Please note: This project is no longer active. The website is kept online for historic purposes only.
If you´re looking for a Linux driver for your Atheros WLAN device, you should continue here .

Ticket #1363 (new defect)

Opened 14 years ago

Last modified 14 years ago

Unable to reach C3/C4 states while wireless is up.

Reported by: ubuntu@volanin.com.br Assigned to:
Priority: minor Milestone:
Component: madwifi: other Version: v0.9.3.1
Keywords: battery powertop powersaving c3 c4 Cc:
Patch is attached: 0 Pending:

Description

While using UBUNTU GUTSY daily build 05-JUN-2007, with powertop 1.5 installed, I noticed that my computer was in the C2 power-saving state for an average of 95%+ of the time, and was not able to reach the battery-friendly C3 and C4 states. After some experimentation, I realized that this problem was caused by the wireless interface just being up, **even when not associated to any Access Point**. Just issuing 'ifconfig ath0 down', or disabling wireless in the Network Manager applet, would allow the computer to reach C3/C4 just fine, increasing my battery time significantly.

My laptop is an Acer 3624, with a built-in Atheros Wi-fi Card.

# uname -a:
Linux ubuntu 2.6.22-6-generic #1 SMP Fri Jun 1 19:24:12 GMT 2007 i686 GNU/Linux


This bug has also been opened in Ubuntu Launchpad: bugs.launchpad.net/ubuntu/+source/linux-source-2.6.22/+bug/118808

For information on processor power-saving states and the powertop utility, go here: www.linuxpowertop.org/powertop.php

Attachments

dmesg.log (21.9 kB) - added by ubuntu@volanin.com.br on 06/05/07 23:54:14.
dmesg output
lspci-vvnn.log (10.4 kB) - added by ubuntu@volanin.com.br on 06/05/07 23:55:16.
lspci -vvnn output

Change History

06/05/07 23:54:14 changed by ubuntu@volanin.com.br

  • attachment dmesg.log added.

dmesg output

06/05/07 23:55:16 changed by ubuntu@volanin.com.br

  • attachment lspci-vvnn.log added.

lspci -vvnn output

06/06/07 06:28:49 changed by mrenzmann

  • priority changed from critical to major.

06/06/07 18:12:09 changed by anonymous

I have this problem too.

SMCWCB-G 03:00.0 Ethernet controller: Atheros Communications, Inc. AR5005G 802.11abg NIC (rev 01)

I'm using SVN r2427. PowerTop? 1.5 shows wifi0 with the highest number of wake ups and it keeps my processor in C2 70-80% of the time. When I disable wifi0 it goes to C3 90% of the time.

07/20/07 10:31:44 changed by ubuntu@volanin.com.br

The current situation of the drivers concerning this bug has improved a lot! I have compiled Madwifi NG r2594 with HAL from 20070719 for Ubuntu Gutsy (Kernel 2.6.22) and the situation is as follows:

The processor stays in C2 for about 55% of the time, and in C3 for 45% of the time. This alone led to an increase of almost half an hour to my battery performance.

You are getting there! Thank you, and keep up the excelent work.

08/15/07 02:37:23 changed by ubuntu@volanin.com.br

This is the current situation, using MadWifi NG r2652.
These modules are compiled against kernel 2.6.22 (from Ubuntu Gutsy).

The powertop captures were made as soon as gnome started.
Before each capture, I waited for the computer to become completely idle.
Check how the C2/C3 states change in each of them.


With the modules LOADED, but wireless disabled in network-manager:

Cn                Avg residency         P-states (frequencies)
C0 (cpu running)        ( 1.8%)
C1                0.0ms ( 0.0%)         1.60 Ghz   (100.0%)
C2                8.5ms ( 0.5%)          600 Mhz   ( 0.0%)
C3                9.2ms (97.7%)          400 Mhz   ( 0.0%)
                                         200 Mhz   ( 0.0%)

Wakeups-from-idle per second : 107.1    interval: 15.0s
no ACPI power usage estimate available

Top causes for wakeups:
  27.2% ( 57.9)       <interrupt> : i915@pci:0000:00:02.0 
  23.3% ( 49.6)         syndaemon : do_nanosleep (hrtimer_wakeup) 
  23.3% ( 49.6)              Xorg : do_setitimer (it_real_fn) 
   5.5% ( 11.7)       <interrupt> : libata 
   5.2% ( 11.0)         nm-applet : schedule_timeout (process_timeout) 
   4.7% ( 10.0)     mixer_applet2 : schedule_timeout (process_timeout)



Keeping the modules LOADED, but enabling wireless:

Cn                Avg residency         P-states (frequencies)
C0 (cpu running)        ( 1.6%)
C1                0.0ms ( 0.0%)         1.60 Ghz   (100.0%)
C2                8.4ms (53.0%)          600 Mhz   ( 0.0%)
C3                7.6ms (45.4%)          400 Mhz   ( 0.0%)
                                         200 Mhz   ( 0.0%)

Wakeups-from-idle per second : 122.8    interval: 15.0s
no ACPI power usage estimate available

Top causes for wakeups:
  24.1% ( 55.7)       <interrupt> : i915@pci:0000:00:02.0 
  21.5% ( 49.7)         syndaemon : do_nanosleep (hrtimer_wakeup) 
  21.5% ( 49.7)              Xorg : do_setitimer (it_real_fn) 
   5.0% ( 11.5)         nm-applet : schedule_timeout (process_timeout) 
   4.7% ( 10.9)       <interrupt> : libata 
   4.3% ( 10.0)     mixer_applet2 : schedule_timeout (process_timeout) 
   4.3% (  9.9)       <interrupt> : wifi0, Intel ICH6 (*** MADWIFI ***)



The situation is much better than when this bug was first posted.
Keep up the good work.
Thank you.

08/15/07 04:13:12 changed by mentor

  • priority changed from major to minor.

Hmmm, it is interesting to note that although we're only listed as 4.3% cause of wake-ups, that seems to translate as a ~50% drop in C3 utilisation...

08/24/07 13:24:21 changed by tumbleweed

I'm using an AR5418 on a MacBook 2007, ubuntu gutsy/amd64, 2.6.22-10 kernel, madwifi trunk/r2659.

I don't see any difference in wake-ups when wifi is off / on in network manager and whether I'm associated or not.

The majority of my wake-ups are USB-related. But that could be because Mac's are so USB-centric.