mbox series

[RESEND,v7,0/2] media: cedrus: Add H264 decoding support

Message ID cover.bfe197a82766125bea21dd26fd3de85893a921ac.1554380738.git-series.maxime.ripard@bootlin.com (mailing list archive)
Headers show
Series media: cedrus: Add H264 decoding support | expand

Message

Maxime Ripard April 4, 2019, 12:26 p.m. UTC
Hi,

Here is a new version of the H264 decoding support in the cedrus
driver.

As you might already know, the cedrus driver relies on the Request
API, and is a reverse engineered driver for the video decoding engine
found on the Allwinner SoCs.

This work has been possible thanks to the work done by the people
behind libvdpau-sunxi found here:
https://github.com/linux-sunxi/libvdpau-sunxi/

I've tested the various ABI using this gdb script:
http://code.bulix.org/jl4se4-505620?raw

And this test script:
http://code.bulix.org/8zle4s-505623?raw

The application compiled is quite trivial:
http://code.bulix.org/e34zp8-505624?raw

The output is:
arm64:	builds/arm64-test-v4l2-h264-structures
	SHA1: 1c48d3868ac9049c6b5efed43a74bf97af710aba
x86:	builds/x86-test-v4l2-h264-structures
	SHA1: 1c48d3868ac9049c6b5efed43a74bf97af710aba
arm:	builds/arm-test-v4l2-h264-structures
	SHA1: 1c48d3868ac9049c6b5efed43a74bf97af710aba
x64:	builds/x64-test-v4l2-h264-structures
	SHA1: 1c48d3868ac9049c6b5efed43a74bf97af710aba

Let me know if there's any flaw using that test setup, or if you have
any comments on the patches.

Maxime

Changes from v6:
  - Rebased on next
  - Renamed the timestamp DPB field to reference_ts
  - Fixed the collision of control type values
  - Removed unused fields
  - Fixed the structure layout that was broken on x86 by reducing the
    num_slices and nal_ref_idc to 16 bits instead of 32

Changes from v5:
  - Made the references to the H264 spec more explicit
  - Added a flag for the IDR pic
  - Fixed typos
  - Rebased on v5.1-rc1

Changes from v4:
  - Changed the luma and chroma weight and offset from s8 to s16
  - Adjusted chroma and luma denominators masks in the driver
  - Casted the luma and chroma offset to prevent an overflow
  - ALways write the interrupt status register
  - Fix a bug in the sram write routine that would write something even if
    the length was 0
  - Make the scaling lists mandatory
  - Made the reference list order explicit in the documentation
  - Made the fact that the slice structure can be an array
  - Renamed the slice format to V4L2_PIX_FMT_H264_SLICE_RAW
  - Rebased on Hans' tag br-v5.1s

Changes from v3:
  - Reintroduced long term reference flag and documented it
  - Reintroduced ref_pic_list_p0/b0/b1 and documented it
  - Documented the DPB flags
  - Treat the scaling matrix as optional in the driver, as documented
  - Free the neighbor buffer
  - Increase the control IDs by a large margin to be safe of collisions
  - Reorder the fields documentation according to the structure layout
  - Change the tag documentation by the timestamp
  - Convert the sram array to size_t
  - Simplify the buffer retrieval from timestamp
  - Rebase

Changes from v2:
  - Simplified _cedrus_write_ref_list as suggested by Jernej
  - Set whether the frame is used as reference using nal_ref_idc
  - Respect chroma_format_idc
  - Fixes for the scaling list and prediction tables
  - Wrote the documentation for the flags
  - Added a bunch of defines to the driver bit fields
  - Reworded the controls and data format descriptions as suggested
    by Hans
  - Reworked the controls' structure field size to avoid padding
  - Removed the long term reference flag
  - Reintroduced the neighbor info buffer
  - Removed the ref_pic_list_p0/b0/b1 arrays that are redundant with the
    one in the DPB
  - used the timestamps instead of tags
  - Rebased on 5.0-rc1

Changes from v1:
  - Rebased on 4.20
  - Did the documentation for the userspace API
  - Used the tags instead of buffer IDs
  - Added a comment to explain why we still needed the swdec trigger
  - Reworked the MV col buffer in order to have one slot per frame
  - Removed the unused neighbor info buffer
  - Made sure to have the same structure offset and alignments across
    32 bits and 64 bits architecture

Maxime Ripard (1):
  media: cedrus: Add H264 decoding support

Pawel Osciak (1):
  media: uapi: Add H264 low-level decoder API compound controls.

 Documentation/media/uapi/v4l/biblio.rst            |   9 +-
 Documentation/media/uapi/v4l/ext-ctrls-codec.rst   | 569 ++++++++++++++-
 Documentation/media/uapi/v4l/pixfmt-compressed.rst |  19 +-
 Documentation/media/uapi/v4l/vidioc-queryctrl.rst  |  30 +-
 Documentation/media/videodev2.h.rst.exceptions     |   5 +-
 drivers/media/v4l2-core/v4l2-ctrls.c               |  42 +-
 drivers/media/v4l2-core/v4l2-ioctl.c               |   1 +-
 drivers/staging/media/sunxi/cedrus/Makefile        |   3 +-
 drivers/staging/media/sunxi/cedrus/cedrus.c        |  31 +-
 drivers/staging/media/sunxi/cedrus/cedrus.h        |  38 +-
 drivers/staging/media/sunxi/cedrus/cedrus_dec.c    |  13 +-
 drivers/staging/media/sunxi/cedrus/cedrus_h264.c   | 574 ++++++++++++++-
 drivers/staging/media/sunxi/cedrus/cedrus_hw.c     |   4 +-
 drivers/staging/media/sunxi/cedrus/cedrus_regs.h   |  91 ++-
 drivers/staging/media/sunxi/cedrus/cedrus_video.c  |   9 +-
 include/media/h264-ctrls.h                         | 192 +++++-
 include/media/v4l2-ctrls.h                         |  13 +-
 include/uapi/linux/videodev2.h                     |   1 +-
 18 files changed, 1641 insertions(+), 3 deletions(-)
 create mode 100644 drivers/staging/media/sunxi/cedrus/cedrus_h264.c
 create mode 100644 include/media/h264-ctrls.h

base-commit: 61de49cb596710b918f7a80839f0b6de2017bc32