Today on the branches

Andy Green andy at openmoko.com
Wed Oct 8 23:12:56 CEST 2008


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

Hi -

I've finally figured out the FIQ problems on stable-tracking (mainly
caused by FTRACE config in the kernel poisoning __attribute__ ((naked))
functions) and Andrzej's find about the pwm patch fixed the slowdown
issue, so tracking branches are coming back to more decent shape.

stable-tracking is rebased to today's upstream git, and I've also put
andy-tracking on top of today's Balaji's branch which boots OK.  Recent
stable branch patches (like Harald's stuff) are also updated onto 2.6.26
and tracking branches.

I think the device tree reordering stuff currently in andy-tracking
needs to go into stable-tracking shortly despite the impact on Glamo
suspend-resume is not solved yet.  It really improves the stability and
correctness of the device relationships in every way except the Glamo
freakage, the longer we put off adding it the worse the divergence with
Balaji's work will get.  Now FIQ is cleared I will be on Glamo resume on
andy-tracking again tomorrow.

You can get an image of today's andy-tracking with critical drivers in
here if you're interested to try it (this is very much "unstable"):

http://people.openmoko.org/andy/uImage-moredrivers-andy-tracking_ee1c5f6978680883.bin

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

iEYEARECAAYFAkjtIlgACgkQOjLpvpq7dMprXACfTBG2lfIlGlTH1jZUq55eOFQM
bzkAn3ZsOZJxw/w6PchNrV5YRHzjnxTG
=KoBd
-----END PGP SIGNATURE-----



More information about the openmoko-kernel mailing list