[thirdparty/linux.git] / drivers / vdpa / vdpa_sim / vdpa_sim.c

// SPDX-License-Identifier: GPL-2.0-only
/*
 * VDPA networking device simulator.
 *
 * Copyright (c) 2020, Red Hat Inc. All rights reserved.
 *     Author: Jason Wang <jasowang@redhat.com>
 *
 */

#include <linux/init.h>
#include <linux/module.h>
#include <linux/device.h>
#include <linux/kernel.h>
#include <linux/fs.h>
#include <linux/poll.h>
#include <linux/slab.h>
#include <linux/sched.h>
#include <linux/wait.h>
#include <linux/uuid.h>
#include <linux/iommu.h>
#include <linux/dma-mapping.h>
#include <linux/sysfs.h>
#include <linux/file.h>
#include <linux/etherdevice.h>
#include <linux/vringh.h>
#include <linux/vdpa.h>
#include <linux/vhost_iotlb.h>
#include <uapi/linux/virtio_config.h>
#include <uapi/linux/virtio_net.h>

#define DRV_VERSION  "0.1"
#define DRV_AUTHOR   "Jason Wang <jasowang@redhat.com>"
#define DRV_DESC     "vDPA Device Simulator"
#define DRV_LICENSE  "GPL v2"

struct vdpasim_virtqueue {
	struct vringh vring;
	struct vringh_kiov iov;
	unsigned short head;
	bool ready;
	u64 desc_addr;
	u64 device_addr;
	u64 driver_addr;
	u32 num;
	void *private;
	irqreturn_t (*cb)(void *data);
};

#define VDPASIM_QUEUE_ALIGN PAGE_SIZE
#define VDPASIM_QUEUE_MAX 256
#define VDPASIM_DEVICE_ID 0x1
#define VDPASIM_VENDOR_ID 0
#define VDPASIM_VQ_NUM 0x2
#define VDPASIM_NAME "vdpasim-netdev"

static u64 vdpasim_features = (1ULL << VIRTIO_F_ANY_LAYOUT) |
			      (1ULL << VIRTIO_F_VERSION_1)  |
			      (1ULL << VIRTIO_F_IOMMU_PLATFORM);

/* State of each vdpasim device */
struct vdpasim {
	struct vdpa_device vdpa;
	struct vdpasim_virtqueue vqs[2];
	struct work_struct work;
	/* spinlock to synchronize virtqueue state */
	spinlock_t lock;
	struct virtio_net_config config;
	struct vhost_iotlb *iommu;
	void *buffer;
	u32 status;
	u32 generation;
	u64 features;
};

static struct vdpasim *vdpasim_dev;

static struct vdpasim *vdpa_to_sim(struct vdpa_device *vdpa)
{
	return container_of(vdpa, struct vdpasim, vdpa);
}

static struct vdpasim *dev_to_sim(struct device *dev)
{
	struct vdpa_device *vdpa = dev_to_vdpa(dev);

	return vdpa_to_sim(vdpa);
}

static void vdpasim_queue_ready(struct vdpasim *vdpasim, unsigned int idx)
{
	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];

	vringh_init_iotlb(&vq->vring, vdpasim_features,
			  VDPASIM_QUEUE_MAX, false,
			  (struct vring_desc *)(uintptr_t)vq->desc_addr,
			  (struct vring_avail *)
			  (uintptr_t)vq->driver_addr,
			  (struct vring_used *)
			  (uintptr_t)vq->device_addr);
}

static void vdpasim_vq_reset(struct vdpasim_virtqueue *vq)
{
	vq->ready = 0;
	vq->desc_addr = 0;
	vq->driver_addr = 0;
	vq->device_addr = 0;
	vq->cb = NULL;
	vq->private = NULL;
	vringh_init_iotlb(&vq->vring, vdpasim_features, VDPASIM_QUEUE_MAX,
			  false, NULL, NULL, NULL);
}

static void vdpasim_reset(struct vdpasim *vdpasim)
{
	int i;

	for (i = 0; i < VDPASIM_VQ_NUM; i++)
		vdpasim_vq_reset(&vdpasim->vqs[i]);

	vhost_iotlb_reset(vdpasim->iommu);

	vdpasim->features = 0;
	vdpasim->status = 0;
	++vdpasim->generation;
}

static void vdpasim_work(struct work_struct *work)
{
	struct vdpasim *vdpasim = container_of(work, struct
						 vdpasim, work);
	struct vdpasim_virtqueue *txq = &vdpasim->vqs[1];
	struct vdpasim_virtqueue *rxq = &vdpasim->vqs[0];
	size_t read, write, total_write;
	int err;
	int pkts = 0;

	spin_lock(&vdpasim->lock);

	if (!(vdpasim->status & VIRTIO_CONFIG_S_DRIVER_OK))
		goto out;

	if (!txq->ready || !rxq->ready)
		goto out;

	while (true) {
		total_write = 0;
		err = vringh_getdesc_iotlb(&txq->vring, &txq->iov, NULL,
					   &txq->head, GFP_ATOMIC);
		if (err <= 0)
			break;

		err = vringh_getdesc_iotlb(&rxq->vring, NULL, &rxq->iov,
					   &rxq->head, GFP_ATOMIC);
		if (err <= 0) {
			vringh_complete_iotlb(&txq->vring, txq->head, 0);
			break;
		}

		while (true) {
			read = vringh_iov_pull_iotlb(&txq->vring, &txq->iov,
						     vdpasim->buffer,
						     PAGE_SIZE);
			if (read <= 0)
				break;

			write = vringh_iov_push_iotlb(&rxq->vring, &rxq->iov,
						      vdpasim->buffer, read);
			if (write <= 0)
				break;

			total_write += write;
		}

		/* Make sure data is wrote before advancing index */
		smp_wmb();

		vringh_complete_iotlb(&txq->vring, txq->head, 0);
		vringh_complete_iotlb(&rxq->vring, rxq->head, total_write);

		/* Make sure used is visible before rasing the interrupt. */
		smp_wmb();

		local_bh_disable();
		if (txq->cb)
			txq->cb(txq->private);
		if (rxq->cb)
			rxq->cb(rxq->private);
		local_bh_enable();

		if (++pkts > 4) {
			schedule_work(&vdpasim->work);
			goto out;
		}
	}

out:
	spin_unlock(&vdpasim->lock);
}

static int dir_to_perm(enum dma_data_direction dir)
{
	int perm = -EFAULT;

	switch (dir) {
	case DMA_FROM_DEVICE:
		perm = VHOST_MAP_WO;
		break;
	case DMA_TO_DEVICE:
		perm = VHOST_MAP_RO;
		break;
	case DMA_BIDIRECTIONAL:
		perm = VHOST_MAP_RW;
		break;
	default:
		break;
	}

	return perm;
}

static dma_addr_t vdpasim_map_page(struct device *dev, struct page *page,
				   unsigned long offset, size_t size,
				   enum dma_data_direction dir,
				   unsigned long attrs)
{
	struct vdpasim *vdpasim = dev_to_sim(dev);
	struct vhost_iotlb *iommu = vdpasim->iommu;
	u64 pa = (page_to_pfn(page) << PAGE_SHIFT) + offset;
	int ret, perm = dir_to_perm(dir);

	if (perm < 0)
		return DMA_MAPPING_ERROR;

	/* For simplicity, use identical mapping to avoid e.g iova
	 * allocator.
	 */
	ret = vhost_iotlb_add_range(iommu, pa, pa + size - 1,
				    pa, dir_to_perm(dir));
	if (ret)
		return DMA_MAPPING_ERROR;

	return (dma_addr_t)(pa);
}

static void vdpasim_unmap_page(struct device *dev, dma_addr_t dma_addr,
			       size_t size, enum dma_data_direction dir,
			       unsigned long attrs)
{
	struct vdpasim *vdpasim = dev_to_sim(dev);
	struct vhost_iotlb *iommu = vdpasim->iommu;

	vhost_iotlb_del_range(iommu, (u64)dma_addr,
			      (u64)dma_addr + size - 1);
}

static void *vdpasim_alloc_coherent(struct device *dev, size_t size,
				    dma_addr_t *dma_addr, gfp_t flag,
				    unsigned long attrs)
{
	struct vdpasim *vdpasim = dev_to_sim(dev);
	struct vhost_iotlb *iommu = vdpasim->iommu;
	void *addr = kmalloc(size, flag);
	int ret;

	if (!addr)
		*dma_addr = DMA_MAPPING_ERROR;
	else {
		u64 pa = virt_to_phys(addr);

		ret = vhost_iotlb_add_range(iommu, (u64)pa,
					    (u64)pa + size - 1,
					    pa, VHOST_MAP_RW);
		if (ret) {
			*dma_addr = DMA_MAPPING_ERROR;
			kfree(addr);
			addr = NULL;
		} else
			*dma_addr = (dma_addr_t)pa;
	}

	return addr;
}

static void vdpasim_free_coherent(struct device *dev, size_t size,
				  void *vaddr, dma_addr_t dma_addr,
				  unsigned long attrs)
{
	struct vdpasim *vdpasim = dev_to_sim(dev);
	struct vhost_iotlb *iommu = vdpasim->iommu;

	vhost_iotlb_del_range(iommu, (u64)dma_addr,
			      (u64)dma_addr + size - 1);
	kfree(phys_to_virt((uintptr_t)dma_addr));
}

static const struct dma_map_ops vdpasim_dma_ops = {
	.map_page = vdpasim_map_page,
	.unmap_page = vdpasim_unmap_page,
	.alloc = vdpasim_alloc_coherent,
	.free = vdpasim_free_coherent,
};

static const struct vdpa_config_ops vdpasim_net_config_ops;

static struct vdpasim *vdpasim_create(void)
{
	struct virtio_net_config *config;
	struct vdpasim *vdpasim;
	struct device *dev;
	int ret = -ENOMEM;

	vdpasim = vdpa_alloc_device(struct vdpasim, vdpa, NULL,
				    &vdpasim_net_config_ops);
	if (!vdpasim)
		goto err_alloc;

	INIT_WORK(&vdpasim->work, vdpasim_work);
	spin_lock_init(&vdpasim->lock);

	dev = &vdpasim->vdpa.dev;
	dev->coherent_dma_mask = DMA_BIT_MASK(64);
	set_dma_ops(dev, &vdpasim_dma_ops);

	vdpasim->iommu = vhost_iotlb_alloc(2048, 0);
	if (!vdpasim->iommu)
		goto err_iommu;

	vdpasim->buffer = kmalloc(PAGE_SIZE, GFP_KERNEL);
	if (!vdpasim->buffer)
		goto err_iommu;

	config = &vdpasim->config;
	config->mtu = 1500;
	config->status = VIRTIO_NET_S_LINK_UP;
	eth_random_addr(config->mac);

	vringh_set_iotlb(&vdpasim->vqs[0].vring, vdpasim->iommu);
	vringh_set_iotlb(&vdpasim->vqs[1].vring, vdpasim->iommu);

	vdpasim->vdpa.dma_dev = dev;
	ret = vdpa_register_device(&vdpasim->vdpa);
	if (ret)
		goto err_iommu;

	return vdpasim;

err_iommu:
	put_device(dev);
err_alloc:
	return ERR_PTR(ret);
}

static int vdpasim_set_vq_address(struct vdpa_device *vdpa, u16 idx,
				  u64 desc_area, u64 driver_area,
				  u64 device_area)
{
	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];

	vq->desc_addr = desc_area;
	vq->driver_addr = driver_area;
	vq->device_addr = device_area;

	return 0;
}

static void vdpasim_set_vq_num(struct vdpa_device *vdpa, u16 idx, u32 num)
{
	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];

	vq->num = num;
}

static void vdpasim_kick_vq(struct vdpa_device *vdpa, u16 idx)
{
	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];

	if (vq->ready)
		schedule_work(&vdpasim->work);
}

static void vdpasim_set_vq_cb(struct vdpa_device *vdpa, u16 idx,
			      struct vdpa_callback *cb)
{
	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];

	vq->cb = cb->callback;
	vq->private = cb->private;
}

static void vdpasim_set_vq_ready(struct vdpa_device *vdpa, u16 idx, bool ready)
{
	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];

	spin_lock(&vdpasim->lock);
	vq->ready = ready;
	if (vq->ready)
		vdpasim_queue_ready(vdpasim, idx);
	spin_unlock(&vdpasim->lock);
}

static bool vdpasim_get_vq_ready(struct vdpa_device *vdpa, u16 idx)
{
	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];

	return vq->ready;
}

static int vdpasim_set_vq_state(struct vdpa_device *vdpa, u16 idx, u64 state)
{
	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];
	struct vringh *vrh = &vq->vring;

	spin_lock(&vdpasim->lock);
	vrh->last_avail_idx = state;
	spin_unlock(&vdpasim->lock);

	return 0;
}

static u64 vdpasim_get_vq_state(struct vdpa_device *vdpa, u16 idx)
{
	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];
	struct vringh *vrh = &vq->vring;

	return vrh->last_avail_idx;
}

static u32 vdpasim_get_vq_align(struct vdpa_device *vdpa)
{
	return VDPASIM_QUEUE_ALIGN;
}

static u64 vdpasim_get_features(struct vdpa_device *vdpa)
{
	return vdpasim_features;
}

static int vdpasim_set_features(struct vdpa_device *vdpa, u64 features)
{
	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);

	/* DMA mapping must be done by driver */
	if (!(features & (1ULL << VIRTIO_F_IOMMU_PLATFORM)))
		return -EINVAL;

	vdpasim->features = features & vdpasim_features;

	return 0;
}

static void vdpasim_set_config_cb(struct vdpa_device *vdpa,
				  struct vdpa_callback *cb)
{
	/* We don't support config interrupt */
}

static u16 vdpasim_get_vq_num_max(struct vdpa_device *vdpa)
{
	return VDPASIM_QUEUE_MAX;
}

static u32 vdpasim_get_device_id(struct vdpa_device *vdpa)
{
	return VDPASIM_DEVICE_ID;
}

static u32 vdpasim_get_vendor_id(struct vdpa_device *vdpa)
{
	return VDPASIM_VENDOR_ID;
}

static u8 vdpasim_get_status(struct vdpa_device *vdpa)
{
	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
	u8 status;

	spin_lock(&vdpasim->lock);
	status = vdpasim->status;
	spin_unlock(&vdpasim->lock);

	return status;
}

static void vdpasim_set_status(struct vdpa_device *vdpa, u8 status)
{
	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);

	spin_lock(&vdpasim->lock);
	vdpasim->status = status;
	if (status == 0)
		vdpasim_reset(vdpasim);
	spin_unlock(&vdpasim->lock);
}

static void vdpasim_get_config(struct vdpa_device *vdpa, unsigned int offset,
			     void *buf, unsigned int len)
{
	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);

	if (offset + len < sizeof(struct virtio_net_config))
		memcpy(buf, &vdpasim->config + offset, len);
}

static void vdpasim_set_config(struct vdpa_device *vdpa, unsigned int offset,
			     const void *buf, unsigned int len)
{
	/* No writable config supportted by vdpasim */
}

static u32 vdpasim_get_generation(struct vdpa_device *vdpa)
{
	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);

	return vdpasim->generation;
}

static int vdpasim_set_map(struct vdpa_device *vdpa,
			   struct vhost_iotlb *iotlb)
{
	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
	struct vhost_iotlb_map *map;
	u64 start = 0ULL, last = 0ULL - 1;
	int ret;

	vhost_iotlb_reset(vdpasim->iommu);

	for (map = vhost_iotlb_itree_first(iotlb, start, last); map;
	     map = vhost_iotlb_itree_next(map, start, last)) {
		ret = vhost_iotlb_add_range(vdpasim->iommu, map->start,
					    map->last, map->addr, map->perm);
		if (ret)
			goto err;
	}
	return 0;

err:
	vhost_iotlb_reset(vdpasim->iommu);
	return ret;
}

static int vdpasim_dma_map(struct vdpa_device *vdpa, u64 iova, u64 size,
			   u64 pa, u32 perm)
{
	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);

	return vhost_iotlb_add_range(vdpasim->iommu, iova,
				     iova + size - 1, pa, perm);
}

static int vdpasim_dma_unmap(struct vdpa_device *vdpa, u64 iova, u64 size)
{
	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);

	vhost_iotlb_del_range(vdpasim->iommu, iova, iova + size - 1);

	return 0;
}

static void vdpasim_free(struct vdpa_device *vdpa)
{
	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);

	cancel_work_sync(&vdpasim->work);
	kfree(vdpasim->buffer);
	if (vdpasim->iommu)
		vhost_iotlb_free(vdpasim->iommu);
}

static const struct vdpa_config_ops vdpasim_net_config_ops = {
	.set_vq_address         = vdpasim_set_vq_address,
	.set_vq_num             = vdpasim_set_vq_num,
	.kick_vq                = vdpasim_kick_vq,
	.set_vq_cb              = vdpasim_set_vq_cb,
	.set_vq_ready           = vdpasim_set_vq_ready,
	.get_vq_ready           = vdpasim_get_vq_ready,
	.set_vq_state           = vdpasim_set_vq_state,
	.get_vq_state           = vdpasim_get_vq_state,
	.get_vq_align           = vdpasim_get_vq_align,
	.get_features           = vdpasim_get_features,
	.set_features           = vdpasim_set_features,
	.set_config_cb          = vdpasim_set_config_cb,
	.get_vq_num_max         = vdpasim_get_vq_num_max,
	.get_device_id          = vdpasim_get_device_id,
	.get_vendor_id          = vdpasim_get_vendor_id,
	.get_status             = vdpasim_get_status,
	.set_status             = vdpasim_set_status,
	.get_config             = vdpasim_get_config,
	.set_config             = vdpasim_set_config,
	.get_generation         = vdpasim_get_generation,
	.set_map                = vdpasim_set_map,
	.dma_map                = vdpasim_dma_map,
	.dma_unmap              = vdpasim_dma_unmap,
	.free                   = vdpasim_free,
};

static int __init vdpasim_dev_init(void)
{
	vdpasim_dev = vdpasim_create();

	if (!IS_ERR(vdpasim_dev))
		return 0;

	return PTR_ERR(vdpasim_dev);
}

static void __exit vdpasim_dev_exit(void)
{
	struct vdpa_device *vdpa = &vdpasim_dev->vdpa;

	vdpa_unregister_device(vdpa);
}

module_init(vdpasim_dev_init)
module_exit(vdpasim_dev_exit)

MODULE_VERSION(DRV_VERSION);
MODULE_LICENSE(DRV_LICENSE);
MODULE_AUTHOR(DRV_AUTHOR);
MODULE_DESCRIPTION(DRV_DESC);
Commit	Line	Data
2c53d0f6 JW	1	// SPDX-License-Identifier: GPL-2.0-only
	2	/*
	3	* VDPA networking device simulator.
	4	*
	5	* Copyright (c) 2020, Red Hat Inc. All rights reserved.
	6	* Author: Jason Wang <jasowang@redhat.com>
	7	*
	8	*/
	9
	10	#include <linux/init.h>
	11	#include <linux/module.h>
	12	#include <linux/device.h>
	13	#include <linux/kernel.h>
	14	#include <linux/fs.h>
	15	#include <linux/poll.h>
	16	#include <linux/slab.h>
	17	#include <linux/sched.h>
	18	#include <linux/wait.h>
	19	#include <linux/uuid.h>
	20	#include <linux/iommu.h>
	21	#include <linux/dma-mapping.h>
	22	#include <linux/sysfs.h>
	23	#include <linux/file.h>
	24	#include <linux/etherdevice.h>
	25	#include <linux/vringh.h>
	26	#include <linux/vdpa.h>
	27	#include <linux/vhost_iotlb.h>
	28	#include <uapi/linux/virtio_config.h>
	29	#include <uapi/linux/virtio_net.h>
	30
	31	#define DRV_VERSION "0.1"
	32	#define DRV_AUTHOR "Jason Wang <jasowang@redhat.com>"
	33	#define DRV_DESC "vDPA Device Simulator"
	34	#define DRV_LICENSE "GPL v2"
	35
	36	struct vdpasim_virtqueue {
	37	struct vringh vring;
	38	struct vringh_kiov iov;
	39	unsigned short head;
	40	bool ready;
	41	u64 desc_addr;
	42	u64 device_addr;
	43	u64 driver_addr;
	44	u32 num;
	45	void *private;
	46	irqreturn_t (cb)(void data);
	47	};
	48
	49	#define VDPASIM_QUEUE_ALIGN PAGE_SIZE
	50	#define VDPASIM_QUEUE_MAX 256
	51	#define VDPASIM_DEVICE_ID 0x1
	52	#define VDPASIM_VENDOR_ID 0
	53	#define VDPASIM_VQ_NUM 0x2
	54	#define VDPASIM_NAME "vdpasim-netdev"
	55
	56	static u64 vdpasim_features = (1ULL << VIRTIO_F_ANY_LAYOUT) \|
	57	(1ULL << VIRTIO_F_VERSION_1) \|
	58	(1ULL << VIRTIO_F_IOMMU_PLATFORM);
	59
	60	/* State of each vdpasim device */
	61	struct vdpasim {
	62	struct vdpa_device vdpa;
	63	struct vdpasim_virtqueue vqs[2];
	64	struct work_struct work;
65	/* spinlock to synchronize virtqueue state */
66	spinlock_t lock;
67	struct virtio_net_config config;
68	struct vhost_iotlb *iommu;
69	void *buffer;
70	u32 status;
71	u32 generation;
72	u64 features;
73	};
74
75	static struct vdpasim *vdpasim_dev;
76
77	static struct vdpasim vdpa_to_sim(struct vdpa_device vdpa)
78	{
79	return container_of(vdpa, struct vdpasim, vdpa);
80	}
81
82	static struct vdpasim dev_to_sim(struct device dev)
83	{
84	struct vdpa_device *vdpa = dev_to_vdpa(dev);
85
86	return vdpa_to_sim(vdpa);
87	}
88
89	static void vdpasim_queue_ready(struct vdpasim *vdpasim, unsigned int idx)
90	{
91	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];
2c53d0f6	92
0f8e3823 Y	93	vringh_init_iotlb(&vq->vring, vdpasim_features,
	94	VDPASIM_QUEUE_MAX, false,
	95	(struct vring_desc *)(uintptr_t)vq->desc_addr,
	96	(struct vring_avail *)
	97	(uintptr_t)vq->driver_addr,
	98	(struct vring_used *)
	99	(uintptr_t)vq->device_addr);
2c53d0f6 JW	100	}
	101
	102	static void vdpasim_vq_reset(struct vdpasim_virtqueue *vq)
	103	{
	104	vq->ready = 0;
	105	vq->desc_addr = 0;
	106	vq->driver_addr = 0;
	107	vq->device_addr = 0;
	108	vq->cb = NULL;
	109	vq->private = NULL;
	110	vringh_init_iotlb(&vq->vring, vdpasim_features, VDPASIM_QUEUE_MAX,
	111	false, NULL, NULL, NULL);
	112	}
	113
	114	static void vdpasim_reset(struct vdpasim *vdpasim)
	115	{
	116	int i;
	117
	118	for (i = 0; i < VDPASIM_VQ_NUM; i++)
	119	vdpasim_vq_reset(&vdpasim->vqs[i]);
	120
	121	vhost_iotlb_reset(vdpasim->iommu);
	122
	123	vdpasim->features = 0;
	124	vdpasim->status = 0;
	125	++vdpasim->generation;
	126	}
	127
	128	static void vdpasim_work(struct work_struct *work)
	129	{
	130	struct vdpasim *vdpasim = container_of(work, struct
	131	vdpasim, work);
	132	struct vdpasim_virtqueue *txq = &vdpasim->vqs[1];
	133	struct vdpasim_virtqueue *rxq = &vdpasim->vqs[0];
	134	size_t read, write, total_write;
	135	int err;
	136	int pkts = 0;
	137
	138	spin_lock(&vdpasim->lock);
	139
	140	if (!(vdpasim->status & VIRTIO_CONFIG_S_DRIVER_OK))
	141	goto out;
	142
	143	if (!txq->ready \|\| !rxq->ready)
	144	goto out;
	145
	146	while (true) {
	147	total_write = 0;
	148	err = vringh_getdesc_iotlb(&txq->vring, &txq->iov, NULL,
	149	&txq->head, GFP_ATOMIC);
	150	if (err <= 0)
	151	break;
	152
	153	err = vringh_getdesc_iotlb(&rxq->vring, NULL, &rxq->iov,
	154	&rxq->head, GFP_ATOMIC);
	155	if (err <= 0) {
	156	vringh_complete_iotlb(&txq->vring, txq->head, 0);
	157	break;
	158	}
	159
	160	while (true) {
	161	read = vringh_iov_pull_iotlb(&txq->vring, &txq->iov,
	162	vdpasim->buffer,
	163	PAGE_SIZE);
164	if (read <= 0)
165	break;
166
167	write = vringh_iov_push_iotlb(&rxq->vring, &rxq->iov,
168	vdpasim->buffer, read);
169	if (write <= 0)
170	break;
171
172	total_write += write;
173	}
174
175	/* Make sure data is wrote before advancing index */
176	smp_wmb();
177
178	vringh_complete_iotlb(&txq->vring, txq->head, 0);
179	vringh_complete_iotlb(&rxq->vring, rxq->head, total_write);
180
181	/* Make sure used is visible before rasing the interrupt. */
182	smp_wmb();
183
184	local_bh_disable();
185	if (txq->cb)
186	txq->cb(txq->private);
187	if (rxq->cb)
188	rxq->cb(rxq->private);
189	local_bh_enable();
190
191	if (++pkts > 4) {
192	schedule_work(&vdpasim->work);
193	goto out;
194	}
195	}
196
197	out:
198	spin_unlock(&vdpasim->lock);
199	}
200
201	static int dir_to_perm(enum dma_data_direction dir)
202	{
203	int perm = -EFAULT;
204
205	switch (dir) {
206	case DMA_FROM_DEVICE:
207	perm = VHOST_MAP_WO;
208	break;
209	case DMA_TO_DEVICE:
210	perm = VHOST_MAP_RO;
211	break;
212	case DMA_BIDIRECTIONAL:
213	perm = VHOST_MAP_RW;
214	break;
215	default:
216	break;
217	}
218
219	return perm;
220	}
221
222	static dma_addr_t vdpasim_map_page(struct device dev, struct page page,
223	unsigned long offset, size_t size,
224	enum dma_data_direction dir,
225	unsigned long attrs)
226	{
227	struct vdpasim *vdpasim = dev_to_sim(dev);
228	struct vhost_iotlb *iommu = vdpasim->iommu;
229	u64 pa = (page_to_pfn(page) << PAGE_SHIFT) + offset;
230	int ret, perm = dir_to_perm(dir);
231
232	if (perm < 0)
233	return DMA_MAPPING_ERROR;
234
235	/* For simplicity, use identical mapping to avoid e.g iova
236	* allocator.
237	*/
238	ret = vhost_iotlb_add_range(iommu, pa, pa + size - 1,
239	pa, dir_to_perm(dir));
240	if (ret)
241	return DMA_MAPPING_ERROR;
242
243	return (dma_addr_t)(pa);
244	}
245
246	static void vdpasim_unmap_page(struct device *dev, dma_addr_t dma_addr,
247	size_t size, enum dma_data_direction dir,
248	unsigned long attrs)
249	{
250	struct vdpasim *vdpasim = dev_to_sim(dev);
251	struct vhost_iotlb *iommu = vdpasim->iommu;
252
253	vhost_iotlb_del_range(iommu, (u64)dma_addr,
254	(u64)dma_addr + size - 1);
255	}
256
257	static void vdpasim_alloc_coherent(struct device dev, size_t size,
258	dma_addr_t *dma_addr, gfp_t flag,
259	unsigned long attrs)
260	{
261	struct vdpasim *vdpasim = dev_to_sim(dev);
262	struct vhost_iotlb *iommu = vdpasim->iommu;
263	void *addr = kmalloc(size, flag);
264	int ret;
265
266	if (!addr)
267	*dma_addr = DMA_MAPPING_ERROR;
268	else {
269	u64 pa = virt_to_phys(addr);
270
271	ret = vhost_iotlb_add_range(iommu, (u64)pa,
272	(u64)pa + size - 1,
273	pa, VHOST_MAP_RW);
274	if (ret) {
275	*dma_addr = DMA_MAPPING_ERROR;
276	kfree(addr);
277	addr = NULL;
278	} else
279	*dma_addr = (dma_addr_t)pa;
280	}
281
282	return addr;
283	}
284
285	static void vdpasim_free_coherent(struct device *dev, size_t size,
286	void *vaddr, dma_addr_t dma_addr,
287	unsigned long attrs)
288	{
289	struct vdpasim *vdpasim = dev_to_sim(dev);
290	struct vhost_iotlb *iommu = vdpasim->iommu;
291
292	vhost_iotlb_del_range(iommu, (u64)dma_addr,
293	(u64)dma_addr + size - 1);
294	kfree(phys_to_virt((uintptr_t)dma_addr));
295	}
296
297	static const struct dma_map_ops vdpasim_dma_ops = {
298	.map_page = vdpasim_map_page,
299	.unmap_page = vdpasim_unmap_page,
300	.alloc = vdpasim_alloc_coherent,
301	.free = vdpasim_free_coherent,
302	};
303
304	static const struct vdpa_config_ops vdpasim_net_config_ops;
305
306	static struct vdpasim *vdpasim_create(void)
307	{
308	struct virtio_net_config *config;
309	struct vdpasim *vdpasim;
310	struct device *dev;
311	int ret = -ENOMEM;
312
313	vdpasim = vdpa_alloc_device(struct vdpasim, vdpa, NULL,
314	&vdpasim_net_config_ops);
315	if (!vdpasim)
316	goto err_alloc;
317
318	INIT_WORK(&vdpasim->work, vdpasim_work);
319	spin_lock_init(&vdpasim->lock);
320
321	dev = &vdpasim->vdpa.dev;
322	dev->coherent_dma_mask = DMA_BIT_MASK(64);
323	set_dma_ops(dev, &vdpasim_dma_ops);
324
325	vdpasim->iommu = vhost_iotlb_alloc(2048, 0);
326	if (!vdpasim->iommu)
327	goto err_iommu;
328
329	vdpasim->buffer = kmalloc(PAGE_SIZE, GFP_KERNEL);
330	if (!vdpasim->buffer)
331	goto err_iommu;
332
333	config = &vdpasim->config;
334	config->mtu = 1500;
335	config->status = VIRTIO_NET_S_LINK_UP;
336	eth_random_addr(config->mac);
337
338	vringh_set_iotlb(&vdpasim->vqs[0].vring, vdpasim->iommu);
339	vringh_set_iotlb(&vdpasim->vqs[1].vring, vdpasim->iommu);
340
341	vdpasim->vdpa.dma_dev = dev;
342	ret = vdpa_register_device(&vdpasim->vdpa);
343	if (ret)
344	goto err_iommu;
345
346	return vdpasim;
347
348	err_iommu:
349	put_device(dev);
350	err_alloc:
351	return ERR_PTR(ret);
352	}
353
354	static int vdpasim_set_vq_address(struct vdpa_device *vdpa, u16 idx,
355	u64 desc_area, u64 driver_area,
356	u64 device_area)
357	{
358	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
359	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];
360
361	vq->desc_addr = desc_area;
362	vq->driver_addr = driver_area;
363	vq->device_addr = device_area;
364
365	return 0;
366	}
367
368	static void vdpasim_set_vq_num(struct vdpa_device *vdpa, u16 idx, u32 num)
369	{
370	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
371	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];
372
373	vq->num = num;
374	}
375
376	static void vdpasim_kick_vq(struct vdpa_device *vdpa, u16 idx)
377	{
378	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
379	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];
380
381	if (vq->ready)
382	schedule_work(&vdpasim->work);
383	}
384
385	static void vdpasim_set_vq_cb(struct vdpa_device *vdpa, u16 idx,
386	struct vdpa_callback *cb)
387	{
388	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
389	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];
390
391	vq->cb = cb->callback;
392	vq->private = cb->private;
393	}
394
395	static void vdpasim_set_vq_ready(struct vdpa_device *vdpa, u16 idx, bool ready)
396	{
397	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
398	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];
399
400	spin_lock(&vdpasim->lock);
401	vq->ready = ready;
402	if (vq->ready)
403	vdpasim_queue_ready(vdpasim, idx);
404	spin_unlock(&vdpasim->lock);
405	}
406
407	static bool vdpasim_get_vq_ready(struct vdpa_device *vdpa, u16 idx)
408	{
409	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
410	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];
411
412	return vq->ready;
413	}
414
415	static int vdpasim_set_vq_state(struct vdpa_device *vdpa, u16 idx, u64 state)
416	{
417	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
418	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];
419	struct vringh *vrh = &vq->vring;
420
421	spin_lock(&vdpasim->lock);
422	vrh->last_avail_idx = state;
423	spin_unlock(&vdpasim->lock);
424
425	return 0;
426	}
427
428	static u64 vdpasim_get_vq_state(struct vdpa_device *vdpa, u16 idx)
429	{
430	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
431	struct vdpasim_virtqueue *vq = &vdpasim->vqs[idx];
432	struct vringh *vrh = &vq->vring;
433
434	return vrh->last_avail_idx;
435	}
436
425a5070	437	static u32 vdpasim_get_vq_align(struct vdpa_device *vdpa)
2c53d0f6 JW	438	{
	439	return VDPASIM_QUEUE_ALIGN;
	440	}
	441
	442	static u64 vdpasim_get_features(struct vdpa_device *vdpa)
	443	{
	444	return vdpasim_features;
	445	}
	446
	447	static int vdpasim_set_features(struct vdpa_device *vdpa, u64 features)
	448	{
	449	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
	450
	451	/* DMA mapping must be done by driver */
	452	if (!(features & (1ULL << VIRTIO_F_IOMMU_PLATFORM)))
	453	return -EINVAL;
	454
	455	vdpasim->features = features & vdpasim_features;
	456
	457	return 0;
	458	}
	459
	460	static void vdpasim_set_config_cb(struct vdpa_device *vdpa,
	461	struct vdpa_callback *cb)
	462	{
	463	/* We don't support config interrupt */
	464	}
	465
	466	static u16 vdpasim_get_vq_num_max(struct vdpa_device *vdpa)
	467	{
	468	return VDPASIM_QUEUE_MAX;
	469	}
	470
	471	static u32 vdpasim_get_device_id(struct vdpa_device *vdpa)
	472	{
	473	return VDPASIM_DEVICE_ID;
	474	}
	475
	476	static u32 vdpasim_get_vendor_id(struct vdpa_device *vdpa)
	477	{
	478	return VDPASIM_VENDOR_ID;
	479	}
	480
	481	static u8 vdpasim_get_status(struct vdpa_device *vdpa)
	482	{
	483	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
	484	u8 status;
	485
	486	spin_lock(&vdpasim->lock);
	487	status = vdpasim->status;
	488	spin_unlock(&vdpasim->lock);
	489
21818ed0	490	return status;
2c53d0f6 JW	491	}
	492
	493	static void vdpasim_set_status(struct vdpa_device *vdpa, u8 status)
	494	{
	495	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
	496
	497	spin_lock(&vdpasim->lock);
	498	vdpasim->status = status;
	499	if (status == 0)
	500	vdpasim_reset(vdpasim);
	501	spin_unlock(&vdpasim->lock);
	502	}
	503
	504	static void vdpasim_get_config(struct vdpa_device *vdpa, unsigned int offset,
	505	void *buf, unsigned int len)
	506	{
	507	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
	508
	509	if (offset + len < sizeof(struct virtio_net_config))
	510	memcpy(buf, &vdpasim->config + offset, len);
	511	}
	512
	513	static void vdpasim_set_config(struct vdpa_device *vdpa, unsigned int offset,
	514	const void *buf, unsigned int len)
	515	{
	516	/* No writable config supportted by vdpasim */
	517	}
	518
	519	static u32 vdpasim_get_generation(struct vdpa_device *vdpa)
	520	{
	521	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
	522
	523	return vdpasim->generation;
	524	}
	525
	526	static int vdpasim_set_map(struct vdpa_device *vdpa,
	527	struct vhost_iotlb *iotlb)
	528	{
	529	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
	530	struct vhost_iotlb_map *map;
	531	u64 start = 0ULL, last = 0ULL - 1;
	532	int ret;
	533
	534	vhost_iotlb_reset(vdpasim->iommu);
	535
	536	for (map = vhost_iotlb_itree_first(iotlb, start, last); map;
	537	map = vhost_iotlb_itree_next(map, start, last)) {
	538	ret = vhost_iotlb_add_range(vdpasim->iommu, map->start,
	539	map->last, map->addr, map->perm);
	540	if (ret)
	541	goto err;
	542	}
	543	return 0;
	544
	545	err:
	546	vhost_iotlb_reset(vdpasim->iommu);
	547	return ret;
	548	}
	549
	550	static int vdpasim_dma_map(struct vdpa_device *vdpa, u64 iova, u64 size,
	551	u64 pa, u32 perm)
	552	{
	553	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
	554
555	return vhost_iotlb_add_range(vdpasim->iommu, iova,
556	iova + size - 1, pa, perm);
557	}
558
559	static int vdpasim_dma_unmap(struct vdpa_device *vdpa, u64 iova, u64 size)
560	{
561	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
562
563	vhost_iotlb_del_range(vdpasim->iommu, iova, iova + size - 1);
564
565	return 0;
566	}
567
568	static void vdpasim_free(struct vdpa_device *vdpa)
569	{
570	struct vdpasim *vdpasim = vdpa_to_sim(vdpa);
571
572	cancel_work_sync(&vdpasim->work);
573	kfree(vdpasim->buffer);
574	if (vdpasim->iommu)
575	vhost_iotlb_free(vdpasim->iommu);
576	}
577
578	static const struct vdpa_config_ops vdpasim_net_config_ops = {
579	.set_vq_address = vdpasim_set_vq_address,
580	.set_vq_num = vdpasim_set_vq_num,
581	.kick_vq = vdpasim_kick_vq,
582	.set_vq_cb = vdpasim_set_vq_cb,
583	.set_vq_ready = vdpasim_set_vq_ready,
584	.get_vq_ready = vdpasim_get_vq_ready,
585	.set_vq_state = vdpasim_set_vq_state,
586	.get_vq_state = vdpasim_get_vq_state,
587	.get_vq_align = vdpasim_get_vq_align,
588	.get_features = vdpasim_get_features,
589	.set_features = vdpasim_set_features,
590	.set_config_cb = vdpasim_set_config_cb,
591	.get_vq_num_max = vdpasim_get_vq_num_max,
592	.get_device_id = vdpasim_get_device_id,
593	.get_vendor_id = vdpasim_get_vendor_id,
594	.get_status = vdpasim_get_status,
595	.set_status = vdpasim_set_status,
596	.get_config = vdpasim_get_config,
597	.set_config = vdpasim_set_config,
598	.get_generation = vdpasim_get_generation,
599	.set_map = vdpasim_set_map,
600	.dma_map = vdpasim_dma_map,
601	.dma_unmap = vdpasim_dma_unmap,
602	.free = vdpasim_free,
603	};
604
605	static int __init vdpasim_dev_init(void)
606	{
607	vdpasim_dev = vdpasim_create();
608
609	if (!IS_ERR(vdpasim_dev))
610	return 0;
611
612	return PTR_ERR(vdpasim_dev);
613	}
614
615	static void __exit vdpasim_dev_exit(void)
616	{
617	struct vdpa_device *vdpa = &vdpasim_dev->vdpa;
618
619	vdpa_unregister_device(vdpa);
620	}
621
622	module_init(vdpasim_dev_init)
623	module_exit(vdpasim_dev_exit)
624
625	MODULE_VERSION(DRV_VERSION);
626	MODULE_LICENSE(DRV_LICENSE);
627	MODULE_AUTHOR(DRV_AUTHOR);
628	MODULE_DESCRIPTION(DRV_DESC);