Hi Sangwook,
On 08/20/2012 12:43 PM, Sangwook Lee wrote:
On 20 August 2012 09:12, Sangwook Lee sangwook.lee@linaro.org wrote:
On 19 August 2012 22:29, Sylwester Nawrocki sylvester.nawrocki@gmail.com wrote:
On 08/03/2012 04:24 PM, Sangwook Lee wrote:
I was thinking about this, but this seems to be is a bit time-consuming because I have to do this just due to lack of s5k4ecgx hardware information. let me try it later once this patch is accepted.
I've converted this driver to use function calls instead of the register arrays. It can be pulled, along with a couple of minor fixes/improvements, from following git tree:
git://linuxtv.org/snawrocki/media.git s5k4ecgx (gitweb: http://git.linuxtv.org/snawrocki/media.git/s5k4ecgx)I don't own any Origen board thus it's untested. Could you give it a try ?
Sorry, It doesn't work. I will send pictures to you by another mail thread. Previously, I tested preview array and found out that
- /*
* FIXME: according to the datasheet,* 0x70000496~ 0x7000049c seems to be only for capture mode,* but without these value, it doesn't work with preview mode.*/Do we need to set those values ?
Yes, after my changes it should be set too. But there is a bug in the registers definitions. Attached patch should correct this, sorry about this oversight. Let me know if there are still any issues. To make sure the I2C write sequences are correct it might be useful to log all calls to s5k4ecgx_write() and compare the logs with original tables.
--
Regards, Sylwester