[meta-ti] Which machine type to use for DaVinci DM6446 and TI816x?

[Brian Hutchinson]

On Wed, Jul 17, 2013 at 11:04 AM, Brian Hutchinson <b.hutchman at gmail.com>wrote:

> On Wed, Jul 17, 2013 at 10:44 AM, Maupin, Chase <chase.maupin at ti.com>wrote:
>
>>  ** **
>>
>> *From:* Brian Hutchinson [mailto:b.hutchman at gmail.com]
>> *Sent:* Wednesday, July 17, 2013 8:26 AM
>> *To:* Maupin, Chase
>> *Cc:* Dmytriyenko, Denys; meta-ti at yoctoproject.org
>>
>> *Subject:* Re: [meta-ti] Which machine type to use for DaVinci DM6446
>> and TI816x?****
>>
>>  ** **
>>
>> On Wed, Jul 17, 2013 at 9:14 AM, Maupin, Chase <chase.maupin at ti.com>
>> wrote:****
>>
>>   ****
>>
>> *From:* meta-ti-bounces at yoctoproject.org [mailto:
>> meta-ti-bounces at yoctoproject.org] *On Behalf Of *Brian Hutchinson
>> *Sent:* Wednesday, July 17, 2013 8:08 AM
>> *To:* Dmytriyenko, Denys
>> *Cc:* meta-ti at yoctoproject.org
>> *Subject:* Re: [meta-ti] Which machine type to use for DaVinci DM6446
>> and TI816x?****
>>
>>  ****
>>
>> On Wed, Jul 10, 2013 at 12:46 PM, Denys Dmytriyenko <denys at ti.com> wrote:
>> ****
>>
>>   ****
>>
>> DM6446 is quite different from OMAP-L138. Both AM180x and DA850 are just
>> derivatives of OMAP-L138.
>>
>> Anyway, DM6446 is no longer supported, as it's a very old platform.****
>>
>>
>>
>> > I've always used Arago or Angstrom for DM6446 support in the past and so
>> > I'm not sure which .conf I should be using with meta-ti.  My gut says
>> use
>> > am1808-evm.conf but I thought I should ask to be sure.****
>>
>> --
>> Denys****
>>
>>   ****
>>
>> So how hard is it to get ti816x machine support moved up into Dylan (or
>> master)?  I've always had machine support for the branches I used in the
>> past ... now I find myself needing Dylan or even master and there is no
>> ti816x support there.****
>>
>>  ****
>>
>> Is it not there just because nobody had hardware to test it?  If so, I'm
>> game to add it with some guidance :)****
>>
>>  ****
>>
>> [CEM] Generally not too hard.  You can often start with the old arago
>> machine and move that in with some cleanup.  Then you will need to
>> move/create the bootloader and kernel recipes as well, again looking for
>> cleanup.  That should give you a basic working system.  The main issue is
>> that someone needs to be signed up to keep those recipes working and up to
>> date.  That is what is missing in most cases is a maitainer.****
>>
>>  ****
>>
>> Even better, for you use I would recommend looking at commit c586bd9a66
>> which is the one that removed these configs and then add back the pieces
>> you need.****
>>
>>  ****
>>
>> ** **
>>
>>  Thanks Chase,  I kind of figured it was a (people) bandwidth problem.
>>  At the moment I'm using a custom board based on the ti816x so I don't need
>> kernel or u-boot (already have those working for the board and since the
>> kernel for this platform was never mainlined, it doesn't have much churn)
>> but simply need packages built using all the Cortex-A8 flags.  I guess I
>> could be lazy and just build for Beagleboard or something since the ABI's
>> would be the same but I thought I would ask how to go about it the right
>> way.  Also, there have been several revs of silicon for that part and I'm
>> about to acquire the most up-to-date EVM so I could add machine support for
>> that.****
>>
>> ** **
>>
>> I'll check out that commit you mentioned and start coming up to speed on
>> what is involved.****
>>
>> ** **
>>
>> [CEM] Sure, ping if you have questions J****
>>
>>
>>
> Will do.  In the mean time ... just to kick the can further down the road
> so I can get "something" built so I can try a rootfs for my system ... I
> did a fresh clone of yocto master, meta-ti master, and meta-openembedded
> master and set machine type to beaglebone.  Bitbake of core-image-minimal
> results in:
>
> hutch at neo:~/yocto_master_beaglebone/poky/build$ bitbake core-image-minimal
> WARNING: Host distribution "Debian-7.1" has not been validated with this
> version of the build system; you may possibly experience unexpected
> failures. It is recommended that you use a tested distribution.
> ERROR: This recipe does not have the LICENSE field set
> (ti-hw-bringup-image)
> ERROR: This recipe does not have the LICENSE field set (cloud9-gfx-image)
> ERROR: This recipe does not have the LICENSE field set (cloud9-image)
> ERROR: This recipe does not have the LICENSE field set (cloud9-gnome-image)
> ERROR: Unable to parse
> /home/hutch/yocto_master_beaglebone/poky/meta-ti/recipes-misc/images/
> ti-hw-bringup-image.bb: Exited with "1"
>                                                                          |
> ETA:  --:--:--
> ERROR: Command execution failed: Exited with 1
>
> Summary: There was 1 WARNING message shown.
> Summary: There were 6 ERROR messages shown, returning a non-zero exit code.
>
> Guess I need to do some searching through the archives to see what this
> new LICENSE stuff is all about.
>
> Regards,
>
> Brian
>
>
Ah, forgot BBMASK ... maybe that is it.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.yoctoproject.org/pipermail/meta-ti/attachments/20130717/321e6ae2/attachment.html>