static void uninitialize(struct kernel_queue *kq) { if (kq->queue->properties.type == KFD_QUEUE_TYPE_HIQ) kq->mqd->destroy_mqd(kq->mqd, kq->queue->mqd, KFD_PREEMPT_TYPE_WAVEFRONT_RESET, KFD_UNMAP_LATENCY_MS, kq->queue->pipe, kq->queue->queue); else if (kq->queue->properties.type == KFD_QUEUE_TYPE_DIQ) kfd_gtt_sa_free(kq->dev, kq->fence_mem_obj); kq->mqd->uninit_mqd(kq->mqd, kq->queue->mqd, kq->queue->mqd_mem_obj); kfd_gtt_sa_free(kq->dev, kq->rptr_mem); kfd_gtt_sa_free(kq->dev, kq->wptr_mem); kq->ops_asic_specific.uninitialize(kq); kfd_gtt_sa_free(kq->dev, kq->pq); kfd_release_kernel_doorbell(kq->dev, kq->queue->properties.doorbell_ptr); uninit_queue(kq->queue); }
static void uninitialize(struct kernel_queue *kq) { BUG_ON(!kq); if (kq->queue->properties.type == KFD_QUEUE_TYPE_HIQ) kq->mqd->destroy_mqd(kq->mqd, NULL, false, QUEUE_PREEMPT_DEFAULT_TIMEOUT_MS, kq->queue->pipe, kq->queue->queue); else if (kq->queue->properties.type == KFD_QUEUE_TYPE_DIQ) kfd_gtt_sa_free(kq->dev, kq->fence_mem_obj); kq->mqd->uninit_mqd(kq->mqd, kq->queue->mqd, kq->queue->mqd_mem_obj); kfd_gtt_sa_free(kq->dev, kq->rptr_mem); kfd_gtt_sa_free(kq->dev, kq->wptr_mem); kq->ops_asic_specific.uninitialize(kq); kfd_gtt_sa_free(kq->dev, kq->pq); kfd_release_kernel_doorbell(kq->dev, kq->queue->properties.doorbell_ptr); uninit_queue(kq->queue); }
static bool initialize(struct kernel_queue *kq, struct kfd_dev *dev, enum kfd_queue_type type, unsigned int queue_size) { struct queue_properties prop; int retval; union PM4_MES_TYPE_3_HEADER nop; BUG_ON(!kq || !dev); BUG_ON(type != KFD_QUEUE_TYPE_DIQ && type != KFD_QUEUE_TYPE_HIQ); pr_debug("amdkfd: In func %s initializing queue type %d size %d\n", __func__, KFD_QUEUE_TYPE_HIQ, queue_size); memset(&prop, 0, sizeof(prop)); memset(&nop, 0, sizeof(nop)); nop.opcode = IT_NOP; nop.type = PM4_TYPE_3; nop.u32all |= PM4_COUNT_ZERO; kq->dev = dev; kq->nop_packet = nop.u32all; switch (type) { case KFD_QUEUE_TYPE_DIQ: case KFD_QUEUE_TYPE_HIQ: kq->mqd = dev->dqm->ops.get_mqd_manager(dev->dqm, KFD_MQD_TYPE_HIQ); break; default: BUG(); break; } if (kq->mqd == NULL) return false; prop.doorbell_ptr = kfd_get_kernel_doorbell(dev, &prop.doorbell_off); if (prop.doorbell_ptr == NULL) { pr_err("amdkfd: error init doorbell"); goto err_get_kernel_doorbell; } retval = kfd_gtt_sa_allocate(dev, queue_size, &kq->pq); if (retval != 0) { pr_err("amdkfd: error init pq queues size (%d)\n", queue_size); goto err_pq_allocate_vidmem; } kq->pq_kernel_addr = kq->pq->cpu_ptr; kq->pq_gpu_addr = kq->pq->gpu_addr; retval = kq->ops_asic_specific.initialize(kq, dev, type, queue_size); if (retval == false) goto err_eop_allocate_vidmem; retval = kfd_gtt_sa_allocate(dev, sizeof(*kq->rptr_kernel), &kq->rptr_mem); if (retval != 0) goto err_rptr_allocate_vidmem; kq->rptr_kernel = kq->rptr_mem->cpu_ptr; kq->rptr_gpu_addr = kq->rptr_mem->gpu_addr; retval = kfd_gtt_sa_allocate(dev, sizeof(*kq->wptr_kernel), &kq->wptr_mem); if (retval != 0) goto err_wptr_allocate_vidmem; kq->wptr_kernel = kq->wptr_mem->cpu_ptr; kq->wptr_gpu_addr = kq->wptr_mem->gpu_addr; memset(kq->pq_kernel_addr, 0, queue_size); memset(kq->rptr_kernel, 0, sizeof(*kq->rptr_kernel)); memset(kq->wptr_kernel, 0, sizeof(*kq->wptr_kernel)); prop.queue_size = queue_size; prop.is_interop = false; prop.priority = 1; prop.queue_percent = 100; prop.type = type; prop.vmid = 0; prop.queue_address = kq->pq_gpu_addr; prop.read_ptr = (uint32_t *) kq->rptr_gpu_addr; prop.write_ptr = (uint32_t *) kq->wptr_gpu_addr; prop.eop_ring_buffer_address = kq->eop_gpu_addr; prop.eop_ring_buffer_size = PAGE_SIZE; if (init_queue(&kq->queue, &prop) != 0) goto err_init_queue; kq->queue->device = dev; kq->queue->process = kfd_get_process(current); retval = kq->mqd->init_mqd(kq->mqd, &kq->queue->mqd, &kq->queue->mqd_mem_obj, &kq->queue->gart_mqd_addr, &kq->queue->properties); if (retval != 0) goto err_init_mqd; /* assign HIQ to HQD */ if (type == KFD_QUEUE_TYPE_HIQ) { pr_debug("assigning hiq to hqd\n"); kq->queue->pipe = KFD_CIK_HIQ_PIPE; kq->queue->queue = KFD_CIK_HIQ_QUEUE; kq->mqd->load_mqd(kq->mqd, kq->queue->mqd, kq->queue->pipe, kq->queue->queue, NULL, NULL); } else { /* allocate fence for DIQ */ retval = kfd_gtt_sa_allocate(dev, sizeof(uint32_t), &kq->fence_mem_obj); if (retval != 0) goto err_alloc_fence; kq->fence_kernel_address = kq->fence_mem_obj->cpu_ptr; kq->fence_gpu_addr = kq->fence_mem_obj->gpu_addr; } print_queue(kq->queue); return true; err_alloc_fence: err_init_mqd: uninit_queue(kq->queue); err_init_queue: kfd_gtt_sa_free(dev, kq->wptr_mem); err_wptr_allocate_vidmem: kfd_gtt_sa_free(dev, kq->rptr_mem); err_rptr_allocate_vidmem: kfd_gtt_sa_free(dev, kq->eop_mem); err_eop_allocate_vidmem: kfd_gtt_sa_free(dev, kq->pq); err_pq_allocate_vidmem: kfd_release_kernel_doorbell(dev, prop.doorbell_ptr); err_get_kernel_doorbell: return false; }