Merge tag 'amd-drm-next-5.19-2022-04-22' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-5.19-2022-04-22:

amdgpu:
- SMU message documentation update
- Misc code cleanups
- Documenation updates
- PSP TA updates
- Runtime PM regression fix
- SR-IOV header cleanup
- Misc fixes

amdkfd:
- TLB flush fixes
- GWS fixes
- CRIU GWS support

radeon:
- Misc code cleanups

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220422150049.5859-1-alexander.deucher@amd.com

4 files changed, 53 insertions, 58 deletions

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 60438193c0c1..ee8b288dd8cc 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -1128,14 +1128,6 @@ err_pdd:
 	return ret;
 }
 
-static bool kfd_flush_tlb_after_unmap(struct kfd_dev *dev)
-{
-	return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2) ||
-		(KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 1) &&
-		dev->adev->sdma.instance[0].fw_version >= 18) ||
-		KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 0);
-}
-
 static int kfd_ioctl_map_memory_to_gpu(struct file *filep,
 					struct kfd_process *p, void *data)
 {
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 acf4f7975850..198672264492 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -130,19 +130,33 @@ void program_sh_mem_settings(struct device_queue_manager *dqm,
 }
 
 static void increment_queue_count(struct device_queue_manager *dqm,
-			enum kfd_queue_type type)
+				  struct qcm_process_device *qpd,
+				  struct queue *q)
 {
 	dqm->active_queue_count++;
-	if (type == KFD_QUEUE_TYPE_COMPUTE || type == KFD_QUEUE_TYPE_DIQ)
+	if (q->properties.type == KFD_QUEUE_TYPE_COMPUTE ||
+	    q->properties.type == KFD_QUEUE_TYPE_DIQ)
 		dqm->active_cp_queue_count++;
+
+	if (q->properties.is_gws) {
+		dqm->gws_queue_count++;
+		qpd->mapped_gws_queue = true;
+	}
 }
 
 static void decrement_queue_count(struct device_queue_manager *dqm,
-			enum kfd_queue_type type)
+				  struct qcm_process_device *qpd,
+				  struct queue *q)
 {
 	dqm->active_queue_count--;
-	if (type == KFD_QUEUE_TYPE_COMPUTE || type == KFD_QUEUE_TYPE_DIQ)
+	if (q->properties.type == KFD_QUEUE_TYPE_COMPUTE ||
+	    q->properties.type == KFD_QUEUE_TYPE_DIQ)
 		dqm->active_cp_queue_count--;
+
+	if (q->properties.is_gws) {
+		dqm->gws_queue_count--;
+		qpd->mapped_gws_queue = false;
+	}
 }
 
 /*
@@ -412,7 +426,7 @@ add_queue_to_list:
 	list_add(&q->list, &qpd->queues_list);
 	qpd->queue_count++;
 	if (q->properties.is_active)
-		increment_queue_count(dqm, q->properties.type);
+		increment_queue_count(dqm, qpd, q);
 
 	/*
 	 * Unconditionally increment this counter, regardless of the queue's
@@ -601,13 +615,8 @@ static int destroy_queue_nocpsch_locked(struct device_queue_manager *dqm,
 		deallocate_vmid(dqm, qpd, q);
 	}
 	qpd->queue_count--;
-	if (q->properties.is_active) {
-		decrement_queue_count(dqm, q->properties.type);
-		if (q->properties.is_gws) {
-			dqm->gws_queue_count--;
-			qpd->mapped_gws_queue = false;
-		}
-	}
+	if (q->properties.is_active)
+		decrement_queue_count(dqm, qpd, q);
 
 	return retval;
 }
@@ -700,12 +709,11 @@ static int update_queue(struct device_queue_manager *dqm, struct queue *q,
 	 * dqm->active_queue_count to determine whether a new runlist must be
 	 * uploaded.
 	 */
-	if (q->properties.is_active && !prev_active)
-		increment_queue_count(dqm, q->properties.type);
-	else if (!q->properties.is_active && prev_active)
-		decrement_queue_count(dqm, q->properties.type);
-
-	if (q->gws && !q->properties.is_gws) {
+	if (q->properties.is_active && !prev_active) {
+		increment_queue_count(dqm, &pdd->qpd, q);
+	} else if (!q->properties.is_active && prev_active) {
+		decrement_queue_count(dqm, &pdd->qpd, q);
+	} else if (q->gws && !q->properties.is_gws) {
 		if (q->properties.is_active) {
 			dqm->gws_queue_count++;
 			pdd->qpd.mapped_gws_queue = true;
@@ -767,11 +775,7 @@ static int evict_process_queues_nocpsch(struct device_queue_manager *dqm,
 		mqd_mgr = dqm->mqd_mgrs[get_mqd_type_from_queue_type(
 				q->properties.type)];
 		q->properties.is_active = false;
-		decrement_queue_count(dqm, q->properties.type);
-		if (q->properties.is_gws) {
-			dqm->gws_queue_count--;
-			qpd->mapped_gws_queue = false;
-		}
+		decrement_queue_count(dqm, qpd, q);
 
 		if (WARN_ONCE(!dqm->sched_running, "Evict when stopped\n"))
 			continue;
@@ -817,7 +821,7 @@ static int evict_process_queues_cpsch(struct device_queue_manager *dqm,
 			continue;
 
 		q->properties.is_active = false;
-		decrement_queue_count(dqm, q->properties.type);
+		decrement_queue_count(dqm, qpd, q);
 	}
 	pdd->last_evict_timestamp = get_jiffies_64();
 	retval = execute_queues_cpsch(dqm,
@@ -888,11 +892,7 @@ static int restore_process_queues_nocpsch(struct device_queue_manager *dqm,
 		mqd_mgr = dqm->mqd_mgrs[get_mqd_type_from_queue_type(
 				q->properties.type)];
 		q->properties.is_active = true;
-		increment_queue_count(dqm, q->properties.type);
-		if (q->properties.is_gws) {
-			dqm->gws_queue_count++;
-			qpd->mapped_gws_queue = true;
-		}
+		increment_queue_count(dqm, qpd, q);
 
 		if (WARN_ONCE(!dqm->sched_running, "Restore when stopped\n"))
 			continue;
@@ -950,7 +950,7 @@ static int restore_process_queues_cpsch(struct device_queue_manager *dqm,
 			continue;
 
 		q->properties.is_active = true;
-		increment_queue_count(dqm, q->properties.type);
+		increment_queue_count(dqm, &pdd->qpd, q);
 	}
 	retval = execute_queues_cpsch(dqm,
 				KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0);
@@ -1378,7 +1378,7 @@ static int create_kernel_queue_cpsch(struct device_queue_manager *dqm,
 			dqm->total_queue_count);
 
 	list_add(&kq->list, &qpd->priv_queue_list);
-	increment_queue_count(dqm, kq->queue->properties.type);
+	increment_queue_count(dqm, qpd, kq->queue);
 	qpd->is_debug = true;
 	execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0);
 	dqm_unlock(dqm);
@@ -1392,7 +1392,7 @@ static void destroy_kernel_queue_cpsch(struct device_queue_manager *dqm,
 {
 	dqm_lock(dqm);
 	list_del(&kq->list);
-	decrement_queue_count(dqm, kq->queue->properties.type);
+	decrement_queue_count(dqm, qpd, kq->queue);
 	qpd->is_debug = false;
 	execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES, 0);
 	/*
@@ -1467,7 +1467,7 @@ static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q,
 	qpd->queue_count++;
 
 	if (q->properties.is_active) {
-		increment_queue_count(dqm, q->properties.type);
+		increment_queue_count(dqm, qpd, q);
 
 		execute_queues_cpsch(dqm,
 				KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0);
@@ -1683,15 +1683,11 @@ static int destroy_queue_cpsch(struct device_queue_manager *dqm,
 	list_del(&q->list);
 	qpd->queue_count--;
 	if (q->properties.is_active) {
-		decrement_queue_count(dqm, q->properties.type);
+		decrement_queue_count(dqm, qpd, q);
 		retval = execute_queues_cpsch(dqm,
 				KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0);
 		if (retval == -ETIME)
 			qpd->reset_wavefronts = true;
-		if (q->properties.is_gws) {
-			dqm->gws_queue_count--;
-			qpd->mapped_gws_queue = false;
-		}
 	}
 
 	/*
@@ -1932,7 +1928,7 @@ static int process_termination_cpsch(struct device_queue_manager *dqm,
 	/* Clean all kernel queues */
 	list_for_each_entry_safe(kq, kq_next, &qpd->priv_queue_list, list) {
 		list_del(&kq->list);
-		decrement_queue_count(dqm, kq->queue->properties.type);
+		decrement_queue_count(dqm, qpd, kq->queue);
 		qpd->is_debug = false;
 		dqm->total_queue_count--;
 		filter = KFD_UNMAP_QUEUES_FILTER_ALL_QUEUES;
@@ -1945,13 +1941,8 @@ static int process_termination_cpsch(struct device_queue_manager *dqm,
 		else if (q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI)
 			deallocate_sdma_queue(dqm, q);
 
-		if (q->properties.is_active) {
-			decrement_queue_count(dqm, q->properties.type);
-			if (q->properties.is_gws) {
-				dqm->gws_queue_count--;
-				qpd->mapped_gws_queue = false;
-			}
-		}
+		if (q->properties.is_active)
+			decrement_queue_count(dqm, qpd, q);
 
 		dqm->total_queue_count--;
 	}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 644f616b776f..bfb3b5c288cb 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -1106,7 +1106,7 @@ struct kfd_criu_queue_priv_data {
 	uint32_t priority;
 	uint32_t q_percent;
 	uint32_t doorbell_id;
-	uint32_t is_gws;
+	uint32_t gws;
 	uint32_t sdma_id;
 	uint32_t eop_ring_buffer_size;
 	uint32_t ctx_save_restore_area_size;
@@ -1329,6 +1329,14 @@ void kfd_signal_poison_consumed_event(struct kfd_dev *dev, u32 pasid);
 
 void kfd_flush_tlb(struct kfd_process_device *pdd, enum TLB_FLUSH_TYPE type);
 
+static inline bool kfd_flush_tlb_after_unmap(struct kfd_dev *dev)
+{
+	return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2) ||
+	       (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 1) &&
+	       dev->adev->sdma.instance[0].fw_version >= 18) ||
+	       KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 0);
+}
+
 bool kfd_is_locked(void);
 
 /* Compute profile */
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
index 6eca9509f2e3..4f58e671d39b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
@@ -636,6 +636,8 @@ static int criu_checkpoint_queue(struct kfd_process_device *pdd,
 	q_data->ctx_save_restore_area_size =
 		q->properties.ctx_save_restore_area_size;
 
+	q_data->gws = !!q->gws;
+
 	ret = pqm_checkpoint_mqd(&pdd->process->pqm, q->properties.queue_id, mqd, ctl_stack);
 	if (ret) {
 		pr_err("Failed checkpoint queue_mqd (%d)\n", ret);
@@ -743,7 +745,6 @@ static void set_queue_properties_from_criu(struct queue_properties *qp,
 					  struct kfd_criu_queue_priv_data *q_data)
 {
 	qp->is_interop = false;
-	qp->is_gws = q_data->is_gws;
 	qp->queue_percent = q_data->q_percent;
 	qp->priority = q_data->priority;
 	qp->queue_address = q_data->q_address;
@@ -826,12 +827,15 @@ int kfd_criu_restore_queue(struct kfd_process *p,
 				NULL);
 	if (ret) {
 		pr_err("Failed to create new queue err:%d\n", ret);
-		ret = -EINVAL;
+		goto exit;
 	}
 
+	if (q_data->gws)
+		ret = pqm_set_gws(&p->pqm, q_data->q_id, pdd->dev->gws);
+
 exit:
 	if (ret)
-		pr_err("Failed to create queue (%d)\n", ret);
+		pr_err("Failed to restore queue (%d)\n", ret);
 	else
 		pr_debug("Queue id %d was restored successfully\n", queue_id);