virtio 有分为guest 中的前端程序和qemu中的后端程序
virtio中的五种前端程序为
virtio-blk:/drivers/block/virtio-blk.c
virtio-net:/drivers/net/virtio-net.c
virtio-pci:/drivers/virtio/virtio-pci.c
virtio-ballon:/drivers/virtio/virtio-ballon.c
virtio-console:/drivers/virtio/virtio-console.c
这五种往下调用/drivers/virtio/virtio.c -> /drivers/virtio/virtio_ring.c
总结一下virtio的flow:guest->qemu->host kernel ->hw
下来我们先看看virtio-blk.c
static int __init init(void)
{
int error;
virtblk_wq = alloc_workqueue("virtio-blk", 0, 0);
if (!virtblk_wq)
return -ENOMEM;
major = register_blkdev(0, "virtblk");
if (major < 0) {
error = major;
goto out_destroy_workqueue;
}
error = register_virtio_driver(&virtio_blk);
if (error)
goto out_unregister_blkdev;
return 0;
out_unregister_blkdev:
unregister_blkdev(major, "virtblk");
out_destroy_workqueue:
destroy_workqueue(virtblk_wq);
return error;
}
在其入口函数中,申请了一个static struct workqueue_struct *virtblk_wq;,新建了一个virtblk的block设备,并调用register_virtio_driver 注册static struct virtio_driver virtio_blk
int register_virtio_driver(struct virtio_driver *driver)
{
/* Catch this early. */
BUG_ON(driver->feature_table_size && !driver->feature_table);
driver->driver.bus = &virtio_bus;
return driver_register(&driver->driver);
}
可见所有的virtio_driver 有自己的总线virtio_bus
当通过qemu启动guest的时候如果指定-device virtio-blk-device。例如:
qemu-system-aarch64 -machine virt -cpu host -kernel Image -drive if=none,file=ubuntu.img,id=fs -device virtio-blk-device,drive=fs -append "console=ttyAMA0 root=/dev/vda1" -nographic -D -d -enable-kvm
就会调用virtio_blk的virtblk_probe函数
在virtblk_probe 函数中,分配一个 struct virtio_blk *vblk;这个代表一个virtio的blk设备
vdev->priv = vblk = kmalloc(sizeof(*vblk), GFP_KERNEL);
并保存到virtio_device的priv变量中。
通过 err = init_vq(vblk);申请一个 struct virtqueue **vqs;
static int init_vq(struct virtio_blk *vblk)
{
int err;
int i;
vq_callback_t **callbacks;
const char **names;
struct virtqueue **vqs;
unsigned short num_vqs;
struct virtio_device *vdev = vblk->vdev;
err = virtio_cread_feature(vdev, VIRTIO_BLK_F_MQ,
struct virtio_blk_config, num_queues,
&num_vqs);
if (err)
num_vqs = 1;
vblk->vqs = kmalloc_array(num_vqs, sizeof(*vblk->vqs), GFP_KERNEL);
if (!vblk->vqs)
return -ENOMEM;
//通过kmalloc_array 申请数组
names = kmalloc_array(num_vqs, sizeof(*names), GFP_KERNEL);
callbacks = kmalloc_array(num_vqs, sizeof(*callbacks), GFP_KERNEL);
vqs = kmalloc_array(num_vqs, sizeof(*vqs), GFP_KERNEL);
if (!names || !callbacks || !vqs) {
err = -ENOMEM;
goto out;
}
//给callback 和name 这两个数组赋值
for (i = 0; i < num_vqs; i++) {
callbacks[i] = virtblk_done;
snprintf(vblk->vqs[i].name, VQ_NAME_LEN, "req.%d", i);
names[i] = vblk->vqs[i].name;
}
//看是否可以找到virtqueues,并将前面赋值的callback和names 写道configuration中
/* Discover virtqueues and write information to configuration. */
err = vdev->config->find_vqs(vdev, num_vqs, vqs, callbacks, names);
if (err)
goto out;
//将 struct virtqueue **vqs; 保存到vblk->vqs[i].vq 中
for (i = 0; i < num_vqs; i++) {
spin_lock_init(&vblk->vqs[i].lock);
vblk->vqs[i].vq = vqs[i];
}
vblk->num_vqs = num_vqs;
out:
kfree(vqs);
kfree(callbacks);
kfree(names);
if (err)
kfree(vblk->vqs);
return err;
}
回到virtblk_probe 中通过 vblk->disk = alloc_disk(1 << PART_BITS); 申请一个gendisk,这里代表的是virtio blk的物理硬盘
分配blk 用的queue q = vblk->disk->queue = blk_mq_init_queue(&vblk->tag_set); 其中对request的操作函数是通过
memset(&vblk->tag_set, 0, sizeof(vblk->tag_set));
vblk->tag_set.ops = &virtio_mq_ops;
来设定的。
static struct blk_mq_ops virtio_mq_ops = {
.queue_rq = virtio_queue_rq,
.complete = virtblk_request_done,
.init_request = virtblk_init_request,
};
因此virtio blk的queue_rq 函数是virtio_queue_rq
static int virtio_queue_rq(struct blk_mq_hw_ctx *hctx,
const struct blk_mq_queue_data *bd)
{
struct virtio_blk *vblk = hctx->queue->queuedata;
struct request *req = bd->rq;
struct virtblk_req *vbr = blk_mq_rq_to_pdu(req);
unsigned long flags;
unsigned int num;
int qid = hctx->queue_num;
int err;
bool notify = false;
BUG_ON(req->nr_phys_segments + 2 > vblk->sg_elems);
vbr->req = req;
blk_mq_start_request(req);
//将vbr添加到sg中
num = blk_rq_map_sg(hctx->queue, vbr->req, vbr->sg);
if (num) {
if (rq_data_dir(vbr->req) == WRITE)
vbr->out_hdr.type |= cpu_to_virtio32(vblk->vdev, VIRTIO_BLK_T_OUT);
else
vbr->out_hdr.type |= cpu_to_virtio32(vblk->vdev, VIRTIO_BLK_T_IN);
}
spin_lock_irqsave(&vblk->vqs[qid].lock, flags);
err = __virtblk_add_req(vblk->vqs[qid].vq, vbr, vbr->sg, num);
if (err) {
virtqueue_kick(vblk->vqs[qid].vq);
blk_mq_stop_hw_queue(hctx);
spin_unlock_irqrestore(&vblk->vqs[qid].lock, flags);
/* Out of mem doesn't actually happen, since we fall back
* to direct descriptors */
if (err == -ENOMEM || err == -ENOSPC)
return BLK_MQ_RQ_QUEUE_BUSY;
return BLK_MQ_RQ_QUEUE_ERROR;
}
if (bd->last && virtqueue_kick_prepare(vblk->vqs[qid].vq))
notify = true;
spin_unlock_irqrestore(&vblk->vqs[qid].lock, flags);
if (notify)
virtqueue_notify(vblk->vqs[qid].vq);
return BLK_MQ_RQ_QUEUE_OK;
}
最终调用virtqueue_kick 来通知qemu(及所谓的后端),当qemu处理完后又调用virtio_mq_ops的virtblk_request_done函数。这样就完成了guest->qemu->host kernel ->hw 一个循环.
给磁盘设置名字virtblk_name_format("vd", index, vblk->disk->disk_name, DISK_NAME_LEN);
从这里看到由于是virto设备,这里是以vd开头的.
给disk的其他成员变量赋值
vblk->disk->major = major;
vblk->disk->first_minor = index_to_minor(index);
vblk->disk->private_data = vblk;
vblk->disk->fops = &virtblk_fops;
vblk->disk->flags |= GENHD_FL_EXT_DEVT;
vblk->index = index;
virtio中的五种前端程序为
virtio-blk:/drivers/block/virtio-blk.c
virtio-net:/drivers/net/virtio-net.c
virtio-pci:/drivers/virtio/virtio-pci.c
virtio-ballon:/drivers/virtio/virtio-ballon.c
virtio-console:/drivers/virtio/virtio-console.c
这五种往下调用/drivers/virtio/virtio.c -> /drivers/virtio/virtio_ring.c
总结一下virtio的flow:guest->qemu->host kernel ->hw
下来我们先看看virtio-blk.c
static int __init init(void)
{
int error;
virtblk_wq = alloc_workqueue("virtio-blk", 0, 0);
if (!virtblk_wq)
return -ENOMEM;
major = register_blkdev(0, "virtblk");
if (major < 0) {
error = major;
goto out_destroy_workqueue;
}
error = register_virtio_driver(&virtio_blk);
if (error)
goto out_unregister_blkdev;
return 0;
out_unregister_blkdev:
unregister_blkdev(major, "virtblk");
out_destroy_workqueue:
destroy_workqueue(virtblk_wq);
return error;
}
在其入口函数中,申请了一个static struct workqueue_struct *virtblk_wq;,新建了一个virtblk的block设备,并调用register_virtio_driver 注册static struct virtio_driver virtio_blk
int register_virtio_driver(struct virtio_driver *driver)
{
/* Catch this early. */
BUG_ON(driver->feature_table_size && !driver->feature_table);
driver->driver.bus = &virtio_bus;
return driver_register(&driver->driver);
}
可见所有的virtio_driver 有自己的总线virtio_bus
当通过qemu启动guest的时候如果指定-device virtio-blk-device。例如:
qemu-system-aarch64 -machine virt -cpu host -kernel Image -drive if=none,file=ubuntu.img,id=fs -device virtio-blk-device,drive=fs -append "console=ttyAMA0 root=/dev/vda1" -nographic -D -d -enable-kvm
就会调用virtio_blk的virtblk_probe函数
在virtblk_probe 函数中,分配一个 struct virtio_blk *vblk;这个代表一个virtio的blk设备
vdev->priv = vblk = kmalloc(sizeof(*vblk), GFP_KERNEL);
并保存到virtio_device的priv变量中。
通过 err = init_vq(vblk);申请一个 struct virtqueue **vqs;
static int init_vq(struct virtio_blk *vblk)
{
int err;
int i;
vq_callback_t **callbacks;
const char **names;
struct virtqueue **vqs;
unsigned short num_vqs;
struct virtio_device *vdev = vblk->vdev;
err = virtio_cread_feature(vdev, VIRTIO_BLK_F_MQ,
struct virtio_blk_config, num_queues,
&num_vqs);
if (err)
num_vqs = 1;
vblk->vqs = kmalloc_array(num_vqs, sizeof(*vblk->vqs), GFP_KERNEL);
if (!vblk->vqs)
return -ENOMEM;
//通过kmalloc_array 申请数组
names = kmalloc_array(num_vqs, sizeof(*names), GFP_KERNEL);
callbacks = kmalloc_array(num_vqs, sizeof(*callbacks), GFP_KERNEL);
vqs = kmalloc_array(num_vqs, sizeof(*vqs), GFP_KERNEL);
if (!names || !callbacks || !vqs) {
err = -ENOMEM;
goto out;
}
//给callback 和name 这两个数组赋值
for (i = 0; i < num_vqs; i++) {
callbacks[i] = virtblk_done;
snprintf(vblk->vqs[i].name, VQ_NAME_LEN, "req.%d", i);
names[i] = vblk->vqs[i].name;
}
//看是否可以找到virtqueues,并将前面赋值的callback和names 写道configuration中
/* Discover virtqueues and write information to configuration. */
err = vdev->config->find_vqs(vdev, num_vqs, vqs, callbacks, names);
if (err)
goto out;
//将 struct virtqueue **vqs; 保存到vblk->vqs[i].vq 中
for (i = 0; i < num_vqs; i++) {
spin_lock_init(&vblk->vqs[i].lock);
vblk->vqs[i].vq = vqs[i];
}
vblk->num_vqs = num_vqs;
out:
kfree(vqs);
kfree(callbacks);
kfree(names);
if (err)
kfree(vblk->vqs);
return err;
}
回到virtblk_probe 中通过 vblk->disk = alloc_disk(1 << PART_BITS); 申请一个gendisk,这里代表的是virtio blk的物理硬盘
分配blk 用的queue q = vblk->disk->queue = blk_mq_init_queue(&vblk->tag_set); 其中对request的操作函数是通过
memset(&vblk->tag_set, 0, sizeof(vblk->tag_set));
vblk->tag_set.ops = &virtio_mq_ops;
来设定的。
static struct blk_mq_ops virtio_mq_ops = {
.queue_rq = virtio_queue_rq,
.complete = virtblk_request_done,
.init_request = virtblk_init_request,
};
因此virtio blk的queue_rq 函数是virtio_queue_rq
static int virtio_queue_rq(struct blk_mq_hw_ctx *hctx,
const struct blk_mq_queue_data *bd)
{
struct virtio_blk *vblk = hctx->queue->queuedata;
struct request *req = bd->rq;
struct virtblk_req *vbr = blk_mq_rq_to_pdu(req);
unsigned long flags;
unsigned int num;
int qid = hctx->queue_num;
int err;
bool notify = false;
BUG_ON(req->nr_phys_segments + 2 > vblk->sg_elems);
vbr->req = req;
blk_mq_start_request(req);
//将vbr添加到sg中
num = blk_rq_map_sg(hctx->queue, vbr->req, vbr->sg);
if (num) {
if (rq_data_dir(vbr->req) == WRITE)
vbr->out_hdr.type |= cpu_to_virtio32(vblk->vdev, VIRTIO_BLK_T_OUT);
else
vbr->out_hdr.type |= cpu_to_virtio32(vblk->vdev, VIRTIO_BLK_T_IN);
}
spin_lock_irqsave(&vblk->vqs[qid].lock, flags);
err = __virtblk_add_req(vblk->vqs[qid].vq, vbr, vbr->sg, num);
if (err) {
virtqueue_kick(vblk->vqs[qid].vq);
blk_mq_stop_hw_queue(hctx);
spin_unlock_irqrestore(&vblk->vqs[qid].lock, flags);
/* Out of mem doesn't actually happen, since we fall back
* to direct descriptors */
if (err == -ENOMEM || err == -ENOSPC)
return BLK_MQ_RQ_QUEUE_BUSY;
return BLK_MQ_RQ_QUEUE_ERROR;
}
if (bd->last && virtqueue_kick_prepare(vblk->vqs[qid].vq))
notify = true;
spin_unlock_irqrestore(&vblk->vqs[qid].lock, flags);
if (notify)
virtqueue_notify(vblk->vqs[qid].vq);
return BLK_MQ_RQ_QUEUE_OK;
}
最终调用virtqueue_kick 来通知qemu(及所谓的后端),当qemu处理完后又调用virtio_mq_ops的virtblk_request_done函数。这样就完成了guest->qemu->host kernel ->hw 一个循环.
给磁盘设置名字virtblk_name_format("vd", index, vblk->disk->disk_name, DISK_NAME_LEN);
从这里看到由于是virto设备,这里是以vd开头的.
给disk的其他成员变量赋值
vblk->disk->major = major;
vblk->disk->first_minor = index_to_minor(index);
vblk->disk->private_data = vblk;
vblk->disk->fops = &virtblk_fops;
vblk->disk->flags |= GENHD_FL_EXT_DEVT;
vblk->index = index;
而在register_virtio_driver->device_register->device_add时会调用 kobject_uevent(&dev->kobj, KOBJ_ADD); 通知user space,新添加了一个device。注意这里的aciton是KOBJ_ADD