Battery Lifetime

[Andy Green]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Somebody in the thread at some point said:

|> Backlight should be fixed for a couple of days now:
|>
|>
http://git.openmoko.org/?p=kernel.git;a=commitdiff;h=db07519c1dfe916bcf9644
|> bfdc4d7c03707a979e
|
| That's good to hear, thanks. Will this be in the latest daily build
along with
| the SD fix?

Yes although how much of "fix" the SD patches are is uncertain.  We have
reason to believe SD_CLK is implicated and that should help, but reports
from folks trying the patches are unclear about it.  I have another more
aggressive and uglier way to come at it if it isn't actually helping.
Anyway today's distribution build should have all that in FWIW.

| If you need this testing then point me at the images and let me know
what you
| need doing. My cell reregistrations seem to do a fair job of provoking
resume
| problems within a few hours on all the images I've tried so far.

I added some really nice diagnostic code in case of OOPS in resume it
will spew dmesg buffer down debug console regardless of resume state
(ie, it doesn't need serial driver resumed) so you can capture the OOPS
trace and resume context.  But nobody saw it triggered yet except me
when I tested it.  (BTW I added another recent patch so if we PANIC, we
flash the top LED, that doesn't happen either.)  So it seems somehow we
deadlock somewhere.  If I can reproduce it, I can debug it using very
gritty methods like moving around printk(); force OOPS or lighting LEDs
at different points... that's not really end user debug technique :-)
But thanks for the offer.

- -Andy
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiAZb4ACgkQOjLpvpq7dMpUpgCcCyDe6lkGCSpXka7PAEur4ttR
588AnR88Xdxofsn3P9g9r9q8HfvD2zJ6
=xht1
-----END PGP SIGNATURE-----