from the acpi-sucks-pooh dept.
Alright it is time for some ranting. After a few frustrating days working on
ACPI code I have to complain about it damn it. We have made progress however
getting there has been painful. For example, when I insert a 100% full battery
into my laptop I get an event, however that event doesn't have the battery bit
set. So in other words the even goes unanswered and the driver therefore does
not know that there is a battery physically inside the system. Now I don't
know if this is a bug in dsdt or in the ACPI implementation but it has set me
back hours in development time. All indications are towards a bug in ACPI but
the handling of the interrupt is so complex that it is really hard to debug.
We'll figure this out over time but it still stinks to waste hours on something
like that. I have observed similar behavior with the power plug as well. All
this happens when the battery is at 100% though; whenever that is not the case
it seems to work. ARGH!
All that aside I do want to mention that Jordan has been making steady progress
on the memory leak situation. This code is not in cvs yet though so please be
a little more patient. I have received a number of bug reports that are
useful, keep them coming just don't expect instant fixes. I will actually
post a bug report that is so good that it should be the norm.