All of lore.kernel.org
 help / color / mirror / Atom feed
From: Miquel Raynal <miquel.raynal@bootlin.com>
To: Maxime Ripard <mripard@kernel.org>
Cc: Thierry Reding <thierry.reding@gmail.com>,
	Sam Ravnborg <sam@ravnborg.org>, David Airlie <airlied@gmail.com>,
	Daniel Vetter <daniel@ffwll.ch>,
	dri-devel@lists.freedesktop.org, Rob Herring <robh+dt@kernel.org>,
	Krzysztof Kozlowski <krzk+dt@kernel.org>,
	devicetree@vger.kernel.org,
	Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Subject: Re: [PATCH 6/7] drm/panel: sitronix-st7789v: Add EDT ET028013DMA panel support
Date: Fri, 16 Jun 2023 12:01:29 +0200	[thread overview]
Message-ID: <20230616120129.75fb00b3@xps-13> (raw)
In-Reply-To: <tr2rmz375jwkwjufhgvb7vbxwqik2w6i276yshh2tbo6dcfhh7@llvt6bigivfy>

Hi Maxime,

mripard@kernel.org wrote on Mon, 12 Jun 2023 10:51:09 +0200:

> On Fri, Jun 09, 2023 at 04:59:50PM +0200, Miquel Raynal wrote:
> > This panel from Emerging Display Technologies Corporation features an
> > ST7789V2 panel inside which is almost identical to what the Sitronix
> > panel driver supports.
> > 
> > In practice, the module physical size is specific, and experiments show
> > that the display will malfunction if any of the following situation
> > occurs:
> > * Pixel clock is above 3MHz
> > * Pixel clock is not inverted
> > I could not properly identify the reasons behind these failures, scope
> > captures show valid input signals.
> > 
> > Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
> > ---
> >  .../gpu/drm/panel/panel-sitronix-st7789v.c    | 34 +++++++++++++++++--
> >  1 file changed, 32 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/panel/panel-sitronix-st7789v.c b/drivers/gpu/drm/panel/panel-sitronix-st7789v.c
> > index 212bccc31804..7de192a3a8aa 100644
> > --- a/drivers/gpu/drm/panel/panel-sitronix-st7789v.c
> > +++ b/drivers/gpu/drm/panel/panel-sitronix-st7789v.c
> > @@ -30,7 +30,8 @@
> >  #define ST7789V_RGBCTRL_RCM(n)			(((n) & 3) << 5)
> >  #define ST7789V_RGBCTRL_VSYNC_HIGH		BIT(3)
> >  #define ST7789V_RGBCTRL_HSYNC_HIGH		BIT(2)
> > -#define ST7789V_RGBCTRL_PCLK_HIGH		BIT(1)
> > +#define ST7789V_RGBCTRL_PCLK_FALLING		BIT(1)
> > +#define ST7789V_RGBCTRL_PCLK_RISING		0
> >  #define ST7789V_RGBCTRL_VBP(n)			((n) & 0x7f)
> >  #define ST7789V_RGBCTRL_HBP(n)			((n) & 0x1f)
> >  
> > @@ -117,6 +118,7 @@ struct st7789v_panel_info {
> >  	u16 width_mm;
> >  	u16 height_mm;
> >  	u32 bus_format;
> > +	u32 bus_flags;
> >  };
> >  
> >  struct st7789v {
> > @@ -175,6 +177,18 @@ static const struct drm_display_mode default_mode = {
> >  	.vtotal = 320 + 8 + 4 + 4,
> >  };
> >  
> > +static const struct drm_display_mode slow_mode = {
> > +	.clock = 3000,
> > +	.hdisplay = 240,
> > +	.hsync_start = 240 + 38,
> > +	.hsync_end = 240 + 38 + 10,
> > +	.htotal = 240 + 38 + 10 + 10,
> > +	.vdisplay = 320,
> > +	.vsync_start = 320 + 8,
> > +	.vsync_end = 320 + 8 + 4,
> > +	.vtotal = 320 + 8 + 4 + 4,
> > +};  
> 
> Why is it supposed to be slow (and compared to what)? It looks like a
> fairly normal mode to me? Or is it because it's refreshed at 30Hz?

The initial support was using 60Hz and for a reason that I do not
understand (scope capture look right), the panel I use is highly
unstable at whatever frequency above 30Hz, so for me it was "slow" :-)
But of course I'll use the panel name to qualify the mode.

> Either way, the ST7789V is a panel controller and can be connected to a
> wide range of panels depending on the model, so it would be better to
> just use the model name there.

Sure.

> >  static int st7789v_get_modes(struct drm_panel *panel,
> >  			     struct drm_connector *connector)
> >  {
> > @@ -197,6 +211,7 @@ static int st7789v_get_modes(struct drm_panel *panel,
> >  
> >  	connector->display_info.width_mm = panel_info->width_mm;
> >  	connector->display_info.height_mm = panel_info->height_mm;
> > +	connector->display_info.bus_flags = panel_info->bus_flags;
> >  	drm_display_info_set_bus_formats(&connector->display_info,
> >  					 &panel_info->bus_format, 1);
> >  
> > @@ -206,8 +221,13 @@ static int st7789v_get_modes(struct drm_panel *panel,
> >  static int st7789v_prepare(struct drm_panel *panel)
> >  {
> >  	struct st7789v *ctx = panel_to_st7789v(panel);
> > +	const struct st7789v_panel_info *panel_info = ctx->panel_info;
> > +	u8 pck = ST7789V_RGBCTRL_PCLK_FALLING;
> >  	int ret;
> >  
> > +	if (panel_info->bus_flags & DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE)
> > +		pck = ST7789V_RGBCTRL_PCLK_RISING;
> > +  
> 
> I guess it could be in a separate commit

I will rebase on top of Sebastian's series who already addressed that
point (as well as many others).

[...]

> >  
> > +static const struct st7789v_panel_info et028013dma_info = {
> > +	.display_mode = &slow_mode,
> > +	.width_mm = 43,
> > +	.height_mm = 58,
> > +	.bus_format = MEDIA_BUS_FMT_RGB666_1X18,
> > +	.bus_flags = DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE,
> > +};
> > +
> >  static const struct of_device_id st7789v_of_match[] = {
> >  	{ .compatible = "sitronix,st7789v", .data = &st7789v_info },
> > +	{ .compatible = "edt,et028013dma", .data = &et028013dma_info },  
> 
> We should sort them by alphabetical order
> 
> Maxime

Ok!

Thanks,
Miquèl

WARNING: multiple messages have this Message-ID (diff)
From: Miquel Raynal <miquel.raynal@bootlin.com>
To: Maxime Ripard <mripard@kernel.org>
Cc: devicetree@vger.kernel.org,
	Thomas Petazzoni <thomas.petazzoni@bootlin.com>,
	Sam Ravnborg <sam@ravnborg.org>,
	dri-devel@lists.freedesktop.org, Rob Herring <robh+dt@kernel.org>,
	Thierry Reding <thierry.reding@gmail.com>,
	Krzysztof Kozlowski <krzk+dt@kernel.org>
Subject: Re: [PATCH 6/7] drm/panel: sitronix-st7789v: Add EDT ET028013DMA panel support
Date: Fri, 16 Jun 2023 12:01:29 +0200	[thread overview]
Message-ID: <20230616120129.75fb00b3@xps-13> (raw)
In-Reply-To: <tr2rmz375jwkwjufhgvb7vbxwqik2w6i276yshh2tbo6dcfhh7@llvt6bigivfy>

Hi Maxime,

mripard@kernel.org wrote on Mon, 12 Jun 2023 10:51:09 +0200:

> On Fri, Jun 09, 2023 at 04:59:50PM +0200, Miquel Raynal wrote:
> > This panel from Emerging Display Technologies Corporation features an
> > ST7789V2 panel inside which is almost identical to what the Sitronix
> > panel driver supports.
> > 
> > In practice, the module physical size is specific, and experiments show
> > that the display will malfunction if any of the following situation
> > occurs:
> > * Pixel clock is above 3MHz
> > * Pixel clock is not inverted
> > I could not properly identify the reasons behind these failures, scope
> > captures show valid input signals.
> > 
> > Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
> > ---
> >  .../gpu/drm/panel/panel-sitronix-st7789v.c    | 34 +++++++++++++++++--
> >  1 file changed, 32 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/panel/panel-sitronix-st7789v.c b/drivers/gpu/drm/panel/panel-sitronix-st7789v.c
> > index 212bccc31804..7de192a3a8aa 100644
> > --- a/drivers/gpu/drm/panel/panel-sitronix-st7789v.c
> > +++ b/drivers/gpu/drm/panel/panel-sitronix-st7789v.c
> > @@ -30,7 +30,8 @@
> >  #define ST7789V_RGBCTRL_RCM(n)			(((n) & 3) << 5)
> >  #define ST7789V_RGBCTRL_VSYNC_HIGH		BIT(3)
> >  #define ST7789V_RGBCTRL_HSYNC_HIGH		BIT(2)
> > -#define ST7789V_RGBCTRL_PCLK_HIGH		BIT(1)
> > +#define ST7789V_RGBCTRL_PCLK_FALLING		BIT(1)
> > +#define ST7789V_RGBCTRL_PCLK_RISING		0
> >  #define ST7789V_RGBCTRL_VBP(n)			((n) & 0x7f)
> >  #define ST7789V_RGBCTRL_HBP(n)			((n) & 0x1f)
> >  
> > @@ -117,6 +118,7 @@ struct st7789v_panel_info {
> >  	u16 width_mm;
> >  	u16 height_mm;
> >  	u32 bus_format;
> > +	u32 bus_flags;
> >  };
> >  
> >  struct st7789v {
> > @@ -175,6 +177,18 @@ static const struct drm_display_mode default_mode = {
> >  	.vtotal = 320 + 8 + 4 + 4,
> >  };
> >  
> > +static const struct drm_display_mode slow_mode = {
> > +	.clock = 3000,
> > +	.hdisplay = 240,
> > +	.hsync_start = 240 + 38,
> > +	.hsync_end = 240 + 38 + 10,
> > +	.htotal = 240 + 38 + 10 + 10,
> > +	.vdisplay = 320,
> > +	.vsync_start = 320 + 8,
> > +	.vsync_end = 320 + 8 + 4,
> > +	.vtotal = 320 + 8 + 4 + 4,
> > +};  
> 
> Why is it supposed to be slow (and compared to what)? It looks like a
> fairly normal mode to me? Or is it because it's refreshed at 30Hz?

The initial support was using 60Hz and for a reason that I do not
understand (scope capture look right), the panel I use is highly
unstable at whatever frequency above 30Hz, so for me it was "slow" :-)
But of course I'll use the panel name to qualify the mode.

> Either way, the ST7789V is a panel controller and can be connected to a
> wide range of panels depending on the model, so it would be better to
> just use the model name there.

Sure.

> >  static int st7789v_get_modes(struct drm_panel *panel,
> >  			     struct drm_connector *connector)
> >  {
> > @@ -197,6 +211,7 @@ static int st7789v_get_modes(struct drm_panel *panel,
> >  
> >  	connector->display_info.width_mm = panel_info->width_mm;
> >  	connector->display_info.height_mm = panel_info->height_mm;
> > +	connector->display_info.bus_flags = panel_info->bus_flags;
> >  	drm_display_info_set_bus_formats(&connector->display_info,
> >  					 &panel_info->bus_format, 1);
> >  
> > @@ -206,8 +221,13 @@ static int st7789v_get_modes(struct drm_panel *panel,
> >  static int st7789v_prepare(struct drm_panel *panel)
> >  {
> >  	struct st7789v *ctx = panel_to_st7789v(panel);
> > +	const struct st7789v_panel_info *panel_info = ctx->panel_info;
> > +	u8 pck = ST7789V_RGBCTRL_PCLK_FALLING;
> >  	int ret;
> >  
> > +	if (panel_info->bus_flags & DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE)
> > +		pck = ST7789V_RGBCTRL_PCLK_RISING;
> > +  
> 
> I guess it could be in a separate commit

I will rebase on top of Sebastian's series who already addressed that
point (as well as many others).

[...]

> >  
> > +static const struct st7789v_panel_info et028013dma_info = {
> > +	.display_mode = &slow_mode,
> > +	.width_mm = 43,
> > +	.height_mm = 58,
> > +	.bus_format = MEDIA_BUS_FMT_RGB666_1X18,
> > +	.bus_flags = DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE,
> > +};
> > +
> >  static const struct of_device_id st7789v_of_match[] = {
> >  	{ .compatible = "sitronix,st7789v", .data = &st7789v_info },
> > +	{ .compatible = "edt,et028013dma", .data = &et028013dma_info },  
> 
> We should sort them by alphabetical order
> 
> Maxime

Ok!

Thanks,
Miquèl

  reply	other threads:[~2023-06-16 10:01 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-06-09 14:59 [PATCH 0/7] drm/panel: sitronix-st7789v: Support ET028013DMA panel Miquel Raynal
2023-06-09 14:59 ` Miquel Raynal
2023-06-09 14:59 ` [PATCH 1/7] drm/panel: sitronix-st7789v: Prevent core spi warnings Miquel Raynal
2023-06-09 14:59   ` Miquel Raynal
2023-06-10 20:06   ` Sam Ravnborg
2023-06-10 20:06     ` Sam Ravnborg
2023-06-13  6:15   ` Michael Riesch
2023-06-13  6:15     ` Michael Riesch
2023-06-13  6:56     ` Miquel Raynal
2023-06-13  6:56       ` Miquel Raynal
2023-06-14 23:22       ` Sebastian Reichel
2023-06-14 23:22         ` Sebastian Reichel
2023-06-15  5:43         ` Sam Ravnborg
2023-06-15  5:43           ` Sam Ravnborg
2023-06-15 12:58           ` Sebastian Reichel
2023-06-15 12:58             ` Sebastian Reichel
2023-06-09 14:59 ` [PATCH 2/7] drm/panel: sitronix-st7789v: Use 9 bits per spi word by default Miquel Raynal
2023-06-09 14:59   ` Miquel Raynal
2023-06-10 20:09   ` Sam Ravnborg
2023-06-10 20:09     ` Sam Ravnborg
2023-06-09 14:59 ` [PATCH 3/7] drm/panel: sitronix-st7789v: Specify the expected bus format Miquel Raynal
2023-06-09 14:59   ` Miquel Raynal
2023-06-10 20:12   ` Sam Ravnborg
2023-06-10 20:12     ` Sam Ravnborg
2023-06-16  9:56     ` Miquel Raynal
2023-06-16  9:56       ` Miquel Raynal
2023-06-12  8:44   ` Maxime Ripard
2023-06-12  8:44     ` Maxime Ripard
2023-06-09 14:59 ` [PATCH 4/7] drm/panel: sitronix-st7789v: Use platform data Miquel Raynal
2023-06-09 14:59   ` Miquel Raynal
2023-06-10 20:15   ` Sam Ravnborg
2023-06-10 20:15     ` Sam Ravnborg
2023-06-09 14:59 ` [PATCH 5/7] dt-bindings: display: st7789v: Add the edt,et028013dma panel compatible Miquel Raynal
2023-06-09 14:59   ` [PATCH 5/7] dt-bindings: display: st7789v: Add the edt, et028013dma " Miquel Raynal
2023-06-09 14:59 ` [PATCH 6/7] drm/panel: sitronix-st7789v: Add EDT ET028013DMA panel support Miquel Raynal
2023-06-09 14:59   ` Miquel Raynal
2023-06-10 20:22   ` Sam Ravnborg
2023-06-10 20:22     ` Sam Ravnborg
2023-06-12  8:51   ` Maxime Ripard
2023-06-12  8:51     ` Maxime Ripard
2023-06-16 10:01     ` Miquel Raynal [this message]
2023-06-16 10:01       ` Miquel Raynal
2023-06-09 14:59 ` [PATCH 7/7] drm/panel: sitronix-st7789v: Check display ID Miquel Raynal
2023-06-09 14:59   ` Miquel Raynal
2023-06-10 20:45   ` Sam Ravnborg
2023-06-10 20:45     ` Sam Ravnborg
2023-06-14 23:27     ` Sebastian Reichel
2023-06-14 23:27       ` Sebastian Reichel
2023-06-16 10:13       ` Miquel Raynal
2023-06-16 10:13         ` Miquel Raynal
2023-06-16 13:38         ` Sebastian Reichel
2023-06-16 13:38           ` Sebastian Reichel
2023-06-13  6:25   ` Michael Riesch

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=20230616120129.75fb00b3@xps-13 \
    --to=miquel.raynal@bootlin.com \
    --cc=airlied@gmail.com \
    --cc=daniel@ffwll.ch \
    --cc=devicetree@vger.kernel.org \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=krzk+dt@kernel.org \
    --cc=mripard@kernel.org \
    --cc=robh+dt@kernel.org \
    --cc=sam@ravnborg.org \
    --cc=thierry.reding@gmail.com \
    --cc=thomas.petazzoni@bootlin.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.