[Intel-gfx] [PATCH v9 10/39] drm/i915: Implement HDCP2.2 receiver authentication

Jani Nikula jani.nikula at linux.intel.com
Thu Dec 20 07:42:36 UTC 2018


On Wed, 19 Dec 2018, "Winkler, Tomas" <tomas.winkler at intel.com> wrote:
>> 
>> On Wed, 19 Dec 2018, "C, Ramalingam" <ramalingam.c at intel.com> wrote:
>> > On 12/19/2018 8:05 PM, Daniel Vetter wrote:
>> >> On Thu, Dec 13, 2018 at 09:31:12AM +0530, Ramalingam C wrote:
>> >>>   struct intel_hdcp {
>> >>> @@ -414,6 +430,24 @@ struct intel_hdcp {
>> >>>   	 */
>> >>>   	u8 content_type;
>> >>>   	struct hdcp_port_data port_data;
>> >>> +
>> >>> +	u8 is_paired;
>> >>> +	u8 is_repeater;
>> >> Make these two bool, will simplify the code a bunch.
>> >
>> > Seems there is a movement for not to use the bool in structures.
>> 
>> No. Please use bools in structs when it makes sense. Avoid bools in structs
>> when you need to care about memory footprint or alignment or packing or
>> the like. This is not one of those cases.
>> 
>> > Thats why I have changed these from bool to u8 from v8 onwards.
>> > Checkpatch also complains on this
>> 
>> Sorry to say, checkpatch is not the authority although we do send out
>> automated checkpatch results.
>
> I believe it was Linus' call to not use  bool in structs at all
> https://lkml.org/lkml/2017/11/21/384

I don't care. That's a valid judgement in the context referenced, but
the conclusion "no bools in structs at all" isn't. In this case, I think
bools are the better option, and anything else makes the code worse.

BR,
Jani.

-- 
Jani Nikula, Intel Open Source Graphics Center


More information about the Intel-gfx mailing list