Re: [PATCH] ACPI: video: Add backlight=native DMI quirk for Dell G15 5515

[Hans de Goede <hdegoede@redhat.com>]

Hi,

On 10/31/22 23:20, Daniel Dadap wrote:
> On Mon, Oct 31, 2022 at 09:51:55PM +0000, Limonciello, Mario wrote:
>> [Public]
>>
>>> -----Original Message-----
>>> From: Hans de Goede <hdegoede@redhat.com>
>>> Sent: Monday, October 31, 2022 16:26
>>> To: Rafael J . Wysocki <rafael@kernel.org>
>>> Cc: Hans de Goede <hdegoede@redhat.com>; Len Brown
>>> <lenb@kernel.org>; linux-acpi@vger.kernel.org; platform-driver-
>>> x86@vger.kernel.org; Daniel Dadap <ddadap@nvidia.com>
>>> Subject: [PATCH] ACPI: video: Add backlight=native DMI quirk for Dell G15
>>> 5515
>>>
>>> The Dell G15 5515 has the WMI interface (and WMI call returns) expected
>>> by the nvidia-wmi-ec-backlight interface. But the backlight class device
>>> registered by the nvidia-wmi-ec-backlight driver does not actually work.
>>>
>>
>> Is there a way to detect this from nvidia-wmi-ec-backlight?  I would think it's
>> cleaner to have that driver return -ENODEV so you end up with just the "working"
>> amdgpu_bl0.
> 
> That would be ideal, although I'm not certain whether this would work
> with the new backlight control type selection model where video.ko
> determines what type of backlight interface the system is supposed to
> have. I am assuming that the below patch is against a kernel predating
> the recent changes that Hans made to rework the backlight subsystem,
> because IIUC it shouldn't be possible for both the EC backlight driver
> and the iGPU's native one to both be active at the same time. I would
> expect that without this patch, there would be no working backlight
> interfaces on this system when running a newer kernel, since
> __acpi_video_get_backlight_type() would return
> acpi_backlight_nvidia_wmi_ec, which means
> acpi_video_backlight_use_native() would return false, and the iGPU
> native driver should not be registering a brightness handler.

The reporter needs to blacklist the nvidia-wmi-ec-backlight module
because with 6.0 they otherwise get both that backlight + the native
amdgpu backlight in /sys/class/backlight and then userspace will
prefer the nvidia-wmi-ec-backlight one which according to the reporter
does not work.

You are correct that with 6.1 the issue is that the native driver
now honors the acpi_video_get_backlight_type() return and no longer
registers. This is fixed by this patch.

The good news is that with 6.1 nvidia-wmi-ec-backlight now also honors
acpi_video_get_backlight_type() so the quirk also avoids the need
to blacklist nvidia-wmi-ec-backlight and with the quirk in 6.1 everything
just works OOTB (with the current BIOS settings for the GPU mode).

Regards,

Hans


>  
>>> The amdgpu_bl0 native GPU backlight class device does actually work,
>>> add a backlight=native DMI quirk for this.
>>>
>>> Cc: Daniel Dadap <ddadap@nvidia.com>
>>> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
>>> ---
>>>  drivers/acpi/video_detect.c | 10 ++++++++++
>>>  1 file changed, 10 insertions(+)
>>>
>>> diff --git a/drivers/acpi/video_detect.c b/drivers/acpi/video_detect.c
>>> index 841f6213b4de..88acc09773bb 100644
>>> --- a/drivers/acpi/video_detect.c
>>> +++ b/drivers/acpi/video_detect.c
>>> @@ -645,6 +645,16 @@ static const struct dmi_system_id
>>> video_detect_dmi_table[] = {
>>>  		},
>>>  	},
>>>
>>> +	/* Models which have nvidia-ec-wmi support, but should not use it.
>>> */
>>> +	{
>>> +	 .callback = video_detect_force_native,
>>> +	 /* Dell G15 5515 */
>>> +	 .matches = {
>>> +		DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
>>> +		DMI_MATCH(DMI_PRODUCT_NAME, "Dell G15 5515"),
>>> +		},
>>> +	},
>>> +
>>>  	/*
>>>  	 * Desktops which falsely report a backlight and which our heuristics
>>>  	 * for this do not catch.
>>> --
>>> 2.37.3
>