Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.37 as permitted sender) client-ip=23.128.96.37;
Message-ID: <6fa280dd-af9c-4171-86b7-cbef91fd9ff8@gmail.com>
Date:   Tue, 31 Oct 2023 20:51:17 -0400
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:115.0) Gecko/20100101
 Thunderbird/115.4.1
Subject: Re: [PATCH drm-misc-next v4] drm/sched: implement dynamic job-flow
 control
Content-Language: en-CA, en-US
To:     Danilo Krummrich <dakr@redhat.com>, airlied@gmail.com,
        daniel@ffwll.ch, matthew.brost@intel.com,
        boris.brezillon@collabora.com, christian.koenig@amd.com,
        faith@gfxstrand.net
Cc:     linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org
References: <20231031002655.38707-1-dakr@redhat.com>
From:   Luben Tuikov <ltuikov89@gmail.com>
Autocrypt: addr=ltuikov89@gmail.com; keydata=
 xjMEZTohOhYJKwYBBAHaRw8BAQdAWSq76k+GsENjDTMVCy9Vr4fAO9Rb57/bPT1APnbnnRHN
 Ikx1YmVuIFR1aWtvdiA8bHR1aWtvdjg5QGdtYWlsLmNvbT7CmQQTFgoAQRYhBJkj7+VmFO9b
 eaAl10wVR5QxozSvBQJlOiE6AhsDBQkJZgGABQsJCAcCAiICBhUKCQgLAgQWAgMBAh4HAheA
 AAoJEEwVR5QxozSvSm4BAOwCpX53DTQhE20FBGlTMqKCOQyJqlMcIQ9SO1qPWX1iAQCv3vfy
 JwktF7REl1yt7IU2Sye1qmQMfJxdt9JMbMNNBs44BGU6IToSCisGAQQBl1UBBQEBB0BT9wSP
 cCE8uGe7FWo8C+nTSyWPXKTx9F0gpEnlqReRBwMBCAfCfgQYFgoAJhYhBJkj7+VmFO9beaAl
 10wVR5QxozSvBQJlOiE6AhsMBQkJZgGAAAoJEEwVR5QxozSvSsYA/2LIFjbxQ2ikbU5S0pKo
 aMDzO9eGz69uNhNWJcvIKJK6AQC9228Mqc1JeZMIyjYWr2HKYHi8S2q2/zHrSZwAWYYwDA==
In-Reply-To: <20231031002655.38707-1-dakr@redhat.com>
Content-Type: multipart/signed; micalg=pgp-sha256;
 protocol="application/pgp-signature";
 boundary="------------VvX20XGxV78GMScBqSWwGhJZ"
Precedence: bulk

This is an OpenPGP/MIME signed message (RFC 4880 and 3156)
--------------VvX20XGxV78GMScBqSWwGhJZ
Content-Type: multipart/mixed; boundary="------------RWsmEOH97E9ast0mTWzsc58q";
 protected-headers="v1"
From: Luben Tuikov <ltuikov89@gmail.com>
To: Danilo Krummrich <dakr@redhat.com>, airlied@gmail.com, daniel@ffwll.ch,
 matthew.brost@intel.com, boris.brezillon@collabora.com,
 christian.koenig@amd.com, faith@gfxstrand.net
Cc: linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org
Message-ID: <6fa280dd-af9c-4171-86b7-cbef91fd9ff8@gmail.com>
Subject: Re: [PATCH drm-misc-next v4] drm/sched: implement dynamic job-flow
 control
References: <20231031002655.38707-1-dakr@redhat.com>
In-Reply-To: <20231031002655.38707-1-dakr@redhat.com>

--------------RWsmEOH97E9ast0mTWzsc58q
Content-Type: multipart/mixed; boundary="------------mRHteIR4lB03PAorEkJH06Fc"

--------------mRHteIR4lB03PAorEkJH06Fc
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Hi,

(PSA: luben.tuikov@amd.com should've bounced :-) I'm removing it from the=
 To: field.)

On 2023-10-30 20:26, Danilo Krummrich wrote:
> Currently, job flow control is implemented simply by limiting the numbe=
r
> of jobs in flight. Therefore, a scheduler is initialized with a credit
> limit that corresponds to the number of jobs which can be sent to the
> hardware.
>=20
> This implies that for each job, drivers need to account for the maximum=

> job size possible in order to not overflow the ring buffer.
>=20
> However, there are drivers, such as Nouveau, where the job size has a
> rather large range. For such drivers it can easily happen that job
> submissions not even filling the ring by 1% can block subsequent
> submissions, which, in the worst case, can lead to the ring run dry.
>=20
> In order to overcome this issue, allow for tracking the actual job size=

> instead of the number of jobs. Therefore, add a field to track a job's
> credit count, which represents the number of credits a job contributes
> to the scheduler's credit limit.
>=20
> Signed-off-by: Danilo Krummrich <dakr@redhat.com>
> ---
> Changes in V2:
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>   - fixed up influence on scheduling fairness due to consideration of a=
 job's
>     size
>     - If we reach a ready entity in drm_sched_select_entity() but can't=
 actually
>       queue a job from it due to size limitations, just give up and go =
to sleep
>       until woken up due to a pending job finishing, rather than contin=
ue to try
>       other entities.
>   - added a callback to dynamically update a job's credits (Boris)
>   - renamed 'units' to 'credits'
>   - fixed commit message and comments as requested by Luben
>=20
> Changes in V3:
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>   - rebased onto V7 of the "DRM scheduler changes for Xe" series by Mat=
t
>   - move up drm_sched_can_queue() instead of adding a forward declarati=
on
>     (Boris)
>   - add a drm_sched_available_credits() helper (Boris)
>   - adjust control flow in drm_sched_rq_select_entity_fifo() to Luben's=
 proposal
>   - re-phrase a few comments and fix a typo (Luben)
>   - change naming of all structures credit fields and function paramete=
rs to the
>     following scheme
>     - drm_sched_job::credits
>     - drm_gpu_scheduler::credit_limit
>     - drm_gpu_scheduler::credit_count
>     - drm_sched_init(..., u32 credit_limit, ...)
>     - drm_sched_job_init(..., u32 credits, ...)
>   - add proper documentation for the scheduler's job-flow control mecha=
nism
>=20
> Changes in V4:
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>   - address Lubens comments regarding documentation
>   - switch to drm_WARN() variants
>   - WARN_ON() drivers passing in zero job credits for both drm_sched_jo=
b_init()
>     and the update_job_credits() callback
>   - don't retry with another runq if job doesn't fit on the ring to pre=
vent
>     priority inversion
>   - rebase onto drm-misc-next (will probably land before Matt's series)=

>=20
> Patch also available in [1].
>=20
> [1] https://gitlab.freedesktop.org/nouvelles/kernel/-/commits/sched-cre=
dits/
> ---
>  Documentation/gpu/drm-mm.rst                  |   6 +
>  drivers/gpu/drm/amd/amdgpu/amdgpu_job.c       |   2 +-
>  drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c  |   2 +-
>  drivers/gpu/drm/lima/lima_sched.c             |   2 +-
>  drivers/gpu/drm/msm/msm_gem_submit.c          |   2 +-
>  drivers/gpu/drm/nouveau/nouveau_sched.c       |   2 +-
>  drivers/gpu/drm/panfrost/panfrost_drv.c       |   2 +-
>  .../gpu/drm/scheduler/gpu_scheduler_trace.h   |   2 +-
>  drivers/gpu/drm/scheduler/sched_entity.c      |   4 +-
>  drivers/gpu/drm/scheduler/sched_main.c        | 148 ++++++++++++++----=

>  drivers/gpu/drm/v3d/v3d_gem.c                 |   2 +-
>  include/drm/gpu_scheduler.h                   |  31 +++-
>  12 files changed, 156 insertions(+), 49 deletions(-)
>=20
> diff --git a/Documentation/gpu/drm-mm.rst b/Documentation/gpu/drm-mm.rs=
t
> index 602010cb6894..acc5901ac840 100644
> --- a/Documentation/gpu/drm-mm.rst
> +++ b/Documentation/gpu/drm-mm.rst
> @@ -552,6 +552,12 @@ Overview
>  .. kernel-doc:: drivers/gpu/drm/scheduler/sched_main.c
>     :doc: Overview
> =20
> +Flow Control
> +------------
> +
> +.. kernel-doc:: drivers/gpu/drm/scheduler/sched_main.c
> +   :doc: Flow Control
> +
>  Scheduler Function References
>  -----------------------------
> =20
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/=
amd/amdgpu/amdgpu_job.c
> index 1f357198533f..62bb7fc7448a 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
> @@ -115,7 +115,7 @@ int amdgpu_job_alloc(struct amdgpu_device *adev, st=
ruct amdgpu_vm *vm,
>  	if (!entity)
>  		return 0;
> =20
> -	return drm_sched_job_init(&(*job)->base, entity, owner);
> +	return drm_sched_job_init(&(*job)->base, entity, 1, owner);
>  }
> =20
>  int amdgpu_job_alloc_with_ib(struct amdgpu_device *adev,
> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c b/drivers/gpu=
/drm/etnaviv/etnaviv_gem_submit.c
> index 2416c526f9b0..3d0f8d182506 100644
> --- a/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c
> +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c
> @@ -535,7 +535,7 @@ int etnaviv_ioctl_gem_submit(struct drm_device *dev=
, void *data,
> =20
>  	ret =3D drm_sched_job_init(&submit->sched_job,
>  				 &ctx->sched_entity[args->pipe],
> -				 submit->ctx);
> +				 1, submit->ctx);
>  	if (ret)
>  		goto err_submit_put;
> =20
> diff --git a/drivers/gpu/drm/lima/lima_sched.c b/drivers/gpu/drm/lima/l=
ima_sched.c
> index 295f0353a02e..0febe8ceb43b 100644
> --- a/drivers/gpu/drm/lima/lima_sched.c
> +++ b/drivers/gpu/drm/lima/lima_sched.c
> @@ -123,7 +123,7 @@ int lima_sched_task_init(struct lima_sched_task *ta=
sk,
>  	for (i =3D 0; i < num_bos; i++)
>  		drm_gem_object_get(&bos[i]->base.base);
> =20
> -	err =3D drm_sched_job_init(&task->base, &context->base, vm);
> +	err =3D drm_sched_job_init(&task->base, &context->base, 1, vm);
>  	if (err) {
>  		kfree(task->bos);
>  		return err;
> diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm=
/msm_gem_submit.c
> index 99744de6c05a..c002cabe7b9c 100644
> --- a/drivers/gpu/drm/msm/msm_gem_submit.c
> +++ b/drivers/gpu/drm/msm/msm_gem_submit.c
> @@ -48,7 +48,7 @@ static struct msm_gem_submit *submit_create(struct dr=
m_device *dev,
>  		return ERR_PTR(ret);
>  	}
> =20
> -	ret =3D drm_sched_job_init(&submit->base, queue->entity, queue);
> +	ret =3D drm_sched_job_init(&submit->base, queue->entity, 1, queue);
>  	if (ret) {
>  		kfree(submit->hw_fence);
>  		kfree(submit);
> diff --git a/drivers/gpu/drm/nouveau/nouveau_sched.c b/drivers/gpu/drm/=
nouveau/nouveau_sched.c
> index e62b04109b2f..2a845f9c34ec 100644
> --- a/drivers/gpu/drm/nouveau/nouveau_sched.c
> +++ b/drivers/gpu/drm/nouveau/nouveau_sched.c
> @@ -89,7 +89,7 @@ nouveau_job_init(struct nouveau_job *job,
> =20
>  	}
> =20
> -	ret =3D drm_sched_job_init(&job->base, &entity->base, NULL);
> +	ret =3D drm_sched_job_init(&job->base, &entity->base, 1, NULL);
>  	if (ret)
>  		goto err_free_chains;
> =20
> diff --git a/drivers/gpu/drm/panfrost/panfrost_drv.c b/drivers/gpu/drm/=
panfrost/panfrost_drv.c
> index b834777b409b..54d1c19bea84 100644
> --- a/drivers/gpu/drm/panfrost/panfrost_drv.c
> +++ b/drivers/gpu/drm/panfrost/panfrost_drv.c
> @@ -274,7 +274,7 @@ static int panfrost_ioctl_submit(struct drm_device =
*dev, void *data,
> =20
>  	ret =3D drm_sched_job_init(&job->base,
>  				 &file_priv->sched_entity[slot],
> -				 NULL);
> +				 1, NULL);
>  	if (ret)
>  		goto out_put_job;
> =20
> diff --git a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h b/drivers/=
gpu/drm/scheduler/gpu_scheduler_trace.h
> index 3143ecaaff86..f8ed093b7356 100644
> --- a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h
> +++ b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h
> @@ -51,7 +51,7 @@ DECLARE_EVENT_CLASS(drm_sched_job,
>  			   __assign_str(name, sched_job->sched->name);
>  			   __entry->job_count =3D spsc_queue_count(&entity->job_queue);
>  			   __entry->hw_job_count =3D atomic_read(
> -				   &sched_job->sched->hw_rq_count);
> +				   &sched_job->sched->credit_count);
>  			   ),
>  	    TP_printk("entity=3D%p, id=3D%llu, fence=3D%p, ring=3D%s, job cou=
nt:%u, hw job count:%d",
>  		      __entry->entity, __entry->id,
> diff --git a/drivers/gpu/drm/scheduler/sched_entity.c b/drivers/gpu/drm=
/scheduler/sched_entity.c
> index 409e4256f6e7..b79e0672b94f 100644
> --- a/drivers/gpu/drm/scheduler/sched_entity.c
> +++ b/drivers/gpu/drm/scheduler/sched_entity.c
> @@ -370,7 +370,7 @@ static void drm_sched_entity_wakeup(struct dma_fenc=
e *f,
>  		container_of(cb, struct drm_sched_entity, cb);
> =20
>  	drm_sched_entity_clear_dep(f, cb);
> -	drm_sched_wakeup_if_can_queue(entity->rq->sched);
> +	drm_sched_wakeup_if_can_queue(entity->rq->sched, entity);
>  }
> =20
>  /**
> @@ -602,7 +602,7 @@ void drm_sched_entity_push_job(struct drm_sched_job=
 *sched_job)
>  		if (drm_sched_policy =3D=3D DRM_SCHED_POLICY_FIFO)
>  			drm_sched_rq_update_fifo(entity, submit_ts);
> =20
> -		drm_sched_wakeup_if_can_queue(entity->rq->sched);
> +		drm_sched_wakeup_if_can_queue(entity->rq->sched, entity);
>  	}
>  }
>  EXPORT_SYMBOL(drm_sched_entity_push_job);
> diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/s=
cheduler/sched_main.c
> index fd755e953487..5dfde149441a 100644
> --- a/drivers/gpu/drm/scheduler/sched_main.c
> +++ b/drivers/gpu/drm/scheduler/sched_main.c
> @@ -48,6 +48,30 @@
>   * through the jobs entity pointer.
>   */
> =20
> +/**
> + * DOC: Flow Control
> + *
> + * The DRM GPU scheduler provides a flow control mechanism to regulate=
 the rate
> + * in which the jobs fetched from scheduler entities are executed.
> + *
> + * In this context the &drm_gpu_scheduler keeps track of a driver spec=
ified
> + * credit limit representing the capacity of this scheduler and a cred=
it count;
> + * every &drm_sched_job carries a driver specified number of credits.
> + *
> + * Once a job is executed (but not yet finished), the job's credits co=
ntribute
> + * to the scheduler's credit count until the job is finished. If by ex=
ecuting
> + * one more job the scheduler's credit count would exceed the schedule=
r's
> + * credit limit, the job won't be executed. Instead, the scheduler wil=
l wait
> + * until the credit count has decreased enough to not overflow its cre=
dit limit.
> + * This implies waiting for previously executed jobs.
> + *
> + * Optionally, drivers may register a callback (update_job_credits) pr=
ovided by
> + * struct drm_sched_backend_ops to update the job's credits dynamicall=
y. The
> + * scheduler executes this callback every time the scheduler considers=
 a job for
> + * execution and subsequently checks whether the job fits the schedule=
r's credit
> + * limit.
> + */

Excellent!

This reads really good right now and is very lucid. Thanks for updating i=
t.

> +
>  #include <linux/kthread.h>
>  #include <linux/wait.h>
>  #include <linux/sched.h>
> @@ -76,6 +100,46 @@ int drm_sched_policy =3D DRM_SCHED_POLICY_FIFO;
>  MODULE_PARM_DESC(sched_policy, "Specify the scheduling policy for enti=
ties on a run-queue, " __stringify(DRM_SCHED_POLICY_RR) " =3D Round Robin=
, " __stringify(DRM_SCHED_POLICY_FIFO) " =3D FIFO (default).");
>  module_param_named(sched_policy, drm_sched_policy, int, 0444);
> =20
> +static u32 drm_sched_available_credits(struct drm_gpu_scheduler *sched=
)
> +{
> +	u32 credits;
> +
> +	drm_WARN_ON(sched, check_sub_overflow(sched->credit_limit,
> +					      atomic_read(&sched->credit_count),
> +					      &credits));
> +
> +	return credits;
> +}
> +
> +/**
> + * drm_sched_can_queue -- Can we queue more to the hardware?
> + * @sched: scheduler instance
> + * @entity: the scheduler entity
> + *
> + * Return true if we can push at least one more job from @entity, fals=
e
> + * otherwise.
> + */
> +static bool drm_sched_can_queue(struct drm_gpu_scheduler *sched,
> +				struct drm_sched_entity *entity)
> +{
> +	struct drm_sched_job *s_job;
> +
> +	s_job =3D to_drm_sched_job(spsc_queue_peek(&entity->job_queue));
> +	if (!s_job)
> +		return false;
> +
> +	if (sched->ops->update_job_credits) {
> +		s_job->credits =3D sched->ops->update_job_credits(s_job);
> +
> +		drm_WARN(sched, !s_job->credits,
> +			 "Jobs with zero credits bypass job-flow control\n");
> +	}
> +
> +	drm_WARN_ON(sched, s_job->credits > sched->credit_limit);
> +
> +	return drm_sched_available_credits(sched) >=3D s_job->credits;
> +}
> +

Yeah, this is good.

>  static __always_inline bool drm_sched_entity_compare_before(struct rb_=
node *a,
>  							    const struct rb_node *b)
>  {
> @@ -187,12 +251,14 @@ void drm_sched_rq_remove_entity(struct drm_sched_=
rq *rq,
>  /**
>   * drm_sched_rq_select_entity_rr - Select an entity which could provid=
e a job to run
>   *
> + * @sched: the gpu scheduler
>   * @rq: scheduler run queue to check.
>   *
>   * Try to find a ready entity, returns NULL if none found.
>   */
>  static struct drm_sched_entity *
> -drm_sched_rq_select_entity_rr(struct drm_sched_rq *rq)
> +drm_sched_rq_select_entity_rr(struct drm_gpu_scheduler *sched,
> +			      struct drm_sched_rq *rq)
>  {
>  	struct drm_sched_entity *entity;
> =20
> @@ -202,6 +268,14 @@ drm_sched_rq_select_entity_rr(struct drm_sched_rq =
*rq)
>  	if (entity) {
>  		list_for_each_entry_continue(entity, &rq->entities, list) {
>  			if (drm_sched_entity_is_ready(entity)) {
> +				/* If we can't queue yet, preserve the current
> +				 * entity in terms of fairness.
> +				 */
> +				if (!drm_sched_can_queue(sched, entity)) {
> +					spin_unlock(&rq->lock);
> +					return ERR_PTR(-ENOSPC);
> +				}
> +
>  				rq->current_entity =3D entity;
>  				reinit_completion(&entity->entity_idle);
>  				spin_unlock(&rq->lock);
> @@ -211,8 +285,15 @@ drm_sched_rq_select_entity_rr(struct drm_sched_rq =
*rq)
>  	}
> =20
>  	list_for_each_entry(entity, &rq->entities, list) {
> -
>  		if (drm_sched_entity_is_ready(entity)) {
> +			/* If we can't queue yet, preserve the current entity in
> +			 * terms of fairness.
> +			 */
> +			if (!drm_sched_can_queue(sched, entity)) {
> +				spin_unlock(&rq->lock);
> +				return ERR_PTR(-ENOSPC);
> +			}
> +
>  			rq->current_entity =3D entity;
>  			reinit_completion(&entity->entity_idle);
>  			spin_unlock(&rq->lock);
> @@ -231,12 +312,14 @@ drm_sched_rq_select_entity_rr(struct drm_sched_rq=
 *rq)
>  /**
>   * drm_sched_rq_select_entity_fifo - Select an entity which provides a=
 job to run
>   *
> + * @sched: the gpu scheduler
>   * @rq: scheduler run queue to check.
>   *
>   * Find oldest waiting ready entity, returns NULL if none found.
>   */
>  static struct drm_sched_entity *
> -drm_sched_rq_select_entity_fifo(struct drm_sched_rq *rq)
> +drm_sched_rq_select_entity_fifo(struct drm_gpu_scheduler *sched,
> +				struct drm_sched_rq *rq)
>  {
>  	struct rb_node *rb;
> =20
> @@ -246,6 +329,14 @@ drm_sched_rq_select_entity_fifo(struct drm_sched_r=
q *rq)
> =20
>  		entity =3D rb_entry(rb, struct drm_sched_entity, rb_tree_node);
>  		if (drm_sched_entity_is_ready(entity)) {
> +			/* If we can't queue yet, preserve the current entity in
> +			 * terms of fairness.
> +			 */
> +			if (!drm_sched_can_queue(sched, entity)) {
> +				spin_unlock(&rq->lock);
> +				return ERR_PTR(-ENOSPC);
> +			}
> +

So, this kinda did this abrupt "return" in v2, then it was fixed fine in =
v3,
and now I see we return and return an error now, which doesn't seem to be=
 used
by anyone. In fact, in drm_sched_select_entity(), we do this,

-	return entity;
+	return IS_ERR(entity) ? NULL : entity;

So, it's perhaps best to leave this as it was in v3, and if in the future=

we need to distinguish between the type of error, then that future patch
would do that and also show how this is used with new code and logic.

There is little value to over-engineer this right now, given that in
the future, the logic may be more esoteric than we can think of.

>  			rq->current_entity =3D entity;
>  			reinit_completion(&entity->entity_idle);
>  			break;
> @@ -267,7 +358,7 @@ static void drm_sched_job_done(struct drm_sched_job=
 *s_job, int result)
>  	struct drm_sched_fence *s_fence =3D s_job->s_fence;
>  	struct drm_gpu_scheduler *sched =3D s_fence->sched;
> =20
> -	atomic_dec(&sched->hw_rq_count);
> +	atomic_sub(s_job->credits, &sched->credit_count);
>  	atomic_dec(sched->score);
> =20
>  	trace_drm_sched_process_job(s_fence);
> @@ -468,7 +559,7 @@ void drm_sched_stop(struct drm_gpu_scheduler *sched=
, struct drm_sched_job *bad)
>  					      &s_job->cb)) {
>  			dma_fence_put(s_job->s_fence->parent);
>  			s_job->s_fence->parent =3D NULL;
> -			atomic_dec(&sched->hw_rq_count);
> +			atomic_sub(s_job->credits, &sched->credit_count);
>  		} else {
>  			/*
>  			 * remove job from pending_list.
> @@ -529,7 +620,7 @@ void drm_sched_start(struct drm_gpu_scheduler *sche=
d, bool full_recovery)
>  	list_for_each_entry_safe(s_job, tmp, &sched->pending_list, list) {
>  		struct dma_fence *fence =3D s_job->s_fence->parent;
> =20
> -		atomic_inc(&sched->hw_rq_count);
> +		atomic_add(s_job->credits, &sched->credit_count);
> =20
>  		if (!full_recovery)
>  			continue;
> @@ -613,6 +704,8 @@ EXPORT_SYMBOL(drm_sched_resubmit_jobs);
>   * drm_sched_job_init - init a scheduler job
>   * @job: scheduler job to init
>   * @entity: scheduler entity to use
> + * @credits: the number of credits this job contributes to the schedul=
ers
> + * credit limit
>   * @owner: job owner for debugging
>   *
>   * Refer to drm_sched_entity_push_job() documentation
> @@ -630,7 +723,7 @@ EXPORT_SYMBOL(drm_sched_resubmit_jobs);
>   */
>  int drm_sched_job_init(struct drm_sched_job *job,
>  		       struct drm_sched_entity *entity,
> -		       void *owner)
> +		       u32 credits, void *owner)
>  {
>  	if (!entity->rq) {
>  		/* This will most likely be followed by missing frames
> @@ -647,6 +740,10 @@ int drm_sched_job_init(struct drm_sched_job *job,
>  		return -ENOMEM;
> =20
>  	INIT_LIST_HEAD(&job->list);
> +	job->credits =3D credits;
> +
> +	drm_WARN(job->sched, !credits,
> +		 "Jobs with zero credits bypass job-flow control\n");
> =20
>  	xa_init_flags(&job->dependencies, XA_FLAGS_ALLOC);
> =20
> @@ -853,27 +950,17 @@ void drm_sched_job_cleanup(struct drm_sched_job *=
job)
>  }
>  EXPORT_SYMBOL(drm_sched_job_cleanup);
> =20
> -/**
> - * drm_sched_can_queue -- Can we queue more to the hardware?
> - * @sched: scheduler instance
> - *
> - * Return true if we can push more jobs to the hw, otherwise false.
> - */
> -static bool drm_sched_can_queue(struct drm_gpu_scheduler *sched)
> -{
> -	return atomic_read(&sched->hw_rq_count) <
> -		sched->hw_submission_limit;
> -}
> -
>  /**
>   * drm_sched_wakeup_if_can_queue - Wake up the scheduler
>   * @sched: scheduler instance
> + * @entity: the scheduler entity
>   *
>   * Wake up the scheduler if we can queue jobs.
>   */
> -void drm_sched_wakeup_if_can_queue(struct drm_gpu_scheduler *sched)
> +void drm_sched_wakeup_if_can_queue(struct drm_gpu_scheduler *sched,
> +				   struct drm_sched_entity *entity)
>  {
> -	if (drm_sched_can_queue(sched))
> +	if (drm_sched_can_queue(sched, entity))
>  		wake_up_interruptible(&sched->wake_up_worker);
>  }
> =20
> @@ -890,19 +977,16 @@ drm_sched_select_entity(struct drm_gpu_scheduler =
*sched)
>  	struct drm_sched_entity *entity;
>  	int i;
> =20
> -	if (!drm_sched_can_queue(sched))
> -		return NULL;
> -
>  	/* Kernel run queue has higher priority than normal run queue*/
>  	for (i =3D sched->num_rqs - 1; i >=3D DRM_SCHED_PRIORITY_MIN; i--) {
>  		entity =3D drm_sched_policy =3D=3D DRM_SCHED_POLICY_FIFO ?
> -			drm_sched_rq_select_entity_fifo(sched->sched_rq[i]) :
> -			drm_sched_rq_select_entity_rr(sched->sched_rq[i]);
> +			drm_sched_rq_select_entity_fifo(sched, sched->sched_rq[i]) :
> +			drm_sched_rq_select_entity_rr(sched, sched->sched_rq[i]);
>  		if (entity)
>  			break;
>  	}
> =20
> -	return entity;
> +	return IS_ERR(entity) ? NULL : entity;

Let's leave this alone as we did in v3. The smaller diff-stat
this patch has, yet achieving the desired behaviour--the better
the code is.

>  }
> =20
>  /**
> @@ -1043,7 +1127,7 @@ static int drm_sched_main(void *param)
> =20
>  		s_fence =3D sched_job->s_fence;
> =20
> -		atomic_inc(&sched->hw_rq_count);
> +		atomic_add(sched_job->credits, &sched->credit_count);
>  		drm_sched_job_begin(sched_job);
> =20
>  		trace_drm_run_job(sched_job, entity);
> @@ -1078,7 +1162,7 @@ static int drm_sched_main(void *param)
>   * @sched: scheduler instance
>   * @ops: backend operations for this scheduler
>   * @num_rqs: number of runqueues, one for each priority, up to DRM_SCH=
ED_PRIORITY_COUNT
> - * @hw_submission: number of hw submissions that can be in flight
> + * @credit_limit: the number of credits this scheduler can hold from a=
ll jobs
>   * @hang_limit: number of times to allow a job to hang before dropping=
 it
>   * @timeout: timeout value in jiffies for the scheduler
>   * @timeout_wq: workqueue to use for timeout work. If NULL, the system=
_wq is
> @@ -1091,14 +1175,14 @@ static int drm_sched_main(void *param)
>   */
>  int drm_sched_init(struct drm_gpu_scheduler *sched,
>  		   const struct drm_sched_backend_ops *ops,
> -		   u32 num_rqs, uint32_t hw_submission, unsigned int hang_limit,
> +		   u32 num_rqs, u32 credit_limit, unsigned int hang_limit,
>  		   long timeout, struct workqueue_struct *timeout_wq,
>  		   atomic_t *score, const char *name, struct device *dev)
>  {
>  	int i, ret;
> =20
>  	sched->ops =3D ops;
> -	sched->hw_submission_limit =3D hw_submission;
> +	sched->credit_limit =3D credit_limit;
>  	sched->name =3D name;
>  	sched->timeout =3D timeout;
>  	sched->timeout_wq =3D timeout_wq ? : system_wq;
> @@ -1140,7 +1224,7 @@ int drm_sched_init(struct drm_gpu_scheduler *sche=
d,
>  	init_waitqueue_head(&sched->job_scheduled);
>  	INIT_LIST_HEAD(&sched->pending_list);
>  	spin_lock_init(&sched->job_list_lock);
> -	atomic_set(&sched->hw_rq_count, 0);
> +	atomic_set(&sched->credit_count, 0);
>  	INIT_DELAYED_WORK(&sched->work_tdr, drm_sched_job_timedout);
>  	atomic_set(&sched->_score, 0);
>  	atomic64_set(&sched->job_id_count, 0);
> diff --git a/drivers/gpu/drm/v3d/v3d_gem.c b/drivers/gpu/drm/v3d/v3d_ge=
m.c
> index afa7d170d1ff..8dc5a7988488 100644
> --- a/drivers/gpu/drm/v3d/v3d_gem.c
> +++ b/drivers/gpu/drm/v3d/v3d_gem.c
> @@ -417,7 +417,7 @@ v3d_job_init(struct v3d_dev *v3d, struct drm_file *=
file_priv,
>  	job->free =3D free;
> =20
>  	ret =3D drm_sched_job_init(&job->base, &v3d_priv->sched_entity[queue]=
,
> -				 v3d_priv);
> +				 1, v3d_priv);
>  	if (ret)
>  		goto fail;
> =20
> diff --git a/include/drm/gpu_scheduler.h b/include/drm/gpu_scheduler.h
> index 653e0eec9743..8bafa81300b2 100644
> --- a/include/drm/gpu_scheduler.h
> +++ b/include/drm/gpu_scheduler.h
> @@ -321,6 +321,7 @@ struct drm_sched_fence *to_drm_sched_fence(struct d=
ma_fence *f);
>   * @sched: the scheduler instance on which this job is scheduled.
>   * @s_fence: contains the fences for the scheduling of job.
>   * @finish_cb: the callback for the finished fence.
> + * @credits: the number of credits this job contributes to the schedul=
er
>   * @work: Helper to reschdeule job kill to different context.
>   * @id: a unique id assigned to each job scheduled on the scheduler.
>   * @karma: increment on every hang caused by this job. If this exceeds=
 the hang
> @@ -340,6 +341,8 @@ struct drm_sched_job {
>  	struct drm_gpu_scheduler	*sched;
>  	struct drm_sched_fence		*s_fence;
> =20
> +	u32				credits;
> +
>  	/*
>  	 * work is used only after finish_cb has been used and will not be
>  	 * accessed anymore.
> @@ -463,13 +466,27 @@ struct drm_sched_backend_ops {
>           * and it's time to clean it up.
>  	 */
>  	void (*free_job)(struct drm_sched_job *sched_job);
> +
> +	/**
> +	 * @update_job_credits: Called once the scheduler is considering this=

> +	 * job for execution.

"once" --> "when", a close cousin of "once", but clearer in code comments=
=2E
It is called in fact as many times as the job is considered to be pushed =
down
to hardware, if we couldn't previously.

> +	 *
> +	 * This callback returns the number of credits this job would take if=


Too many repetitions of "this". Instead of "this job", say "the job".

> +	 * pushed to the hardware. Drivers may use this to dynamically update=

> +	 * the job's credit count. For instance, deduct the number of credits=

> +	 * for already signalled native fences.
> +	 *
> +	 * This callback is optional.
> +	 */

Yeah, this is really good and well said. It's very clear and lucid. Thank=
s
for updating it.

> +	u32 (*update_job_credits)(struct drm_sched_job *sched_job);
>  };
> =20
>  /**
>   * struct drm_gpu_scheduler - scheduler instance-specific data
>   *
>   * @ops: backend operations provided by the driver.
> - * @hw_submission_limit: the max size of the hardware queue.
> + * @credit_limit: the credit limit of this scheduler
> + * @credit_count: the current credit count of this scheduler
>   * @timeout: the time after which a job is removed from the scheduler.=

>   * @name: name of the ring for which this scheduler is being used.
>   * @num_rqs: Number of run-queues. This is at most DRM_SCHED_PRIORITY_=
COUNT,
> @@ -480,7 +497,6 @@ struct drm_sched_backend_ops {
>   * @job_scheduled: once @drm_sched_entity_do_release is called the sch=
eduler
>   *                 waits on this wait queue until all the scheduled jo=
bs are
>   *                 finished.
> - * @hw_rq_count: the number of jobs currently in the hardware queue.
>   * @job_id_count: used to assign unique id to the each job.
>   * @timeout_wq: workqueue used to queue @work_tdr
>   * @work_tdr: schedules a delayed call to @drm_sched_job_timedout afte=
r the
> @@ -500,14 +516,14 @@ struct drm_sched_backend_ops {
>   */
>  struct drm_gpu_scheduler {
>  	const struct drm_sched_backend_ops	*ops;
> -	uint32_t			hw_submission_limit;
> +	u32				credit_limit;
> +	atomic_t			credit_count;
>  	long				timeout;
>  	const char			*name;
>  	u32                             num_rqs;
>  	struct drm_sched_rq             **sched_rq;
>  	wait_queue_head_t		wake_up_worker;
>  	wait_queue_head_t		job_scheduled;
> -	atomic_t			hw_rq_count;
>  	atomic64_t			job_id_count;
>  	struct workqueue_struct		*timeout_wq;
>  	struct delayed_work		work_tdr;
> @@ -524,14 +540,14 @@ struct drm_gpu_scheduler {
> =20
>  int drm_sched_init(struct drm_gpu_scheduler *sched,
>  		   const struct drm_sched_backend_ops *ops,
> -		   u32 num_rqs, uint32_t hw_submission, unsigned int hang_limit,
> +		   u32 num_rqs, u32 credit_limit, unsigned int hang_limit,
>  		   long timeout, struct workqueue_struct *timeout_wq,
>  		   atomic_t *score, const char *name, struct device *dev);
> =20
>  void drm_sched_fini(struct drm_gpu_scheduler *sched);
>  int drm_sched_job_init(struct drm_sched_job *job,
>  		       struct drm_sched_entity *entity,
> -		       void *owner);
> +		       u32 credits, void *owner);
>  void drm_sched_job_arm(struct drm_sched_job *job);
>  int drm_sched_job_add_dependency(struct drm_sched_job *job,
>  				 struct dma_fence *fence);
> @@ -552,7 +568,8 @@ void drm_sched_entity_modify_sched(struct drm_sched=
_entity *entity,
>                                     unsigned int num_sched_list);
> =20
>  void drm_sched_job_cleanup(struct drm_sched_job *job);
> -void drm_sched_wakeup_if_can_queue(struct drm_gpu_scheduler *sched);
> +void drm_sched_wakeup_if_can_queue(struct drm_gpu_scheduler *sched,
> +				   struct drm_sched_entity *entity);
>  void drm_sched_stop(struct drm_gpu_scheduler *sched, struct drm_sched_=
job *bad);
>  void drm_sched_start(struct drm_gpu_scheduler *sched, bool full_recove=
ry);
>  void drm_sched_resubmit_jobs(struct drm_gpu_scheduler *sched);
>=20
> base-commit: b2139fb5051554a7f297e4ad584ef1bc26c76d5d

Ah, thanks for adding base!

So, yeah--we're almost there!
--=20
Regards,
Luben

--------------mRHteIR4lB03PAorEkJH06Fc
Content-Type: application/pgp-keys; name="OpenPGP_0x4C15479431A334AF.asc"
Content-Disposition: attachment; filename="OpenPGP_0x4C15479431A334AF.asc"
Content-Description: OpenPGP public key
Content-Transfer-Encoding: quoted-printable

-----BEGIN PGP PUBLIC KEY BLOCK-----
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=3D=3D
=3DqCaZ
-----END PGP PUBLIC KEY BLOCK-----

--------------mRHteIR4lB03PAorEkJH06Fc--

--------------RWsmEOH97E9ast0mTWzsc58q--

--------------VvX20XGxV78GMScBqSWwGhJZ
Content-Type: application/pgp-signature; name="OpenPGP_signature.asc"
Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="OpenPGP_signature.asc"

-----BEGIN PGP SIGNATURE-----

wnsEABYIACMWIQSZI+/lZhTvW3mgJddMFUeUMaM0rwUCZUGhBQUDAAAAAAAKCRBMFUeUMaM0r7P0
AQCb4Ufdn6DfpGHLD0H4v7iVnr1Ye5wHQbJRfL9w527r+AEAyhfdh+CJYmLS9vzyU7Dp55KvL2Xd
ihdyL5mf7EJwcgk=
=XtBg
-----END PGP SIGNATURE-----

--------------VvX20XGxV78GMScBqSWwGhJZ--