CS35L56 is designed for SDCA and a generic SDCA driver would
know nothing about these chip-specific registers. So the
firmware sets up the SDW TX mixer registers to whatever audio
is relevant on a specific system.
This means that the driver cannot assume the initial values
of these registers. But Linux has ALSA controls to configure
routing, so the registers can be patched to silicon default and
the ALSA controls used to select what audio to feed back to the
host capture path.
humm, which has the precedence then?
a) the values set by firmware
b) the default values set by the driver?
Also if the firmware touches those registers shouldn't they be marked as
'volatile'?
The firmware was designed to work with Windows, so it looks a bit
strange if you are coming at it from ALSA. There's not really any
defined 'precedence'. The firmware will setup the feedback monitor paths
to something that satisfies SDCA and Windows expectations.
We don't care about that in Linux, the firmware on the Intel DSP
probably isn't running the same algorithms for Linux, and we have ALSA
controls to configure those paths. So we patch the mixers back to their
silicon defaults and take over complete control of them.
The firmware only writes them during its power-up sequence so they
will only change when we are rebooting the firmware or coming out of
low-power standby, which is under the control of the driver. When that
happens regmap will re-apply the patch and then sync up the registers
again. The firmware won't touch them after boot, so we can avoid having
to mark them volatile (which would mean implementing our own manual
caching of the settings).
ok, thanks for the explanations.
