lkml.org 
[lkml]   [2013]   [Mar]   [15]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 3/3] ACPI video: Fix applying indexed initial brightness value.
On 03/15/2013 04:55 PM, Danny Baumann wrote:
> Hi,
>
> >> +static unsigned long long
>>> +acpi_video_index_to_level(struct acpi_video_device *device,
>>> + unsigned long long index)
>>> +{
>>> + if (device->brightness->flags._BCL_reversed)
>>> + index = device->brightness->count - 3 - index;
>>> +
>>> + return device->brightness->levels[index + 2];
>>> +}
>>
>> What about making this function also take care of the
>> bqc_offset_aml_bug_workaround? so that this function serves more like a
>> conversion from raw value to fixed value, the function name can perhaps
>> be named as: acpi_video_fix_bqc_value, or whatever you think is more
>> appropriate.
>
> Makes sense to me. How about acpi_video_bqc_to_level? I'd suggest
> acpi_video_bqc_value_to_level, but that makes it hard to keep the 80
> char limit in acpi_video_device_lcd_get_level_current.

Right, so let's go with acpi_video_bqc_to_level.

>
>>> @@ -742,9 +749,7 @@ acpi_video_init_brightness(struct acpi_video_device *device)
>>> }
>>> }
>>> } else {
>>> - if (br->flags._BCL_reversed)
>>> - level_old = (br->count - 1) - level_old;
>>> - level = br->levels[level_old];
>>> + level = acpi_video_index_to_level(device, level_old);
>>
>> And here, that new function should be used, which also takes care of the
>> offset_aml_bug problem(though in theory, the two problems may not happen
>> on the same BIOS table).
>
> But that new function (whatever it's named) is already used here?

Yes, I mean the new function that also takes care of offset_aml_bug :-)

>
> BTW, shouldn't the use_bios_initial_backlight also be respected for the
> BQC-returns-index case? Currently it's only used for the
> BQC-returns-level case.

Definitely, we should care that.

>
>> And the acpi_video_device_lcd_get_level_current's param init can
>> probably be renamed as raw, meaning if raw value is desired or fixed
>> value, but it's not a big deal.
>
> Agreed. I'll send a new patch set (this time with signed-off-by) once I
> get opinion on the question above.

Cool, thanks.

-Aaron



\
 
 \ /
  Last update: 2013-03-15 10:44    [W:0.033 / U:0.096 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site