Many people were hoping bug 15707 would bring resolution, but still no joy. So here goes again: For me the laptop is a Toshiba L645. The battery shows as not present in /proc/acpi/battery/info and the ac adapter status reports either online or offline but never changes when its plugged or unplugged. The two workarounds are to customize the DSDT, or patch the ec driver as shown in bug 15707 or even better as in ACPCIA bug 875. This laptop comes preinstalled with windows 7 and battery/acad works in that. And to answer a question from bug 15707, windows xp does handle the battery and ac functions correctly, so windows has been "orphan _reg" capable for a long time. Hope this helps, any information or testing I'd be more than happy to provide.
Created attachment 69782 [details] acpidump
Created attachment 69792 [details] dmesg
Created attachment 69802 [details] dmidecode
I have the similiar issue on a System76 laptop. Everything works fine on 2.6.38 kernel but on 3.0 onwards, AC Adapter unplugged/plugged does not trigger any events. Battery shows as present and does not change state - plugged or unplugged /sys/devices/LNXSYSTM:00/device:00/PNP0C0A:00/power_supply/BAT0 vendor: Clevo CO. model: M740TX power supply: yes updated: Thu Feb 9 08:26:45 2012 (8 seconds ago) has history: yes has statistics: yes
I have this same issue on two different laptops, a Latitude E6410 and a Latitude E6420. Both use different Intel chipsets. Previously, this bug was hit or miss, sometimes it would detect the battery and sometimes not. Now, the problem is much worse, the battery is almost never detected. Now while I'm using my laptops, I have no idea how much battery is left, until it literally starts flashing red on the laptop and then it dies.
Battery was not detected also on my Toshiba Satellite L755-161 notebook by kernel versions < 3.3. The reason was that for the Toshiba ACPI the methods _STA and _BST of device BAT1 (Battery) return the current value only if flag BTIN has been set. This flag is set by method BSTA of device BAT1. BSTA is called by methods _Q09 and _Q20 of device EC (Embedded Control). Starting with kernel version 3.3.0, this has been fixed. I also have a patch for kernel versions 3.1 and 3.2 (calling BSTA directly or via _Qxx), but I have not submitted it to kernel developers: Why bother when the problem has been fixed?
With yours, was your battery detected sometimes? With both of mine, it will sometimes be detected, and sometimes not. It's much worse right now than it was, with my battery being detected literally around 1 in every 15 boots give or take.
With kernel versions < 3.3 the battery was not detected on boot. The battery was detected and correctly described only after one of the following actions was taken: 1. Open the lid of the notebook 2. Unplug and plug the external power ("AC") 3. Start the system without the battery. Shutdown and start again with battery. The first two are to be done with a running system, of course. The third only shows the battery for the first start after the shutdown. These actions trigger the ACPI methods _Q09 or _Q20 and so the battery gets "visible" for methods _STA and _BST.
I have a Toshiba L650 laptop with Fedora16 with kernel- 3.3.2-6 . Facing the same issue of batter not being detected at all. Kindly let me know if any more info is needed on this that might help.
Hi All: Please check whether the following patch fix your issue. https://bugzilla.kernel.org/attachment.cgi?id=87901
ping...
Since no response more than one month, Close the bug.