On Mon, Jul 18, 2011 at 1:21 PM, Raju, Sundaram [off-list ref] wrote:

quoted

Maybe a new api to pass fixed-format variable-length encoded message
to the DMAC drivers?
Which could be interpreted by DMAC drivers to extract all the needed xfer
parameters from the 'header' section and instructions to program the xfers
in the DMAC from the variable length body of the 'message' buffer.
It might sound complicated but we can have helpers to make the job easy.
Btw, the regular single/sg-list xfers could also be expressed by this method.

Do you expect this variable length body of the message to be DMAC
independent? I don't think so. In that case how is this different from what
we have here already?

Yes, this whould be DMAC independent.

If it can be DMAC independent, can you illustrate more on how this can
be done? But the point to note is, if this can be made DMAC independent
then the control command we have also can be made DMAC independent
by generalizing the configuration structure passed to it.

The 'header' I suggest, would in fact be a structure body, only an extra pointer
would point to the 'instructions' to convey actual location and sizes
of mico-xfers.
I don't think it is possible to have general definition of such transfers fully
within a structure. If I understand what you ask.

I have just posted an RFC. I kept the terms same so that it is easier
to understand.
Please have a look. You are CC'ed too.

Thanks,
Jassi