Re: [Intel-gfx] [PATCH 2/5] drm/i915: Create a kernel context for GGTT updates

[Nirmoy Das <nirmoy.das@linux.intel.com>]

On 9/7/2023 3:41 PM, Andi Shyti wrote:
> Hi Nirmoy,
>
> [...]
>
>> +	/* mark the bind context's availability status */
>> +	bool bind_context_ready;
> Do we need some locking here?

bind_context_ready is changed in suspend/resume or after mod probe so no need to
worry about locking here.


>
>>   	/**
>>   	 * pinned_contexts_list: List of pinned contexts. This list is only
>> diff --git a/drivers/gpu/drm/i915/gt/intel_gt.c b/drivers/gpu/drm/i915/gt/intel_gt.c
>> index 449f0b7fc843..cd0ff1597db9 100644
>> --- a/drivers/gpu/drm/i915/gt/intel_gt.c
>> +++ b/drivers/gpu/drm/i915/gt/intel_gt.c
>> @@ -1019,3 +1019,21 @@ enum i915_map_type intel_gt_coherent_map_type(struct intel_gt *gt,
>>   	else
>>   		return I915_MAP_WC;
>>   }
>> +
>> +void intel_gt_bind_context_set_ready(struct intel_gt *gt, bool ready)
>> +{
>> +	struct intel_engine_cs *engine = gt->engine[BCS0];
>> +
>> +	if (engine && engine->bind_context)
>> +		engine->bind_context_ready = ready;
>> +}
>> +
>> +bool intel_gt_is_bind_context_ready(struct intel_gt *gt)
>> +{
>> +	struct intel_engine_cs *engine = gt->engine[BCS0];
>> +
>> +	if (engine)
>> +		return engine->bind_context_ready;
>> +
>> +	return false;
>> +}
>> diff --git a/drivers/gpu/drm/i915/gt/intel_gt.h b/drivers/gpu/drm/i915/gt/intel_gt.h
>> index 239848bcb2a4..9e70e625cebc 100644
>> --- a/drivers/gpu/drm/i915/gt/intel_gt.h
>> +++ b/drivers/gpu/drm/i915/gt/intel_gt.h
>> @@ -180,4 +180,6 @@ enum i915_map_type intel_gt_coherent_map_type(struct intel_gt *gt,
>>   					      struct drm_i915_gem_object *obj,
>>   					      bool always_coherent);
>>   
>> +void intel_gt_bind_context_set_ready(struct intel_gt *gt, bool ready);
>> +bool intel_gt_is_bind_context_ready(struct intel_gt *gt);
> Can you put all this part in patch 4 or make it a separate patch?


This patch contains the new member intel_engine_cs->bind_context_ready 
which is why I added these two function in this.

Probably it makes sense to move only i915_ggtt_require_binder() out of 
this patch ?


Regards,

Nirmoy

>
> Andi