RE: RFC: bus configuration setup for sub-devices

["Hans Verkuil" <hverkuil@xs4all.nl>]

>>>
>>> Up to now I usually saw the master-slave relationship defined as per
>>> whether the protocol is "master" or "slave," which always was used from
>>> the PoV of the bridge. I.e., even in a camera datasheet a phrase like
>>> "supports master-parallel mode" means supports a mode in which the
>>> bridge
>>> is a master and the camera is a slave. So, maybe it is better instead
>>> of
>>a
>>> .is_master flag to use a .master_mode flag?
>>
>>Sounds reasonable. I'll check a few datasheets myself to see what
>>terminology
>>they use.
>
> Master/Slave is always confusing to me. In VPFE, it can act as master
> (when it output sync signal and pixel clock) and slave (when it get sync
> signal from sensor/decoder). We use VPFE as slave and sensor/decoder will
> provide the pixel clock and sync signal. Please confirm if this is what
> master_mode flag means.

That's correct: the master provides the pixel clock signal. I'm not sure
if it also means that the syncs are provided by the master. Do you know?

Regards,

         Hans

-- 
Hans Verkuil - video4linux developer - sponsored by TANDBERG