Red Hat Bugzilla – Bug 62943
APM doesn't handle second battery in Sony Athlon laptop
Last modified: 2007-04-18 12:41:45 EDT
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (compatible; Konqueror/3.0.0; Linux)
Description of problem:
Sony Vaio PCG-FXA36 Athlon laptop computer (All Sony FXA series notebooks
thus far have the same chipset, differing only in CPU, RAM and hard drive.)
APM battery monitoring works properly when the first battery is used. It
detects when AC power is plugged in or not, and it properly detects the amount
of battery remaining.
However, when the second battery is added (during runtime or while powered
off) apm is unable to show the actual power remaining. It is stuck somewhere
between 97-100% until the second battery is completely drained. After that
point the percentage heads towards zero as it drains battery #1. I can see
which battery it is currently draining by the LEDs below my notebook screen.
Version-Release number of selected component (if applicable):
Skipjack beta 2
Steps to Reproduce:
1. Use Sony FXA laptop without AC power with #1 and #2 battery.
2. APM power meter stuck near 100% until battery #2 is fully drained.
3. APM percentage goes down properly while discharging battery #1.
APM properly displays battery percentage only when battery #1 is used, or
battery #2 is completely drained.
APM should get data as a sum of battery #1 and battery #2 when both batteries
are in the computer. ACPI is able to see both battery's status, but ACPI
isn't default in Red Hat Linux.
This misbehaved identically in Red Hat Linux 7.2 and Mandrake 8.2.
This sounds like a very broken APM implementation in the BIOS delivering wrong data to the
kernel. Assigning to kernel so it can be blacklisted.
Please do not blacklist this chipset. The battery indicator works 100% with a
single battery. The presence of the second battery just makes it think that the
first battery is full (until the second battery is drained).
Also, email@example.com posted to Skipjack-list about his Intel Dell Inspiron 8100
laptop seeing similar behavior. This may be a more general problem that may be
fixable. I'm hoping he will post here soon with his additional details.
I wouldn't be surprised if the APM spec doesn't take 2 batteries into account.
The "time left" comes from the bios and the bios ought to take both batteries
DEFERRED for now? I understand this may have to wait until ACPI is complete
and stable. =(