В Wed, 05 Mar 2014 10:43:42 +0100 Clemens Ladisch clemens@ladisch.de пишет:

Hi Roman,

I backporting the bugfixes in your patches to the stable kernels. However, I'm unable to see what was changed to enable the headphone outputs. What particular thing did the old driver wrong?

Regards, Clemens

Hi Clemens,

The most important thing was to switch the capture input to the I2S channel 1 (The patch #13). Read the CS4245 datasheet carefully: 4.11 "The MUTEC pin becomes active during power-up initialization, reset, and muting if the MCLK2 to LRCK2 ratio is incorrect in Asynchronous Mode or the MCLK1 to LRCK2 ratio is incorrect in Synchronous Mode, and during power-down."

Also there was other minor fixes. I had problems when I wrote values to the CS4245 SPI memory at the starting point (had no any effect), and I rewrote the SPI communication functions. Was there any bug or the CS4245 was just muted - I can't exactly recall.