Re: [PATCH] backlight: led_bl: fix initial power state

[Sam Ravnborg]

Hi Daniel,

On Wed, Jul 05, 2023 at 03:07:31PM +0100, Daniel Thompson wrote:
> On Tue, Jul 04, 2023 at 07:07:31PM +0200, Sam Ravnborg wrote:
> > Hi Daniel,
> >
> > > > @@ -200,8 +200,8 @@ static int led_bl_probe(struct platform_device *pdev)
> > > >  	props.type = BACKLIGHT_RAW;
> > > >  	props.max_brightness = priv->max_brightness;
> > > >  	props.brightness = priv->default_brightness;
> > > > -	props.power = (priv->default_brightness > 0) ? FB_BLANK_POWERDOWN :
> > > > -		      FB_BLANK_UNBLANK;
> > > > +	props.power = (priv->default_brightness > 0) ? FB_BLANK_UNBLANK :
> > > > +		      FB_BLANK_POWERDOWN;
> > >
> > > The logic was wrong before but I think will still be wrong after the
> > > change too (e.g. the bogus logic is probably avoiding backlight flicker
> > > in some use cases).
> > >
> > > The logic here needs to be similar to what pwm_bl.c implements in
> > > pwm_backlight_initial_power_state(). Whilst it might be better
> > > to implement this in led_bl_get_leds() let me show what I mean
> > > in code that fits in the current line:
> > >
> > > 	/*
> > > 	 * Activate the backlight if the LEDs are already lit *or*
> > > 	 * there is no phandle link (meaning the backlight power
> > > 	 * state cannot be synced with the display state).
> > > 	 */
> > > 	props.power = (active_at_boot || !dev->node->phandle) ?
> > > 			FB_BLANK_UNBLANK : FB_BLANK_POWERDOWN;
> > >
> > The following code does the same using helpers:
> >
> > 	if (active_at_boot || !dev->node->phandle))
> > 		backlight_enable(bd);
> > 	else
> > 		backlight_disable(bd);
> >
> > The code needs to execute after backlight_device_register() so maybe not
> > so great an idea?!?
> 
> It would introduce a need for a bunch of new locks since userspace
> could get in between device creation and the call to the helpers.
I thought we were safe while in the probe function, but I have been
fooled by the driver model before.

> 
> Additionally, it is even correct for a driver to forcefully set
> fb_blank to powerdown during the probe? All current drivers set
> fb_blank to unblank during the probe.
fb_blank is more or less unused. I thought that Lee applied the patch set
to eliminate most users, but I see that this is not the case.
I need to resend one day.
Some (at least one) drivers update .power after registering the device, so if this
is racy then these drivers could use some care.

Anyway, looking at how many drivers access backlight_properties direct is
is futile to try to avoid it. So the approach you suggest seems the best
way to do it.

	Sam