Thomas Zimmermann <tzimmermann@xxxxxxx> writes:
> Hi
>
> Am 20.02.24 um 10:17 schrieb Javier Martinez Canillas:
>> Thomas Zimmermann <tzimmermann@xxxxxxx> writes:
>>
>> Hello Thomas,
>>
>>> Framebuffer drivers for devices with dedicated backlight are supposed
>>> to set struct fb_info.bl_dev to the backlight's respective device. Use
>>> the value to match backlight and framebuffer in the backlight core code.
>>>
>>> Signed-off-by: Thomas Zimmermann <tzimmermann@xxxxxxx>
>>> ---
>>> drivers/video/backlight/backlight.c | 9 +++++++--
>>> 1 file changed, 7 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/video/backlight/backlight.c b/drivers/video/backlight/backlight.c
>>> index 86e1cdc8e3697..48844a4f28ad3 100644
>>> --- a/drivers/video/backlight/backlight.c
>>> +++ b/drivers/video/backlight/backlight.c
>>> @@ -98,7 +98,8 @@ static int fb_notifier_callback(struct notifier_block *self,
>>> {
>>> struct backlight_device *bd;
>>> struct fb_event *evdata = data;
>>> - int node = evdata->info->node;
>>> + struct fb_info *info = evdata->info;
>>> + int node = info->node;
>>> int fb_blank = 0;
>>>
>>> /* If we aren't interested in this event, skip it immediately ... */
>>> @@ -110,8 +111,12 @@ static int fb_notifier_callback(struct notifier_block *self,
>>>
>>> if (!bd->ops)
>>> goto out;
>>> - if (bd->ops->check_fb && !bd->ops->check_fb(bd, evdata->info))
>>> + else if (bd->ops->check_fb && !bd->ops->check_fb(bd, info))
>>> goto out;
>>> +#if IS_ENABLED(CONFIG_FB_BACKLIGHT)
>>> + else if (info->bl_dev && info->bl_dev != bd)
>> If the driver doesn't provide a struct backlight_ops .check_fb callback, I
>> think that having an info->bl_dev should be mandatory ? Or at least maybe
>> there should be a warning if info->bl_dev isn't set ?
>
> bl_dev can only be used for display drivers that set an explicit
> backlight device; otherwise it's NULL. There seem to be systems where
> backlight and display are distinct. And the docs for check_fb say that
> by default the backlight matches against any display. I tried to keep
> this semantics by silently succeeding if neither check_fb nor bl_dev
> have bene set.
>
>>
>> The would be a driver bug, right ?
>
> I assume that some systems create the backlight instance from platform
> data or DT and the display driver has no means of knowing about it.
>
Ok. I thought that in that case a (platform specific) .check_fb callback
would have to be provided then. But if the semantic is that none could be
missing, then I guess is OK to silently succeeding.
I wonder if at least a debug printout is worth it. But maybe a follow-up.
> Best regards
> Thomas
>
--
Best regards,
Javier Martinez Canillas
Core Platforms
Red Hat
