On Mon, Jun 1, 2009 at 5:14 PM, Yusuf Caglar AKYUZ
<caglarak...@gmail.com> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Markus Rechberger wrote:
>
> [...]
>
>> I got the dvsdk already work with a newer compiler although, my
>> problem is that the target system
>> uses the Ingenient Codecs and was built with OABI/EABI 0 and Linux
>> 2.6.18. In order to be able to
>> use the DVSDK and the gstreamer_ti framework I would need the cmem and
>> dsplinkk driver for the# same system as well as the userspace
>> libraries with OABI/EABI0.
>> The main reason why I cannot change the base system is the audio driver...
>>
>
> So you want to run EABI binaries in an OABI environment? To my
> understanding this is not possible...
>
I know it's not possible.. that's why I'd need DVSDK OABI binaries ..
is there any way to get
those? As for the kernel modules I saw the uClib objects for DVSDK 1.3
are in the OABI format,
but the other libraries are in EABI4 format, and the system which I
got is not compiled for EABI4.
>> The original delivered box shows:
>> bound device '0-001b' to driver 'OMAP+TLV320AIC23 codec'
>> bound device 'audio-i2c' to driver 'audio-i2c'
>> bound device '0-001a' to driver 'WM8955L I2C Driver'
>> bound device 'davinci-audio.0' to driver 'davinci-audio'
>>
>> but with the DVSDK the i2c map shows:
>> 0 1 2 3 4 5 6 7 8 9 a b c d e f
>> 00: -- -- -- -- -- -- -- 0a -- -- -- -- --
>> 10: -- -- -- -- -- -- -- -- -- -- 1a -- -- -- -- --
>> 20: -- -- -- 23 -- -- -- -- -- -- -- -- -- -- -- --
>> 30: -- -- -- -- -- -- -- -- -- 39 UU -- -- 3d -- --
>> 40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
>> 50: 50 51 52 53 54 55 56 57 -- -- -- -- -- -- -- --
>> 60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 6f
>> 70: -- -- -- -- -- -- -- --
>>
>> so the audio codec doesn't seem to be visible on the i2c bus
>> The WM8955L specs are freely available, writing a driver for it wouldn't
>> be a problem, although I'm concerned about how to enable the AIC23
>> codec on it...
>>
>
> I think both have the same address, 0x1a. Looking at aic23
> datasheet, its address can be 0x1a or 0x1b depending on the CS pin.
> Any change that CS pin is software controlled on your board and low
> currently?
>
yes I saw that too but I'm confused about the WM8955L driver sitting
on 0x1a already
bound device '0-001a' to driver 'WM8955L I2C Driver'
bound device 'davinci-audio.0' to driver 'davinci-audio'
Do you know such components can be enabled/disabled on the dm6446? I
tried to play around
with the GPIO configuration and i2c gpio expander on 0x3a of the
dm6446 initcode but without success.
I also tested the latest kernel with that board, there's a difference
in the i2c map between
2.6.18mvl and the latest git kernel:
2.6.18mvl:
0 1 2 3 4 5 6 7 8 9 a b c d e f
00: -- -- -- -- -- -- -- 0a -- -- -- -- --
10: -- -- -- -- -- -- -- -- -- -- 1a -- -- -- -- --
20: -- -- -- 23 -- -- -- -- -- -- -- -- -- -- -- --
30: -- -- -- -- -- -- -- -- -- 39 UU -- -- 3d -- --
40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
50: 50 51 52 53 54 55 56 57 -- -- -- -- -- -- -- --
60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 6f
70: -- -- -- -- -- -- -- --
latest git kernel:
0 1 2 3 4 5 6 7 8 9 a b c d e f
00: -- -- -- -- -- 08 -- -- -- -- -- -- --
10: -- -- -- -- -- -- -- -- -- -- 1a -- -- -- -- --
20: -- -- -- UU -- -- -- -- -- -- -- -- -- -- -- --
30: -- -- -- -- -- -- -- -- -- UU -- -- -- 3d -- --
40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
50: UU 51 52 53 54 55 56 57 -- -- -- -- -- -- -- --
60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 6f
70: -- -- -- -- -- -- -- --
thanks for your response!
Markus
_______________________________________________
Davinci-linux-open-source mailing list
Davinci-linux-open-source@linux.davincidsp.com
http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
