AMD-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Kuehling, Felix" <Felix.Kuehling-5C7GfCeVMHo@public.gmane.org>
To: "Zeng, Oak" <Oak.Zeng-5C7GfCeVMHo@public.gmane.org>,
	"amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org"
	<amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org>
Subject: Re: [PATCH 3/5] drm/amdkfd: Refactor create_queue_nocpsch
Date: Mon, 3 Jun 2019 22:04:58 +0000	[thread overview]
Message-ID: <10b7f416-df5e-6fa2-930f-71af79e67974@amd.com> (raw)
In-Reply-To: <1559584248-12115-3-git-send-email-Oak.Zeng-5C7GfCeVMHo@public.gmane.org>

On 2019-06-03 1:51 p.m., Zeng, Oak wrote:
> This is prepare work to fix a circular lock dependency.
> No logic change
>
> Change-Id: I4e0ee918260e7780de972dd71f4ce787b4f6dde9
> Signed-off-by: Oak Zeng <Oak.Zeng@amd.com>
> ---
>   .../gpu/drm/amd/amdkfd/kfd_device_queue_manager.c  | 171 +++++++--------------
>   1 file changed, 58 insertions(+), 113 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
> index dc1a70b..06654de 100644
> --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
> @@ -42,10 +42,6 @@
>   static int set_pasid_vmid_mapping(struct device_queue_manager *dqm,
>   					unsigned int pasid, unsigned int vmid);
>   
> -static int create_compute_queue_nocpsch(struct device_queue_manager *dqm,
> -					struct queue *q,
> -					struct qcm_process_device *qpd);
> -
>   static int execute_queues_cpsch(struct device_queue_manager *dqm,
>   				enum kfd_unmap_queues_filter filter,
>   				uint32_t filter_param);
> @@ -55,13 +51,14 @@ static int unmap_queues_cpsch(struct device_queue_manager *dqm,
>   
>   static int map_queues_cpsch(struct device_queue_manager *dqm);
>   
> -static int create_sdma_queue_nocpsch(struct device_queue_manager *dqm,
> -					struct queue *q,
> -					struct qcm_process_device *qpd);
> -
>   static void deallocate_sdma_queue(struct device_queue_manager *dqm,
>   				struct queue *q);
>   
> +static inline void deallocate_hqd(struct device_queue_manager *dqm,
> +				struct queue *q);
> +static int allocate_hqd(struct device_queue_manager *dqm, struct queue *q);
> +static int allocate_sdma_queue(struct device_queue_manager *dqm,
> +				struct queue *q);
>   static void kfd_process_hw_exception(struct work_struct *work);
>   
>   static inline
> @@ -269,6 +266,7 @@ static int create_queue_nocpsch(struct device_queue_manager *dqm,
>   				struct queue *q,
>   				struct qcm_process_device *qpd)
>   {
> +	struct mqd_manager *mqd_mgr;
>   	int retval;
>   
>   	print_queue(q);
> @@ -300,18 +298,46 @@ static int create_queue_nocpsch(struct device_queue_manager *dqm,
>   	q->properties.tba_addr = qpd->tba_addr;
>   	q->properties.tma_addr = qpd->tma_addr;
>   
> +	mqd_mgr = dqm->mqd_mgrs[get_mqd_type_from_queue_type(
> +			q->properties.type)];
> +	if (q->properties.type == KFD_QUEUE_TYPE_COMPUTE) {
> +		retval = allocate_hqd(dqm, q);
> +		if (retval)
> +			goto deallocate_vmid;
> +	} else if (q->properties.type == KFD_QUEUE_TYPE_SDMA ||
> +		q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) {
> +		retval = allocate_sdma_queue(dqm, q);
> +		if (retval)
> +			goto deallocate_vmid;
> +		dqm->asic_ops.init_sdma_vm(dqm, q, qpd);
> +	}
> +
> +	retval = allocate_doorbell(qpd, q);
> +	if (retval)
> +		goto out_deallocate_hqd;
> +
> +	retval = mqd_mgr->init_mqd(mqd_mgr, &q->mqd, &q->mqd_mem_obj,
> +				&q->gart_mqd_addr, &q->properties);
> +	if (retval)
> +		goto out_deallocate_doorbell;
> +
> +	pr_debug("Loading mqd to hqd on pipe %d, queue %d\n",
> +			q->pipe, q->queue);

This message only makes sense for compute queues. Maybe move this up to 
just after allocate_hqd, because that's what assignes q->pipe and q->queue.


> +
>   	if (q->properties.type == KFD_QUEUE_TYPE_COMPUTE)
> -		retval = create_compute_queue_nocpsch(dqm, q, qpd);
> -	else if (q->properties.type == KFD_QUEUE_TYPE_SDMA ||
> -			q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI)
> -		retval = create_sdma_queue_nocpsch(dqm, q, qpd);
> -	else
> -		retval = -EINVAL;
> +		dqm->dev->kfd2kgd->set_scratch_backing_va(
> +			dqm->dev->kgd, qpd->sh_hidden_private_base, qpd->vmid);

Not sure why this is here. I think it would make more sense in 
allocate_vmid. Normally this is called when the scratch-backing VA is 
set in the ioctl function. Unless no VMID has been assigned to the 
process yet. So setting this when the VMID is assigned in allocate_vmid 
seems to make the most sense.


>   
> -	if (retval) {
> -		if (list_empty(&qpd->queues_list))
> -			deallocate_vmid(dqm, qpd, q);
> -		goto out_unlock;
> +	if (q->properties.is_active) {
> +
> +		if (WARN(q->process->mm != current->mm,
> +					"should only run in user thread"))
> +			retval = -EFAULT;
> +		else
> +			retval = mqd_mgr->load_mqd(mqd_mgr, q->mqd, q->pipe,
> +					q->queue, &q->properties, current->mm);
> +		if (retval)
> +			goto out_uninit_mqd;
>   	}
>   
>   	list_add(&q->list, &qpd->queues_list);
> @@ -331,7 +357,21 @@ static int create_queue_nocpsch(struct device_queue_manager *dqm,
>   	dqm->total_queue_count++;
>   	pr_debug("Total of %d queues are accountable so far\n",
>   			dqm->total_queue_count);
> +	goto out_unlock;
>   
> +out_uninit_mqd:
> +	mqd_mgr->uninit_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
> +out_deallocate_doorbell:
> +	deallocate_doorbell(qpd, q);
> +out_deallocate_hqd:
> +	if (q->properties.type == KFD_QUEUE_TYPE_COMPUTE)
> +		deallocate_hqd(dqm, q);
> +	else if (q->properties.type == KFD_QUEUE_TYPE_SDMA ||
> +		q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI)
> +		deallocate_sdma_queue(dqm, q);
> +deallocate_vmid:
> +	if (list_empty(&qpd->queues_list))
> +		deallocate_vmid(dqm, qpd, q);
>   out_unlock:
>   	dqm_unlock(dqm);
>   	return retval;
> @@ -377,58 +417,6 @@ static inline void deallocate_hqd(struct device_queue_manager *dqm,
>   	dqm->allocated_queues[q->pipe] |= (1 << q->queue);
>   }
>   
> -static int create_compute_queue_nocpsch(struct device_queue_manager *dqm,
> -					struct queue *q,
> -					struct qcm_process_device *qpd)
> -{
> -	struct mqd_manager *mqd_mgr;
> -	int retval;
> -
> -	mqd_mgr = dqm->mqd_mgrs[KFD_MQD_TYPE_COMPUTE];
> -
> -	retval = allocate_hqd(dqm, q);
> -	if (retval)
> -		return retval;
> -
> -	retval = allocate_doorbell(qpd, q);
> -	if (retval)
> -		goto out_deallocate_hqd;
> -
> -	retval = mqd_mgr->init_mqd(mqd_mgr, &q->mqd, &q->mqd_mem_obj,
> -				&q->gart_mqd_addr, &q->properties);
> -	if (retval)
> -		goto out_deallocate_doorbell;
> -
> -	pr_debug("Loading mqd to hqd on pipe %d, queue %d\n",
> -			q->pipe, q->queue);
> -
> -	dqm->dev->kfd2kgd->set_scratch_backing_va(
> -			dqm->dev->kgd, qpd->sh_hidden_private_base, qpd->vmid);
> -
> -	if (!q->properties.is_active)
> -		return 0;
> -
> -	if (WARN(q->process->mm != current->mm,
> -		 "should only run in user thread"))
> -		retval = -EFAULT;
> -	else
> -		retval = mqd_mgr->load_mqd(mqd_mgr, q->mqd, q->pipe, q->queue,
> -					   &q->properties, current->mm);
> -	if (retval)
> -		goto out_uninit_mqd;
> -
> -	return 0;
> -
> -out_uninit_mqd:
> -	mqd_mgr->uninit_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
> -out_deallocate_doorbell:
> -	deallocate_doorbell(qpd, q);
> -out_deallocate_hqd:
> -	deallocate_hqd(dqm, q);
> -
> -	return retval;
> -}
> -
>   /* Access to DQM has to be locked before calling destroy_queue_nocpsch_locked
>    * to avoid asynchronized access
>    */
> @@ -967,49 +955,6 @@ static void deallocate_sdma_queue(struct device_queue_manager *dqm,
>   	}
>   }
>   
> -static int create_sdma_queue_nocpsch(struct device_queue_manager *dqm,
> -					struct queue *q,
> -					struct qcm_process_device *qpd)
> -{
> -	struct mqd_manager *mqd_mgr;
> -	int retval;
> -
> -	mqd_mgr = dqm->mqd_mgrs[KFD_MQD_TYPE_SDMA];
> -
> -	retval = allocate_sdma_queue(dqm, q);
> -	if (retval)
> -		return retval;
> -
> -	retval = allocate_doorbell(qpd, q);
> -	if (retval)
> -		goto out_deallocate_sdma_queue;
> -
> -	dqm->asic_ops.init_sdma_vm(dqm, q, qpd);
> -	retval = mqd_mgr->init_mqd(mqd_mgr, &q->mqd, &q->mqd_mem_obj,
> -				&q->gart_mqd_addr, &q->properties);
> -	if (retval)
> -		goto out_deallocate_doorbell;
> -
> -	if (!q->properties.is_active)
> -		return 0;
> -
> -	retval = mqd_mgr->load_mqd(mqd_mgr, q->mqd, 0, 0, &q->properties,
> -				current->mm);
> -	if (retval)
> -		goto out_uninit_mqd;
> -
> -	return 0;
> -
> -out_uninit_mqd:
> -	mqd_mgr->uninit_mqd(mqd_mgr, q->mqd, q->mqd_mem_obj);
> -out_deallocate_doorbell:
> -	deallocate_doorbell(qpd, q);
> -out_deallocate_sdma_queue:
> -	deallocate_sdma_queue(dqm, q);
> -
> -	return retval;
> -}
> -
>   /*
>    * Device Queue Manager implementation for cp scheduler
>    */
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

  parent reply	other threads:[~2019-06-03 22:04 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-06-03 17:50 [PATCH 1/5] drm/amdkfd: Only initialize sdma vm for sdma queues Zeng, Oak
     [not found] ` <1559584248-12115-1-git-send-email-Oak.Zeng-5C7GfCeVMHo@public.gmane.org>
2019-06-03 17:51   ` [PATCH 2/5] drm/amdkfd: Only load sdma mqd when queue is active Zeng, Oak
2019-06-03 17:51   ` [PATCH 3/5] drm/amdkfd: Refactor create_queue_nocpsch Zeng, Oak
     [not found]     ` <1559584248-12115-3-git-send-email-Oak.Zeng-5C7GfCeVMHo@public.gmane.org>
2019-06-03 22:04       ` Kuehling, Felix [this message]
2019-06-03 17:51   ` [PATCH 4/5] drm/amdkfd: Fix a circular lock dependency Zeng, Oak
     [not found]     ` <1559584248-12115-4-git-send-email-Oak.Zeng-5C7GfCeVMHo@public.gmane.org>
2019-06-03 22:27       ` Kuehling, Felix
     [not found]         ` <5e896419-e14b-41b2-17fe-10eaee1f314c-5C7GfCeVMHo@public.gmane.org>
2019-06-04  1:08           ` Zeng, Oak
2019-06-03 17:51   ` [PATCH 5/5] drm/amdkfd: Fix sdma queue allocate race condition Zeng, Oak

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=10b7f416-df5e-6fa2-930f-71af79e67974@amd.com \
    --to=felix.kuehling-5c7gfcevmho@public.gmane.org \
    --cc=Oak.Zeng-5C7GfCeVMHo@public.gmane.org \
    --cc=amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox