Re: [RESEND PATCH v5 2/2] drm/bridge: hx8837: add a Himax HX8837 display controller driver

[Lubomir Rintel]

Hello Sam,

On Fri, Oct 16, 2020 at 10:07:34PM +0200, Sam Ravnborg wrote:
> Hi Lubomir.
> 
> On Sat, Sep 26, 2020 at 02:07:19AM +0200, Lubomir Rintel wrote:
> > Himax HX8837 is used to drive the LCD panel on OLPC platforms.
> > 
> > It controls the panel backlight and is able to refresh it when the LCD
> > controller (and the rest of the plaform) is powered off.
> > 
> > It also converts regular RGB color data from the LCDC so that it looks
> > reasonable on the OLPC LCD panel with a monochromatic layer on top of a
> > layer that can either reflect light (b/w sunlight readable mode) or light
> > pattern of red, green and blue pixels.
> > 
> > At this point, the driver is rather basic. The self-refresh mode is not
> > supported. There's no way of independently controlling the color swizzling,
> > antialiasing or b/w conversion, but it probably isn't too useful either.
> > 
> > There's another driver for the same hardware on OLPC XO-1.5 and XO-1.75
> > in drivers/staging/olpc_dcon. The display on that hardware doesn't utilize
> > DRM, so this driver doesn't replace the other one yet.
> > 
> > Signed-off-by: Lubomir Rintel <lkundrak@xxxxx>
> 
> A little feedback follows.
> 
> 	Sam
> 
> > 
> > ---
> > Changes since v3:
> > - Added this patch, in place of a driver derived from
> >   drivers/staging/olpc_dcon. Compared to the previous one this
> >   implements the bare minimum, without the fancy stuff such as
> >   self-refresh that need more work/thinking.
> > 
> >  drivers/gpu/drm/bridge/Kconfig        |  13 ++
> >  drivers/gpu/drm/bridge/Makefile       |   1 +
> >  drivers/gpu/drm/bridge/himax-hx8837.c | 325 ++++++++++++++++++++++++++
> >  3 files changed, 339 insertions(+)
> >  create mode 100644 drivers/gpu/drm/bridge/himax-hx8837.c
> > 
> > diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
> > index ef91646441b16..6a923dd56c1d5 100644
> > --- a/drivers/gpu/drm/bridge/Kconfig
> > +++ b/drivers/gpu/drm/bridge/Kconfig
> > @@ -48,6 +48,19 @@ config DRM_DISPLAY_CONNECTOR
> >  	  on ARM-based platforms. Saying Y here when this driver is not needed
> >  	  will not cause any issue.
> >  
> > +config DRM_HIMAX_HX8837
> > +        tristate "HiMax HX8837 OLPC Display Controller"
> > +	depends on OF
> > +	depends on OLPC || ARCH_MMP || COMPILE_TEST
> > +	select DRM_KMS_HELPER
> > +        select BACKLIGHT_LCD_SUPPORT
> > +        select BACKLIGHT_CLASS_DEVICE
> > +        help
> > +          Enable support for HiMax HX8837 Display Controller as found in the
> > +          OLPC XO laptops.
> > +
> > +          If your laptop doesn't have green ears, say "N"
> 
> There is a mixture of tabs and spaces for indent - use tabs only (and
> tabs + 2 spaces for the help text).
> 
> 
> > +
> >  config DRM_LONTIUM_LT9611
> >  	tristate "Lontium LT9611 DSI/HDMI bridge"
> >  	select SND_SOC_HDMI_CODEC if SND_SOC
> > diff --git a/drivers/gpu/drm/bridge/Makefile b/drivers/gpu/drm/bridge/Makefile
> > index 2b3aff104e466..21f72df3260db 100644
> > --- a/drivers/gpu/drm/bridge/Makefile
> > +++ b/drivers/gpu/drm/bridge/Makefile
> > @@ -2,6 +2,7 @@
> >  obj-$(CONFIG_DRM_CDNS_DSI) += cdns-dsi.o
> >  obj-$(CONFIG_DRM_CHRONTEL_CH7033) += chrontel-ch7033.o
> >  obj-$(CONFIG_DRM_DISPLAY_CONNECTOR) += display-connector.o
> > +obj-$(CONFIG_DRM_HIMAX_HX8837) += himax-hx8837.o
> Please add in alphabetical order.
> 
> >  obj-$(CONFIG_DRM_LONTIUM_LT9611) += lontium-lt9611.o
> >  obj-$(CONFIG_DRM_LVDS_CODEC) += lvds-codec.o
> >  obj-$(CONFIG_DRM_MEGACHIPS_STDPXXXX_GE_B850V3_FW) += megachips-stdpxxxx-ge-b850v3-fw.o
> > diff --git a/drivers/gpu/drm/bridge/himax-hx8837.c b/drivers/gpu/drm/bridge/himax-hx8837.c
> > new file mode 100644
> > index 0000000000000..1e97fcb8ce505
> > --- /dev/null
> > +++ b/drivers/gpu/drm/bridge/himax-hx8837.c
> > @@ -0,0 +1,325 @@
> > +// SPDX-License-Identifier: GPL-2.0-or-later
> > +/*
> > + * HiMax HX8837 Display Controller Driver
> > + *
> > + * Datasheet: http://wiki.laptop.org/images/0/09/DCON_datasheet_HX8837-A.pdf
> > + *
> > + * Copyright (C) 2020 Lubomir Rintel
> > + */
> > +
> > +#include <linux/gpio/consumer.h>
> > +#include <linux/module.h>
> > +#include <linux/regmap.h>
> > +
> > +#include <drm/drm_atomic_helper.h>
> > +#include <drm/drm_panel.h>
> > +#include <drm/drm_bridge.h>
> > +#include <drm/drm_of.h>
> > +#include <drm/drm_print.h>
> > +#include <drm/drm_probe_helper.h>
> In blocks are good but please add them in alphabetical order.
> 
> > +
> > +#define bridge_to_hx8837_priv(x) \
> > +	container_of(x, struct hx8837_priv, bridge)
> > +
> > +/* DCON registers */
> > +enum {
> > +	DCON_REG_ID		= 0x00,
> > +	DCON_REG_MODE		= 0x01,
> > +	DCON_REG_HRES		= 0x02,
> > +	DCON_REG_HTOTAL		= 0x03,
> > +	DCON_REG_HSYNC_WIDTH	= 0x04,
> > +	DCON_REG_VRES		= 0x05,
> > +	DCON_REG_VTOTAL		= 0x06,
> > +	DCON_REG_VSYNC_WIDTH	= 0x07,
> > +	DCON_REG_TIMEOUT	= 0x08,
> > +	DCON_REG_SCAN_INT	= 0x09,
> > +	DCON_REG_BRIGHT		= 0x0a,
> > +	DCON_REG_MEM_OPT_A	= 0x41,
> > +	DCON_REG_MEM_OPT_B	= 0x42,
> > +};
> > +
> > +/* DCON_REG_MODE */
> > +enum {
> > +	MODE_PASSTHRU		= BIT(0),
> > +	MODE_SLEEP		= BIT(1),
> > +	MODE_SLEEP_AUTO		= BIT(2),
> > +	MODE_BL_ENABLE		= BIT(3),
> > +	MODE_BLANK		= BIT(4),
> > +	MODE_CSWIZZLE		= BIT(5),
> > +	MODE_COL_AA		= BIT(6),
> > +	MODE_MONO_LUMA		= BIT(7),
> > +	MODE_SCAN_INT		= BIT(8),
> > +	MODE_CLOCKDIV		= BIT(9),
> > +	MODE_DEBUG		= BIT(14),
> > +	MODE_SELFTEST		= BIT(15),
> > +};
> > +
> > +struct hx8837_priv {
> > +	struct regmap *regmap;
> > +	struct gpio_desc *load_gpio;
> > +
> > +	struct drm_bridge *panel_bridge;
> > +	struct drm_panel *panel;
> No need for the panel pointer - use a local variable in the probe
> function 
> 
> > +	struct drm_bridge bridge;
> > +};
> > +
> > +static int hx8837_bridge_attach(struct drm_bridge *bridge,
> > +				enum drm_bridge_attach_flags flags)
> > +{
> > +	struct hx8837_priv *priv = bridge_to_hx8837_priv(bridge);
> > +
> > +	return drm_bridge_attach(bridge->encoder, priv->panel_bridge,
> > +				 bridge, flags);
> > +}
> > +
> > +static enum drm_mode_status hx8837_bridge_mode_valid(
> > +				struct drm_bridge *bridge,
> > +				const struct drm_display_info *info,
> > +				const struct drm_display_mode *mode)
> > +{
> > +	if (mode->hdisplay > 0xffff)
> > +		return MODE_BAD_HVALUE;
> > +	if (mode->htotal > 0xffff)
> > +		return MODE_BAD_HVALUE;
> > +	if (mode->hsync_start - mode->hdisplay > 0xff)
> > +		return MODE_HBLANK_WIDE;
> > +	if (mode->hsync_end - mode->hsync_start > 0xff)
> > +		return MODE_HSYNC_WIDE;
> > +	if (mode->vdisplay > 0xffff)
> > +		return MODE_BAD_VVALUE;
> > +	if (mode->vtotal > 0xffff)
> > +		return MODE_BAD_VVALUE;
> > +	if (mode->vsync_start - mode->vdisplay > 0xff)
> > +		return MODE_VBLANK_WIDE;
> > +	if (mode->vsync_end - mode->vsync_start > 0xff)
> > +		return MODE_VSYNC_WIDE;
> > +
> > +	return MODE_OK;
> > +}
> > +
> > +static void hx8837_bridge_disable(struct drm_bridge *bridge)
> > +{
> > +	struct hx8837_priv *priv = bridge_to_hx8837_priv(bridge);
> > +	int ret;
> > +
> > +	ret = gpiod_direction_output(priv->load_gpio, 0);
> > +	if (ret)
> > +		DRM_ERROR("error enabling the dcon load: %d\n", ret);
> The driver uses both dev_xxx() and DRM_XXX for logging.
> Please stick to one way to do logging - I prefer for bridge drivers
> the dev_* variants.
> 
> > +
> > +	ret = regmap_update_bits(priv->regmap, DCON_REG_MODE,
> > +					       MODE_PASSTHRU |
> > +					       MODE_SLEEP,
> > +					       MODE_PASSTHRU |
> > +					       MODE_SLEEP);
> > +	if (ret)
> > +		DRM_ERROR("error disabling the dcon: %d\n", ret);
> > +}
> > +
> > +static void hx8837_bridge_enable(struct drm_bridge *bridge)
> > +{
> > +	struct hx8837_priv *priv = bridge_to_hx8837_priv(bridge);
> > +	int ret;
> > +
> > +	ret = regmap_update_bits(priv->regmap, DCON_REG_MODE,
> > +					       MODE_PASSTHRU |
> > +					       MODE_SLEEP |
> > +					       MODE_SLEEP_AUTO |
> > +					       MODE_BLANK |
> > +					       MODE_SCAN_INT |
> > +					       MODE_CLOCKDIV |
> > +					       MODE_DEBUG |
> > +					       MODE_SELFTEST,
> > +					       MODE_PASSTHRU);
> > +	if (ret)
> > +		DRM_ERROR("error enabling the dcon: %d\n", ret);
> > +
> > +	ret = gpiod_direction_output(priv->load_gpio, 1);
> > +	if (ret)
> > +		DRM_ERROR("error enabling the dcon load: %d\n", ret);
> > +}
> > +
> > +static void hx8837_bridge_mode_set(struct drm_bridge *bridge,
> > +			   const struct drm_display_mode *mode,
> > +			   const struct drm_display_mode *adjusted_mode)
> > +{
> > +	struct hx8837_priv *priv = bridge_to_hx8837_priv(bridge);
> > +
> > +	regmap_write(priv->regmap, DCON_REG_HRES, mode->hdisplay);
> > +	regmap_write(priv->regmap, DCON_REG_HTOTAL, mode->htotal);
> > +	regmap_write(priv->regmap, DCON_REG_HSYNC_WIDTH,
> > +			(mode->hsync_start - mode->hdisplay) << 8 |
> > +			(mode->hsync_end - mode->hsync_start));
> > +	regmap_write(priv->regmap, DCON_REG_VRES, mode->vdisplay);
> > +	regmap_write(priv->regmap, DCON_REG_VTOTAL, mode->vtotal);
> > +	regmap_write(priv->regmap, DCON_REG_VSYNC_WIDTH,
> > +			(mode->vsync_start - mode->vdisplay) << 8 |
> > +			(mode->vsync_end - mode->vsync_start));
> > +}
> > +
> > +static const struct drm_bridge_funcs hx8837_bridge_funcs = {
> > +	.attach = hx8837_bridge_attach,
> > +	.mode_valid = hx8837_bridge_mode_valid,
> > +	.disable = hx8837_bridge_disable,
> > +	.enable = hx8837_bridge_enable,
> > +	.mode_set = hx8837_bridge_mode_set,
> > +};
> > +
> > +static int hx8837_bl_update_status(struct backlight_device *bl)
> > +{
> > +	struct hx8837_priv *priv = bl_get_data(bl);
> > +	unsigned int val;
> > +	int ret;
> > +
> > +	ret = regmap_update_bits(priv->regmap, DCON_REG_BRIGHT,
> > +					       0x000f,
> > +					       bl->props.brightness);
> 
> Use backlight_get_brightness() to get the brightness.
> This will also make sure 0 is returned when backlight is off so the
> logic a few lines down is correct.

I'm not sure I understand this one. I'm wondering if you could help me out
with it before I follow up with v4.

Currently I read in the current brightness level in probe() (which
prevents struct backlight_properties, below, from being const) and the
nthe brightness is entirely in control of the driver via
update_status().

What would I need get_brightness() for? We know that whatever the driver
set is the current level. It doesn't seem to be called on backlight
device registration so it doesn't make the readin in probe()
unnecessary either.

> > +	if (ret) {
> > +		dev_err(&bl->dev, "error setting the backlight: %d\n", ret);
> > +		return ret;
> > +	}
> > +
> > +	if (bl->props.brightness)
> > +		val = MODE_CSWIZZLE | MODE_COL_AA;
> > +	else
> > +		val = MODE_MONO_LUMA;
> > +
> > +	ret = regmap_update_bits(priv->regmap, DCON_REG_MODE,
> > +					       MODE_CSWIZZLE |
> > +					       MODE_COL_AA |
> > +					       MODE_MONO_LUMA,
> > +					       val);
> > +	if (ret) {
> > +		dev_err(&bl->dev, "error setting color mode: %d\n", ret);
> > +		return ret;
> > +	}
> > +
> > +	return 0;
> > +}
> > +
> > +static const struct backlight_ops hx8837_bl_ops = {
> > +	.update_status = hx8837_bl_update_status,
> > +};
> > +
> > +static const struct regmap_config hx8837_regmap_config = {
> > +	.reg_bits = 8,
> > +	.val_bits = 16,
> > +	.max_register = 0x4c,
> > +	.val_format_endian = REGMAP_ENDIAN_LITTLE,
> > +};
> > +
> > +static int hx8837_probe(struct i2c_client *client,
> > +			const struct i2c_device_id *id)
> > +{
> > +	struct backlight_properties bl_props = {
> > +		.type = BACKLIGHT_RAW,
> > +		.max_brightness = 0xf,
> > +	};
> Make it const
> Maybe use a constant for the max value as it is also used as a mask
> later.
> 
> > +	struct device *dev = &client->dev;
> > +	struct backlight_device *bl;
> > +	struct hx8837_priv *priv;
> > +	unsigned int val;
> > +	int ret;
> > +
> > +	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
> > +	if (!priv)
> > +		return -ENOMEM;
> > +
> > +	dev_set_drvdata(dev, priv);
> > +
> > +	priv->load_gpio = devm_gpiod_get(dev, "load", GPIOD_ASIS);
> > +	if (IS_ERR(priv->load_gpio))
> > +		return PTR_ERR(priv->load_gpio);
> > +
> > +	ret = drm_of_find_panel_or_bridge(dev->of_node, 1, -1,
> > +					  &priv->panel, NULL);
> > +	if (ret)
> > +		return ret;
> > +
> > +	if (priv->panel->backlight) {
> > +		dev_err(dev, "the panel already has a backlight controller\n");
> > +		return -ENODEV;
> > +	}
> > +
> > +	priv->panel_bridge = devm_drm_panel_bridge_add(dev, priv->panel);
> > +	if (IS_ERR(priv->panel_bridge))
> > +		return PTR_ERR(priv->panel_bridge);
> > +
> > +	priv->regmap = devm_regmap_init_i2c(client, &hx8837_regmap_config);
> > +	if (IS_ERR(priv->regmap)) {
> > +		dev_err(dev, "regmap init failed\n");
> > +		return PTR_ERR(priv->regmap);
> > +	}
> > +
> > +	ret = regmap_read(priv->regmap, DCON_REG_ID, &val);
> > +	if (ret < 0) {
> > +		dev_err(dev, "error reading the model id: %d\n", ret);
> > +		return ret;
> > +	}
> > +	if ((val & 0xff00) != 0xdc00) {
> > +		dev_err(dev, "the device is not a hx8837\n");
> > +		return -ENODEV;
> > +	}
> > +
> > +	ret = regmap_read(priv->regmap, DCON_REG_BRIGHT, &val);
> > +	if (ret < 0) {
> > +		dev_err(&bl->dev, "error getting the backlight: %d\n", ret);
> > +		return ret;
> > +	}
> > +	bl_props.brightness = val & 0xf;
> > +
> > +	bl = devm_backlight_device_register(dev, dev_name(dev), dev, priv,
> > +					    &hx8837_bl_ops, &bl_props);
> > +	if (IS_ERR(bl)) {
> > +		dev_err(dev, "failed to register backlight\n");
> > +		return PTR_ERR(bl);
> > +	}
> > +
> > +	priv->panel->backlight = bl;
> > +
> > +	INIT_LIST_HEAD(&priv->bridge.list);
> > +	priv->bridge.funcs = &hx8837_bridge_funcs;
> > +	priv->bridge.of_node = dev->of_node;
> > +	drm_bridge_add(&priv->bridge);
> > +
> > +	dev_info(dev, "HiMax HX8837 Display Controller Driver\n");
> > +	return 0;
> > +}
> > +
> > +static int hx8837_remove(struct i2c_client *client)
> > +{
> > +	struct device *dev = &client->dev;
> > +	struct hx8837_priv *priv = dev_get_drvdata(dev);
> > +
> > +	drm_bridge_remove(&priv->bridge);
> 
> > +	priv->panel->backlight = NULL;
> I cannot see why this is needed.
> IF this is needed we have a bug in the core stuff we need to fix.
> 
> > +
> > +	return 0;
> > +}
> > +
> > +static const struct of_device_id hx8837_dt_ids[] = {
> > +	{ .compatible = "himax,hx8837", },
> > +	{ }
> We often see { /* sentinel */ },
> in these cases.
> 
> > +};
> > +MODULE_DEVICE_TABLE(of, hx8837_dt_ids);
> > +
> > +static const struct i2c_device_id hx8837_ids[] = {
> > +	{ "hx8837", 0 },
> > +	{ }
> Likewise
> 
> > +};
> > +MODULE_DEVICE_TABLE(i2c, hx8837_ids);
> > +
> > +static struct i2c_driver hx8837_driver = {
> > +	.probe = hx8837_probe,
> > +	.remove = hx8837_remove,
> > +	.driver = {
> > +		.name = "hx8837",
> > +		.of_match_table = of_match_ptr(hx8837_dt_ids),
> > +	},
> > +	.id_table = hx8837_ids,
> > +};
> > +
> > +module_i2c_driver(hx8837_driver);
> > +
> > +MODULE_AUTHOR("Lubomir Rintel <lkundrak@xxxxx>");
> > +MODULE_DESCRIPTION("HiMax HX8837 Display Controller Driver");
> > +MODULE_LICENSE("GPL");

Thanks
Lubo