diff mbox series

[3/3] drm/amdgpu: use new scheduler accounting

Message ID 20230615115630.164098-3-christian.koenig@amd.com (mailing list archive)
State New, archived
Headers show
Series [1/3] drm/scheduler: implement hw time accounting | expand

Commit Message

Christian König June 15, 2023, 11:56 a.m. UTC 
Instead of implementing this ourself.

Signed-off-by: Christian König <christian.koenig@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 52 ++++---------------------
 1 file changed, 8 insertions(+), 44 deletions(-)

Comments

Luben Tuikov June 15, 2023, 6:46 p.m. UTC | #1 
On 2023-06-15 07:56, Christian König wrote:
> Instead of implementing this ourself.

Spellcheck: "ourselves".

Acked-by: Luben Tuikov <luben.tuikov@amd.com>

Regards,
Luben

> 
> Signed-off-by: Christian König <christian.koenig@amd.com>
> ---
>  drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 52 ++++---------------------
>  1 file changed, 8 insertions(+), 44 deletions(-)
> 
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
> index 1445e030d788..f787a9b06d62 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
> @@ -163,41 +163,6 @@ static unsigned int amdgpu_ctx_get_hw_prio(struct amdgpu_ctx *ctx, u32 hw_ip)
>  	return hw_prio;
>  }
>  
> -/* Calculate the time spend on the hw */
> -static ktime_t amdgpu_ctx_fence_time(struct dma_fence *fence)
> -{
> -	struct drm_sched_fence *s_fence;
> -
> -	if (!fence)
> -		return ns_to_ktime(0);
> -
> -	/* When the fence is not even scheduled it can't have spend time */
> -	s_fence = to_drm_sched_fence(fence);
> -	if (!test_bit(DMA_FENCE_FLAG_TIMESTAMP_BIT, &s_fence->scheduled.flags))
> -		return ns_to_ktime(0);
> -
> -	/* When it is still running account how much already spend */
> -	if (!test_bit(DMA_FENCE_FLAG_TIMESTAMP_BIT, &s_fence->finished.flags))
> -		return ktime_sub(ktime_get(), s_fence->scheduled.timestamp);
> -
> -	return ktime_sub(s_fence->finished.timestamp,
> -			 s_fence->scheduled.timestamp);
> -}
> -
> -static ktime_t amdgpu_ctx_entity_time(struct amdgpu_ctx *ctx,
> -				      struct amdgpu_ctx_entity *centity)
> -{
> -	ktime_t res = ns_to_ktime(0);
> -	uint32_t i;
> -
> -	spin_lock(&ctx->ring_lock);
> -	for (i = 0; i < amdgpu_sched_jobs; i++) {
> -		res = ktime_add(res, amdgpu_ctx_fence_time(centity->fences[i]));
> -	}
> -	spin_unlock(&ctx->ring_lock);
> -	return res;
> -}
> -
>  static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, u32 hw_ip,
>  				  const u32 ring)
>  {
> @@ -257,16 +222,15 @@ static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, u32 hw_ip,
>  
>  static ktime_t amdgpu_ctx_fini_entity(struct amdgpu_ctx_entity *entity)
>  {
> -	ktime_t res = ns_to_ktime(0);
> +	ktime_t res;
>  	int i;
>  
>  	if (!entity)
> -		return res;
> +		return ns_to_ktime(0);
>  
> -	for (i = 0; i < amdgpu_sched_jobs; ++i) {
> -		res = ktime_add(res, amdgpu_ctx_fence_time(entity->fences[i]));
> +	for (i = 0; i < amdgpu_sched_jobs; ++i)
>  		dma_fence_put(entity->fences[i]);
> -	}
> +	res = drm_sched_entity_time_spend(&entity->entity);
>  	drm_sched_entity_destroy(&entity->entity);
>  	kfree(entity);
>  	return res;
> @@ -718,9 +682,6 @@ uint64_t amdgpu_ctx_add_fence(struct amdgpu_ctx *ctx,
>  	centity->sequence++;
>  	spin_unlock(&ctx->ring_lock);
>  
> -	atomic64_add(ktime_to_ns(amdgpu_ctx_fence_time(other)),
> -		     &ctx->mgr->time_spend[centity->hw_ip]);
> -
>  	dma_fence_put(other);
>  	return seq;
>  }
> @@ -900,12 +861,15 @@ void amdgpu_ctx_mgr_usage(struct amdgpu_ctx_mgr *mgr,
>  		for (hw_ip = 0; hw_ip < AMDGPU_HW_IP_NUM; ++hw_ip) {
>  			for (i = 0; i < amdgpu_ctx_num_entities[hw_ip]; ++i) {
>  				struct amdgpu_ctx_entity *centity;
> +				struct drm_sched_entity *entity;
>  				ktime_t spend;
>  
>  				centity = ctx->entities[hw_ip][i];
>  				if (!centity)
>  					continue;
> -				spend = amdgpu_ctx_entity_time(ctx, centity);
> +
> +				entity = &centity->entity;
> +				spend = drm_sched_entity_time_spend(entity);
>  				usage[hw_ip] = ktime_add(usage[hw_ip], spend);
>  			}
>  		}
diff mbox series

Patch

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
index 1445e030d788..f787a9b06d62 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
@@ -163,41 +163,6 @@  static unsigned int amdgpu_ctx_get_hw_prio(struct amdgpu_ctx *ctx, u32 hw_ip)
 	return hw_prio;
 }
 
-/* Calculate the time spend on the hw */
-static ktime_t amdgpu_ctx_fence_time(struct dma_fence *fence)
-{
-	struct drm_sched_fence *s_fence;
-
-	if (!fence)
-		return ns_to_ktime(0);
-
-	/* When the fence is not even scheduled it can't have spend time */
-	s_fence = to_drm_sched_fence(fence);
-	if (!test_bit(DMA_FENCE_FLAG_TIMESTAMP_BIT, &s_fence->scheduled.flags))
-		return ns_to_ktime(0);
-
-	/* When it is still running account how much already spend */
-	if (!test_bit(DMA_FENCE_FLAG_TIMESTAMP_BIT, &s_fence->finished.flags))
-		return ktime_sub(ktime_get(), s_fence->scheduled.timestamp);
-
-	return ktime_sub(s_fence->finished.timestamp,
-			 s_fence->scheduled.timestamp);
-}
-
-static ktime_t amdgpu_ctx_entity_time(struct amdgpu_ctx *ctx,
-				      struct amdgpu_ctx_entity *centity)
-{
-	ktime_t res = ns_to_ktime(0);
-	uint32_t i;
-
-	spin_lock(&ctx->ring_lock);
-	for (i = 0; i < amdgpu_sched_jobs; i++) {
-		res = ktime_add(res, amdgpu_ctx_fence_time(centity->fences[i]));
-	}
-	spin_unlock(&ctx->ring_lock);
-	return res;
-}
-
 static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, u32 hw_ip,
 				  const u32 ring)
 {
@@ -257,16 +222,15 @@  static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, u32 hw_ip,
 
 static ktime_t amdgpu_ctx_fini_entity(struct amdgpu_ctx_entity *entity)
 {
-	ktime_t res = ns_to_ktime(0);
+	ktime_t res;
 	int i;
 
 	if (!entity)
-		return res;
+		return ns_to_ktime(0);
 
-	for (i = 0; i < amdgpu_sched_jobs; ++i) {
-		res = ktime_add(res, amdgpu_ctx_fence_time(entity->fences[i]));
+	for (i = 0; i < amdgpu_sched_jobs; ++i)
 		dma_fence_put(entity->fences[i]);
-	}
+	res = drm_sched_entity_time_spend(&entity->entity);
 	drm_sched_entity_destroy(&entity->entity);
 	kfree(entity);
 	return res;
@@ -718,9 +682,6 @@  uint64_t amdgpu_ctx_add_fence(struct amdgpu_ctx *ctx,
 	centity->sequence++;
 	spin_unlock(&ctx->ring_lock);
 
-	atomic64_add(ktime_to_ns(amdgpu_ctx_fence_time(other)),
-		     &ctx->mgr->time_spend[centity->hw_ip]);
-
 	dma_fence_put(other);
 	return seq;
 }
@@ -900,12 +861,15 @@  void amdgpu_ctx_mgr_usage(struct amdgpu_ctx_mgr *mgr,
 		for (hw_ip = 0; hw_ip < AMDGPU_HW_IP_NUM; ++hw_ip) {
 			for (i = 0; i < amdgpu_ctx_num_entities[hw_ip]; ++i) {
 				struct amdgpu_ctx_entity *centity;
+				struct drm_sched_entity *entity;
 				ktime_t spend;
 
 				centity = ctx->entities[hw_ip][i];
 				if (!centity)
 					continue;
-				spend = amdgpu_ctx_entity_time(ctx, centity);
+
+				entity = &centity->entity;
+				spend = drm_sched_entity_time_spend(entity);
 				usage[hw_ip] = ktime_add(usage[hw_ip], spend);
 			}
 		}