On 01/12/2017 03:43 PM, Steve Longerbeam wrote:
On 01/12/2017 03:22 PM, Steve Longerbeam wrote:
and since my PRPVF entity roles
up the VDIC internally, it is actually receiving from the VDIC
channel.
So unless you think we should have a distinct VDIC entity, I would
like
to keep this
the way it is.
Yes, I think VDIC should be separated out of PRPVF. What do you think
about splitting the IC PRP into three parts?
PRP could have one input pad connected to either CSI0, CSI1, or VDIC,
and two output pads connected to PRPVF and PRPENC, respectively. This
would even allow to have the PRP describe the downscale and PRPVF and
PRPENC describe the bilinear upscale part of the IC.
Actually, how about the following:
PRP would have one input pad coming from CSI0, CSI1, or VDIC. But
instead of another level of indirection with two more PRPENC and PRPVF
entities, PRP would instead have two output pads, one for PRPVF output
and one for PRPENC output.
Both output pads could be activated if the input is connected to CSI0
or CSI1.
And only the PRPVF output can be activated if the input is from VDIC.
Actually that proved too difficult. I went with your original idea.
Branch that
implements this is imx-media-staging-md-prp. The media dot graph looks good
but I have not tested yet. I'll start testing it tomorrow.
Steve
--
To unsubscribe from this list: send the line "unsubscribe linux-media" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
