[Intel-gfx] [PATCH v7 2/4] drm/i915/perf: stop using the kernel context

[Lionel Landwerlin]

On 28/04/2020 13:19, Chris Wilson wrote:
> Quoting Lionel Landwerlin (2020-04-28 11:08:14)
>> @@ -2837,6 +2851,7 @@ static int i915_oa_stream_init(struct i915_perf_stream *stream,
>>   {
>>          struct drm_i915_private *i915 = stream->perf->i915;
>>          struct i915_perf *perf = stream->perf;
>> +       struct intel_timeline *timeline;
>>          int format_size;
>>          int ret;
>>   
>> @@ -2946,10 +2961,30 @@ static int i915_oa_stream_init(struct i915_perf_stream *stream,
>>   
>>          stream->ops = &i915_oa_stream_ops;
>>   
>> +       timeline = intel_timeline_create(stream->engine->gt, NULL);
>> +       if (IS_ERR(timeline)) {
>> +               ret = PTR_ERR(timeline);
>> +               goto err_timeline;
>> +       }
>> +
>> +       stream->config_context = intel_context_create(stream->engine);
>> +       if (IS_ERR(stream->config_context)) {
>> +               intel_timeline_put(timeline);
>> +               ret = PTR_ERR(stream->config_context);
>> +               goto err_timeline;
>> +       }
>> +
>> +       stream->config_context->sseu = props->sseu;
>> +       stream->config_context->timeline = timeline;
> Timeline?
>
>> +
>> +       ret = intel_context_pin(stream->config_context);
> One is created for the context here [via intel_context_alloc].


I don't see it :

static struct intel_context *intel_context_alloc(void)
{
         return kmem_cache_zalloc(global.slab_ce, GFP_KERNEL);
}



>
> I was intrigued as I thought you might have had something special planned
> with multiple contexts sharing a timeline
>
> Do we need to permanently pin the context? We can fail the OA config
> change request, right?


You're right, it doesn't need to pinned all the time.


Thanks,


-Lionel


> -Chris