[GUTSY] ACPI: battery state reporting errors since 2.6.22-9

Bug #129388 reported by Guillaume Pujol
26
This bug affects 1 person
Affects Status Importance Assigned to Milestone
linux-source-2.6.22 (Ubuntu)
Invalid
Medium
Unassigned

Bug Description

Since my Gutsy update to linux-image-2.6.22-9-generic (2.6.22-9.20) and the related packages (, ACPI doesn't report my battery state properly.
This is on a MacBook Pro Core2Duo from november 2006.
The battery is detected but is reported as empty (altougth the hardware indicator reports it as fully loaded).
If I plug or unplug the power cable, these events are reported correctly.

Output of '$ acpi':
     Battery 1: charging, 0%, charging at zero rate - will never fully charge.

$ cat /proc/acpi/battery/BAT0/info
present: yes
design capacity: 0 mWh
last full capacity: 0 mWh
battery technology: rechargeable
design voltage: 0 mV
design capacity warning: 250 mWh
design capacity low: 100 mWh
capacity granularity 1: 10 mWh
capacity granularity 2: 10 mWh
model number:
serial number:
battery type:
OEM info:

$ cat /proc/acpi/battery/BAT0/state
present: yes
capacity state: ok
charging state: charging
present rate: 0 mW
remaining capacity: 0 mWh
present voltage: 0 mV

ACPI worked flawlessly in the previous version of linux-image (2.6.22-8.18).

Regards,

Tags: macbookpro
Revision history for this message
Guillaume Pujol (guill-p) wrote :
Revision history for this message
Guillaume Pujol (guill-p) wrote :
Revision history for this message
Guillaume Pujol (guill-p) wrote :
Revision history for this message
Guillaume Pujol (guill-p) wrote :
Revision history for this message
Guillaume Pujol (guill-p) wrote :

$ uname -a
Linux guillaume-laptop 2.6.22-9-generic #1 SMP Mon Jul 30 18:00:27 GMT 2007 i686 GNU/Linux

Revision history for this message
Guillaume Pujol (guill-p) wrote :

Battery level came back automagically this morning (UTC time).
In the meantime, I have reboobed, dist-upgraded (nothing kernel or ACPI related was updated), and booted MacOS X.

I will report back later if this problem appears again.

Revision history for this message
Guillaume Pujol (guill-p) wrote :

Battery level disappeared again after a reboot yesterday. Rebooted Linux several times and it did not reappear.
Booting MacOS X and then rebooting to Linux made it work again.
So it seems to be related to some registers that MacOS X initialize and Linux does not. The weird thing is that I don't remember having this problem with linux-image <=2.6.22-8.

What informations can I provide to debug this problem ?

Changed in linux-source-2.6.22:
assignee: nobody → ubuntu-kernel-acpi
importance: Undecided → Medium
status: New → Triaged
Revision history for this message
Benoit Grégoire (benoitg) wrote :

I have the exact same problem. I've never seen it with linux-image <=2.6.22-8 either.

Revision history for this message
Gabriel Ambuehl (gabriel-ambuehl) wrote :

I'm sometimes told my Latitude D830 has no battery in it which is clearly bogus. Never seen this with Feisty.

Revision history for this message
Guillaume Pujol (guill-p) wrote :

It's been about a month and many reboots now and the problem did not occur again. If Benoit can confirm, I think we can consider this bug as FIXED.

And for Gabriel: If I understand your problem correctly, ACPI doesn't detect your battery at all. This is not the same issue (the battery IS detected but not its loading state). Maybe you should report a different bug.

Regards,

Revision history for this message
Gabriel Ambuehl (gabriel-ambuehl) wrote : Re: [Bug 129388] Re: [GUTSY] ACPI: battery state reporting errors since 2.6.22-9

Well I would have to check what exactly ACPI does detect (right now the
battery is there, so that's hard). In guidance power manager, I get the slots
for both batteries, but without any info about it (not sure if it says not
present or 0% just now). But yes, I'll file another report when I get more
data.

Revision history for this message
Benoit Grégoire (benoitg) wrote :

I concurr, the problem seems to have cleared sometimes in the last month. This bug can be closed.

Revision history for this message
Benjamin Otte (Company) (otte) wrote :

I still don't have battery information on my macbook (not macbook pro, just macbook). This is with the latest Gutsy packages.

Revision history for this message
Pepie 34 (romain-herault+launchpad) wrote :

I have the same behaviour with the last gutsy

It seems to be a regression for 1 weeks now

Macbook pro
2.6.22-12-generic

Revision history for this message
Xavier Tardivel (the-frog45) wrote :

I don't think this is related to macbooks only.

I got similar problems on an ASUS laptop ( A4G ). It worked flawlessly under 2.6.20 and is totally broken under 2.6.22-14-generic.

The battery level indicator returns "fully charged" while displaying the battery being at 12%. ( acpi cmd returns the same thing. ) Maybe a new bug should be opened related to "laptops" rather than macbooks.

Revision history for this message
Fran Aquino (fj-aquino) wrote :

The same for my LG E500 SP16B. It's also an Intel Santa Rosa based laptop (MSI mobo as in EX600), so I think the problem is around it. Quite annoying for me because the laptop switches its energy status (plugged/onbattery) each 2 minutes approximately, and the screen dims/brightens continuously.

Revision history for this message
MarkEdwards (shades66) wrote :

I'm having the same problem on my MSI GX700. As Fran said above I too have the screen constantly going dim/bright and for now have disables gnome-power-manager which seems to of solved the problem for now except that I have no battery info. I have run the "acpi -V" command a few times and I get a mixture of

 >acpi -V
     Battery 1: charged, 100%
     Thermal 1: ok, 41.0 degrees C
  AC Adapter 1: on-line

> acpi -V
     Battery 1: charged, 0%
     Thermal 1: ok, 48.0 degrees C
  AC Adapter 1: on-line

>acpi -V
     Battery 1: charging, 0%, 64:00:00 until charged
     Thermal 1: ok, 48.0 degrees C
  AC Adapter 1: on-line

>acpi -V
     Battery 1: charged, 100%
     Thermal 1: ok, 37.0 degrees C
  AC Adapter 1: on-line

   all these were within a few seconds of each other so I suspect that this problem also effects the "Thermal 1" into too!?!? (Again as Fran says above it occurs every two minutes or so but can be forced to happen by using the CD drive or flipping back and forth through the windows with all the GL features enabled)

cheers
Mark
PS. I'm using a new install of ubuntu with the 2.6.22-14-generic kernel

Revision history for this message
Motin (motin) wrote :

When Randall Kennedy blogged about his 5-day long "30-day Ubuntu Plunge", having this ACPI bug as the definitive reason to leave Ubuntu the same day (http://weblog.infoworld.com/enterprisedesktop/archives/2007/11/the_ubuntu_plun_4.html), Scott James Remnant informingly wrote this comment to the post:

>We do obviously have quite a large number of laptops in our testing centre, and perform regular testing throughout the development cycle on this hardware.
>
>Unfortunately our expertise is limited to those vendors that are willing to work with us to certify their hardware. This doesn't just mean the marque on the lid, but necessarily includes the manufacturers of most of the major components of the machine.
>
>If you are interested, please e-mail me so that we can at least get some detailed information about your problem machines so that we at least know what doesn't work -- it may be that there's some component there that we don't test, or are unable to.
>
>(I'm going to randomly guess, with only the information in your articles, that the problem is because both laptops have nVidia graphics cards. Unfortunately only Intel and ATI currently provide the necessary resources to Linux developers to support their hardware; nVidia prefer to maintain their drivers themselves. The fact that your failure mode is a black screen supports the notion that it is their closed-source driver failing to resume, since you clearly have no video.)
>
>The rigours of a time-based release cycle mean that sometimes we have to accept regressions outside of our supported/certified harware set to ensure that the release can be made on time. To counter this, we have regular "Long Term Support" releases that have these kinds of problems more specifically targeted -- although again, we can still only target issues that we are able to fix; if a hardware vendor will not work with us, then there is little that we can do.
>Posted by: Scott James Remnant at November 11, 2007 10:27 PM

Revision history for this message
Motin (motin) wrote :

Bah - wrong bug report! The above comment was for https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.22/+bug/136387
Moderator please remove these two comments from me - Thanks!

Revision history for this message
impact (impact-atlas) wrote :

MarkEdwards: try alternating between acpi -V and sudo acpi -V, the output gets even weirder.

On a GX700, AC Adapter disconnected, typing in console one command after another. Each command shows different results.

impact@msi1719:~$ acpi -V
     Thermal 1: ok, 48.0 degrees C
  AC Adapter 1: on-line
impact@msi1719:~$ sudo acpi -V
[sudo] password for impact:
     Thermal 1: ok, 48.0 degrees C
  AC Adapter 1: off-line
impact@msi1719:~$ sudo acpi -V
     Thermal 1: ok, 48.0 degrees C
  AC Adapter 1: off-line
impact@msi1719:~$ acpi -V
     Battery 1: charging, 87%, 02:24:31 until charged
     Thermal 1: ok, 48.0 degrees C
  AC Adapter 1: on-line
impact@msi1719:~$ acpi -V
     Thermal 1: ok, 48.0 degrees C
  AC Adapter 1: off-line
impact@msi1719:~$ sudo acpi -V
     Thermal 1: ok, 48.0 degrees C
  AC Adapter 1: off-line
impact@msi1719:~$ acpi -V
     Battery 1: charging, 42%, 11:09:09 until charged
     Thermal 1: ok, 48.0 degrees C
  AC Adapter 1: on-line
impact@msi1719:~$ sudo acpi -V
     Battery 1: charging, 56%, 08:26:00 until charged
     Thermal 1: ok, 48.0 degrees C
  AC Adapter 1: off-line
impact@msi1719:~$ acpi -V
     Thermal 1: ok, 48.0 degrees C
  AC Adapter 1: off-line

Just to reiterate, AC Adapter was disconnected all the time.

Revision history for this message
Ricardo Pérez López (ricardo) wrote :

I've the same problem with my LG E500-U.AP49B, but the battery issue appears only when the battery charge level is "low", i.e. there's < 20 min remaining.

Revision history for this message
Ricardo Pérez López (ricardo) wrote :

ricardo@shoggoth:~$ acpi -V
     Battery 1: charging, 55%, 14:24:33 until charged
     Thermal 1: ok, 94.0 degrees C
  AC Adapter 1: on-line
ricardo@shoggoth:~$ acpi -V
     Thermal 1: ok, 94.0 degrees C
  AC Adapter 1: off-line
ricardo@shoggoth:~$ acpi -V
     Thermal 1: ok, 94.0 degrees C
  AC Adapter 1: off-line
ricardo@shoggoth:~$ acpi -V
     Thermal 1: ok, 94.0 degrees C
  AC Adapter 1: off-line
ricardo@shoggoth:~$ acpi -V
     Thermal 1: ok, 94.0 degrees C
  AC Adapter 1: off-line
ricardo@shoggoth:~$ acpi -V
     Thermal 1: ok, 94.0 degrees C
  AC Adapter 1: on-line
ricardo@shoggoth:~$ acpi -V
     Battery 1: charging, 74%, 08:02:49 until charged
     Thermal 1: ok, 94.0 degrees C
  AC Adapter 1: on-line
ricardo@shoggoth:~$

Revision history for this message
Tangui (tanguim) wrote :

I had the same problem. It is due to a hardware bug.

I posted the solution here : https://help.ubuntu.com/community/MacBookPro#head-5064fd602c07a77f22bba3b2a5e31335acb16ccd

Revision history for this message
Tangui (tanguim) wrote :

It is a hardware problem. To reset acpi of the mac :
    * Shut down your Ubuntu
    * Disconnect the AC Adapter
    * Remove the computer's battery
    * Press the power button for 5 seconds
    * Reconnect the battery and the AC Adapter
    * Turn your computer on

Changed in linux-source-2.6.22:
status: Triaged → Invalid
Curtis Hovey (sinzui)
Changed in linux-source-2.6.22 (Ubuntu):
assignee: Registry Administrators (registry) → nobody
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers