linux-fbdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] backlight: rave-sp: don't touch initial state and register with correct device
@ 2019-04-29 15:29 Lucas Stach
  2019-05-02 10:33 ` Daniel Thompson
  0 siblings, 1 reply; 4+ messages in thread
From: Lucas Stach @ 2019-04-29 15:29 UTC (permalink / raw)
  To: Lee Jones, Daniel Thompson, Jingoo Han
  Cc: linux-fbdev, kernel, dri-devel, patchwork-lst

This way the backlight can be referenced through its device node and
enabling/disabling can be managed through the panel driver.

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
---
 drivers/video/backlight/rave-sp-backlight.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/video/backlight/rave-sp-backlight.c b/drivers/video/backlight/rave-sp-backlight.c
index 462f14a1b19d..d296bfcf4396 100644
--- a/drivers/video/backlight/rave-sp-backlight.c
+++ b/drivers/video/backlight/rave-sp-backlight.c
@@ -48,15 +48,13 @@ static int rave_sp_backlight_probe(struct platform_device *pdev)
 	struct device *dev = &pdev->dev;
 	struct backlight_device *bd;
 
-	bd = devm_backlight_device_register(dev, pdev->name, dev->parent,
+	bd = devm_backlight_device_register(dev, pdev->name, dev,
 					    dev_get_drvdata(dev->parent),
 					    &rave_sp_backlight_ops,
 					    &rave_sp_backlight_props);
 	if (IS_ERR(bd))
 		return PTR_ERR(bd);
 
-	backlight_update_status(bd);
-
 	return 0;
 }
 
-- 
2.20.1

^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH] backlight: rave-sp: don't touch initial state and register with correct device
  2019-04-29 15:29 [PATCH] backlight: rave-sp: don't touch initial state and register with correct device Lucas Stach
@ 2019-05-02 10:33 ` Daniel Thompson
  2019-05-02 12:01   ` Lucas Stach
  0 siblings, 1 reply; 4+ messages in thread
From: Daniel Thompson @ 2019-05-02 10:33 UTC (permalink / raw)
  To: Lucas Stach, Lee Jones, Jingoo Han
  Cc: linux-fbdev, kernel, dri-devel, patchwork-lst

On 29/04/2019 16:29, Lucas Stach wrote:
> This way the backlight can be referenced through its device node and
> enabling/disabling can be managed through the panel driver.

Is it possible to implement something similar to 
pwm_backlight_initial_power_state() to handle this?

backlight drivers already suffer from too much diversity so I prefer 
things like this to align behaviour with the (fairly heavilyly used) PWM 
driver if possible.


Daniel.


> Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
> ---
>   drivers/video/backlight/rave-sp-backlight.c | 4 +---
>   1 file changed, 1 insertion(+), 3 deletions(-)
> 
> diff --git a/drivers/video/backlight/rave-sp-backlight.c b/drivers/video/backlight/rave-sp-backlight.c
> index 462f14a1b19d..d296bfcf4396 100644
> --- a/drivers/video/backlight/rave-sp-backlight.c
> +++ b/drivers/video/backlight/rave-sp-backlight.c
> @@ -48,15 +48,13 @@ static int rave_sp_backlight_probe(struct platform_device *pdev)
>   	struct device *dev = &pdev->dev;
>   	struct backlight_device *bd;
>   
> -	bd = devm_backlight_device_register(dev, pdev->name, dev->parent,
> +	bd = devm_backlight_device_register(dev, pdev->name, dev,
>   					    dev_get_drvdata(dev->parent),
>   					    &rave_sp_backlight_ops,
>   					    &rave_sp_backlight_props);
>   	if (IS_ERR(bd))
>   		return PTR_ERR(bd);
>   
> -	backlight_update_status(bd);
> -
>   	return 0;
>   }
>   
> 

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] backlight: rave-sp: don't touch initial state and register with correct device
  2019-05-02 10:33 ` Daniel Thompson
@ 2019-05-02 12:01   ` Lucas Stach
  2019-05-03 13:29     ` Daniel Thompson
  0 siblings, 1 reply; 4+ messages in thread
From: Lucas Stach @ 2019-05-02 12:01 UTC (permalink / raw)
  To: Daniel Thompson, Lee Jones, Jingoo Han
  Cc: linux-fbdev, kernel, dri-devel, patchwork-lst

Hi Daniel,

Am Donnerstag, den 02.05.2019, 11:33 +0100 schrieb Daniel Thompson:
> On 29/04/2019 16:29, Lucas Stach wrote:
> > This way the backlight can be referenced through its device node and
> > enabling/disabling can be managed through the panel driver.
> 
> Is it possible to implement something similar to 
> pwm_backlight_initial_power_state() to handle this?

I'm not aware of any protocol to the RAVE-SP that would allow to read
back the backlight state. AFAICS the backlight is implemented as a
unidirectional protocol.

Regards,
Lucas

> backlight drivers already suffer from too much diversity so I prefer 
> things like this to align behaviour with the (fairly heavilyly used) PWM 
> driver if possible.
> 
> 
> Daniel.
> 
> 
> > > > Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
> > ---
> >   drivers/video/backlight/rave-sp-backlight.c | 4 +---
> >   1 file changed, 1 insertion(+), 3 deletions(-)
> > 
> > diff --git a/drivers/video/backlight/rave-sp-backlight.c b/drivers/video/backlight/rave-sp-backlight.c
> > index 462f14a1b19d..d296bfcf4396 100644
> > --- a/drivers/video/backlight/rave-sp-backlight.c
> > +++ b/drivers/video/backlight/rave-sp-backlight.c
> > @@ -48,15 +48,13 @@ static int rave_sp_backlight_probe(struct platform_device *pdev)
> > > >   	struct device *dev = &pdev->dev;
> > > >   	struct backlight_device *bd;
> >   
> > > > -	bd = devm_backlight_device_register(dev, pdev->name, dev->parent,
> > > > +	bd = devm_backlight_device_register(dev, pdev->name, dev,
> > > >   					    dev_get_drvdata(dev->parent),
> > > >   					    &rave_sp_backlight_ops,
> > > >   					    &rave_sp_backlight_props);
> > > >   	if (IS_ERR(bd))
> > > >   		return PTR_ERR(bd);
> >   
> > > > -	backlight_update_status(bd);
> > -
> > > >   	return 0;
> >   }
> >   
> > 
> 
> 

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] backlight: rave-sp: don't touch initial state and register with correct device
  2019-05-02 12:01   ` Lucas Stach
@ 2019-05-03 13:29     ` Daniel Thompson
  0 siblings, 0 replies; 4+ messages in thread
From: Daniel Thompson @ 2019-05-03 13:29 UTC (permalink / raw)
  To: Lucas Stach
  Cc: linux-fbdev, Jingoo Han, dri-devel, patchwork-lst, kernel,
	Lee Jones

On Thu, May 02, 2019 at 02:01:45PM +0200, Lucas Stach wrote:
> Hi Daniel,
> 
> Am Donnerstag, den 02.05.2019, 11:33 +0100 schrieb Daniel Thompson:
> > On 29/04/2019 16:29, Lucas Stach wrote:
> > > This way the backlight can be referenced through its device node and
> > > enabling/disabling can be managed through the panel driver.
> > 
> > Is it possible to implement something similar to 
> > pwm_backlight_initial_power_state() to handle this?
> 
> I'm not aware of any protocol to the RAVE-SP that would allow to read
> back the backlight state. AFAICS the backlight is implemented as a
> unidirectional protocol.

That function does two things... one is read the current power state to
ensure meta-data is correctly up to date which cannot be replicated.
The other is to choose different behaviour depending on whether the
backlight is standalone or has a phandle link to another device.

Does the second part apply to the rave-sp?


Daniel.



> > backlight drivers already suffer from too much diversity so I prefer 
> > things like this to align behaviour with the (fairly heavilyly used) PWM 
> > driver if possible.
> > 
> > 
> > Daniel.
> > 
> > 
> > > > > Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
> > > ---
> > >   drivers/video/backlight/rave-sp-backlight.c | 4 +---
> > >   1 file changed, 1 insertion(+), 3 deletions(-)
> > > 
> > > diff --git a/drivers/video/backlight/rave-sp-backlight.c b/drivers/video/backlight/rave-sp-backlight.c
> > > index 462f14a1b19d..d296bfcf4396 100644
> > > --- a/drivers/video/backlight/rave-sp-backlight.c
> > > +++ b/drivers/video/backlight/rave-sp-backlight.c
> > > @@ -48,15 +48,13 @@ static int rave_sp_backlight_probe(struct platform_device *pdev)
> > > > >   	struct device *dev = &pdev->dev;
> > > > >   	struct backlight_device *bd;
> > >   
> > > > > -	bd = devm_backlight_device_register(dev, pdev->name, dev->parent,
> > > > > +	bd = devm_backlight_device_register(dev, pdev->name, dev,
> > > > >   					    dev_get_drvdata(dev->parent),
> > > > >   					    &rave_sp_backlight_ops,
> > > > >   					    &rave_sp_backlight_props);
> > > > >   	if (IS_ERR(bd))
> > > > >   		return PTR_ERR(bd);
> > >   
> > > > > -	backlight_update_status(bd);
> > > -
> > > > >   	return 0;
> > >   }
> > >   
> > > 
> > 
> > 

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2019-05-03 13:29 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2019-04-29 15:29 [PATCH] backlight: rave-sp: don't touch initial state and register with correct device Lucas Stach
2019-05-02 10:33 ` Daniel Thompson
2019-05-02 12:01   ` Lucas Stach
2019-05-03 13:29     ` Daniel Thompson

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).