The panel common bindings provide a gpios-reset property. Let's support it in the simple driver.
Two fields are added to the panel description structure: the time to assert the reset and the time to wait right after before starting to interact with it in any manner. In case these default values are not filled but the GPIO is present in the DT, default values are applied.
Signed-off-by: Miquel Raynal miquel.raynal@bootlin.com ---
Changes since v1: * Add two parameters in the panel description structure. * Ensure the reset is asserted the right amount of time and the deasserted before continuing if a reset GPIO is given.
drivers/gpu/drm/panel/panel-simple.c | 32 +++++++++++++++++++++++++++- 1 file changed, 31 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c index 28fa6ba7b767..ac6f6b5d200d 100644 --- a/drivers/gpu/drm/panel/panel-simple.c +++ b/drivers/gpu/drm/panel/panel-simple.c @@ -38,6 +38,9 @@ #include <drm/drm_mipi_dsi.h> #include <drm/drm_panel.h>
+#define MIN_DEFAULT_RESET_US 10 +#define MIN_DEFAULT_WAIT_US 10 + /** * @modes: Pointer to array of fixed modes appropriate for this panel. If * only one mode then this can just be the address of this the mode. @@ -94,6 +97,10 @@ struct panel_desc {
u32 bus_format; u32 bus_flags; + + /* Minimum reset duration and wait period after it in us */ + u32 reset_time; + u32 reset_wait; };
struct panel_simple { @@ -109,6 +116,7 @@ struct panel_simple { struct i2c_adapter *ddc;
struct gpio_desc *enable_gpio; + struct gpio_desc *reset_gpio;
struct drm_display_mode override_mode; }; @@ -432,12 +440,34 @@ static int panel_simple_probe(struct device *dev, const struct panel_desc *desc) if (IS_ERR(panel->supply)) return PTR_ERR(panel->supply);
+ panel->reset_gpio = devm_gpiod_get_optional(dev, "reset", + GPIOD_OUT_HIGH); + if (IS_ERR(panel->reset_gpio)) { + err = PTR_ERR(panel->reset_gpio); + if (err != -EPROBE_DEFER) + dev_err(dev, "failed to request reset pin: %d\n", err); + return err; + } else if (panel->reset_gpio) { + u32 reset_time = panel->desc->reset_time; + u32 reset_wait = panel->desc->reset_wait; + + if (!reset_time) + reset_time = MIN_DEFAULT_RESET_US; + + if (!reset_wait) + reset_wait = MIN_DEFAULT_WAIT_US; + + usleep_range(reset_time, 2 * reset_time); + gpiod_set_value_cansleep(panel->reset_gpio, 0); + usleep_range(reset_wait, 2 * reset_wait); + } + panel->enable_gpio = devm_gpiod_get_optional(dev, "enable", GPIOD_OUT_LOW); if (IS_ERR(panel->enable_gpio)) { err = PTR_ERR(panel->enable_gpio); if (err != -EPROBE_DEFER) - dev_err(dev, "failed to request GPIO: %d\n", err); + dev_err(dev, "failed to request enable pin: %d\n", err); return err; }
Hi Miquel
On Tue, Dec 24, 2019 at 03:21:34PM +0100, Miquel Raynal wrote:
The panel common bindings provide a gpios-reset property. Let's support it in the simple driver.
Two fields are added to the panel description structure: the time to assert the reset and the time to wait right after before starting to interact with it in any manner. In case these default values are not filled but the GPIO is present in the DT, default values are applied.
Wehn we discussed this the last time you wrote:
""" my hardware is:
LVDS IP <----------> LVDS to RGB bridge <------------> Panel
While there is a simple "RGB to LVDS" bridge driver, there is none doing the work the other way around. In my case, the bridge has a reset pin.
As until now there is no way to represent the "LVDS to RGB" bridge and because the bindings already document such reset pin, I decided to add support for it in the simple panel driver. """
Based on the information provided it seems that the correct way is to add a "LVDS to RGB bridge" and then let the bridge handle the reset functionality.
It is obviously much more code to do it this way but then other panels using the same type of brigde have the same functionality without adding bridge functionality to the panel.
Sam
Signed-off-by: Miquel Raynal miquel.raynal@bootlin.com
Changes since v1:
- Add two parameters in the panel description structure.
- Ensure the reset is asserted the right amount of time and the deasserted before continuing if a reset GPIO is given.
drivers/gpu/drm/panel/panel-simple.c | 32 +++++++++++++++++++++++++++- 1 file changed, 31 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c index 28fa6ba7b767..ac6f6b5d200d 100644 --- a/drivers/gpu/drm/panel/panel-simple.c +++ b/drivers/gpu/drm/panel/panel-simple.c @@ -38,6 +38,9 @@ #include <drm/drm_mipi_dsi.h> #include <drm/drm_panel.h>
+#define MIN_DEFAULT_RESET_US 10 +#define MIN_DEFAULT_WAIT_US 10
/**
- @modes: Pointer to array of fixed modes appropriate for this panel. If
only one mode then this can just be the address of this the mode.@@ -94,6 +97,10 @@ struct panel_desc {
u32 bus_format; u32 bus_flags;
- /* Minimum reset duration and wait period after it in us */
- u32 reset_time;
- u32 reset_wait;
};
struct panel_simple { @@ -109,6 +116,7 @@ struct panel_simple { struct i2c_adapter *ddc;
struct gpio_desc *enable_gpio;
struct gpio_desc *reset_gpio;
struct drm_display_mode override_mode;
}; @@ -432,12 +440,34 @@ static int panel_simple_probe(struct device *dev, const struct panel_desc *desc) if (IS_ERR(panel->supply)) return PTR_ERR(panel->supply);
- panel->reset_gpio = devm_gpiod_get_optional(dev, "reset",
GPIOD_OUT_HIGH);- if (IS_ERR(panel->reset_gpio)) {
err = PTR_ERR(panel->reset_gpio);if (err != -EPROBE_DEFER)dev_err(dev, "failed to request reset pin: %d\n", err);return err;- } else if (panel->reset_gpio) {
u32 reset_time = panel->desc->reset_time;u32 reset_wait = panel->desc->reset_wait;if (!reset_time)reset_time = MIN_DEFAULT_RESET_US;if (!reset_wait)reset_wait = MIN_DEFAULT_WAIT_US;usleep_range(reset_time, 2 * reset_time);gpiod_set_value_cansleep(panel->reset_gpio, 0);usleep_range(reset_wait, 2 * reset_wait);- }
- panel->enable_gpio = devm_gpiod_get_optional(dev, "enable", GPIOD_OUT_LOW); if (IS_ERR(panel->enable_gpio)) { err = PTR_ERR(panel->enable_gpio); if (err != -EPROBE_DEFER)
dev_err(dev, "failed to request GPIO: %d\n", err);
dev_err(dev, "failed to request enable pin: %d\n", err);-- 2.20.1
Hi Sam,
Sam Ravnborg sam@ravnborg.org wrote on Thu, 2 Jan 2020 18:27:00 +0100:
Hi Miquel
On Tue, Dec 24, 2019 at 03:21:34PM +0100, Miquel Raynal wrote:
The panel common bindings provide a gpios-reset property. Let's support it in the simple driver.
Two fields are added to the panel description structure: the time to assert the reset and the time to wait right after before starting to interact with it in any manner. In case these default values are not filled but the GPIO is present in the DT, default values are applied.
Wehn we discussed this the last time you wrote:
""" my hardware is:
LVDS IP <----------> LVDS to RGB bridge <------------> Panel
While there is a simple "RGB to LVDS" bridge driver, there is none doing the work the other way around. In my case, the bridge has a reset pin.
As until now there is no way to represent the "LVDS to RGB" bridge and because the bindings already document such reset pin, I decided to add support for it in the simple panel driver. """
Based on the information provided it seems that the correct way is to add a "LVDS to RGB bridge" and then let the bridge handle the reset functionality.
This I agree, but we are talking about my current situation.
It is obviously much more code to do it this way but then other panels using the same type of brigde have the same functionality without adding bridge functionality to the panel.
This, I do not fully agree as bindings for the panel reset already exist and we could have a reset on both: the bridge and the panel. I choose to use a wrong (private) DT representation because I am not willing to add an LVDS->RGB bridge: as you say, it is much more work to do. But, IMHO, this is not related to the patch. If you consider this patch wrong because a panel cannot have a reset, then it should be stated clearly and maybe removed from the bindings?
Anyway if you think this change can't be useful, let's put it aside.
Thanks for your time, Miquèl
Hi Sam,
you've asked in another thread [1] if there is any known simple panel that requires a reset. We have a Densitron DMT070WSNLCMI-1E Panel (compatible to avic,tm070ddh03) with some reset timing requirements, [2] Page 20. So it would be nice to see this patch accepted.
[1] https://patchwork.kernel.org/patch/11292207/ [2] https://www.densitron.com/sites/default/files/2019-09/DMT070WSNLCMI-1E%20Rev...
Regards, Wadim
On 06.01.20 10:10, Miquel Raynal wrote:
Hi Sam,
Sam Ravnborg sam@ravnborg.org wrote on Thu, 2 Jan 2020 18:27:00 +0100:
Hi Miquel
On Tue, Dec 24, 2019 at 03:21:34PM +0100, Miquel Raynal wrote:
The panel common bindings provide a gpios-reset property. Let's support it in the simple driver.
Two fields are added to the panel description structure: the time to assert the reset and the time to wait right after before starting to interact with it in any manner. In case these default values are not filled but the GPIO is present in the DT, default values are applied.
Wehn we discussed this the last time you wrote:
""" my hardware is:
LVDS IP <----------> LVDS to RGB bridge <------------> Panel
While there is a simple "RGB to LVDS" bridge driver, there is none doing the work the other way around. In my case, the bridge has a reset pin.
As until now there is no way to represent the "LVDS to RGB" bridge and because the bindings already document such reset pin, I decided to add support for it in the simple panel driver. """
Based on the information provided it seems that the correct way is to add a "LVDS to RGB bridge" and then let the bridge handle the reset functionality.
This I agree, but we are talking about my current situation.
It is obviously much more code to do it this way but then other panels using the same type of brigde have the same functionality without adding bridge functionality to the panel.
This, I do not fully agree as bindings for the panel reset already exist and we could have a reset on both: the bridge and the panel. I choose to use a wrong (private) DT representation because I am not willing to add an LVDS->RGB bridge: as you say, it is much more work to do. But, IMHO, this is not related to the patch. If you consider this patch wrong because a panel cannot have a reset, then it should be stated clearly and maybe removed from the bindings?
Anyway if you think this change can't be useful, let's put it aside.
Thanks for your time, Miquèl
dri-devel mailing list dri-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/dri-devel
dri-devel@lists.freedesktop.org
-
Miquel Raynal -
Sam Ravnborg -
Wadim Egorov