diff options
| -rw-r--r-- | Documentation/userspace-api/media/v4l/pixfmt-packed-yuv.rst | 14 |
1 files changed, 14 insertions, 0 deletions
diff --git a/Documentation/userspace-api/media/v4l/pixfmt-packed-yuv.rst b/Documentation/userspace-api/media/v4l/pixfmt-packed-yuv.rst index 6f9f9d63a4ad..eb551b57557e 100644 --- a/Documentation/userspace-api/media/v4l/pixfmt-packed-yuv.rst +++ b/Documentation/userspace-api/media/v4l/pixfmt-packed-yuv.rst @@ -25,6 +25,14 @@ components and thus differ in how they interlave the three components. These formats do not subsample the chroma components and store each pixels as a full triplet of Y, Cb and Cr values. +The next table lists the packed YUV 4:4:4 formats with less than 8 bits per +component. They are named based on the order of the Y, Cb and Cr components as +seen in a 16-bit word, which is then stored in memory in little endian byte +order, and on the number of bits for each component. For instance the YUV565 +format stores a pixel in a 16-bit word [15:0] laid out at as [Y'\ :sub:`4-0` +Cb\ :sub:`5-0` Cr\ :sub:`4-0`], and stored in memory in two bytes, +[Cb\ :sub:`2-0` Cr\ :sub:`4-0`] followed by [Y'\ :sub:`4-0` Cb\ :sub:`5-3`]. + .. raw:: latex \begingroup @@ -145,6 +153,12 @@ full triplet of Y, Cb and Cr values. <overlay>` or :ref:`Video Output Overlay <osd>`. +The next table lists the packed YUV 4:4:4 formats with 8 bits per component. +They are named based on the order of the Y, Cb and Cr components as stored in +memory, and on the total number of bits per pixel. For instance, the VUYX32 +format stores a pixel with Cr\ :sub:`7-0` in the first byte, Cb\ :sub:`7-0` in +the second byte and Y'\ :sub:`7-0` in the third byte. + .. flat-table:: Packed YUV Image Formats (8bpc) :header-rows: 1 :stub-columns: 0 |