ARM Sunxi Platform Development
 help / color / mirror / Atom feed
From: "Jernej Škrabec" <jernej.skrabec@gmail.com>
To: mripard@kernel.org, paul.kocialkowski@bootlin.com,
	Hans Verkuil <hverkuil-cisco@xs4all.nl>
Cc: mchehab@kernel.org, gregkh@linuxfoundation.org, wens@csie.org,
	samuel@sholland.org, linux-media@vger.kernel.org,
	linux-staging@lists.linux.dev,
	linux-arm-kernel@lists.infradead.org,
	linux-sunxi@lists.linux.dev, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2 10/11] media: cedrus: Adjust buffer size based on control values
Date: Fri, 04 Nov 2022 19:10:52 +0100	[thread overview]
Message-ID: <1908973.PYKUYFuaPT@jernej-laptop> (raw)
In-Reply-To: <76c3fc1b-e90f-748c-39ee-74be93fd6811@xs4all.nl>

Dne petek, 04. november 2022 ob 16:27:29 CET je Hans Verkuil napisal(a):
> Hi Jernej,
> 
> On 02/11/2022 19:08, Jernej Skrabec wrote:
> > In some cases decoding engine needs extra space in capture buffers. This
> > is the case for decoding 10-bit HEVC frames into 8-bit capture format.
> > This commit only adds infrastructure for such cases.
> > 
> > Signed-off-by: Jernej Skrabec <jernej.skrabec@gmail.com>
> > ---
> > 
> >  drivers/staging/media/sunxi/cedrus/cedrus.c       | 14 ++++++++++++++
> >  drivers/staging/media/sunxi/cedrus/cedrus.h       |  2 ++
> >  drivers/staging/media/sunxi/cedrus/cedrus_video.c |  4 ++++
> >  3 files changed, 20 insertions(+)
> > 
> > diff --git a/drivers/staging/media/sunxi/cedrus/cedrus.c
> > b/drivers/staging/media/sunxi/cedrus/cedrus.c index
> > 6a2c08928613..c36999e47591 100644
> > --- a/drivers/staging/media/sunxi/cedrus/cedrus.c
> > +++ b/drivers/staging/media/sunxi/cedrus/cedrus.c
> > @@ -68,8 +68,22 @@ static int cedrus_try_ctrl(struct v4l2_ctrl *ctrl)
> > 
> >  	return 0;
> >  
> >  }
> > 
> > +static int cedrus_s_ctrl(struct v4l2_ctrl *ctrl)
> > +{
> > +	struct cedrus_ctx *ctx = container_of(ctrl->handler,
> > +					      struct 
cedrus_ctx, hdl);
> > +	struct vb2_queue *vq = v4l2_m2m_get_vq(ctx->fh.m2m_ctx,
> > +					       
V4L2_BUF_TYPE_VIDEO_CAPTURE);
> > +
> > +	if (!vb2_is_busy(vq) && !vb2_is_streaming(vq))
> 
> You can drop the !vb2_is_streaming part. If you are streaming, then
> it will also be 'busy'.
> 
> > +		cedrus_reset_cap_format(ctx);
> 
> This is weird. I would expect that this is only called for specific
> controls, since presumably not all will change the sizeimage value. It
> looks like it applies only to V4L2_CID_STATELESS_HEVC_SPS, so I see no
> reason why you would call cedrus_reset_cap_format() for other controls as
> well.

It's more general approach. It's harmless for other cases, still within specs. 
Anyway, I can make it more specific.

> 
> And what happens if someone makes such a change *while streaming*?
> Shouldn't cedrus_try_ctrl detect that and fail with -EBUSY in that case?

SPS can also be changed during streaming, so we can't outright deny setting it 
during streaming. Also, if it changes from 10 to 8, that also works, since 
capture buffers are still big enough. Only change from 8 to 10 during streaming 
won't work and that check can be added.

> 
> Regardless, this is unexpected behavior, so some explanatory comments
> would be useful.
> 
> I'm also not sure whether it isn't better to add cedrus_s_ctrl in the next
> patch, I think it would be more understandable when seen with an actual
> control.

I guess I can move logic to cedrus_try_ctrl(), where some of this logic 
already exists and it would be just a little bit expanded and more selective.

Best regards,
Jernej

> 
> Regards,
> 
> 	Hans
> 
> > +
> > +	return 0;
> > +}
> > +
> > 
> >  static const struct v4l2_ctrl_ops cedrus_ctrl_ops = {
> >  
> >  	.try_ctrl = cedrus_try_ctrl,
> > 
> > +	.s_ctrl = cedrus_s_ctrl,
> > 
> >  };
> >  
> >  static const struct cedrus_control cedrus_controls[] = {
> > 
> > diff --git a/drivers/staging/media/sunxi/cedrus/cedrus.h
> > b/drivers/staging/media/sunxi/cedrus/cedrus.h index
> > 5904294f3108..774fe8048ce3 100644
> > --- a/drivers/staging/media/sunxi/cedrus/cedrus.h
> > +++ b/drivers/staging/media/sunxi/cedrus/cedrus.h
> > @@ -162,6 +162,8 @@ struct cedrus_dec_ops {
> > 
> >  	int (*start)(struct cedrus_ctx *ctx);
> >  	void (*stop)(struct cedrus_ctx *ctx);
> >  	void (*trigger)(struct cedrus_ctx *ctx);
> > 
> > +	unsigned int (*extra_cap_size)(struct cedrus_ctx *ctx,
> > +				       struct v4l2_pix_format 
*pix_fmt);
> > 
> >  };
> >  
> >  struct cedrus_variant {
> > 
> > diff --git a/drivers/staging/media/sunxi/cedrus/cedrus_video.c
> > b/drivers/staging/media/sunxi/cedrus/cedrus_video.c index
> > dec5d3ae4871..dc67940f1ade 100644
> > --- a/drivers/staging/media/sunxi/cedrus/cedrus_video.c
> > +++ b/drivers/staging/media/sunxi/cedrus/cedrus_video.c
> > @@ -250,6 +250,10 @@ static int cedrus_try_fmt_vid_cap_p(struct cedrus_ctx
> > *ctx,> 
> >  	pix_fmt->height = ctx->src_fmt.height;
> >  	cedrus_prepare_format(pix_fmt);
> > 
> > +	if (ctx->current_codec->extra_cap_size)
> > +		pix_fmt->sizeimage +=
> > +			ctx->current_codec->extra_cap_size(ctx, 
pix_fmt);
> > +
> > 
> >  	return 0;
> >  
> >  }





  reply	other threads:[~2022-11-04 18:10 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-11-02 18:07 [PATCH v2 00/11] media: cedrus: Format handling improvements and 10-bit HEVC support Jernej Skrabec
2022-11-02 18:08 ` [PATCH v2 01/11] media: cedrus: remove superfluous call Jernej Skrabec
2022-11-02 18:08 ` [PATCH v2 02/11] media: cedrus: Add format reset helpers Jernej Skrabec
2022-11-02 18:08 ` [PATCH v2 03/11] media: cedrus: use helper to set default formats Jernej Skrabec
2022-11-02 18:08 ` [PATCH v2 04/11] media: cedrus: Add helper for checking capabilities Jernej Skrabec
2022-11-02 18:08 ` [PATCH v2 05/11] media: cedrus: Filter controls based on capability Jernej Skrabec
2022-11-02 18:08 ` [PATCH v2 06/11] media: cedrus: set codec ops immediately Jernej Skrabec
2022-11-02 18:08 ` [PATCH v2 07/11] media: cedrus: Remove cedrus_codec enum Jernej Skrabec
2022-11-02 18:08 ` [PATCH v2 08/11] media: cedrus: prefer untiled capture format Jernej Skrabec
2022-11-02 18:08 ` [PATCH v2 09/11] media: cedrus: initialize controls a bit later Jernej Skrabec
2022-11-02 18:08 ` [PATCH v2 10/11] media: cedrus: Adjust buffer size based on control values Jernej Skrabec
2022-11-04 15:27   ` Hans Verkuil
2022-11-04 18:10     ` Jernej Škrabec [this message]
2022-11-02 18:08 ` [PATCH v2 11/11] media: cedrus: h265: Support decoding 10-bit frames Jernej Skrabec
2022-11-04 15:31 ` [PATCH v2 00/11] media: cedrus: Format handling improvements and 10-bit HEVC support Hans Verkuil

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=1908973.PYKUYFuaPT@jernej-laptop \
    --to=jernej.skrabec@gmail.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=hverkuil-cisco@xs4all.nl \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-staging@lists.linux.dev \
    --cc=linux-sunxi@lists.linux.dev \
    --cc=mchehab@kernel.org \
    --cc=mripard@kernel.org \
    --cc=paul.kocialkowski@bootlin.com \
    --cc=samuel@sholland.org \
    --cc=wens@csie.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