EEPC 901: rt2x00pci WLAN will not work after suspend if aspm was active during suspend

[chris]

Since I've upgraded to 10.10 on my eepc 901 I have problems reconnecting to WIFI after suspending the computer. Looking the system log I get this message:

Nov 28 13:48:29 netbook kernel: [ 3118.380090] phy0 -> rt2x00pci_regbusy_read: Error - Indirect register access failed
: offset=0x0000101c, value=0xffffffff
Nov 28 13:48:29 netbook kernel: [ 3118.380624] phy0 -> rt2x00pci_regbusy_read: Error - Indirect register access failed
: offset=0x0000101c, value=0xffffffff
Nov 28 13:48:29 netbook kernel: [ 3118.381189] phy0 -> rt2x00pci_regbusy_read: Error - Indirect register access failed
: offset=0x0000101c, value=0xffffffff
Nov 28 13:48:29 netbook kernel: [ 3118.382305] phy0 -> rt2x00pci_regbusy_read: Error - Indirect register access failed
: offset=0x0000101c, value=0xffffffff
Nov 28 13:48:29 netbook kernel: [ 3118.388613] phy0 -> rt2x00pci_regbusy_read: Error - Indirect register access failed
: offset=0x0000101c, value=0xffffffff
Nov 28 13:48:29 netbook kernel: [ 3118.389185] phy0 -> rt2x00pci_regbusy_read: Error - Indirect register access failed
: offset=0x0000101c, value=0xffffffff
Nov 28 13:48:29 netbook kernel: [ 3118.390099] phy0 -> rt2x00pci_regbusy_read: Error - Indirect register access failed
: offset=0x0000101c, value=0xffffffff
Nov 28 13:48:30 netbook NetworkManager[795]: <warn> Activation (wlan0/wireless): association took too long.
Nov 28 13:48:30 netbook NetworkManager[795]: <info> (wlan0): device state change: 5 -> 6 (reason 0)
Nov 28 13:48:30 netbook NetworkManager[795]: <warn> Activation (wlan0/wireless): asking for new secrets
Nov 28 13:48:30 netbook NetworkManager[795]: <info> (wlan0): supplicant connection state: scanning -> disconnected
Nov 28 13:48:34 netbook NetworkManager[795]: <info> (wlan0): device state change: 6 -> 9 (reason 7)
Nov 28 13:48:34 netbook NetworkManager[795]: <warn> Activation (wlan0) failed for access point

The only way to connect again is to reboot :-(

Thanks for your help or for passing it on to whoever would be able to fix this bug ...

[Marius B. Kotsbak (mariusko)]

Some workarounds are described in this thread:

http://ubuntuforums.org/showthread.php?t=1587107

[chris (info-spuhler)]

This workaround uses rt2860sta instead of rt2xxpci. Unfortunately rt2860sta also has problems reconnecting after sleep. Is there any chance rt2xxpci will get fixed at some point or am I better off changing to an older ubuntu or another distro?

[Wolfgang Kufner (wolfgangkufner)]

Hi Chris,

since the rt2x00 project is alive and well and making progress it is reasonable to assume that it will be fixed at some point.

As a first step to getting there please update your bug report by running "apport-cli -u 689004" on the command line.

Thanks,
Wolfgang

[Wolfgang Kufner (wolfgangkufner)]

Could be a power save bug. Could you try if this also happens if you have it plugged in (which should prevent power save mode, I hope).

[chris (info-spuhler)]

Wolfgang,
good call. Yes, in fact it works fine when its plugged in. I only have the problem when running on battery.

[Wolfgang Kufner (wolfgangkufner)]

Good. Next step is finding out which power saving measure is broken. First guess is /usr/lib/pm-utils/power.d/wireless. To disable it do:
sudo touch /etc/pm/power.d/wireless

You now should have an empty file named wireless without the executable bit set in /etc/pm/power.d. Check with:
ls -lh /etc/pm/power.d/
You should see this file:
-rw-r--r-- 1 root root 0 2010-12-31 18:53 wireless

Can you reconnect after suspend on battery now?

[Wolfgang Kufner (wolfgangkufner)]

Before suspending do:

sudo iwconfig wlan0 power off
to shut power management off. The "wireless" file just keeps power management from being activated automatically on battery, but it might be already on.

Or just reboot once before trying.

Check with:
iwconfig
whether power management is off or on.

[chris (info-spuhler)]

Ok, I've followed the steps. iwconfig shows power managment off. The problem still persists when unplugged

[Wolfgang Kufner (wolfgangkufner)]

Not wireless powersaving, then. So let's try PCIe bus power saving:

sudo touch /etc/pm/power.d/pcie_aspm
Same procedure as in comment #6, but this time disabling pcie_aspm. (There is man pm-powersave for the curious, btw.)

Shut down. Boot. Try suspending again.

[chris (info-spuhler)]

Hi Wolfgang,
sorry for the delay. I just tried it and it seems to be working with the PCIe power saving off! Thanks!

So does this mean the notebook will use more power in sleep mode because the PCIe bus now keeps running?

[Pascal Desroches (pascal-desroches)]

My problem is very similar, and since there has been activity here recently I choose to give you details about my problem.

I have a Sony Vaio with Ubuntu 10.10. I've also used Ubuntu 10.04, Xubuntu 10.10, Ubuntu 9.10, Linux Mint 8, 9 and 10, all of them 32-bit as well as 64-bit...

I've always experienced this issue: I use "suspend" a lot and when resuming I have a 50 % chance of having a functional wireless. As expressed here, there is no way to turn it back on (at least I can assure you I have tried numerous times after hours of searches on Google) except to restart the computer.

My wireless LAN is "Intel® PRO/Wireless 3945ABG Network Connection".

Several people have tried to help me at www.ubuntuforums.org but with no luck =/

Any help would be greatly appreciated. Thanks!

[Pascal Desroches (pascal-desroches)]

Oh and my problem occurs while plugged in. On battery I don't know what happens, the computer is never unplugged!

[Wolfgang Kufner (wolfgangkufner)]

Hi chris,

aspm[1] is there to save power in normal use whenever activity on the bus is low enough. AFAIR an example saving of 0.8W has been given for one PCIe bus. AFAIK the PCIe bus is powered off completely during sleep (S3 state). So you should loose work time on battery, but not sleep time.

A possible next step to try could be switching off aspm just before suspending as part off the suspend process and reenable it after resume.

One thing that would be helpful would be to run "apport-cli -u 689004" on the command line to get more detailed information about your system into this report. This will help answer the question which systems exactly are plagued by that bug so that it might be possible to find an out of the box solution for those systems for ubuntu 11.04 "natty".

Thanks

[1] http://en.wikipedia.org/wiki/Active_State_Power_Management

[Wolfgang Kufner (wolfgangkufner)]

Hi Pascal,

I do have to assume that the above measures did not work for you. In that case the only advice I can give you for now is to file a nice bug report with "ubuntu-bug linux". This will collect a lot of logs so try hard to trigger the buggy behavior before you do this. That way the logs will be more helpful.

Thanks

[Pascal Desroches (pascal-desroches)]

Thanks!

[Jeremy Foshee (jeremyfoshee)]

Hi chris,

Please be sure to confirm this issue exists with the latest development release of Ubuntu. ISO CD images are available from http://cdimage.ubuntu.com/daily/current/ . However, note you can only test Suspend, not Hibernate, when using a LiveCD. If the issue remains, 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 689004

Also, please be sure to take a look at https://wiki.ubuntu.com/DebuggingKernelSuspendHibernateResume . If you can provide any additional information outlined there it would be much appreciated.

Additionally, if you could try to reproduce this with the upstream mainline kernel 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. Thanks in advance.

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

[Launchpad Janitor (janitor)]

[Expired for linux (Ubuntu) because there has been no activity for 60 days.]