Commit Graph

11 Commits

Author SHA1 Message Date
Benjamin Gaignard
5423e2d220 media: verisilicon: Add reference buffer compression feature
Reference frame compression is a feature added in the G2 decoder to
compress frame buffers so that the bandwidth of storing/loading
reference frames can be reduced, especially with high resolution decoded
streams.

The impact of compressed frames is confirmed when using perf to monitor
the number of memory accesses with or without the compression feature.
The following command:

perf stat -a -e \
imx8_ddr0/cycles/,imx8_ddr0/read-cycles/,imx8_ddr0/write-cycles/ \
gst-launch-1.0 filesrc \
location=Jockey_3840x2160_120fps_420_8bit_HEVC_RAW.hevc ! queue ! \
h265parse ! v4l2slh265dec ! video/x-raw,format=NV12 ! fakesink

Gives us these results without the compression feature:
Performance counter stats for 'system wide':

        1711300345      imx8_ddr0/cycles/
         892207924      imx8_ddr0/read-cycles/
        1291785864      imx8_ddr0/write-cycles/

      13.760048353 seconds time elapsed

With the compression feature:
Performance counter stats for 'system wide':

         274526799      imx8_ddr0/cycles/
         453120194      imx8_ddr0/read-cycles/
         833391434      imx8_ddr0/write-cycles/

      18.257831534 seconds time elapsed

As expected the number of read/write cycles are really lower when
compression is used.

Since storing the compression data requires more memory a module
parameter named 'hevc_use_compression' is used to enable/disable
this feature and, by default, compression isn't used.

Enabling the compression feature means that the output-frames of the
decoder
are stored with a specific compression pixel-format. Since this
pixel format is unknown, this patch restrains the compression feature
usage to the cases where post-processor pixel-formats (NV12 or NV15)
are selected by the applications.

The Fluster compliance HEVC test suite score is still 141/147 with this
patch.

Signed-off-by: Benjamin Gaignard <benjamin.gaignard@collabora.com>
Tested-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
Signed-off-by: Sebastian Fricke <sebastian.fricke@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
2024-08-28 10:01:10 +02:00
Hugues Fruchet
b0fb9ca30a media: hantro: add support for STM32MP25 VENC
Add support for STM32MP25 VENC video hardware encoder.
Support of JPEG encoding.
VENC has its own reset/clock/irq.

Reviewed-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
Signed-off-by: Hugues Fruchet <hugues.fruchet@foss.st.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
2024-02-01 11:12:04 +01:00
Hugues Fruchet
46c4dffb74 media: hantro: add support for STM32MP25 VDEC
Add support for STM32MP25 VDEC video hardware decoder.
Support of H264/VP8 decoding.
No post-processor support.
VDEC has its own reset/clock/irq.

Successfully tested up to full HD.

Reviewed-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
Signed-off-by: Hugues Fruchet <hugues.fruchet@foss.st.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
2024-02-01 11:11:33 +01:00
Benjamin Gaignard
3eeaee737d media: verisilicon: g2: Use common helpers to compute chroma and mv offsets
HEVC and VP9 are running on the same hardware and share the same
chroma and motion vectors offset constraint.
Create common helpers functions for these computation.
Source and destination buffer height may not be the same because
alignment constraint are different so use destination height to
compute chroma offset because we target this buffer as hardware
output.
To be able to use the helpers in both VP9 HEVC code remove dec_params
and use context->bit_depth instead.

Signed-off-by: Benjamin Gaignard <benjamin.gaignard@collabora.com>
Reviewed-by: Andrzej Pietrasiewicz <andrzej.p@collabora.com>
CC: Ezequiel Garcia <ezequiel@vanguardiasur.com.ar>
CC: Philipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
2023-11-23 12:44:10 +01:00
Benjamin Gaignard
26711491a8 media: verisilicon: Refactor postprocessor to store more buffers
Since vb2 queue can store more than VB2_MAX_FRAME buffers, the
postprocessor buffer storage must be capable to store more buffers too.
Change static dec_q array to allocated array to be capable to store
up to queue 'max_num_buffers'.
Keep allocating queue 'num_buffers' at queue setup time but also allows
to allocate postprocessors buffers on the fly.

Signed-off-by: Benjamin Gaignard <benjamin.gaignard@collabora.com>
Reviewed-by: Andrzej Pietrasiewicz <andrzej.p@collabora.com>
CC: Ezequiel Garcia <ezequiel@vanguardiasur.com.ar>
CC: Philipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
2023-11-23 12:42:56 +01:00
Benjamin Gaignard
003afda97c media: verisilicon: Enable AV1 decoder on rk3588
Add rk3588 AV1 decoder to Hantro variant.
The hardware support image from 64x64 up to 7680x4320
by steps of 16 pixels.

Signed-off-by: Benjamin Gaignard <benjamin.gaignard@collabora.com>
Reviewed-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
2023-06-09 16:19:49 +01:00
Benjamin Gaignard
d8ebe59e7b media: verisilicon: Add film grain feature to AV1 driver
Film grain feature add "old style" grain noise on decoded streams.
Grain noise is applied after decoding by the postprocessor.
The level of grain is based on gaussian sequence.

Signed-off-by: Benjamin Gaignard <benjamin.gaignard@collabora.com>
Reviewed-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
2023-06-09 16:19:27 +01:00
Benjamin Gaignard
727a400686 media: verisilicon: Add Rockchip AV1 decoder
Implement AV1 stateless decoder for rockchip VPU981.
It decode 8 and 10 bits AV1 bitstreams.
AV1 scaling feature is done by the postprocessor.

Signed-off-by: Benjamin Gaignard <benjamin.gaignard@collabora.com>
Reviewed-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
2023-06-09 16:18:46 +01:00
Benjamin Gaignard
c0d0e579db media: verisilicon: Add AV1 entropy helpers
AV1 hardware decoder needs entropy parameters to decode frames.
They are computed from various arrays defined in AV1 section
"9.4. Default CDF tables".

Add helpers functions to init, store and get these parameters.

Signed-off-by: Benjamin Gaignard <benjamin.gaignard@collabora.com>
Reviewed-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
2023-06-09 16:17:20 +01:00
Benjamin Gaignard
1b9ef2744c media: verisilicon: Compute motion vectors size for AV1 frames
Compute the additional space required to store motion vectors at
the end of the frames buffers.

Signed-off-by: Benjamin Gaignard <benjamin.gaignard@collabora.com>
Reviewed-by: Nicolas Dufresne <nicolas.dufresne@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
2023-06-09 16:16:02 +01:00
Ezequiel Garcia
fbb6c848dd media: destage Hantro VPU driver
The Hantro mainline driver has been used in production
since several years and was only kept as a staging driver
due the stateless CODEC controls.

Now that all the stateless CODEC controls have been moved
out of staging, graduate the driver as well.

Signed-off-by: Ezequiel Garcia <ezequiel@vanguardiasur.com.ar>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
2022-08-31 10:23:50 +02:00