Re: [alsa-devel] [PATCH] [v2] ASoC: cs4270: use the built-in register cache support

Mark Brown wrote:

Note that the return value of the read functions is an unsigned int rather than an int.

Oh yeah, I didn't see that. So returning any negative number is broken.

So I don't understand the philosophy here. Are we going to propagate errors during read operations, or not? It seems that sometimes ASoC does, and sometimes it doesn't, and sometimes it doesn't even know what it wants (e.g. the snd_soc_X_X_read functions).