suspend/resume test

Andy Green andy at
Sun Dec 7 21:35:18 CET 2008

Hash: SHA1

Somebody in the thread at some point said:
| Andy Green wrote:
| ...
|> Yes but why does apmd stall?  Is it something that we use in rootfs that
|> we ship?  If so what do we use it for, several people proposed to throw
|> out all support for apm in kernel and if we didn't already we probably
|> will.
| Who's "we"?  There are multiple completely unique rootfs' out there in
| common use today.  Interestingly, one more totally unique rootfs was
| just introduced by the Openmoko kernel developers themselves, not more
| than a couple of weeks ago!
| That's not a problem I guess; diversity is good!  But it does indicate
| that any assumption that "if it's not in the rootfs that *we* ship, then
| it doesn't matter" may not be a good thing.  After all, as we've already
| observed, what if Android decides to use APM?  :-D Best not to be too
| hasty to throw it out!

What it impacts is what we should test against.  It's fine there are an
uncontrolled number of rootfs out there with random things in, but it
means I don't have time to test against them and rely on the users of
the rootfs to inform about problems.

It looks like Jeremy is planning to be "internal tester" for kernel on
whatever rootfs he prefers so I guess I will be hearing more about
trouble from whatever that is using.  If we ship stuff with apmd then
"throw out the broken apmd you seem to have in your rootfs" won't do as
a solution.

Does Android need APM or is this just FUD?

| (BTW, I'm no fan of APM myself, I'm just pointing out that there may be
| others, who are not on this list, who might be disappointed and
| surprised at some date to learn that APM has been summarily dumped.)

The other week folks were chanting "off with its head", it's a good job
I don't feel a need to please everyone at the same time.

- -Andy
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora -


More information about the openmoko-kernel mailing list