dm3730

dm3730

[Raffaele Recalcati]

I'm sorry for so easy question!
It seems that if I compiling for beagleboard I get a 2.6.39 running
also on dm3730 cpu.
It's not clear to me how can dsp codec work correctly at the same time
with omap3530 and with dm3730.

Thx,
Raffaele

Re: dm3730

[Kridner, Jason]

On Nov 19, 2011, at 3:07 AM, "Raffaele Recalcati" <lamiaposta71@gmail.com> wrote:

> I'm sorry for so easy question!
> It seems that if I compiling for beagleboard I get a 2.6.39 running
> also on dm3730 cpu.

I recommend moving on to OE-core and meta-ti if you are doing new development. That is 3.0 today, but I expect we will bring it in-line with the BeagleBone kernel. 

> It's not clear to me how can dsp codec work correctly at the same time
> with omap3530 and with dm3730.

Same DSP?  There are fairly few differences between those two devices.  Mostly a process shrink and incremental improvements. 

> 
> Thx,
> Raffaele
> 
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel

Re: dm3730

[Enrico]

On Sat, Nov 19, 2011 at 9:06 AM, Raffaele Recalcati
<lamiaposta71@gmail.com> wrote:
> I'm sorry for so easy question!
> It seems that if I compiling for beagleboard I get a 2.6.39 running
> also on dm3730 cpu.
> It's not clear to me how can dsp codec work correctly at the same time
> with omap3530 and with dm3730.

If i'm not wrong the dsp is the same, just running at a higher frequency.

Anyway i use the same codec server (cs.x64P) on omap3530 and dm3730
and it works correctly, i tested h264enc and mpeg4enc.

Enrico

Re: dm3730

[Raffaele Recalcati]

On Sat, Nov 19, 2011 at 12:51 PM, Kridner, Jason <jdk@ti.com> wrote:
> On Nov 19, 2011, at 3:07 AM, "Raffaele Recalcati" <lamiaposta71@gmail.com> wrote:
>
>> I'm sorry for so easy question!
>> It seems that if I compiling for beagleboard I get a 2.6.39 running
>> also on dm3730 cpu.
>
> I recommend moving on to OE-core and meta-ti if you are doing new development. That is 3.0 today, but I expect we will bring it in-line with the BeagleBone kernel.

Thx,
monday I'll try.
If I'll have some problems I'll ask.

>> It's not clear to me how can dsp codec work correctly at the same time
>> with omap3530 and with dm3730.
>
> Same DSP?  There are fairly few differences between those two devices.  Mostly a process shrink and incremental improvements.

Nice.

Raffaele

Re: dm3730

[Raffaele Recalcati]

Hi,

On Sat, Nov 19, 2011 at 12:51 PM, Kridner, Jason <jdk@ti.com> wrote:
> On Nov 19, 2011, at 3:07 AM, "Raffaele Recalcati" <lamiaposta71@gmail.com> wrote:
>
>> I'm sorry for so easy question!
>> It seems that if I compiling for beagleboard I get a 2.6.39 running
>> also on dm3730 cpu.
>
> I recommend moving on to OE-core and meta-ti if you are doing new development. That is 3.0 > today, but I expect we will bring it in-line with the BeagleBone kernel.

It's ok.
I have the new monster to compile.
I have followed https://wiki.yoctoproject.org/wiki/OpenEmbedded-Core
setup and started, without changing anything, a compilation of
'bitbake core-image-minimal', to test my setup.

Waiting it ends can I go on preparing the setup for beagle-xm with
GLES and gstreamer-ti running?
Is maybe the right starting point
https://github.com/koenkooi/Angstrom-integration-layer.git?

Maybe at http://beagleboard.org/esc/boston2011/beagleboard101 there
should be a presentation about that, but the link is broken.

Thx,
Raffaele



>> It's not clear to me how can dsp codec work correctly at the same time
>> with omap3530 and with dm3730.
>
> Same DSP?  There are fairly few differences between those two devices.  Mostly a process shrink and incremental improvements.
>
>>
>> Thx,
>> Raffaele
>>
>> _______________________________________________
>> Openembedded-devel mailing list
>> Openembedded-devel@lists.openembedded.org
>> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
>
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
>



-- 
www.opensurf.it

Re: dm3730

[Koen Kooi]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Op 21-11-11 16:38, Raffaele Recalcati schreef:
> Hi,
> 
> On Sat, Nov 19, 2011 at 12:51 PM, Kridner, Jason <jdk@ti.com> wrote:
>> On Nov 19, 2011, at 3:07 AM, "Raffaele Recalcati"
>> <lamiaposta71@gmail.com> wrote:
>> 
>>> I'm sorry for so easy question! It seems that if I compiling for
>>> beagleboard I get a 2.6.39 running also on dm3730 cpu.
>> 
>> I recommend moving on to OE-core and meta-ti if you are doing new
>> development. That is 3.0 > today, but I expect we will bring it in-line
>> with the BeagleBone kernel.
> 
> It's ok. I have the new monster to compile. I have followed
> https://wiki.yoctoproject.org/wiki/OpenEmbedded-Core setup and started,
> without changing anything, a compilation of 'bitbake core-image-minimal',
> to test my setup.
> 
> Waiting it ends can I go on preparing the setup for beagle-xm with GLES
> and gstreamer-ti running?

No, that has no beagleboard support at all, only qemu machines. Worse, if
you follow the offical yocto instructions you'll get only broken and/or
incomplete beagleboard support.

> Is maybe the right starting point 
> https://github.com/koenkooi/Angstrom-integration-layer.git?

How about just going to the angstrom website, clicking on 'developers' and
following those instructions? That's what the README in the meta-ti layers
says you should do.






-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (Darwin)
Comment: GPGTools - http://gpgtools.org

iEYEARECAAYFAk7Kd/QACgkQMkyGM64RGpFMnACfQf8/o4rvpqSN4d775u17wuT0
1R8AnRSZwWTeNCuI3ZdxpAZMnkOL+bO0
=BzrX
-----END PGP SIGNATURE-----

Re: dm3730

[Eric Bénard]

Hi,

Le 21/11/2011 16:38, Raffaele Recalcati a écrit :
> It's ok.
> I have the new monster to compile.
> I have followed https://wiki.yoctoproject.org/wiki/OpenEmbedded-Core
> setup and started, without changing anything, a compilation of
> 'bitbake core-image-minimal', to test my setup.
>
> Waiting it ends can I go on preparing the setup for beagle-xm with
> GLES and gstreamer-ti running?
> Is maybe the right starting point
> https://github.com/koenkooi/Angstrom-integration-layer.git?
>
http://lists.linuxtogo.org/pipermail/openembedded-devel/2011-November/036249.html

Eric

Re: dm3730

[Paul Eggleton]

On Monday 21 November 2011 17:10:30 Koen Kooi wrote:
> Worse, if you follow the offical yocto instructions you'll get only broken
> and/or incomplete beagleboard support.

Rather than vague assertions perhaps you would care to elaborate on what's 
broken and/or incomplete about our beagleboard support? If the implication is 
that it won't work at all, then that is simply not true.

Cheers,
Paul

-- 

Paul Eggleton
Intel Open Source Technology Centre

Re: dm3730

[Koen Kooi]

[-- Attachment #1: Type: text/plain, Size: 813 bytes --]


Op 21 nov. 2011, om 17:47 heeft Paul Eggleton het volgende geschreven:

> On Monday 21 November 2011 17:10:30 Koen Kooi wrote:
>> Worse, if you follow the offical yocto instructions you'll get only broken
>> and/or incomplete beagleboard support.
> 
> Rather than vague assertions perhaps you would care to elaborate on what's 
> broken and/or incomplete about our beagleboard support? If the implication is 
> that it won't work at all, then that is simply not true.

For starters:

* no DSP support
* no SGX support
* no LCD support
* incomplete expansionboard support
* broken powermanagement

So if you only need serial port and audio yocto is all you need. Given that this thread was about DSP support the implication that the yocto support is broken and/or incomplete still stands.



[-- Attachment #2: Message signed with OpenPGP using GPGMail --]
[-- Type: application/pgp-signature, Size: 169 bytes --]

Re: dm3730

[Koen Kooi]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Op 21-11-11 17:47, Paul Eggleton schreef:
> On Monday 21 November 2011 17:10:30 Koen Kooi wrote:
>> Worse, if you follow the offical yocto instructions you'll get only
>> broken and/or incomplete beagleboard support.
> 
> Rather than vague assertions perhaps you would care to elaborate on
> what's broken and/or incomplete about our beagleboard support?

I hope the irony of yocto people asking other people what their own layers
actually support isn't lost on the people involved.





-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (Darwin)
Comment: GPGTools - http://gpgtools.org

iEYEARECAAYFAk7Kg30ACgkQMkyGM64RGpF5sgCgkUco7Wj80qUCnamqLz7FU5+M
bWEAnjtm7MTJypCoPalBO0IEOkbH8j7W
=NC/u
-----END PGP SIGNATURE-----

Re: dm3730

[Paul Eggleton]

On Monday 21 November 2011 17:59:41 Koen Kooi wrote:
> Op 21-11-11 17:47, Paul Eggleton schreef:
> > On Monday 21 November 2011 17:10:30 Koen Kooi wrote:
> >> Worse, if you follow the offical yocto instructions you'll get only
> >> broken and/or incomplete beagleboard support.
> > 
> > Rather than vague assertions perhaps you would care to elaborate on
> > what's broken and/or incomplete about our beagleboard support?
> 
> I hope the irony of yocto people asking other people what their own layers
> actually support isn't lost on the people involved.

There's no irony at all. I personally am not an expert on the beagleboard. 
Given that you *are* however, perhaps you might have simply said - as you 
point out, in the context of this discussion - that the Yocto project doesn't 
include DSP support for the beagleboard? That would have been enough. Almost 
stating that it's "broken" is going too far, in my opinion.

Cheers,
Paul

-- 

Paul Eggleton
Intel Open Source Technology Centre

Re: dm3730

[Peter Bigot]

On Mon, Nov 21, 2011 at 11:06 AM, Paul Eggleton
<paul.eggleton@linux.intel.com> wrote:
> On Monday 21 November 2011 17:59:41 Koen Kooi wrote:
>> Op 21-11-11 17:47, Paul Eggleton schreef:
>> > On Monday 21 November 2011 17:10:30 Koen Kooi wrote:
>> >> Worse, if you follow the offical yocto instructions you'll get only
>> >> broken and/or incomplete beagleboard support.
>> >
>> > Rather than vague assertions perhaps you would care to elaborate on
>> > what's broken and/or incomplete about our beagleboard support?
>>
>> I hope the irony of yocto people asking other people what their own layers
>> actually support isn't lost on the people involved.
>
> There's no irony at all. I personally am not an expert on the beagleboard.
> Given that you *are* however, perhaps you might have simply said - as you
> point out, in the context of this discussion - that the Yocto project doesn't
> include DSP support for the beagleboard?

oe-core+meta-ti will not support dsp for new developers either, at
least not until ti-cgt6x is updated to correspond to the
ti_cgt_c6000_6.1.17_setup_linux_x86.bin currently available from TI,
or somebody makes the old binary available.

Peter

Re: dm3730

[Raffaele Recalcati]

Hi,
it seems really better the oe-core architecture.

> oe-core+meta-ti will not support dsp for new developers either, at
> least not until ti-cgt6x is updated to correspond to the
> ti_cgt_c6000_6.1.17_setup_linux_x86.bin currently available from TI,
> or somebody makes the old binary available.

I hope it will be fixed.
I'm starting now and I find this message:

MACHINE=beagleboard ./oebb.sh bitbake virtual/kernel
..
Layers present in repository:
openembedded-core/meta-skeleton
openembedded-core/meta
..
Using /home/sources/.oe/environment-oecore to setup needed variables.
It is recommended to do '. /home/sources/.oe/environment-oecore' and
run 'bitbake something' without using ./oebb.sh as wrapper

I can not consider this message?

Thx,
Raffaele

Re: dm3730

[Raffaele Recalcati]

On Tue, Nov 22, 2011 at 8:52 AM, Raffaele Recalcati
<lamiaposta71@gmail.com> wrote:
> Hi,
> it seems really better the oe-core architecture.
>
>> oe-core+meta-ti will not support dsp for new developers either, at
>> least not until ti-cgt6x is updated to correspond to the
>> ti_cgt_c6000_6.1.17_setup_linux_x86.bin currently available from TI,
>> or somebody makes the old binary available.
>
> I hope it will be fixed.
> I'm starting now and I find this message:
>
> MACHINE=beagleboard ./oebb.sh bitbake virtual/kernel
> ..
> Layers present in repository:
> openembedded-core/meta-skeleton
> openembedded-core/meta
> ..
> Using /home/sources/.oe/environment-oecore to setup needed variables.
> It is recommended to do '. /home/sources/.oe/environment-oecore' and
> run 'bitbake something' without using ./oebb.sh as wrapper
>
> I can not consider this message?
>
> Thx,
> Raffaele
>

Finally I have 3.0.10+ kernel running on beagleboard.
Thank you very much.
Now I'm trying to find the way to have qt+GLES and gstreamer-ti running.

Raffaele