[Intel-gfx] [PATCH] drm/i915/guc: Keep the ctx_pool_vaddr mapped, for easy access

[Oscar Mateo]

On 02/09/2017 01:01 PM, Chris Wilson wrote:
> On Thu, Feb 09, 2017 at 02:23:39AM -0800, Oscar Mateo wrote:
>> @@ -116,15 +114,8 @@ static int guc_update_doorbell_id(struct intel_guc *guc,
>>   	}
>>   
>>   	/* Update the GuC's idea of the doorbell ID */
>> -	len = sg_pcopy_to_buffer(sg->sgl, sg->nents, &desc, sizeof(desc),
>> -			     sizeof(desc) * client->ctx_index);
>> -	if (len != sizeof(desc))
>> -		return -EFAULT;
>> -	desc.db_id = new_id;
>> -	len = sg_pcopy_from_buffer(sg->sgl, sg->nents, &desc, sizeof(desc),
>> -			     sizeof(desc) * client->ctx_index);
>> -	if (len != sizeof(desc))
>> -		return -EFAULT;
>> +	desc = guc->ctx_pool_vaddr + sizeof(*desc) * client->ctx_index;
> Repeated quite a few times.
>
> static inline struct guc_context_desc *
> guc_context_desc(struct intel_guc *guc,
> 		 struct i915_guc_client *client)
> {
> 	return (guc->ctx_pool_vaddr +
> 		sizeof(struct guc_context_desc) * client->ctx_index);
>
> Or your preference
> }
It's repeated only three times, just one line long and it doesn't look 
like we are going to need more instances, so I didn't see the need to 
make it common code, but I'm OK with changing it.
> Lots of little nitpicks I could make. *summons Joonas.
> -Chris
Adding Joonas in CC, so that the summon spell is more effective.

-- Oscar