Re: [PATCH 14/20] drm/i915/gt: Track timeline activeness in enter/exit

[Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>]

On 18/07/2019 08:00, Chris Wilson wrote:
> Lift moving the timeline to/from the active_list on enter/exit in order
> to shorten the active tracking span in comparison to the existing
> pin/unpin.
> 
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> ---
>   drivers/gpu/drm/i915/gem/i915_gem_pm.c        |  1 -
>   drivers/gpu/drm/i915/gt/intel_context.c       |  2 +
>   drivers/gpu/drm/i915/gt/intel_engine_pm.c     |  1 +
>   drivers/gpu/drm/i915/gt/intel_lrc.c           |  4 +
>   drivers/gpu/drm/i915/gt/intel_timeline.c      | 98 +++++++------------
>   drivers/gpu/drm/i915/gt/intel_timeline.h      |  3 +-
>   .../gpu/drm/i915/gt/intel_timeline_types.h    |  1 +
>   drivers/gpu/drm/i915/gt/selftest_timeline.c   |  2 -
>   8 files changed, 46 insertions(+), 66 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pm.c b/drivers/gpu/drm/i915/gem/i915_gem_pm.c
> index 8faf262278ae..195ee6eedac0 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_pm.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_pm.c
> @@ -39,7 +39,6 @@ static void i915_gem_park(struct drm_i915_private *i915)
>   		i915_gem_batch_pool_fini(&engine->batch_pool);
>   	}
>   
> -	intel_timelines_park(i915);
>   	i915_vma_parked(i915);
>   
>   	i915_globals_park();
> diff --git a/drivers/gpu/drm/i915/gt/intel_context.c b/drivers/gpu/drm/i915/gt/intel_context.c
> index 9830edda1ade..87c84cc0f658 100644
> --- a/drivers/gpu/drm/i915/gt/intel_context.c
> +++ b/drivers/gpu/drm/i915/gt/intel_context.c
> @@ -242,10 +242,12 @@ int __init i915_global_context_init(void)
>   void intel_context_enter_engine(struct intel_context *ce)
>   {
>   	intel_engine_pm_get(ce->engine);
> +	intel_timeline_enter(ce->ring->timeline);
>   }
>   
>   void intel_context_exit_engine(struct intel_context *ce)
>   {
> +	intel_timeline_exit(ce->ring->timeline);
>   	intel_engine_pm_put(ce->engine);
>   }
>   
> diff --git a/drivers/gpu/drm/i915/gt/intel_engine_pm.c b/drivers/gpu/drm/i915/gt/intel_engine_pm.c
> index e74fbf04a68d..072f65e6a09e 100644
> --- a/drivers/gpu/drm/i915/gt/intel_engine_pm.c
> +++ b/drivers/gpu/drm/i915/gt/intel_engine_pm.c
> @@ -89,6 +89,7 @@ static bool switch_to_kernel_context(struct intel_engine_cs *engine)
>   
>   	/* Check again on the next retirement. */
>   	engine->wakeref_serial = engine->serial + 1;
> +	intel_timeline_enter(rq->timeline);
>   
>   	i915_request_add_barriers(rq);
>   	__i915_request_commit(rq);
> diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
> index 884dfc1cb033..aceb990ae3b9 100644
> --- a/drivers/gpu/drm/i915/gt/intel_lrc.c
> +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
> @@ -3253,6 +3253,8 @@ static void virtual_context_enter(struct intel_context *ce)
>   
>   	for (n = 0; n < ve->num_siblings; n++)
>   		intel_engine_pm_get(ve->siblings[n]);
> +
> +	intel_timeline_enter(ce->ring->timeline);

Here we couldn't enter all sibling contexts instead? Would be a bit 
wasteful I guess. And there must be a place where it is already done. 
But can't be on picking the engine, where is it?

>   }
>   
>   static void virtual_context_exit(struct intel_context *ce)
> @@ -3260,6 +3262,8 @@ static void virtual_context_exit(struct intel_context *ce)
>   	struct virtual_engine *ve = container_of(ce, typeof(*ve), context);
>   	unsigned int n;
>   
> +	intel_timeline_exit(ce->ring->timeline);
> +
>   	for (n = 0; n < ve->num_siblings; n++)
>   		intel_engine_pm_put(ve->siblings[n]);
>   }
> diff --git a/drivers/gpu/drm/i915/gt/intel_timeline.c b/drivers/gpu/drm/i915/gt/intel_timeline.c
> index 6daa9eb59e19..4af0b9801d91 100644
> --- a/drivers/gpu/drm/i915/gt/intel_timeline.c
> +++ b/drivers/gpu/drm/i915/gt/intel_timeline.c
> @@ -278,64 +278,11 @@ void intel_timelines_init(struct drm_i915_private *i915)
>   	timelines_init(&i915->gt);
>   }
>   
> -static void timeline_add_to_active(struct intel_timeline *tl)
> -{
> -	struct intel_gt_timelines *gt = &tl->gt->timelines;
> -
> -	mutex_lock(&gt->mutex);
> -	list_add(&tl->link, &gt->active_list);
> -	mutex_unlock(&gt->mutex);
> -}
> -
> -static void timeline_remove_from_active(struct intel_timeline *tl)
> -{
> -	struct intel_gt_timelines *gt = &tl->gt->timelines;
> -
> -	mutex_lock(&gt->mutex);
> -	list_del(&tl->link);
> -	mutex_unlock(&gt->mutex);
> -}
> -
> -static void timelines_park(struct intel_gt *gt)
> -{
> -	struct intel_gt_timelines *timelines = &gt->timelines;
> -	struct intel_timeline *timeline;
> -
> -	mutex_lock(&timelines->mutex);
> -	list_for_each_entry(timeline, &timelines->active_list, link) {
> -		/*
> -		 * All known fences are completed so we can scrap
> -		 * the current sync point tracking and start afresh,
> -		 * any attempt to wait upon a previous sync point
> -		 * will be skipped as the fence was signaled.
> -		 */
> -		i915_syncmap_free(&timeline->sync);
> -	}
> -	mutex_unlock(&timelines->mutex);
> -}
> -
> -/**
> - * intel_timelines_park - called when the driver idles
> - * @i915: the drm_i915_private device
> - *
> - * When the driver is completely idle, we know that all of our sync points
> - * have been signaled and our tracking is then entirely redundant. Any request
> - * to wait upon an older sync point will be completed instantly as we know
> - * the fence is signaled and therefore we will not even look them up in the
> - * sync point map.
> - */
> -void intel_timelines_park(struct drm_i915_private *i915)
> -{
> -	timelines_park(&i915->gt);
> -}
> -
>   void intel_timeline_fini(struct intel_timeline *timeline)
>   {
>   	GEM_BUG_ON(timeline->pin_count);
>   	GEM_BUG_ON(!list_empty(&timeline->requests));
>   
> -	i915_syncmap_free(&timeline->sync);
> -
>   	if (timeline->hwsp_cacheline)
>   		cacheline_free(timeline->hwsp_cacheline);
>   	else
> @@ -370,6 +317,7 @@ int intel_timeline_pin(struct intel_timeline *tl)
>   	if (tl->pin_count++)
>   		return 0;
>   	GEM_BUG_ON(!tl->pin_count);
> +	GEM_BUG_ON(tl->active_count);
>   
>   	err = i915_vma_pin(tl->hwsp_ggtt, 0, 0, PIN_GLOBAL | PIN_HIGH);
>   	if (err)
> @@ -380,7 +328,6 @@ int intel_timeline_pin(struct intel_timeline *tl)
>   		offset_in_page(tl->hwsp_offset);
>   
>   	cacheline_acquire(tl->hwsp_cacheline);
> -	timeline_add_to_active(tl);
>   
>   	return 0;
>   
> @@ -389,6 +336,40 @@ int intel_timeline_pin(struct intel_timeline *tl)
>   	return err;
>   }
>   
> +void intel_timeline_enter(struct intel_timeline *tl)
> +{
> +	struct intel_gt_timelines *timelines = &tl->gt->timelines;
> +
> +	GEM_BUG_ON(!tl->pin_count);
> +	if (tl->active_count++)
> +		return;
> +	GEM_BUG_ON(!tl->active_count); /* overflow? */
> +
> +	mutex_lock(&timelines->mutex);
> +	list_add(&tl->link, &timelines->active_list);
> +	mutex_unlock(&timelines->mutex);
> +}
> +
> +void intel_timeline_exit(struct intel_timeline *tl)
> +{
> +	struct intel_gt_timelines *timelines = &tl->gt->timelines;
> +
> +	GEM_BUG_ON(!tl->active_count);
> +	if (--tl->active_count)
> +		return;
> +
> +	mutex_lock(&timelines->mutex);
> +	list_del(&tl->link);
> +	mutex_unlock(&timelines->mutex);

So we end up with one lock protecting tl->active_count and another for 
the list of active timelines?

> +
> +	/*
> +	 * Since this timeline is idle, all bariers upon which we were waiting
> +	 * must also be complete and so we can discard the last used barriers
> +	 * without loss of information.
> +	 */
> +	i915_syncmap_free(&tl->sync);
> +}
> +
>   static u32 timeline_advance(struct intel_timeline *tl)
>   {
>   	GEM_BUG_ON(!tl->pin_count);
> @@ -546,16 +527,9 @@ void intel_timeline_unpin(struct intel_timeline *tl)
>   	if (--tl->pin_count)
>   		return;
>   
> -	timeline_remove_from_active(tl);
> +	GEM_BUG_ON(tl->active_count);
>   	cacheline_release(tl->hwsp_cacheline);
>   
> -	/*
> -	 * Since this timeline is idle, all bariers upon which we were waiting
> -	 * must also be complete and so we can discard the last used barriers
> -	 * without loss of information.
> -	 */
> -	i915_syncmap_free(&tl->sync);
> -
>   	__i915_vma_unpin(tl->hwsp_ggtt);
>   }
>   
> diff --git a/drivers/gpu/drm/i915/gt/intel_timeline.h b/drivers/gpu/drm/i915/gt/intel_timeline.h
> index e08cebf64833..f583af1ba18d 100644
> --- a/drivers/gpu/drm/i915/gt/intel_timeline.h
> +++ b/drivers/gpu/drm/i915/gt/intel_timeline.h
> @@ -77,9 +77,11 @@ static inline bool intel_timeline_sync_is_later(struct intel_timeline *tl,
>   }
>   
>   int intel_timeline_pin(struct intel_timeline *tl);
> +void intel_timeline_enter(struct intel_timeline *tl);
>   int intel_timeline_get_seqno(struct intel_timeline *tl,
>   			     struct i915_request *rq,
>   			     u32 *seqno);
> +void intel_timeline_exit(struct intel_timeline *tl);
>   void intel_timeline_unpin(struct intel_timeline *tl);
>   
>   int intel_timeline_read_hwsp(struct i915_request *from,
> @@ -87,7 +89,6 @@ int intel_timeline_read_hwsp(struct i915_request *from,
>   			     u32 *hwsp_offset);
>   
>   void intel_timelines_init(struct drm_i915_private *i915);
> -void intel_timelines_park(struct drm_i915_private *i915);
>   void intel_timelines_fini(struct drm_i915_private *i915);
>   
>   #endif
> diff --git a/drivers/gpu/drm/i915/gt/intel_timeline_types.h b/drivers/gpu/drm/i915/gt/intel_timeline_types.h
> index 9a71aea7a338..b820ee76b7f5 100644
> --- a/drivers/gpu/drm/i915/gt/intel_timeline_types.h
> +++ b/drivers/gpu/drm/i915/gt/intel_timeline_types.h
> @@ -58,6 +58,7 @@ struct intel_timeline {
>   	 */
>   	struct i915_syncmap *sync;
>   
> +	unsigned int active_count;

Is it becoming non-obvious what is pin_count and what is active_count? I 
suggest some comments dropped along the members here.

>   	struct list_head link;
>   	struct intel_gt *gt;
>   
> diff --git a/drivers/gpu/drm/i915/gt/selftest_timeline.c b/drivers/gpu/drm/i915/gt/selftest_timeline.c
> index f0a840030382..d54113697745 100644
> --- a/drivers/gpu/drm/i915/gt/selftest_timeline.c
> +++ b/drivers/gpu/drm/i915/gt/selftest_timeline.c
> @@ -816,8 +816,6 @@ static int live_hwsp_recycle(void *arg)
>   
>   			if (err)
>   				goto out;
> -
> -			intel_timelines_park(i915); /* Encourage recycling! */
>   		} while (!__igt_timeout(end_time, NULL));
>   	}
>   
> 

Regards,

Tvrtko
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx