Re: [PATCH v8 1/4] media: v4l: Add definition for the Aspeed JPEG format

From: Jammy Huang
Date: Mon Jan 17 2022 - 04:38:31 EST


Hi Hans,

On 2022/1/17 下午 05:28, Hans Verkuil wrote:
Hi Jammy,

On 1/17/22 08:18, Jammy Huang wrote:
On 2022/1/17 上午 10:24, Joel Stanley wrote:
On Mon, 17 Jan 2022 at 02:00, Jammy Huang <jammy_huang@xxxxxxxxxxxxxx> wrote:
Hi Hans,

On 2022/1/14 下午 04:11, Hans Verkuil wrote:
On 24/12/2021 02:27, Jammy Huang wrote:
This introduces support for the Aspeed JPEG format, where the new frame
can refer to previous frame to reduce the amount of compressed data. The
concept is similar to I/P frame of video compression. It will compare the
new frame with previous one to decide which macroblock's data is
changed, and only the changed macroblocks will be compressed.

This Aspeed JPEG format is used by the video engine on Aspeed platforms,
which is generally adapted for remote KVM.

Signed-off-by: Jammy Huang <jammy_huang@xxxxxxxxxxxxxx>
---
v8:
    - Add decoder information for aspeed-jpeg
v7:
    - Add more information for aspeed-jpeg
v6:
    - Update description for new format, aspeed-jpeg, in Documentation.
v5:
    - no update
v4:
    - new
---
   .../media/uapi/v4l/pixfmt-reserved.rst          | 17 +++++++++++++++++
   drivers/media/v4l2-core/v4l2-ioctl.c            |  1 +
   include/uapi/linux/videodev2.h                  |  1 +
   3 files changed, 19 insertions(+)

diff --git a/Documentation/media/uapi/v4l/pixfmt-reserved.rst b/Documentation/media/uapi/v4l/pixfmt-reserved.rst
This is the wrong file! It should be:

Documentation/userspace-api/media/v4l/pixfmt-reserved.rst
Thanks, I just used git format-patch to generate the cover-letter and
didn't notice this.

It looks like the file, pixfmt-reserved.rst, has different path in
different Linux kernel,

* 5.4:           Documentation/media/uapi/v4l/pixfmt-reserved.rst

* 5.10/5.15: Documentation/userspace-api/media/v4l/pixfmt-reserved.rst

5.4 is the one I based to submit the patches.

Could you suggest the kernel that I should based to submit these patches??
You should always submit based on the latest release.

There are sometimes exceptions where you will base your patches on
something even newer, but generally you should grab the latest.

Be sure to re-test on the latest kernel before submitting.
OK, thank you.
I recommend waiting until later this week. I hope to have a PR for other outstanding
aspeed patches ready in a few days. It is easiest if you base your v9 on top of that
series, this to avoid merge conflicts.

Sure, no problem. Thanks for your help.


Regards,

Jammy


Regards,

Hans

Cheers,

Joel

I will need to change the number of V4L2_CID_USER_ASPEED_BASE per different

linux kernel as well.

Regards,

       Hans

index b2cd155e691b..1d0dc8d86ed7 100644
--- a/Documentation/media/uapi/v4l/pixfmt-reserved.rst
+++ b/Documentation/media/uapi/v4l/pixfmt-reserved.rst
@@ -264,6 +264,23 @@ please make a proposal on the linux-media mailing list.
      of tiles, resulting in 32-aligned resolutions for the luminance plane
      and 16-aligned resolutions for the chrominance plane (with 2x2
      subsampling).
+    * .. _V4L2-PIX-FMT-AJPG:
+
+      - ``V4L2_PIX_FMT_AJPG``
+      - 'AJPG'
+      - ASPEED JPEG format used by the aspeed-video driver on Aspeed platforms,
+        which is generally adapted for remote KVM.
+        On each frame compression, I will compare the new frame with previous
+        one to decide which macroblock's data is changed, and only the changed
+        macroblocks will be compressed.
+
+        The implementation is based on AST2600 A3 datasheet, revision 0.9, which
+        is not publicly available. Or you can reference Video stream data format
+        – ASPEED mode compression of SDK_User_Guide which available on
+        AspeedTech-BMC/openbmc/releases.
+
+        Decoder's implementation can be found here,
+        `https://github.com/AspeedTech-BMC/aspeed_codec/ <https://github.com/AspeedTech-BMC/aspeed_codec/>`__

   .. tabularcolumns:: |p{6.6cm}|p{2.2cm}|p{8.7cm}|

diff --git a/drivers/media/v4l2-core/v4l2-ioctl.c b/drivers/media/v4l2-core/v4l2-ioctl.c
index 24db33f803c0..00dde01d2f97 100644
--- a/drivers/media/v4l2-core/v4l2-ioctl.c
+++ b/drivers/media/v4l2-core/v4l2-ioctl.c
@@ -1378,6 +1378,7 @@ static void v4l_fill_fmtdesc(struct v4l2_fmtdesc *fmt)
              case V4L2_PIX_FMT_S5C_UYVY_JPG: descr = "S5C73MX interleaved UYVY/JPEG"; break;
              case V4L2_PIX_FMT_MT21C:        descr = "Mediatek Compressed Format"; break;
              case V4L2_PIX_FMT_SUNXI_TILED_NV12: descr = "Sunxi Tiled NV12 Format"; break;
+            case V4L2_PIX_FMT_AJPG:         descr = "Aspeed JPEG"; break;
              default:
                      if (fmt->description[0])
                              return;
diff --git a/include/uapi/linux/videodev2.h b/include/uapi/linux/videodev2.h
index 3210b3c82a4a..994eb6155ea9 100644
--- a/include/uapi/linux/videodev2.h
+++ b/include/uapi/linux/videodev2.h
@@ -726,6 +726,7 @@ struct v4l2_pix_format {
   #define V4L2_PIX_FMT_INZI     v4l2_fourcc('I', 'N', 'Z', 'I') /* Intel Planar Greyscale 10-bit and Depth 16-bit */
   #define V4L2_PIX_FMT_SUNXI_TILED_NV12 v4l2_fourcc('S', 'T', '1', '2') /* Sunxi Tiled NV12 Format */
   #define V4L2_PIX_FMT_CNF4     v4l2_fourcc('C', 'N', 'F', '4') /* Intel 4-bit packed depth confidence information */
+#define V4L2_PIX_FMT_AJPG     v4l2_fourcc('A', 'J', 'P', 'G') /* Aspeed JPEG */

   /* 10bit raw bayer packed, 32 bytes for every 25 pixels, last LSB 6 bits unused */
   #define V4L2_PIX_FMT_IPU3_SBGGR10  v4l2_fourcc('i', 'p', '3', 'b') /* IPU3 packed 10-bit BGGR bayer */
--
Best Regards
Jammy