[oe] [meta-browser][PATCH v2] chromium-wayland: update to 53.0.2785.14

Tue Jan 31 18:24:55 UTC 2017

Hi again,

Otavio already merged the v2 of the update patch (thanks!), so I'll send
the change we are discussing as a separate patch.

Best,
-- 
Jacobo Aragunde
Software Engineer at Igalia

On 31/01/17 09:56, Herve Jourdain wrote:
> Hi Jacobo, Khem,
> 
> I'm traveling, so not able to access my build system for a few more days.
> I didn't check Julien's patch yet, so as long as it does at least what mine does, it should be OK for me.
> I'm open to whatever solution, whether I regenerate a patch or Jacobo integrates it in his.
> One thing I will do, though, is test the resulting recipe on my build platform, to see if it's still working in my configuration.
> 
> Cheers,
> Herve
> 
>> On 31 Jan 2017, at 03:04, Jacobo Aragunde Pérez <jaragunde at igalia.com> wrote:
>>
>>> On 30/01/17 19:36, Khem Raj wrote:
>>> On Mon, Jan 30, 2017 at 10:33 AM, Jacobo Aragunde Pérez
>>> <jaragunde at igalia.com> wrote:
>>>> Hi Khem Raj,
>>>>
>>>> sorry, I didn't understand your response. Do you prefer I send a new
>>>> version of the rebase patch, including Herve's fix? In that case, Herve,
>>>> I think I could use jbriance's upstream patch so you don't need to
>>>> generate it, if you all agree. I'd use the one that didn't get merged
>>>> [1], because we need the .gyp part too.
>>>>
>>>> Future rebases will require changes to use GN, at that point we should
>>>> probably get rid of this custom patch. That's a problem for another day,
>>>> though :)
>>>
>>> I was thinking both of Herve and you will send your patches
>>> independently after rework but if you can integrate both that would be
>>> wonderful.
>>>
>>
>> I definitely can, it's easy enough by picking Julien's diff from
>> Chromium code review. I've updated my patch and just triggered a build
>> to make sure everything is in place, it will compile overnight. I just
>> don't want to "steal" Herve's idea, so I'll wait for his blessing :)
>>
>> Best,
>> -- 
>> Jacobo Aragunde
>> Software Engineer at Igalia
>>
>>>>
>>>> [1] https://codereview.chromium.org/2239833003/#ps1
>>>>
>>>> Thank you all,
>>>> --
>>>> Jacobo Aragunde
>>>> Software Engineer at Igalia
>>>>
>>>>
>>>>> On 29/01/17 02:13, Khem Raj wrote:
>>>>> Hi Herve
>>>>>
>>>>> Yes please send a V2 with suggested changes.
>>>>>
>>>>>> On Sat, Jan 28, 2017 at 7:05 AM, Herve Jourdain <herve.jourdain at neuf.fr> wrote:
>>>>>> Hi Khem,
>>>>>>
>>>>>> Shall I generate a proper patch with the proper header? Or should Jacobo add that to his patch?
>>>>>> I'm not even sure the patch is accepted on master yet.
>>>>>>
>>>>>> Cheers,
>>>>>> Herve
>>>>>>
>>>>>>> On 28 Jan 2017, at 01:51, Khem Raj <raj.khem at gmail.com> wrote:
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On 1/27/17 6:06 AM, Julien Brianceau (jbriance) wrote:
>>>>>>>>> -----Original Message-----
>>>>>>>>> From: openembedded-devel-bounces at lists.openembedded.org
>>>>>>>>> [mailto:openembedded-devel-bounces at lists.openembedded.org] On Behalf Of
>>>>>>>>> Khem Raj
>>>>>>>>> Sent: vendredi 27 janvier 2017 02:47
>>>>>>>>> To: openembedded-devel at lists.openembedded.org
>>>>>>>>> Subject: Re: [oe] [meta-browser][PATCH v2] chromium-wayland: update to
>>>>>>>>> 53.0.2785.14
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>> On 1/25/17 9:50 PM, Herve Jourdain wrote:
>>>>>>>>>> Hi Jacob,
>>>>>>>>>>
>>>>>>>>>> I've been able to compile chromium-wayland 53.0.2785.14 on
>>>>>>>>> RaspberryPi, with Wayland only (no X11), BUT I had to add an additional
>>>>>>>>> patch for it to work, or else I get an error in
>>>>>>>>> ui/events/keycodes/keyboard_code_conversion_xkb.cc.
>>>>>>>>>> I'm using VC4, which has a DRM/KMS interface (it seems that DRM
>>>>>>>>> interface is required by ozone-wayland), but this patch seems to not be
>>>>>>>>> related to RaspberryPi itself.
>>>>>>>>>> Basically, it seems that USE_XKBCOMMON is not defined when compiling
>>>>>>>>> keycodes.
>>>>>>>>>>
>>>>>>>>>> Is it something that you saw as well? Or do you have a different
>>>>>>>>> setting/experience?
>>>>>>>>>>
>>>>>>>>>> Just for reference, here is the patch I need (right now, I'm
>>>>>>>>> triggering it from meta-raspberrypi, but I believe it may be needed in
>>>>>>>>> meta-browser).
>>>>>>>>>>
>>>>>>>>>> --- a/ui/events/keycodes/events_keycodes.gyp    2017-01-23
>>>>>>>>> 19:12:01.767271256 +0800
>>>>>>>>>> +++ b/ui/events/keycodes/events_keycodes.gyp    2017-01-23
>>>>>>>>> 19:12:11.503048216 +0800
>>>>>>>>>> @@ -21,6 +21,9 @@
>>>>>>>>>>        'scoped_xkb.h',
>>>>>>>>>>        'xkb_keysym.h',
>>>>>>>>>>      ],
>>>>>>>>>> +      'defines': [
>>>>>>>>>> +        'USE_XKBCOMMON',
>>>>>>>>>
>>>>>>>>> This seems fine to me. It should also be proposed upstream as well I
>>>>>>>>> think.
>>>>>>>>>
>>>>>>>>
>>>>>>>> For the record, I've already tried to upstream a similar change (see Patch Set 1 of review https://codereview.chromium.org/2239833003/), but only the GN part has been accepted as gyp support has been removed in Chromium 54.
>>>>>>>>
>>>>>>>
>>>>>>> thats we should make a note of this in patch header. We will have to
>>>>>>> remove it or replace it as well when we upgrade chromium.
>>>>>>>> Julien
>>>>>>>>
>> -- 
>> _______________________________________________
>> Openembedded-devel mailing list
>> Openembedded-devel at lists.openembedded.org
>> http://lists.openembedded.org/mailman/listinfo/openembedded-devel
> 