11 May
2011
11 May
'11
11:20 a.m.
On Tue, May 10, 2011 at 10:13:18PM +0100, Mark Brown wrote:
On Tue, Mar 22, 2011 at 10:37:03AM +0000, Dimitris Papastamos wrote:
- switch (codec->control_type) {
- case SND_SOC_I2C:
ret = i2c_master_send(codec->control_data, data, len);break;- case SND_SOC_SPI:
ret = do_spi_write(codec->control_data, data, len);break;Hrm, I now realise that this doesn't actually do what I'd expect - it only writes out the provided data, there's no inclusion of the register address. I'd been mislead by the fact that it is taking in the register address as an argument and an overly hasty review it seems. That's a little too raw for me, we should be prefixing the data with the base register address.
Yes, I was going to send a patch for this because I need it for syncing the block based rbtree cache.