DPDK中断管理

DPDK中断管理

DPDK中的网卡都是PCI设备，多数中断都属于

PCI_MSIX类型的中断。

中断执行流程跟踪：
kernel uio驱动uio_interrupt -> igb_uio注册的回调函数igbuio_pci_irqhandler -> kernel uio驱动调用 uio_event_notify -> 唤醒DPDK中断线程(epoll机制) -> 网卡驱动注册的回调(rte_intr_callback_register注册的) -> 用户注册的回调(rte_eth_dev_callback_register)。
主要连接点：打开的/dev/uiox文件fd描述符。

igb_uio的创建

下面是igb_uio创建过程部分代码。

static int
igbuio_pci_probe(struct pci_dev *dev, const struct pci_device_id *id)
{
   
   
     /* 省略代码 */
     
	/* fill uio infos */
	udev->info.name = "igb_uio";
	udev->info.version = "0.1";
	udev->info.handler = igbuio_pci_irqhandler;
	udev->info.irqcontrol = igbuio_pci_irqcontrol;

	switch (igbuio_intr_mode_preferred) {
   
   
	case RTE_INTR_MODE_MSIX:
		/* Only 1 msi-x vector needed */
		msix_entry.entry = 0;
		if (pci_enable_msix(dev, &msix_entry, 1) == 0) {
   
   
			dev_dbg(&dev->dev, "using MSI-X");
			udev->info.irq = msix_entry.vector;
			udev->mode = RTE_INTR_MODE_MSIX;
			break;
		}
		/* fall back to INTX */
	case RTE_INTR_MODE_LEGACY:
		if (pci_intx_mask_supported(dev)) {
   
   
			dev_dbg(&dev->dev, "using INTX");
			udev->info.irq_flags = IRQF_SHARED;
			udev->info.irq = dev->irq;
			udev->mode = RTE_INTR_MODE_LEGACY;
			break;
		}
	
	default:
		dev_err(&dev->dev, "invalid IRQ mode %u",
			igbuio_intr_mode_preferred);
		err = -EINVAL;
		goto fail_release_iomem;
	}
	
     /* 省略代码 */
     
	/* register uio driver */
	err = uio_register_device(&dev->dev, &udev->info);
	if (err != 0)
		goto fail_remove_group;


	dev_info(&dev->dev, "uio device registered with irq %lx\n",
		 udev->info.irq);

	return 0;

这里创建了一个uio设备，uio设备的中断回调函数是igbuio_pci_irqhandler，当pci设备发生了中断，就会调用此函数。

static irqreturn_t
igbuio_pci_irqhandler(int irq, struct uio_info *info)
{
   
   
	struct rte_uio_pci_dev *udev = igbuio_get_uio_pci_dev(info);

	/* Legacy mode need to mask in hardware */
	if (udev->mode == RTE_INTR_MODE_LEGACY &&
	    !pci_check_and_mask_intx(udev->pdev))
		return IRQ_NONE;

	/* Message signal mode, no share IRQ and automasked */
	return IRQ_HANDLED;
}

linux中uio设备的创建

先看一下uio设备注册函数，文件driver/uio/uio.c

/* use a define to avoid include chaining to get THIS_MODULE */
#define uio_register_device(parent, info) \
	__uio_register_device(THIS_MODULE, parent, info)

int __uio_register_device(struct module *owner,
			  struct device *parent,
			  struct uio_info *info)
{
   
   
	struct uio_device *idev;
	int ret = 0;

	if (!parent || !info || !info->name || !info->version)
		return -EINVAL;

	info->uio_dev = NULL;

	idev = devm_kzalloc(parent, sizeof(*idev), GFP_KERNEL);
	if (!idev) {
   
   
		return -ENOMEM;
	}

	idev->owner = owner;
	idev->info = info;
	init_waitqueue_head(&idev->wait);
	atomic_set(&idev->event, 0);

	ret = uio_get_minor(idev);
	if (ret)
		return ret;

	idev->dev = device_create(&uio_class, parent,
				  MKDEV(uio_major, idev->minor), idev,
				  "uio%d", idev->minor);
	if (IS_ERR(idev->dev)) {
   
   
		printk(KERN_ERR "UIO: device register failed\n");
		ret = PTR_ERR(idev->dev);
		goto err_device_create;
	}

	ret = uio_dev_add_attributes(idev);
	if (ret)
		goto err_uio_dev_add_attributes;

	info->uio_dev = idev;

	if (info->irq && (info->irq != UIO_IRQ_CUSTOM)) {
   
   
		/*
		 * Note that we deliberately don't use devm_request_irq
		 * here. The parent module can unregister the UIO device
		 * and call pci_disable_msi, which requires that this
		 * irq has been freed. However, the device may have open
		 * FDs at the time of unregister and therefore may not be
		 * freed until they are released.
		 */
		ret = request_irq(info->irq, uio_interrupt,
				  info->irq_flags, info->name, idev);
		if (ret)
			goto err_request_irq;
	}

	return 0;

err_request_irq:
	uio_dev_del_attributes(idev