[Intel-gfx] [PATCH v3 01/14] drm: add HDMI 2.0 VIC support for AVI info-frames

[Sharma, Shashank]

Hello Thierry,

Thanks for the comments.

In fact, that was the plan earlier, but the problem is, this function is 
being called from several drivers, and not all of them have
the drm_connector readily available with their caller function. For few 
drivers, I might have to go up two to three levels in stack,
to get and pass the drm_connector, which means, I have to modify those 
drivers up to two-to-three levels.
It dint seem right, so we decided to call it with a bool.

Regards
Shashank
On 6/15/2017 1:50 AM, Thierry Reding wrote:
> On Wed, Jun 14, 2017 at 11:17:32PM +0530, Shashank Sharma wrote:
> [...]
>> diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
>> index 2e55599..d312fe1 100644
>> --- a/drivers/gpu/drm/drm_edid.c
>> +++ b/drivers/gpu/drm/drm_edid.c
>> @@ -4334,12 +4334,14 @@ EXPORT_SYMBOL(drm_set_preferred_mode);
>>    *                                              data from a DRM display mode
>>    * @frame: HDMI AVI infoframe
>>    * @mode: DRM display mode
>> + * @is_hdmi2_sink: Sink is HDMI 2.0 compliant
>>    *
>>    * Return: 0 on success or a negative error code on failure.
>>    */
>>   int
>>   drm_hdmi_avi_infoframe_from_display_mode(struct hdmi_avi_infoframe *frame,
>> -					 const struct drm_display_mode *mode)
>> +					 const struct drm_display_mode *mode,
>> +					 bool is_hdmi2_sink)
> Perhaps instead of passing a boolean flag here we could pass a const
> struct drm_hdmi_info *. That way the checks could be made more explicit
> and it becomes more obvious what to pass into the function. Also, as it
> is every driver will have to derive is_hdmi2_sink by itself, and they're
> likely to have to do the same thing as i915 anyway.
>
> Thierry