Skip to content

RFC Better handle pads for tuning/decoder part of the devices

Mauro Carvalho Chehab linux-media 1 version New

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 RFC Latest4 patches 0 repliesNew

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/4media: v4l2: remove VBI output paddrivers/media/dvb-frontends/au8522_decoder.c drivers/media/i2c/saa7115.c drivers/media/i2c/tvp5150.c +3 more+1 -70 replies
2/4media: v4l2: taint pads with the signal types for consumer devicesdrivers/media/dvb-frontends/au8522_decoder.c drivers/media/i2c/msp3400-driver.c drivers/media/i2c/saa7115.c +6 more+54 -00 replies
3/4v4l2-mc: switch it to use the new approach to setup pipelinesdrivers/media/v4l2-core/v4l2-mc.c+71 -160 replies
4/4media: dvb: use signals to discover padsdrivers/media/dvb-core/dvbdev.c+33 -40 replies

Reviews trailers per patch

No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.

Read the v1 thread

Cover letter [PATCH RFC 0/4] Better handle pads for tuning/decoder part of the devices

At PC consumer devices, it is very common that the bridge same driver to be attached to different types of tuners and demods. We need a way for the Kernel to properly identify what kind of signal is provided by each PAD, in order to properly setup the pipelines.

The previous approach were to hardcode a fixed number of PADs for all elements of the same type. This is not good, as different devices may actually have a different number of pads.

It was acceptable in the past, as there were a promisse of adding "soon" a properties API that would allow to identify the type for each PADs, but this was never merged (or even a patchset got submitted).

So, replace this approach by another one: add a "taint" mark to pads that contain different types of signals.

I tried to minimize the number of signals, in order to make it simpler to convert from the past way. However, I'm not inspired today to give names to the signals each pad contain. So, feel free to give better suggestions if this one doesn't fit too well.

For now, this is just a RFC, compile-tested only, as the main goal here is to discuss about an approach. Once I get enough feedback, I'll do some tests.

Mauro Carvalho Chehab (4):
media: v4l2: remove VBI output pad
media: v4l2: taint pads with the signal types for consumer devices
v4l2-mc: switch it to use the new approach to setup pipelines
media: dvb: use signals to discover pads

 drivers/media/dvb-core/dvbdev.c              | 37 ++++++++-
 drivers/media/dvb-frontends/au8522_decoder.c |  4 +-
 drivers/media/i2c/msp3400-driver.c           |  2 +
 drivers/media/i2c/saa7115.c                  |  3 +-
 drivers/media/i2c/tvp5150.c                  |  3 +-
 drivers/media/pci/saa7134/saa7134-core.c     |  3 +-
 drivers/media/tuners/si2157.c                |  3 +
 drivers/media/usb/dvb-usb-v2/mxl111sf.c      |  2 +
 drivers/media/v4l2-core/tuner-core.c         |  5 ++
 drivers/media/v4l2-core/v4l2-mc.c            | 87 ++++++++++++++++----
 include/media/media-entity.h                 | 33 ++++++++
 include/media/v4l2-mc.h                      |  2 -
 12 files changed, 158 insertions(+), 26 deletions(-)
Signature · 1 line
2.17.1