public inbox for linux-rockchip@lists.infradead.org
 help / color / mirror / Atom feed
From: Andrzej Pietrasiewicz <andrzej.p@collabora.com>
To: Alex Bee <knaerzche@gmail.com>,
	linux-media@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org, linux-rockchip@lists.infradead.org,
	linux-staging@lists.linux.dev
Cc: Benjamin Gaignard <benjamin.gaignard@collabora.com>,
	Boris Brezillon <boris.brezillon@collabora.com>,
	Ezequiel Garcia <ezequiel@vanguardiasur.com.ar>,
	Fabio Estevam <festevam@gmail.com>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Hans Verkuil <hverkuil-cisco@xs4all.nl>,
	Heiko Stuebner <heiko@sntech.de>,
	Jernej Skrabec <jernej.skrabec@gmail.com>,
	Mauro Carvalho Chehab <mchehab@kernel.org>,
	Nicolas Dufresne <nicolas.dufresne@collabora.com>,
	NXP Linux Team <linux-imx@nxp.com>,
	Pengutronix Kernel Team <kernel@pengutronix.de>,
	Philipp Zabel <p.zabel@pengutronix.de>,
	Sascha Hauer <s.hauer@pengutronix.de>,
	Shawn Guo <shawnguo@kernel.org>,
	kernel@collabora.com, Ezequiel Garcia <ezequiel@collabora.com>,
	Adrian Ratiu <adrian.ratiu@collabora.com>
Subject: Re: [PATCH v7 07/11] media: rkvdec: Add the VP9 backend
Date: Wed, 20 Oct 2021 15:07:11 +0200	[thread overview]
Message-ID: <91ba5098-2528-1e63-3a1a-b908db8d6f2a@collabora.com> (raw)
In-Reply-To: <966b04a7-421a-a592-2e17-ea5ecdb76b00@gmail.com>

Hi Alex,

W dniu 20.10.2021 o 01:24, Alex Bee pisze:
> Hi Andrzej,
> 
> Am 29.09.21 um 18:04 schrieb Andrzej Pietrasiewicz:
>> From: Boris Brezillon <boris.brezillon@collabora.com>
>>
>> The Rockchip VDEC supports VP9 profile 0 up to 4096x2304@30fps. Add
>> a backend for this new format.
>>
>> Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
>> Signed-off-by: Ezequiel Garcia <ezequiel@collabora.com>
>> Signed-off-by: Adrian Ratiu <adrian.ratiu@collabora.com>
>> Co-developed-by: Andrzej Pietrasiewicz <andrzej.p@collabora.com>
>> Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@collabora.com>
>> ---
>>   drivers/staging/media/rkvdec/Kconfig      |    1 +
>>   drivers/staging/media/rkvdec/Makefile     |    2 +-
>>   drivers/staging/media/rkvdec/rkvdec-vp9.c | 1078 +++++++++++++++++++++
>>   drivers/staging/media/rkvdec/rkvdec.c     |   52 +-
>>   drivers/staging/media/rkvdec/rkvdec.h     |   12 +-
>>   5 files changed, 1137 insertions(+), 8 deletions(-)
>>   create mode 100644 drivers/staging/media/rkvdec/rkvdec-vp9.c

<snip>

>> diff --git a/drivers/staging/media/rkvdec/rkvdec.c 
>> b/drivers/staging/media/rkvdec/rkvdec.c
>> index 7131156c1f2c..6aa8aca66547 100644
>> --- a/drivers/staging/media/rkvdec/rkvdec.c
>> +++ b/drivers/staging/media/rkvdec/rkvdec.c
>> @@ -99,10 +99,30 @@ static const struct rkvdec_ctrls rkvdec_h264_ctrls = {
>>       .num_ctrls = ARRAY_SIZE(rkvdec_h264_ctrl_descs),
>>   };
>> -static const u32 rkvdec_h264_decoded_fmts[] = {
>> +static const u32 rkvdec_h264_vp9_decoded_fmts[] = {
>>       V4L2_PIX_FMT_NV12,
> 
> For H.264 rkvdec HW supports additional formats: V4L2_PIX_FMT_NV15, 
> V4L2_PIX_FMT_NV16 and V4L2_PIX_FMT_NV20. Not all of those are upstreamed yet and 
> thus not supported by rkvdec driver - but I think we should introduce a seperate 
> rkvdec_vp9_decoded_fmts already a this point. (To avoid unnecessary diff 
> afterwards)

I will do it if I get to re-spinning the series for other reasons.

> 
>>   };
>> +static const struct rkvdec_ctrl_desc rkvdec_vp9_ctrl_descs[] = {
>> +    {
>> +        .cfg.id = V4L2_CID_STATELESS_VP9_FRAME,
>> +    },
>> +    {
>> +        .cfg.id = V4L2_CID_STATELESS_VP9_COMPRESSED_HDR,
>> +    },
>> +    {
>> +        .cfg.id = V4L2_CID_MPEG_VIDEO_VP9_PROFILE,
>> +        .cfg.min = V4L2_MPEG_VIDEO_VP9_PROFILE_0,
>> +        .cfg.max = V4L2_MPEG_VIDEO_VP9_PROFILE_0,
>> +        .cfg.def = V4L2_MPEG_VIDEO_VP9_PROFILE_0,
>> +    },
>> +};
>> +
>> +static const struct rkvdec_ctrls rkvdec_vp9_ctrls = {
>> +    .ctrls = rkvdec_vp9_ctrl_descs,
>> +    .num_ctrls = ARRAY_SIZE(rkvdec_vp9_ctrl_descs),
>> +};
>> +
>>   static const struct rkvdec_coded_fmt_desc rkvdec_coded_fmts[] = {
>>       {
>>           .fourcc = V4L2_PIX_FMT_H264_SLICE,
>> @@ -116,8 +136,23 @@ static const struct rkvdec_coded_fmt_desc 
>> rkvdec_coded_fmts[] = {
>>           },
>>           .ctrls = &rkvdec_h264_ctrls,
>>           .ops = &rkvdec_h264_fmt_ops,
>> -        .num_decoded_fmts = ARRAY_SIZE(rkvdec_h264_decoded_fmts),
>> -        .decoded_fmts = rkvdec_h264_decoded_fmts,
>> +        .num_decoded_fmts = ARRAY_SIZE(rkvdec_h264_vp9_decoded_fmts),
>> +        .decoded_fmts = rkvdec_h264_vp9_decoded_fmts,
>> +    },
>> +    {
>> +        .fourcc = V4L2_PIX_FMT_VP9_FRAME,
>> +        .frmsize = {
>> +            .min_width = 64,
>> +            .max_width = 4096,
>> +            .step_width = 64,
>> +            .min_height = 64,
>> +            .max_height = 2304,
>> +            .step_height = 64,
>> +        },
> I checked (available) documentation and couldn't find any hint to the 
> .step_width and .step_height, but I'm not sure that's correct: taking
> this values here neither framesize of 3840x2160 nor 1280x720 would be possible - 
> but the HW seems to have no problem with those, i.e. decoding works fine.
> Given the output format is the same as the (only) currently supported H.264 
> output format (NV12) and those steps are usually for alignment purposes need by 
> the HW , I strongly guess .step_height and .step_width are the same as 
> V4L2_PIX_FMT_H264_SLICE has.
> 

Aren't these used primarily by v4l2_apply_frmsize_constraints()? Doesn't
this merely mean that even though userspace requests, say, 48x48,
it will get 64x64 instead?

I tried decoding a 720p video with gstreamer and it worked fine
(I got a properly sized 1280x720 output).

Regards,

Andrzej

_______________________________________________
Linux-rockchip mailing list
Linux-rockchip@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-rockchip

  reply	other threads:[~2021-10-20 13:07 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-29 16:04 [PATCH v7 00/11] VP9 codec V4L2 control interface Andrzej Pietrasiewicz
2021-09-29 16:04 ` [PATCH v7 01/11] hantro: postproc: Fix motion vector space size Andrzej Pietrasiewicz
2021-09-29 16:04 ` [PATCH v7 02/11] hantro: postproc: Introduce struct hantro_postproc_ops Andrzej Pietrasiewicz
2021-09-29 16:04 ` [PATCH v7 03/11] hantro: Simplify postprocessor Andrzej Pietrasiewicz
2021-09-29 16:04 ` [PATCH v7 04/11] hantro: Add quirk for NV12/NV12_4L4 capture format Andrzej Pietrasiewicz
2021-09-29 16:04 ` [PATCH v7 05/11] media: uapi: Add VP9 stateless decoder controls Andrzej Pietrasiewicz
2021-09-29 16:04 ` [PATCH v7 06/11] media: Add VP9 v4l2 library Andrzej Pietrasiewicz
2021-09-29 16:04 ` [PATCH v7 07/11] media: rkvdec: Add the VP9 backend Andrzej Pietrasiewicz
2021-10-08 10:30   ` Chen-Yu Tsai
2021-10-19 23:24   ` Alex Bee
2021-10-20 13:07     ` Andrzej Pietrasiewicz [this message]
2021-09-29 16:04 ` [PATCH v7 08/11] media: hantro: Rename registers Andrzej Pietrasiewicz
2021-09-29 16:04 ` [PATCH v7 09/11] media: hantro: Prepare for other G2 codecs Andrzej Pietrasiewicz
2021-09-29 16:04 ` [PATCH v7 10/11] media: hantro: Support VP9 on the G2 core Andrzej Pietrasiewicz
2021-09-29 16:04 ` [PATCH v7 11/11] media: hantro: Support NV12 " Andrzej Pietrasiewicz
2021-10-14 17:42   ` Jernej Škrabec
2021-10-15 17:19     ` Andrzej Pietrasiewicz
2021-10-19 16:38       ` Jernej Škrabec
2021-10-20 11:06         ` Ezequiel Garcia
2021-10-20 15:04           ` Jernej Škrabec
2021-10-20 15:25             ` Ezequiel Garcia
2021-10-21 15:36               ` Jernej Škrabec
2021-10-19 17:55 ` [PATCH v7 00/11] VP9 codec V4L2 control interface Ezequiel Garcia
2021-11-11 14:44 ` Hans Verkuil
2021-11-12 15:27   ` Nicolas Dufresne
2021-11-15 12:56     ` Andrzej Pietrasiewicz
2021-11-15 13:09       ` Andrzej Pietrasiewicz
2021-11-15 15:07 ` Hans Verkuil
2021-11-15 17:14   ` Andrzej Pietrasiewicz
2021-11-15 21:16     ` Hans Verkuil
2021-11-16  8:09       ` Andrzej Pietrasiewicz
2021-11-16  8:21         ` Hans Verkuil
2021-11-16 13:14           ` Andrzej Pietrasiewicz
2021-11-17  9:59             ` Hans Verkuil
2021-11-17 10:49               ` Andrzej Pietrasiewicz
2021-11-17 10:51                 ` Andrzej Pietrasiewicz
2021-11-17 11:33                   ` Andrzej Pietrasiewicz

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=91ba5098-2528-1e63-3a1a-b908db8d6f2a@collabora.com \
    --to=andrzej.p@collabora.com \
    --cc=adrian.ratiu@collabora.com \
    --cc=benjamin.gaignard@collabora.com \
    --cc=boris.brezillon@collabora.com \
    --cc=ezequiel@collabora.com \
    --cc=ezequiel@vanguardiasur.com.ar \
    --cc=festevam@gmail.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=heiko@sntech.de \
    --cc=hverkuil-cisco@xs4all.nl \
    --cc=jernej.skrabec@gmail.com \
    --cc=kernel@collabora.com \
    --cc=kernel@pengutronix.de \
    --cc=knaerzche@gmail.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-imx@nxp.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-rockchip@lists.infradead.org \
    --cc=linux-staging@lists.linux.dev \
    --cc=mchehab@kernel.org \
    --cc=nicolas.dufresne@collabora.com \
    --cc=p.zabel@pengutronix.de \
    --cc=s.hauer@pengutronix.de \
    --cc=shawnguo@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox