Re: [Angstrom-devel] [PATCH] set default tune for Overo to cortexa8hf-neon

2013-05-28 Thread Khem Raj

On May 28, 2013, at 10:42 PM, Andreas Müller  
wrote:

> On Wed, May 29, 2013 at 3:20 AM, Khem Raj  wrote:
>> 
>> On May 26, 2013, at 2:41 PM, Khem Raj  wrote:
>> 
>>> On Thu, May 23, 2013 at 12:36 AM, Andreas Müller
>>>  wrote:
 On Thu, May 23, 2013 at 9:17 AM, Khem Raj  wrote:
> 
> On May 22, 2013, at 8:16 AM, Cliff Brake  wrote:
> 
>> there is a missing DEPENDS to gconf.  After I built gconf, I get:
>> 
>> https://gist.github.com/cbrake/5628386
>> 
> 
> 
> well I think this here
> http://src.chromium.org/viewvc/chrome?revision=135388&view=revision
> 
 Thanks for the hint - I will look into that (am working on chrome
 anyway - for gcc 4.8 fix).
 
 Still I would like to have tests on my images with hardfp before
 applying - I don't want to open another WIP with unpredictable
 efforts.
>>> 
>>> Please test out.
>>> 
>>> http://patchwork.openembedded.org/patch/50553/
>>> http://patchwork.openembedded.org/patch/50555/
>>> 
>>> I have compile tested the new chromium v29 recipe on 
>>> hardfp/beaglebone/gcc-4.8
>>> 
>>> may be we should get rid of older chromium recipes in there but that can be
>>> done once v29 works out ok.
>>> 
>>> Let me know if you run into issues.
>> 
>> 
>> 
>> Andreas, Cliff
>> 
>> Now that above patches are applied to meta-browser, can one of you test out 
>> the hardfp build on overo ? I don't have hardware myself to try it out.
>> I would like to pull the above patch in question into angstrom.
>> 
> I have run tested most of my images with this patch applied - I don't
> see any issues. Especially chromium 29.x builds (also for gcc 4.8) and
> runs fine - thanks for taking care.
> 
> The only issue I see after this patch is broken TI's libgles. Since
> this is useless for many years anyway (just running demos at
> predefined size and 'mouse eating' is not what I expect when I ask for
> gles/egl support), we can take this patch in. I don't expect a working
> gles support from TI anyway.
> 

Thanks, patch is installed.


___
Angstrom-distro-devel mailing list
Angstrom-distro-devel@linuxtogo.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/angstrom-distro-devel


Re: [Angstrom-devel] [PATCH] set default tune for Overo to cortexa8hf-neon

2013-05-28 Thread Andreas Müller
On Wed, May 29, 2013 at 3:20 AM, Khem Raj  wrote:
>
> On May 26, 2013, at 2:41 PM, Khem Raj  wrote:
>
>> On Thu, May 23, 2013 at 12:36 AM, Andreas Müller
>>  wrote:
>>> On Thu, May 23, 2013 at 9:17 AM, Khem Raj  wrote:

 On May 22, 2013, at 8:16 AM, Cliff Brake  wrote:

> there is a missing DEPENDS to gconf.  After I built gconf, I get:
>
> https://gist.github.com/cbrake/5628386
>


 well I think this here
 http://src.chromium.org/viewvc/chrome?revision=135388&view=revision

>>> Thanks for the hint - I will look into that (am working on chrome
>>> anyway - for gcc 4.8 fix).
>>>
>>> Still I would like to have tests on my images with hardfp before
>>> applying - I don't want to open another WIP with unpredictable
>>> efforts.
>>
>> Please test out.
>>
>> http://patchwork.openembedded.org/patch/50553/
>> http://patchwork.openembedded.org/patch/50555/
>>
>> I have compile tested the new chromium v29 recipe on 
>> hardfp/beaglebone/gcc-4.8
>>
>> may be we should get rid of older chromium recipes in there but that can be
>> done once v29 works out ok.
>>
>> Let me know if you run into issues.
>
>
>
> Andreas, Cliff
>
> Now that above patches are applied to meta-browser, can one of you test out 
> the hardfp build on overo ? I don't have hardware myself to try it out.
> I would like to pull the above patch in question into angstrom.
>
I have run tested most of my images with this patch applied - I don't
see any issues. Especially chromium 29.x builds (also for gcc 4.8) and
runs fine - thanks for taking care.

The only issue I see after this patch is broken TI's libgles. Since
this is useless for many years anyway (just running demos at
predefined size and 'mouse eating' is not what I expect when I ask for
gles/egl support), we can take this patch in. I don't expect a working
gles support from TI anyway.

Andreas

___
Angstrom-distro-devel mailing list
Angstrom-distro-devel@linuxtogo.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/angstrom-distro-devel


Re: [Angstrom-devel] [PATCH] set default tune for Overo to cortexa8hf-neon

2013-05-28 Thread Khem Raj

On May 26, 2013, at 2:41 PM, Khem Raj  wrote:

> On Thu, May 23, 2013 at 12:36 AM, Andreas Müller
>  wrote:
>> On Thu, May 23, 2013 at 9:17 AM, Khem Raj  wrote:
>>> 
>>> On May 22, 2013, at 8:16 AM, Cliff Brake  wrote:
>>> 
 there is a missing DEPENDS to gconf.  After I built gconf, I get:
 
 https://gist.github.com/cbrake/5628386
 
>>> 
>>> 
>>> well I think this here
>>> http://src.chromium.org/viewvc/chrome?revision=135388&view=revision
>>> 
>> Thanks for the hint - I will look into that (am working on chrome
>> anyway - for gcc 4.8 fix).
>> 
>> Still I would like to have tests on my images with hardfp before
>> applying - I don't want to open another WIP with unpredictable
>> efforts.
> 
> Please test out.
> 
> http://patchwork.openembedded.org/patch/50553/
> http://patchwork.openembedded.org/patch/50555/
> 
> I have compile tested the new chromium v29 recipe on hardfp/beaglebone/gcc-4.8
> 
> may be we should get rid of older chromium recipes in there but that can be
> done once v29 works out ok.
> 
> Let me know if you run into issues.



Andreas, Cliff

Now that above patches are applied to meta-browser, can one of you test out the 
hardfp build on overo ? I don't have hardware myself to try it out.
I would like to pull the above patch in question into angstrom.

-Khem


___
Angstrom-distro-devel mailing list
Angstrom-distro-devel@linuxtogo.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/angstrom-distro-devel


Re: [Angstrom-devel] [PATCH] set default tune for Overo to cortexa8hf-neon

2013-05-26 Thread Khem Raj
On Thu, May 23, 2013 at 12:36 AM, Andreas Müller
 wrote:
> On Thu, May 23, 2013 at 9:17 AM, Khem Raj  wrote:
>>
>> On May 22, 2013, at 8:16 AM, Cliff Brake  wrote:
>>
>>> there is a missing DEPENDS to gconf.  After I built gconf, I get:
>>>
>>> https://gist.github.com/cbrake/5628386
>>>
>>
>>
>> well I think this here
>> http://src.chromium.org/viewvc/chrome?revision=135388&view=revision
>>
> Thanks for the hint - I will look into that (am working on chrome
> anyway - for gcc 4.8 fix).
>
> Still I would like to have tests on my images with hardfp before
> applying - I don't want to open another WIP with unpredictable
> efforts.

Please test out.

http://patchwork.openembedded.org/patch/50553/
http://patchwork.openembedded.org/patch/50555/

I have compile tested the new chromium v29 recipe on hardfp/beaglebone/gcc-4.8

may be we should get rid of older chromium recipes in there but that can be
done once v29 works out ok.

Let me know if you run into issues.

___
Angstrom-distro-devel mailing list
Angstrom-distro-devel@linuxtogo.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/angstrom-distro-devel


Re: [Angstrom-devel] [PATCH] set default tune for Overo to cortexa8hf-neon

2013-05-23 Thread Andreas Müller
On Thu, May 23, 2013 at 9:17 AM, Khem Raj  wrote:
>
> On May 22, 2013, at 8:16 AM, Cliff Brake  wrote:
>
>> there is a missing DEPENDS to gconf.  After I built gconf, I get:
>>
>> https://gist.github.com/cbrake/5628386
>>
>
>
> well I think this here
> http://src.chromium.org/viewvc/chrome?revision=135388&view=revision
>
Thanks for the hint - I will look into that (am working on chrome
anyway - for gcc 4.8 fix).

Still I would like to have tests on my images with hardfp before
applying - I don't want to open another WIP with unpredictable
efforts.

Andreas

___
Angstrom-distro-devel mailing list
Angstrom-distro-devel@linuxtogo.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/angstrom-distro-devel


Re: [Angstrom-devel] [PATCH] set default tune for Overo to cortexa8hf-neon

2013-05-23 Thread Khem Raj

On May 22, 2013, at 8:16 AM, Cliff Brake  wrote:

> there is a missing DEPENDS to gconf.  After I built gconf, I get:
> 
> https://gist.github.com/cbrake/5628386
> 


well I think this here 
http://src.chromium.org/viewvc/chrome?revision=135388&view=revision

should have fixed it.

it should be fixed in recipe how float ABI is passed to chromium build 

> Thanks,
> Cliff
> 
> 
> 
> On Wed, May 22, 2013 at 10:44 AM, Andreas Müller <
> schnitzelt...@googlemail.com> wrote:
> 
>> On Wed, May 22, 2013 at 4:37 PM, Cliff Brake 
>> wrote:
>>> From: Cliff Brake 
>>> 
>>> Signed-off-by: Cliff Brake 
>>> ---
>>> conf/distro/include/arm-defaults.inc | 1 +
>>> 1 file changed, 1 insertion(+)
>>> 
>>> diff --git a/conf/distro/include/arm-defaults.inc
>> b/conf/distro/include/arm-defaults.inc
>>> index 6e29332..89ed530 100644
>>> --- a/conf/distro/include/arm-defaults.inc
>>> +++ b/conf/distro/include/arm-defaults.inc
>>> @@ -1,2 +1,3 @@
>>> DEFAULTTUNE_beagleboard = "cortexa8hf-neon"
>>> DEFAULTTUNE_beaglebone = "cortexa8hf-neon"
>>> +DEFAULTTUNE_overo = "cortexa8hf-neon"
>>> --
>>> 1.8.2.3
>>> 
>>> 
>> I got the information (never tested by myself) that hardfp causes
>> chromium build to fail. Can you test that, because I would like to
>> keep that working.
>> 
>> Andreas
>> 
>> ___
>> Angstrom-distro-devel mailing list
>> Angstrom-distro-devel@linuxtogo.org
>> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/angstrom-distro-devel
>> 
> ___
> Angstrom-distro-devel mailing list
> Angstrom-distro-devel@linuxtogo.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/angstrom-distro-devel


___
Angstrom-distro-devel mailing list
Angstrom-distro-devel@linuxtogo.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/angstrom-distro-devel


Re: [Angstrom-devel] [PATCH] set default tune for Overo to cortexa8hf-neon

2013-05-22 Thread Cliff Brake
I had no concrete reason, other than to emulate the bb builds.

Thanks,
Cliff


On Wed, May 22, 2013 at 11:41 AM, Andreas Müller <
schnitzelt...@googlemail.com> wrote:

> On Wed, May 22, 2013 at 5:16 PM, Cliff Brake 
> wrote:
> > there is a missing DEPENDS to gconf.  After I built gconf, I get:
> >
> > https://gist.github.com/cbrake/5628386
> >
> > Thanks,
> > Cliff
> Can we leave softfp until it works then. AFAIK there is no measurable
> performance increase to expect. Or do you have other reasons why we
> should change?
>
> Andreas
>
> ___
> Angstrom-distro-devel mailing list
> Angstrom-distro-devel@linuxtogo.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/angstrom-distro-devel
>



-- 
=
http://bec-systems.com
___
Angstrom-distro-devel mailing list
Angstrom-distro-devel@linuxtogo.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/angstrom-distro-devel


Re: [Angstrom-devel] [PATCH] set default tune for Overo to cortexa8hf-neon

2013-05-22 Thread Andreas Müller
On Wed, May 22, 2013 at 5:16 PM, Cliff Brake  wrote:
> there is a missing DEPENDS to gconf.  After I built gconf, I get:
>
> https://gist.github.com/cbrake/5628386
>
> Thanks,
> Cliff
Can we leave softfp until it works then. AFAIK there is no measurable
performance increase to expect. Or do you have other reasons why we
should change?

Andreas

___
Angstrom-distro-devel mailing list
Angstrom-distro-devel@linuxtogo.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/angstrom-distro-devel


Re: [Angstrom-devel] [PATCH] set default tune for Overo to cortexa8hf-neon

2013-05-22 Thread Cliff Brake
there is a missing DEPENDS to gconf.  After I built gconf, I get:

https://gist.github.com/cbrake/5628386

Thanks,
Cliff



On Wed, May 22, 2013 at 10:44 AM, Andreas Müller <
schnitzelt...@googlemail.com> wrote:

> On Wed, May 22, 2013 at 4:37 PM, Cliff Brake 
> wrote:
> > From: Cliff Brake 
> >
> > Signed-off-by: Cliff Brake 
> > ---
> >  conf/distro/include/arm-defaults.inc | 1 +
> >  1 file changed, 1 insertion(+)
> >
> > diff --git a/conf/distro/include/arm-defaults.inc
> b/conf/distro/include/arm-defaults.inc
> > index 6e29332..89ed530 100644
> > --- a/conf/distro/include/arm-defaults.inc
> > +++ b/conf/distro/include/arm-defaults.inc
> > @@ -1,2 +1,3 @@
> >  DEFAULTTUNE_beagleboard = "cortexa8hf-neon"
> >  DEFAULTTUNE_beaglebone = "cortexa8hf-neon"
> > +DEFAULTTUNE_overo = "cortexa8hf-neon"
> > --
> > 1.8.2.3
> >
> >
> I got the information (never tested by myself) that hardfp causes
> chromium build to fail. Can you test that, because I would like to
> keep that working.
>
> Andreas
>
> ___
> Angstrom-distro-devel mailing list
> Angstrom-distro-devel@linuxtogo.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/angstrom-distro-devel
>
___
Angstrom-distro-devel mailing list
Angstrom-distro-devel@linuxtogo.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/angstrom-distro-devel


Re: [Angstrom-devel] [PATCH] set default tune for Overo to cortexa8hf-neon

2013-05-22 Thread Andreas Müller
On Wed, May 22, 2013 at 4:37 PM, Cliff Brake  wrote:
> From: Cliff Brake 
>
> Signed-off-by: Cliff Brake 
> ---
>  conf/distro/include/arm-defaults.inc | 1 +
>  1 file changed, 1 insertion(+)
>
> diff --git a/conf/distro/include/arm-defaults.inc 
> b/conf/distro/include/arm-defaults.inc
> index 6e29332..89ed530 100644
> --- a/conf/distro/include/arm-defaults.inc
> +++ b/conf/distro/include/arm-defaults.inc
> @@ -1,2 +1,3 @@
>  DEFAULTTUNE_beagleboard = "cortexa8hf-neon"
>  DEFAULTTUNE_beaglebone = "cortexa8hf-neon"
> +DEFAULTTUNE_overo = "cortexa8hf-neon"
> --
> 1.8.2.3
>
>
I got the information (never tested by myself) that hardfp causes
chromium build to fail. Can you test that, because I would like to
keep that working.

Andreas

___
Angstrom-distro-devel mailing list
Angstrom-distro-devel@linuxtogo.org
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/angstrom-distro-devel