Re: [meta-fsl-arm][PATCH v3 00/11] Freescale 3.10.17-1.0.0-beta - part 1

[Eric Nelson <eric.nelson@boundarydevices.com>]

Thanks Otavio,

On 01/25/2014 11:56 AM, Otavio Salvador wrote:
> Hello Eric,
>
> On Sat, Jan 25, 2014 at 1:02 AM, Eric Nelson
> <eric.nelson@boundarydevices.com> wrote:
>> On 01/24/2014 01:37 PM, Otavio Salvador wrote:
>>>
>>> Hello Eric,
>>>
>>> On Fri, Jan 24, 2014 at 3:18 PM, Eric Nelson
>>> <eric.nelson@boundarydevices.com
>>> <mailto:eric.nelson@boundarydevices.com>> wrote:
>>>   > On 01/23/2014 06:47 PM, Otavio Salvador wrote:
>>>   >> I did some small fixes that I found during test at O.S. Systems'
>>>   >> autobuilder and I dropped some fixes that were pushed to master (as
>>>   >> you've probably noticed from the e-mail to the mailing list).
>>>   >>
>>>   >> I am awaiting people's feedback for the patches as this upgrades
>>>   >> package versions and we should have a better test coverage.
>>>   >>
>>>   >> Please give a try in the 'master-next' branch and comment in anything
>>>   >> that could be improved in the patches. All comments are welcome...
>>>   >>
>>
>>>   >> <snip>
>>
>>>   >>
>>>   >
>>>   > I compile-tested the series against MACHINE="nitrogen6x" and
>>>   > MACHINE="nit6xlte" with target image "fsl-image-gui".
>>>
>>> Thank you.
>>>
>>>   > The only compile-time issue was with clocks.sh and the imx_test recipe
>>>   > as noted separately.
>>>
>>> Please confirm you have the current patchset; you may have used the v2
>>> were this was indeed the case. The v3,
>>> http://patchwork.openembedded.org/patch/65673/, should be fine.
>>>
>>
>> Maybe.
>>
>> I'm having trouble piecing this together, but I'm running master-next
>> directly.
>>
>> Unfortunately, the patch version gets dropped on the way into git
>> and I'm not seeing a V2 in either the ML archive or patchwork.
>>
>> I'm also not finding a "--- changed in V3" comment anywhere.
>
> Yes; it is a little bit hard for me to keep track of the patchset (as
> we usually do in U-Boot and like) as I've been using Gerrit to review
> those patches internally before putting them in the mailing list and
> patman requires also those extra header in the commit log.
>

Cool. So my testing was against the V3 patch set.

> I will try to think how to improve it for future submissions ...
>
> I keep master-next updated with the very last revision. I just did it
> again after pushing your patch.
>
>>>   > Run-time testing showed that the GPU binaries still need to be
>>>   > updated from 1_3.10.9-1.0.0-hfp-r0, so X wouldn't start.
>>>
>>> Hum this is not expected as I didn't include GPU changes on this. Can
>>> you send the Xorg.0.log?
>>>
>> I'll send this in the A.M.. I don't have access to it at the moment.
>>
>> This was expected though, right?
>>
>> We haven't been able to get even the frame-buffer versions of the
>> libraries to operate against the 3.10.9 libraries and presumed an
>> ABI break here.
>
> If you are running 3.10.17 this is expected as it uses p13 GPU
> version. Otherwise it ought to work just fine.
>

Yep. I was building 3.10.17-beta in preparation for submitting
patches for linux-boundary.

Let me know when you have a GPU patch. I'll be happy to test.

Regards,


Eric