Hi, On Mon, Jun 1, 2009 at 4:47 PM, Yusuf Caglar AKYUZ <caglarak...@gmail.com> wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Markus Rechberger wrote: > > [...] > >> 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. >> > > AFAIR all the DVSDKs I saw were compiled with EABI toolchain. In > fact MV toolchain was eabi. But I don't know for sure. > > [...] >> >> 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. >> > > If you codec's CS line is connected to a GPIO on your board then you > can control that GPIO from user space. I guess this is valid for i2c > expanders as well. For instance to control GPIO 5: > > $cd /sys/class/gpio > $echo 5 > export > $cd gpio5 > $echo out > direction > $echo 1/0 > value >
interesting didn't know about the sysfs interface, I wrote a kernel module for this before.. Although it did not work either; no "visible" change in the i2c map. >> 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: -- -- -- -- -- -- -- -- >> > > They seem consistent. I guess you have another chip(max7310???) > which is addressed via GPIO. I don't think anything there is > anything wrong with I2C setup, other than chip addresses. > as far as I have seen there's no max* chip on it. I asked the company for the linux sourcecode, they told me that they will prepare it... let's see if it's true. Thanks for your help! best regards, Markus _______________________________________________ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
