v4l: Documentation: Serial busses use parallel mbus codes
Subsystems (2)
PATCH 1/1 v4l: Documentation: Serial busses use parallel mbus codes
When the support for serial busses was introduced in V4L2, it was decided to use the existing parallel bus media bus pixel codes to describe them. While this was a practical choice at the time, it necessitates choosing which one of the many parallel mbus pixel codes to use, for on the serial busses these formats are effectively all equivalent.
The practice has always been to use the pixel code that describes a bus that transfers a single sample per clock. Document this.
--- Documentation/media/kapi/csi2.rst | 7 +++++++ Documentation/media/uapi/v4l/subdev-formats.rst | 4 ++++ 2 files changed, 11 insertions(+)
Documentation/media/kapi/csi2.rst +7 -0
@@ -72,3 +72,10 @@ the transmitter up by using the :c:type:`v4l2_subdev_core_ops`->s_power() callback. This may take place either indirectly by using :c:func:`v4l2_pipeline_pm_use` or directly. + +Formats +------- + +The media bus pixel codes document parallel formats. Should the pixel data be +transported over a serial bus, the media bus pixel code that describes a +parallel format that transfers a sample on a single clock cycle is used.
Documentation/media/uapi/v4l/subdev-formats.rst +4 -0
@@ -85,6 +85,10 @@ formats in memory (a raw Bayer image won't be magically converted to JPEG just by storing it to memory), there is no one-to-one correspondence between them. +The media bus pixel codes document parallel formats. Should the pixel data be +transported over a serial bus, the media bus pixel code that describes a +parallel format that transfers a sample on a single clock cycle is used. + Packed RGB Formats ^^^^^^^^^^^^^^^^^^
Signature · 1 line
2.20.1
Quoted 35 lines · Sakari Ailus
When the support for serial busses was introduced in V4L2, it was decided to use the existing parallel bus media bus pixel codes to describe them. While this was a practical choice at the time, it necessitates choosing which one of the many parallel mbus pixel codes to use, for on the serial busses these formats are effectively all equivalent.
The practice has always been to use the pixel code that describes a bus that transfers a single sample per clock. Document this.
Signed-off-by Sakari Ailus <[email protected]>--- Documentation/media/kapi/csi2.rst | 7 +++++++ Documentation/media/uapi/v4l/subdev-formats.rst | 4 ++++ 2 files changed, 11 insertions(+)Documentation/media/kapi/csi2.rst +7
-0@@ -72,3 +72,10 @@ the transmitter up by using the :c:type:`v4l2_subdev_core_ops`->s_power() callback. This may take place either indirectly by using :c:func:`v4l2_pipeline_pm_use` or directly. + +Formats +------- + +The media bus pixel codes document parallel formats. Should the pixel data be +transported over a serial bus, the media bus pixel code that describes a +parallel format that transfers a sample on a single clock cycle is used.Documentation/media/uapi/v4l/subdev-formats.rst +0
-0
Documentation/media/uapi/v4l/subdev-formats.rst formats in memory (a raw Bayer image won't be magically converted to · line 90
Quoted 6 lines
Documentation/media/uapi/v4l/subdev-formats.rst +3
-0@@ -85,6 +85,10 @@ formats in memory (a raw Bayer image won't be magically converted to JPEG just by storing it to memory), there is no one-to-one correspondence between them. +The media bus pixel codes document parallel formats. Should the pixel data be +transported over a serial bus, the media bus pixel code that describes a +parallel format that transfers a sample on a single clock cycle is used.
I would add an example here along the lines of using MEDIA_BUS_FMT_BGR888_1X24 instead of MEDIA_BUS_FMT_BGR888_3X8.
Possibly in csi2.rst as well.
Regards,
Hans
Quoted 5 lines
Documentation/media/uapi/v4l/subdev-formats.rst +1
-0+ Packed RGB Formats ^^^^^^^^^^^^^^^^^^