linux-media.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Hans Verkuil <hverkuil@xs4all.nl>
To: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>,
	linux-media@vger.kernel.org, laurent.pinchart@ideasonboard.com
Cc: helen.koike@collabora.com, ezequiel@collabora.com,
	kernel@collabora.com, dafna3@gmail.com,
	sakari.ailus@linux.intel.com, linux-rockchip@lists.infradead.org,
	mchehab@kernel.org, tfiga@chromium.org,
	skhan@linuxfoundation.org, p.zabel@pengutronix.de
Subject: Re: [RFC v4 7/8] media: vivid: Add support to the CSC API
Date: Mon, 8 Jun 2020 12:10:46 +0200	[thread overview]
Message-ID: <faedc5f3-d872-ef5c-b503-26de899be09e@xs4all.nl> (raw)
In-Reply-To: <20200605172625.19777-8-dafna.hirschfeld@collabora.com>

On 05/06/2020 19:26, Dafna Hirschfeld wrote:
> The CSC API (Colorspace conversion) allows userspace to try
> to configure the ycbcr/hsv_enc function and the quantization
> for capture devices. This patch adds support to the CSC API
> in vivid.
> Using the CSC API, userspace is allowed to do the following:
> 
> 1. Set the ycbcr_enc function for YUV formats.
> 2. Set the hsv_enc function for HSV formats
> 3. Set the quantization for YUV and RGB formats.

I just realized something. We excluded colorspace and xfer_func from the CSC API
for now since there are no drivers that support those conversions. But actually,
that's not true since vivid *does* support this unless it is in loopback mode
(then no CSC conversion is supported).

So if dev->loop_video is true, then disable CSC conversion in vivid.

Otherwise CSC can be fully enabled for all four colorimetry fields.

So I think we should add flags to signal this for colorspace and xfer_func as well
and implement this in vivid.

Sorry about this, I should have realized this (much) earlier.

Regards,

	Hans

> 
> Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
> ---
>  .../media/test-drivers/vivid/vivid-vid-cap.c  | 49 +++++++++++++++++--
>  .../test-drivers/vivid/vivid-vid-common.c     | 20 ++++++++
>  2 files changed, 65 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/media/test-drivers/vivid/vivid-vid-cap.c b/drivers/media/test-drivers/vivid/vivid-vid-cap.c
> index e94beef008c8..8a43d7ebe53f 100644
> --- a/drivers/media/test-drivers/vivid/vivid-vid-cap.c
> +++ b/drivers/media/test-drivers/vivid/vivid-vid-cap.c
> @@ -549,6 +549,29 @@ int vivid_g_fmt_vid_cap(struct file *file, void *priv,
>  	return 0;
>  }
>  
> +static bool vivid_is_hsv_enc_valid(__u8 hsv_enc)
> +{
> +	if (hsv_enc == V4L2_HSV_ENC_180 || hsv_enc == V4L2_HSV_ENC_256)
> +		return true;
> +	return false;
> +}
> +
> +static bool vivid_is_ycbcr_enc_valid(__u8 ycbcr_enc)
> +{
> +	/* V4L2_YCBCR_ENC_SMPTE240M is the last ycbcr_enc enum */
> +	if (ycbcr_enc && ycbcr_enc <= V4L2_YCBCR_ENC_SMPTE240M)
> +		return true;
> +	return false;
> +}
> +
> +static bool vivid_is_quant_valid(__u8 quantization)
> +{
> +	if (quantization == V4L2_QUANTIZATION_FULL_RANGE ||
> +	    quantization == V4L2_QUANTIZATION_LIM_RANGE)
> +		return true;
> +	return false;
> +}
> +
>  int vivid_try_fmt_vid_cap(struct file *file, void *priv,
>  			struct v4l2_format *f)
>  {
> @@ -560,6 +583,7 @@ int vivid_try_fmt_vid_cap(struct file *file, void *priv,
>  	unsigned factor = 1;
>  	unsigned w, h;
>  	unsigned p;
> +	bool user_set_csc = !!(mp->flags & V4L2_PIX_FMT_FLAG_SET_CSC);
>  
>  	fmt = vivid_get_format(dev, mp->pixelformat);
>  	if (!fmt) {
> @@ -634,12 +658,23 @@ int vivid_try_fmt_vid_cap(struct file *file, void *priv,
>  			(fmt->bit_depth[0] / fmt->vdownsampling[0]);
>  
>  	mp->colorspace = vivid_colorspace_cap(dev);
> -	if (fmt->color_enc == TGP_COLOR_ENC_HSV)
> -		mp->hsv_enc = vivid_hsv_enc_cap(dev);
> -	else
> +	if (fmt->color_enc == TGP_COLOR_ENC_HSV) {
> +		if (!user_set_csc || !vivid_is_hsv_enc_valid(mp->hsv_enc))
> +			mp->hsv_enc = vivid_hsv_enc_cap(dev);
> +	} else if (fmt->color_enc == TGP_COLOR_ENC_YCBCR) {
> +		if (!user_set_csc || !vivid_is_ycbcr_enc_valid(mp->ycbcr_enc))
> +			mp->ycbcr_enc = vivid_ycbcr_enc_cap(dev);
> +	} else {
>  		mp->ycbcr_enc = vivid_ycbcr_enc_cap(dev);
> +	}
>  	mp->xfer_func = vivid_xfer_func_cap(dev);
> -	mp->quantization = vivid_quantization_cap(dev);
> +	if (fmt->color_enc == TGP_COLOR_ENC_YCBCR ||
> +	    fmt->color_enc == TGP_COLOR_ENC_RGB) {
> +		if (!user_set_csc || !vivid_is_quant_valid(mp->quantization))
> +			mp->quantization = vivid_quantization_cap(dev);
> +	} else {
> +		mp->quantization = vivid_quantization_cap(dev);
> +	}
>  	memset(mp->reserved, 0, sizeof(mp->reserved));
>  	return 0;
>  }
> @@ -769,6 +804,12 @@ int vivid_s_fmt_vid_cap(struct file *file, void *priv,
>  	if (vivid_is_sdtv_cap(dev))
>  		dev->tv_field_cap = mp->field;
>  	tpg_update_mv_step(&dev->tpg);
> +	dev->tpg.quantization = mp->quantization;
> +	if (dev->fmt_cap->color_enc == TGP_COLOR_ENC_YCBCR)
> +		dev->tpg.ycbcr_enc = mp->ycbcr_enc;
> +	else
> +		dev->tpg.hsv_enc = mp->hsv_enc;
> +
>  	return 0;
>  }
>  
> diff --git a/drivers/media/test-drivers/vivid/vivid-vid-common.c b/drivers/media/test-drivers/vivid/vivid-vid-common.c
> index 76b0be670ebb..19aacb180e67 100644
> --- a/drivers/media/test-drivers/vivid/vivid-vid-common.c
> +++ b/drivers/media/test-drivers/vivid/vivid-vid-common.c
> @@ -920,6 +920,26 @@ int vivid_enum_fmt_vid(struct file *file, void  *priv,
>  	fmt = &vivid_formats[f->index];
>  
>  	f->pixelformat = fmt->fourcc;
> +
> +	if (f->type != V4L2_BUF_TYPE_VIDEO_CAPTURE &&
> +	    f->type != V4L2_BUF_TYPE_VIDEO_CAPTURE_MPLANE)
> +		return 0;
> +	/*
> +	 * For capture devices, we support the CSC API.
> +	 * We allow userspace to:
> +	 * 1. set the ycbcr_enc on yuv format
> +	 * 2. set the hsv_enc on hsv format
> +	 * 3. set the quantization on yuv and rgb formats
> +	 */
> +	if (fmt->color_enc == TGP_COLOR_ENC_YCBCR) {
> +		f->flags |= V4L2_FMT_FLAG_CSC_YCBCR_ENC;
> +		f->flags |= V4L2_FMT_FLAG_CSC_QUANTIZATION;
> +	} else if (fmt->color_enc == TGP_COLOR_ENC_HSV) {
> +		f->flags |= V4L2_FMT_FLAG_CSC_HSV_ENC;
> +	} else if (fmt->color_enc == TGP_COLOR_ENC_RGB) {
> +		f->flags |= V4L2_FMT_FLAG_CSC_QUANTIZATION;
> +	}
> +
>  	return 0;
>  }
>  
> 


  reply	other threads:[~2020-06-08 10:10 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-06-05 17:26 [RFC v4 0/8] v4l2: add support for colorspace conversion API (CSC) for video capture and subdevices Dafna Hirschfeld
2020-06-05 17:26 ` [RFC v4 1/8] media: staging: rkisp1: rsz: supported formats are the isp's src formats, not sink formats Dafna Hirschfeld
2020-06-08 11:33   ` Hans Verkuil
2020-06-05 17:26 ` [RFC v4 2/8] media: staging: rkisp1: rsz: set default format if the given format is not RKISP1_DIR_SRC Dafna Hirschfeld
2020-06-08 11:31   ` Hans Verkuil
2020-06-05 17:26 ` [RFC v4 3/8] media: Documentation: v4l: move table of v4l2_pix_format(_mplane) flags to pixfmt-v4l2.rst Dafna Hirschfeld
2020-06-25 23:29   ` Helen Koike
2020-06-05 17:26 ` [RFC v4 4/8] v4l2: add support for colorspace conversion API (CSC) for video capture and subdevices Dafna Hirschfeld
2020-06-08 10:00   ` Hans Verkuil
2020-06-26 12:22     ` Philipp Zabel
2020-06-30 15:23       ` Dafna Hirschfeld
2020-07-01 13:21         ` Philipp Zabel
2020-06-25 23:29   ` Helen Koike
2020-06-26  7:13     ` Hans Verkuil
2020-06-30 16:36     ` Dafna Hirschfeld
2020-06-05 17:26 ` [RFC v4 5/8] media: staging: rkisp1: allow quantization conversion from userspace for isp source pad Dafna Hirschfeld
2020-06-06  7:28   ` Dafna Hirschfeld
2020-06-25 23:52   ` Helen Koike
2020-06-26  7:07     ` Dafna Hirschfeld
2020-06-05 17:26 ` [RFC v4 6/8] media: staging: rkisp1: validate quantization matching in link_validate callbacks Dafna Hirschfeld
2020-06-05 17:26 ` [RFC v4 7/8] media: vivid: Add support to the CSC API Dafna Hirschfeld
2020-06-08 10:10   ` Hans Verkuil [this message]
2020-06-08 16:34     ` Dafna Hirschfeld
2020-06-09 11:51       ` Hans Verkuil
2020-06-05 17:26 ` [RFC v4 8/8] media: v4l2: add support for the CSC API for hsv_enc on mediabus Dafna Hirschfeld
2020-06-26 12:24   ` Philipp Zabel

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=faedc5f3-d872-ef5c-b503-26de899be09e@xs4all.nl \
    --to=hverkuil@xs4all.nl \
    --cc=dafna.hirschfeld@collabora.com \
    --cc=dafna3@gmail.com \
    --cc=ezequiel@collabora.com \
    --cc=helen.koike@collabora.com \
    --cc=kernel@collabora.com \
    --cc=laurent.pinchart@ideasonboard.com \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-rockchip@lists.infradead.org \
    --cc=mchehab@kernel.org \
    --cc=p.zabel@pengutronix.de \
    --cc=sakari.ailus@linux.intel.com \
    --cc=skhan@linuxfoundation.org \
    --cc=tfiga@chromium.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;
as well as URLs for NNTP newsgroup(s).