TI omap4 2.6.35 branch

[Ricardo Salveti de Araujo]

On Fri, 2010-08-27 at 12:34 +0800, Bryan Wu wrote:
> Sorry for missing to post it to our mail list.
> 
> On Fri, Aug 27, 2010 at 12:28 PM, Bryan Wu <bryan.wu at canonical.com> wrote:
> > Hiya,
> >
> > I got the TI OMAP4 2.6.35 based kernel branch [1] yesterday and have
> > already prepared a branch [2] which includes all TI patches and our
> > Ubuntu stuff.
> >
> > How?
> > ====
> >  * Checkout our latest Maverick master branch which is at Ubuntu-2.6.35-19.25
> >  * Revert 5 conflict patches [3] in our master branch, most of those
> > patches are OMAP3 related but will cause conflict when rebasing
> >  * Checkout TI's branch [1]
> >  * Rebase [1] to our master and solve some simple conflicts when rebasing
> >  * Set up our debian and debian.ti-omap4 packaging stuff, which is
> > synced from debian.master and our old 2.6.34 debian.ti-omap4
> >  * Sync up config options from 2.6.35 omap4 config
> >  * Fix some building issues such as package name, etc.
> >  * Finally got kernel package built and it can boots up on my OMAP4
> > Pandaboard ES1.0. I don't have ES2.0, we need more testing
> >
> > Issues:
> > =====
> >  * For our Ubuntu usage, we need to set CONFIG_SND_SOC_ALL_CODECS=m.
> > But currently, if we built the modules, we will got a building error:
> >    ubuntu-maverick/sound/soc/codecs/ak4535.c: In function
> > 'ak4535_set_bias_level':
> >    ubuntu-maverick/sound/soc/codecs/ak4535.c:402: error: 'struct
> > snd_soc_codec' has no member named 'bias_level'
> >   So I simply disabled CONFIG_SND_SOC_ALL_CODECS, we got no sound in
> > current kernel. I've already pinged sebjan from TI to ask help from
> > their devel team.
> >   We can find the same issue in TI's own L24.9 release. I guess some
> > ASoC patches were missed
> >
> >  * perf tools cann't be cross compiled, I've posted this questions to
> > Tim and kernel team mail list. And so far it can be built in schroot,
> > so I believe it is a issue of CodeSourcery 2010q1 toolchain.
> >
> > Test:
> > ====
> >  * It boots on my ES1.0 board with a minimal Maverick rootfs.
> >  * Need to test on ES2.0 board (rsalveti helped to tested)
> >   http://paste.ubuntu.com/484321/
> >   USB doesn't work.
> 
> Sorry, it is tested on ES1.0 by rsalveti.

Here's my test with ES2.0:
http://paste.ubuntu.com/484332/

See that the USB is working fine, and I'm also able to get 1GB \o/

Just need to use the proper X-Loader
(http://people.canonical.com/~rsalveti/maverick/boot/es2/MLO-1gb from
http://gitorious.org/~robclark/pandaboard/robclarks-x-loader/commits/omap4_panda_es2.0-1gb)

> >  * Need to test with our Maverick daily image

I can easily test when we get our daily image working :-) We're working
hard on that, so we expect to have something for tomorrow or the
beginning of next week.

Also, they worked on different remaining issues (as I noticed from the
channel). Mostly the ASoC and display, as they were trying to get the
proper patches to handle the HDMI disconnect/connect while booting (like
http://gitorious.org/~robclark/pandaboard/robclarks-kernel-omap4/commit/8fcc6ef6170a470ed5380f47e043182360e8237c to substitute http://kernel.ubuntu.com/git?p=roc/ubuntu-maverick.git;a=commit;h=560118846ca54b81e6498433e6d4398f628866d2)

So I'd say that in one day or two we'll be ready to jump on the next
kernel.

Let me know if you need any other test with ES2.0, I can help you on
that.

And thanks for working on this.

Cheers,
-- 
Ricardo Salveti de Araujo