linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andrzej Hajda <a.hajda@samsung.com>
To: Jonas Karlman <jonas@kwiboo.se>,
	"Laurent.pinchart@ideasonboard.com" 
	<Laurent.pinchart@ideasonboard.com>
Cc: "jernej.skrabec@siol.net" <jernej.skrabec@siol.net>,
	"narmstrong@baylibre.com" <narmstrong@baylibre.com>,
	"khilman@baylibre.com" <khilman@baylibre.com>,
	"zhengyang@rock-chips.com" <zhengyang@rock-chips.com>,
	"maxime.ripard@bootlin.com" <maxime.ripard@bootlin.com>,
	"wens@csie.org" <wens@csie.org>,
	"hjc@rock-chips.com" <hjc@rock-chips.com>,
	"heiko@sntech.de" <heiko@sntech.de>,
	"dri-devel@lists.freedesktop.org"
	<dri-devel@lists.freedesktop.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 1/4] drm/bridge: dw-hdmi: Add Dynamic Range and Mastering InfoFrame support
Date: Mon, 24 Jun 2019 15:40:24 +0200	[thread overview]
Message-ID: <a832e26c-811b-7e75-abe9-7c2cdeeb72fb@samsung.com> (raw)
In-Reply-To: <VI1PR03MB4206B9AF66443E35946BD000AC1C0@VI1PR03MB4206.eurprd03.prod.outlook.com>

On 26.05.2019 23:19, Jonas Karlman wrote:
> Add support for configuring Dynamic Range and Mastering InfoFrame from
> the hdr_output_metadata connector property.
>
> This patch adds a drm_infoframe flag to dw_hdmi_plat_data that platform drivers
> use to signal when Dynamic Range and Mastering infoframes is supported.
> This flag is needed because Amlogic GXBB and GXL report same DW-HDMI version,
> and only GXL support DRM InfoFrame.
>
> These changes were based on work done by Zheng Yang <zhengyang@rock-chips.com>
> to support DRM InfoFrame on the Rockchip 4.4 BSP kernel at [1] and [2]
>
> [1] https://github.com/rockchip-linux/kernel/tree/develop-4.4
> [2] https://github.com/rockchip-linux/kernel/commit/d1943fde81ff41d7cca87f4a42f03992e90bddd5
>
> Cc: Zheng Yang <zhengyang@rock-chips.com>
> Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
> ---
>  drivers/gpu/drm/bridge/synopsys/dw-hdmi.c | 109 ++++++++++++++++++++++
>  drivers/gpu/drm/bridge/synopsys/dw-hdmi.h |  37 ++++++++
>  include/drm/bridge/dw_hdmi.h              |   1 +
>  3 files changed, 147 insertions(+)
>
> diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
> index 284ce59be8f8..801bbbd732fd 100644
> --- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
> +++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c
> @@ -24,6 +24,7 @@
>  
>  #include <drm/drm_of.h>
>  #include <drm/drmP.h>
> +#include <drm/drm_atomic.h>
>  #include <drm/drm_atomic_helper.h>
>  #include <drm/drm_edid.h>
>  #include <drm/drm_encoder_slave.h>
> @@ -1592,6 +1593,78 @@ static void hdmi_config_vendor_specific_infoframe(struct dw_hdmi *hdmi,
>  			HDMI_FC_DATAUTO0_VSD_MASK);
>  }
>  
> +#define HDR_LSB(n) ((n) & 0xff)
> +#define HDR_MSB(n) (((n) & 0xff00) >> 8)
> +
> +static void hdmi_config_drm_infoframe(struct dw_hdmi *hdmi)
> +{
> +	const struct drm_connector_state *conn_state = hdmi->connector.state;
> +	struct hdmi_drm_infoframe frame;
> +	int ret;
> +
> +	if (hdmi->version < 0x200a || !hdmi->plat_data->drm_infoframe)


Why do we need this double check? Here and in dw_hdmi_bridge_attach.

Wouldn't be enough to test drm_infoframe?


> +		return;
> +
> +	hdmi_modb(hdmi, HDMI_FC_PACKET_TX_EN_DRM_DISABLE,
> +		  HDMI_FC_PACKET_TX_EN_DRM_MASK, HDMI_FC_PACKET_TX_EN);
> +
> +	ret = drm_hdmi_infoframe_set_hdr_metadata(&frame, conn_state);
> +	if (ret < 0)
> +		return;
> +
> +	ret = hdmi_drm_infoframe_check(&frame);
> +	if (WARN_ON(ret))
> +		return;
> +
> +	hdmi_writeb(hdmi, frame.version, HDMI_FC_DRM_HB0);
> +	hdmi_writeb(hdmi, frame.length, HDMI_FC_DRM_HB1);
> +	hdmi_writeb(hdmi, frame.eotf, HDMI_FC_DRM_PB0);
> +	hdmi_writeb(hdmi, frame.metadata_type, HDMI_FC_DRM_PB1);
> +	hdmi_writeb(hdmi, HDR_LSB(frame.display_primaries[0].x),
> +		    HDMI_FC_DRM_PB2);
> +	hdmi_writeb(hdmi, HDR_MSB(frame.display_primaries[0].x),
> +		    HDMI_FC_DRM_PB3);


Adding two-byte version hdmi_writew would allow to remove multiple lines
here, but since register names and their relative location reflects
structure of raw frame you should be able to replace this
long/boring/error-prone sequence of writes with something like:

len = hdmi_infoframe_pack(frame, buf, len);

if (len < 0)

    ...

for (i = 0; i < len; ++i)

    hdmi_writeb(hdmi, buf[i], HDMI_FC_DRM_HB0 + i);


... or something similar.


> +	hdmi_writeb(hdmi, HDR_LSB(frame.display_primaries[0].y),
> +		    HDMI_FC_DRM_PB4);
> +	hdmi_writeb(hdmi, HDR_MSB(frame.display_primaries[0].y),
> +		    HDMI_FC_DRM_PB5);
> +	hdmi_writeb(hdmi, HDR_LSB(frame.display_primaries[1].x),
> +		    HDMI_FC_DRM_PB6);
> +	hdmi_writeb(hdmi, HDR_MSB(frame.display_primaries[1].x),
> +		    HDMI_FC_DRM_PB7);
> +	hdmi_writeb(hdmi, HDR_LSB(frame.display_primaries[1].y),
> +		    HDMI_FC_DRM_PB8);
> +	hdmi_writeb(hdmi, HDR_MSB(frame.display_primaries[1].y),
> +		    HDMI_FC_DRM_PB9);
> +	hdmi_writeb(hdmi, HDR_LSB(frame.display_primaries[2].x),
> +		    HDMI_FC_DRM_PB10);
> +	hdmi_writeb(hdmi, HDR_MSB(frame.display_primaries[2].x),
> +		    HDMI_FC_DRM_PB11);
> +	hdmi_writeb(hdmi, HDR_LSB(frame.display_primaries[2].y),
> +		    HDMI_FC_DRM_PB12);
> +	hdmi_writeb(hdmi, HDR_MSB(frame.display_primaries[2].y),
> +		    HDMI_FC_DRM_PB13);
> +	hdmi_writeb(hdmi, HDR_LSB(frame.white_point.x), HDMI_FC_DRM_PB14);
> +	hdmi_writeb(hdmi, HDR_MSB(frame.white_point.x), HDMI_FC_DRM_PB15);
> +	hdmi_writeb(hdmi, HDR_LSB(frame.white_point.y), HDMI_FC_DRM_PB16);
> +	hdmi_writeb(hdmi, HDR_MSB(frame.white_point.y), HDMI_FC_DRM_PB17);
> +	hdmi_writeb(hdmi, HDR_LSB(frame.max_display_mastering_luminance),
> +		    HDMI_FC_DRM_PB18);
> +	hdmi_writeb(hdmi, HDR_MSB(frame.max_display_mastering_luminance),
> +		    HDMI_FC_DRM_PB19);
> +	hdmi_writeb(hdmi, HDR_LSB(frame.min_display_mastering_luminance),
> +		    HDMI_FC_DRM_PB20);
> +	hdmi_writeb(hdmi, HDR_MSB(frame.min_display_mastering_luminance),
> +		    HDMI_FC_DRM_PB21);
> +	hdmi_writeb(hdmi, HDR_LSB(frame.max_cll), HDMI_FC_DRM_PB22);
> +	hdmi_writeb(hdmi, HDR_MSB(frame.max_cll), HDMI_FC_DRM_PB23);
> +	hdmi_writeb(hdmi, HDR_LSB(frame.max_fall), HDMI_FC_DRM_PB24);
> +	hdmi_writeb(hdmi, HDR_MSB(frame.max_fall), HDMI_FC_DRM_PB25);
> +	hdmi_writeb(hdmi, 1, HDMI_FC_DRM_UP);
> +	hdmi_modb(hdmi, HDMI_FC_PACKET_TX_EN_DRM_ENABLE,
> +		  HDMI_FC_PACKET_TX_EN_DRM_MASK, HDMI_FC_PACKET_TX_EN);
> +}
> +
>  static void hdmi_av_composer(struct dw_hdmi *hdmi,
>  			     const struct drm_display_mode *mode)
>  {
> @@ -1963,6 +2036,7 @@ static int dw_hdmi_setup(struct dw_hdmi *hdmi, struct drm_display_mode *mode)
>  		/* HDMI Initialization Step F - Configure AVI InfoFrame */
>  		hdmi_config_AVI(hdmi, mode);
>  		hdmi_config_vendor_specific_infoframe(hdmi, mode);
> +		hdmi_config_drm_infoframe(hdmi);
>  	} else {
>  		dev_dbg(hdmi->dev, "%s DVI mode\n", __func__);
>  	}
> @@ -2139,6 +2213,36 @@ static int dw_hdmi_connector_get_modes(struct drm_connector *connector)
>  	return ret;
>  }
>  
> +static bool blob_equal(const struct drm_property_blob *a,
> +		       const struct drm_property_blob *b)


Quite generic function, but since there are no other users(?) it can
stay here.


> +{
> +	if (a && b)
> +		return a->length == b->length &&
> +			!memcmp(a->data, b->data, a->length);
> +
> +	return !a == !b;


Ugly code (but it is probably matter of taste), maybe sth like this:

    if (!a || !b)

        return a != b;

    if (a->length != b->length)

        return false;

    return !memcmp(a->data, b->data, a->length);


> +}
> +
> +static int dw_hdmi_connector_atomic_check(struct drm_connector *conn,
> +	struct drm_connector_state *new_state)
> +{
> +	struct drm_connector_state *old_state =
> +		drm_atomic_get_old_connector_state(new_state->state, conn);
> +	struct drm_crtc_state *crtc_state;
> +
> +	if (!new_state->crtc)
> +		return 0;
> +
> +	crtc_state = drm_atomic_get_new_crtc_state(new_state->state,
> +						   new_state->crtc);
> +
> +	if (!blob_equal(new_state->hdr_output_metadata,
> +			old_state->hdr_output_metadata))
> +		crtc_state->mode_changed = true;
> +
> +	return 0;
> +}
> +
>  static void dw_hdmi_connector_force(struct drm_connector *connector)
>  {
>  	struct dw_hdmi *hdmi = container_of(connector, struct dw_hdmi,
> @@ -2163,6 +2267,7 @@ static const struct drm_connector_funcs dw_hdmi_connector_funcs = {
>  
>  static const struct drm_connector_helper_funcs dw_hdmi_connector_helper_funcs = {
>  	.get_modes = dw_hdmi_connector_get_modes,
> +	.atomic_check = dw_hdmi_connector_atomic_check,
>  };
>  
>  static int dw_hdmi_bridge_attach(struct drm_bridge *bridge)
> @@ -2179,6 +2284,10 @@ static int dw_hdmi_bridge_attach(struct drm_bridge *bridge)
>  	drm_connector_init(bridge->dev, connector, &dw_hdmi_connector_funcs,
>  			   DRM_MODE_CONNECTOR_HDMIA);
>  
> +	if (hdmi->version >= 0x200a && hdmi->plat_data->drm_infoframe)
> +		drm_object_attach_property(&connector->base,
> +			connector->dev->mode_config.hdr_output_metadata_property, 0);
> +
>  	drm_connector_attach_encoder(connector, encoder);
>  
>  	return 0;
> diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.h b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.h
> index 3f3c616eba97..d4efbec14f68 100644
> --- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.h
> +++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.h
> @@ -256,6 +256,7 @@
>  #define HDMI_FC_POL2                            0x10DB
>  #define HDMI_FC_PRCONF                          0x10E0
>  #define HDMI_FC_SCRAMBLER_CTRL                  0x10E1
> +#define HDMI_FC_PACKET_TX_EN                    0x10E3
>  
>  #define HDMI_FC_GMD_STAT                        0x1100
>  #define HDMI_FC_GMD_EN                          0x1101
> @@ -291,6 +292,37 @@
>  #define HDMI_FC_GMD_PB26                        0x111F
>  #define HDMI_FC_GMD_PB27                        0x1120
>  
> +#define HDMI_FC_DRM_UP                          0x1167
> +#define HDMI_FC_DRM_HB0                         0x1168
> +#define HDMI_FC_DRM_HB1                         0x1169
> +#define HDMI_FC_DRM_PB0                         0x116A
> +#define HDMI_FC_DRM_PB1                         0x116B
> +#define HDMI_FC_DRM_PB2                         0x116C
> +#define HDMI_FC_DRM_PB3                         0x116D
> +#define HDMI_FC_DRM_PB4                         0x116E
> +#define HDMI_FC_DRM_PB5                         0x116F
> +#define HDMI_FC_DRM_PB6                         0x1170
> +#define HDMI_FC_DRM_PB7                         0x1171
> +#define HDMI_FC_DRM_PB8                         0x1172
> +#define HDMI_FC_DRM_PB9                         0x1173
> +#define HDMI_FC_DRM_PB10                        0x1174
> +#define HDMI_FC_DRM_PB11                        0x1175
> +#define HDMI_FC_DRM_PB12                        0x1176
> +#define HDMI_FC_DRM_PB13                        0x1177
> +#define HDMI_FC_DRM_PB14                        0x1178
> +#define HDMI_FC_DRM_PB15                        0x1179
> +#define HDMI_FC_DRM_PB16                        0x117A
> +#define HDMI_FC_DRM_PB17                        0x117B
> +#define HDMI_FC_DRM_PB18                        0x117C
> +#define HDMI_FC_DRM_PB19                        0x117D
> +#define HDMI_FC_DRM_PB20                        0x117E
> +#define HDMI_FC_DRM_PB21                        0x117F
> +#define HDMI_FC_DRM_PB22                        0x1180
> +#define HDMI_FC_DRM_PB23                        0x1181
> +#define HDMI_FC_DRM_PB24                        0x1182
> +#define HDMI_FC_DRM_PB25                        0x1183
> +#define HDMI_FC_DRM_PB26                        0x1184
> +
>  #define HDMI_FC_DBGFORCE                        0x1200
>  #define HDMI_FC_DBGAUD0CH0                      0x1201
>  #define HDMI_FC_DBGAUD1CH0                      0x1202
> @@ -746,6 +778,11 @@ enum {
>  	HDMI_FC_PRCONF_OUTPUT_PR_FACTOR_MASK = 0x0F,
>  	HDMI_FC_PRCONF_OUTPUT_PR_FACTOR_OFFSET = 0,
>  
> +/* FC_PACKET_TX_EN field values */
> +	HDMI_FC_PACKET_TX_EN_DRM_MASK = 0x80,
> +	HDMI_FC_PACKET_TX_EN_DRM_ENABLE = 0x80,
> +	HDMI_FC_PACKET_TX_EN_DRM_DISABLE = 0x00,
> +
>  /* FC_AVICONF0-FC_AVICONF3 field values */
>  	HDMI_FC_AVICONF0_PIX_FMT_MASK = 0x03,
>  	HDMI_FC_AVICONF0_PIX_FMT_RGB = 0x00,
> diff --git a/include/drm/bridge/dw_hdmi.h b/include/drm/bridge/dw_hdmi.h
> index 0f0e82638fbe..04d1ec60f218 100644
> --- a/include/drm/bridge/dw_hdmi.h
> +++ b/include/drm/bridge/dw_hdmi.h
> @@ -131,6 +131,7 @@ struct dw_hdmi_plat_data {
>  	unsigned long input_bus_format;
>  	unsigned long input_bus_encoding;
>  	bool ycbcr_420_allowed;
> +	bool drm_infoframe;


maybe sth less confusing, for example: use_drm_infoframe?


Regards

Andrzej



>  
>  	/* Vendor PHY support */
>  	const struct dw_hdmi_phy_ops *phy_ops;



  parent reply	other threads:[~2019-06-24 13:40 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-05-26 21:18 [PATCH 0/4] drm/bridge: dw-hdmi: Add support for HDR metadata Jonas Karlman
2019-05-26 21:19 ` [PATCH 1/4] drm/bridge: dw-hdmi: Add Dynamic Range and Mastering InfoFrame support Jonas Karlman
2019-06-05 12:57   ` Neil Armstrong
2019-06-24 13:40   ` Andrzej Hajda [this message]
2019-05-26 21:20 ` [PATCH 2/4] drm/rockchip: Enable DRM InfoFrame support on RK3328 and RK3399 Jonas Karlman
2019-06-14  9:29   ` Heiko Stuebner
2019-06-24 14:43   ` Andrzej Hajda
2019-05-26 21:20 ` [PATCH 3/4] drm/meson: Enable DRM InfoFrame support on GXL, GXM and G12A Jonas Karlman
2019-06-05 12:55   ` Neil Armstrong
2019-06-24 14:59   ` Andrzej Hajda
2019-06-26  9:16     ` Neil Armstrong
2019-05-26 21:20 ` [PATCH 4/4] drm/sun4i: Enable DRM InfoFrame support on H6 Jonas Karlman
2019-06-24 15:03   ` Andrzej Hajda
2019-06-24 15:05     ` Jernej Škrabec
2019-06-24 15:49       ` Andrzej Hajda
2019-06-24 15:56         ` Chen-Yu Tsai
2019-06-24 16:03           ` Jernej Škrabec
2019-06-24 16:07             ` Chen-Yu Tsai
2019-06-24 16:59               ` Andrzej Hajda
2019-05-28 17:03 ` [PATCH 0/4] drm/bridge: dw-hdmi: Add support for HDR metadata Jernej Škrabec
2019-06-20 14:40 ` Neil Armstrong
2019-06-21  9:01   ` Daniel Vetter
2019-06-23 23:30     ` Laurent Pinchart
2019-06-24  8:19       ` Neil Armstrong
2019-06-24  8:42         ` Daniel Vetter
2019-06-24 11:16         ` Laurent Pinchart
2019-06-24 15:27           ` Andrzej Hajda
2019-09-18  8:05 ` Neil Armstrong
2019-09-18 16:39   ` Jonas Karlman

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=a832e26c-811b-7e75-abe9-7c2cdeeb72fb@samsung.com \
    --to=a.hajda@samsung.com \
    --cc=Laurent.pinchart@ideasonboard.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=heiko@sntech.de \
    --cc=hjc@rock-chips.com \
    --cc=jernej.skrabec@siol.net \
    --cc=jonas@kwiboo.se \
    --cc=khilman@baylibre.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=maxime.ripard@bootlin.com \
    --cc=narmstrong@baylibre.com \
    --cc=wens@csie.org \
    --cc=zhengyang@rock-chips.com \
    /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).