[PATCH 01/13] ASoC: Intel: Add catpt device

Cezary Rojewski cezary.rojewski at intel.com
Tue Aug 11 12:04:39 CEST 2020


On 2020-08-08 1:50 PM, Andy Shevchenko wrote:
> On Fri, Aug 07, 2020 at 01:06:37PM +0200, Cezary Rojewski wrote:
>> Declare base structures, registers and device routines for the catpt
>> solution. Catpt deprecates and is a direct replacement for
>> sound/soc/intel/haswell. Supports Lynxpoint and Wildcat Point both.
> 
> 
> ...
> 
>> +struct catpt_mregion {
>> +	u32 start;
>> +	u32 end;
>> +	bool busy;
>> +	struct list_head node;
>> +};
> 
> I'm wondering if struct resource can be used instead.
> 
> You know that you may introduce a new type of resource if you want to.
> 
>> +static inline size_t catpt_mregion_size(const struct catpt_mregion *reg)
>> +{
>> +	return reg->end - reg->start + 1;
>> +}
>> +
>> +/* True if region r1 intersects region r2 */
>> +static inline bool catpt_mregion_intersects(struct catpt_mregion *r1,
>> +					    struct catpt_mregion *r2)
>> +{
>> +	return (r1->start >= r2->start && r1->start <= r2->end) ||
>> +	       (r1->end >= r2->start && r1->end <= r2->end);
>> +}
>> +
>> +static inline bool catpt_mregion_intersecting(struct catpt_mregion *r1,
>> +					      struct catpt_mregion *r2,
>> +					      struct catpt_mregion *ret)
>> +{
>> +	if (!catpt_mregion_intersects(r1, r2))
>> +		return false;
>> +	ret->start = max(r1->start, r2->start);
>> +	ret->end = min(r1->end, r2->end);
>> +	return true;
>> +}
> 
> Yeah, it reminds the existing resource infrastructure. Why to repeat it?
> 

As mentioned in the followup message:
https://www.spinics.net/lists/alsa-devel/msg113563.html
(Resource management section)

I opted out of 'struct resource' usage due to differences in its layout 
and preferred usage. Perhaps I shouldn't have.

I've embraced 'struct resource' fully in v2. Your suggestion cascaded 
into several other changes and some things were made redundant with 
removal of struct catpt_mbank and catpt_mregion. Change log has been 
added in v2 cover-letter describing consequences of said change.

Thanks,
Czarek


More information about the Alsa-devel mailing list