sudo lspci -vvnn) are you using and what channel are channel/frequency are the AP you are trying to connect to running ?
from a terminal run and find your SSID, if its broadcasting the SSID : nm-tool
if it's not broadcasting, but slightingly more complicated as you need to know the MAC addess : iwlist wlan0 scan

If you are running above channel 6 or 2437 Mhz on the AP, then you could try changing the APs channel below 7 (if you have admin access). This has solved my problem on a couple of installations, since this is a known bug in the ath5k module, which doesn't seem to get much attention in the upstream linux-kernel-bugs 14511 (see the top of this bug for a link).

Revision history for this message
Christiansen (happylinux) wrote :

This bug is solved by using the upstream compat-wireless-2010-04-19.tar.bz2, and also works fine using compat-wireless-2010-05-22.tar.bz2.

Any chance that this upstream update could get into Lucid updates somehow ?

Revision history for this message
ashgtx (ashwinhgtx-yahoo) wrote :

@Christiansen

I've managed to compile and load madwifi after blacklisting ath5k. This solves the problem. For now.

Will this bug be rectified if I update to the latest kernel?

Revision history for this message
Christiansen (happylinux) wrote :

@ashgtx

Nope, madwifi is out of K/Ubuntu for good I think, and ath5k despite all the bug are the one non-hackers are bound to from the 9.10 release. The good news is that the ath5k module in compat-wireless-2010-04-19.tar.bz2 from: http://wireless.kernel.org/download/compat-wireless-2.6/
is easier and faster to compile than madwifi (just remember to do a "./scripts/driver-select ath5k" first), and that this particular bug (and probably many other ath5k's) finally been solved.

Right now the only way to get this updated driver without compiling, is to use the K/Ubuntu mainline and precompiled 2.6.36 RC1 kernel image (and the 2 header) packages for your CPU architecture from: http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.35-rc1-lucid/
The driver will according to the upstream kernel guys never come to the 2.6.32 kernel, which Lucid is based upon.

What I think we_could_hope_for, is that someone will put this an updated compat-wireless stack in the repository as the "linux-backports-modules-wireless-lucid-generic" package (right now referencing to a NOT fixed/updated "linux-backports-modules-wireless-2.6.32-22-generic" package).

Revision history for this message
alcarola (alcarola) wrote :

I got tired of the problems with ath5k, such as freezing my computer (90% sure ath5k was the culprit), network manger roaming to none etc. and installed the windows driver under ndiswrapper. Worked like a charm -- why didn't I think of this immediately?

Instructions in the last post in the following thread: http://ubuntuforums.org/archive/index.php/t-429537.htm

Maybe this can be of some help to someone else as well.

(Sorry for crossposting this comment to several discussion threads.)

tags: added: kernel-needs-review kernel-net
Revision history for this message
Chase Douglas (chasedouglas) wrote :

Please try installing the latest linux-backports-modules-wireless package. It currently includes driver updates from 2.6.34, which may include a fix for this issue.

Thanks!

tags: added: kernel-reviewed
removed: kernel-needs-review
Revision history for this message
Christiansen (happylinux) wrote :

Nope, the modules that fix this particular bug will first be included in the 2.6.35 kernel - please have a look at the last comment on the upstream kernel bug.

BTW. I've been running the Ubuntus mainstream kernel 2.6.35 RC1 and RC 3 on Lucid for a while and they solve this bug. On another box I've been using the compat-wireless dated 2010-04-19 and 2010-06-15, both work but the latest have absolutely best roaming capabilities.

Changed in linux:
status: Confirmed → Fix Released
Changed in linux:
importance: Unknown → Medium
Revision history for this message
Brad Figg (brad-figg) wrote : Unsupported series, setting status to "Won't Fix".

This bug was filed against a series that is no longer supported and so is being marked as Won't Fix. If this issue still exists in a supported series, please file a new bug.

This change has been made by an automated script, maintained by the Ubuntu Kernel Team.

Changed in linux (Ubuntu):
status: Triaged → Won't Fix
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.