Ondrej Pindroch wrote:
Thank you for you answer.
I want to use buffers of VPBE as output of previewer and than perhaps
output of resizer to avoid memcpy operation. Previewing task will be
synchronized with VPBE.I have realized that it is not possible to do it
with FB driver I need to use V4L2 to get offsets for previewer.
Yes, FB allocates its own mem that you have to memcpy into, but V4L allows you
to pass your own pointer (USERPTR).
The requirement is to reduce latency. Now I am using pipe of CCDC,
PREVIEWER(for demosaicing), memcpy and FB. And the latency is about 160ms.
Best regards
Let me get this correct: you want to take the output from the
previewer engine in the CCDC module, and drive the VPBE display
device with the same memory buffers?
Although this is theoretically possible, it does not make sense from
an implementation perspective – the CCDC module operates on an input
clock that is absolutely asynchronous (and possibly at a different
rate) with the display output clock.
In that regard, the VPBE hardware will continually pick-up at its
native display rate data from that memory location; since the
production of data into the buffer by the CCDC is not in ‘sync’ with
the VPBE, you’ll definitely get tearing at the VPBE output.
You prevent tearing if you use more than one buffer, CCDC captures into one
buffer while another one is displayed.
_______________________________________________
Davinci-linux-open-source mailing list
[email protected]
http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
