linux实现底半部的机制主要有tasklet、workqueue、softirq和线程化irq。
1.tasklettasklet的使用较为简单,它的执行上下文是软中断,所以在tasklet中不能睡眠,它的执行时机通常是中断顶半部返回的时候。我们只需要定义tasklet及其处理函数,并将两者关联起来即可,例如:
1 void my_tasklet_func(unsigned long); /* 定义一个处理函数 */ 2 DECLARE_TASKLET(my_tasklet, my_tasklet_func, data); /* 定义一个tasklet结构my_tasklet,与my_tasklet_func(data)函数相关联 */
代码 DECLEARE_TASKLET(my_tasklet, my_tasklet_func, data)实现了定义名称为my_tasklet的tasklet,并将其与my_tasklet_func()这个函数绑定,而传入这个函数的参数为data。在需要调度tasklet的时候引用一个tasklet_schedule()函数就能使系统在适当的时候进行调度运行:
tasklet_schedule(&my_tasklet);
使用tasklet作为底半部处理中断的设备驱动程序模板如下:
struct xxx_struct { int xxx_irq; ... ... }; static unsigned long data; /* 定义 tasklet 和底半部函数并将它们关联 */ DECLARE_TASKLET(xxx_tasklet, xxx_do_tasklet, (unsigned long)&data); /* 中断处理底半部 */ void xxx_do_tasklet(unsigned long data) { struct xxx_struct *pdata = (void *)*(unsigned long *)data; ... } /* 中断处理顶半部 */ irqreturn_t xxx_interrupt(int irq, void *dev_id) { ... data = (unsigned long)dev_id; tasklet_schedule(&xxx_tasklet); ... } /* 设备驱动模块探测函数 */ static int xxx_probe(struct platform_device *pdev) { ... struct xxx_struct *pdata = devm_kzalloc(&pdev->dev, sizeof(struct xxx_struct), GFP_KERNEL); if (!pdata) { dev_err(&pdev->dev, "Out of memory\n"); return -ENOMEM; } platform_set_drvdata(pdev, pdata); /* 申请中断 */ result = request_irq(pdata->xxx_irq, xxx_interrupt, 0, "xxx", pdata); ...
return IRQ_HANDLED; } /* 设备驱动模块remove函数 */ static int xxx_remove(struct platforn_device *pdev) { struct xxx_struct *pdata = platform_get_drvdata(pdev); ... /* 释放中断 */ free_irq(pdata->xxx_irq, NULL); ... }
上述程序在模块加载函数中申请中断,并在模块卸载函数中释放它。对应于xxx_irq的中断处理程序被设置为xxx_interrupt()函数,在这个函数中,tasklet_schedule(&xxx_tasklet)调度被定义的tasklet函数xxx_do_tasklet()在适当的时候执行。
2.工作队列工作队列的使用方法和tasklet非常相似,但是工作队列的执行上下文是内核线程,因此可以调度和睡眠。下面的代码用于定义一个工作队列和一个底半部执行函数:
struct work_struct my_wq; /* 定义一个工作队列 */ void my_wq_func(struct work_struct *work); /* 定义一个处理函数 */
通过INIT_WORK()可以初始化这个工作队列并将工作队列与处理函数绑定:
INIT_WORK(&my_wq, my_wq_func); /* 初始化工作队列并将其与处理函数绑定 */
与tasklet_schedule()对应的用于调度工作队列执行的函数为 schedule_work(),如:
schedule_work(&my_wq); /* 调度工作队列执行 */
使用工作队列处理中断底半部的设备驱动程序模板代码如下:
struct xxx_struct { int xxx_irq; struct work_struct xxx_wq; /* 定义工作队列 */ ... ... }; /* 中断处理底半部 */ void xxx_do_work(struct work_struct *work) { struct xxx_struct *pdata = container_of(work, struct xxx_struct, xxx_wq); ... } /* 中断处理顶半部 */ irqreturn_t xxx_interrupt(int irq, void *dev_id) { struct xxx_struct *pdata = (struct xxx_struct *)dev_id; ... schedule_work(&pdata->xxx_wq); ... return IRQ_HANDLED; } /* 设备驱动模块探测函数 */ int xxx_probe(struct platform_device *pdev) { ... struct xxx_struct *pdata = devm_kzalloc(&pdev->dev, sizeof(struct xxx_struct ), GFP_KERNEL); /*申请中断*/ result = request_irq(pdata->xxx_irq, xxx_interrupt, 0, "xxx", pdata); ... /* 初始化工作队列 */ INIT_WORK(&pdata->xxx_wq, xxx_do_work); platform_set_drvdata(pdev, pdata); ... } /* 设备驱动模块remove函数 */ int xxx_remove(struct platform_device *pdev) { struct xxx_struct *pdata = platform_get_drvdata(pdev); ...
cancel_work_sync(&pdata->xxx_wq); /* 释放中断 */ free_irq(pdata->xxx_irq, NULL); ... }
工作队列早期的实现是在每个CPU核上创建一个worker内核线程,所有在这个核上调度的工作都在该worker线程中执行,其并发性显然差强人意。在linux 2.6.36以后,转而实现了 “Concurrency-managed workqueues”,简称"cmwq",cmwq会自动维护工作队列的线程池以提高并发性,同时保持了API的向后兼容。
延时工作队列1. 数据结构delayed_work用于处理延迟执行
struct delayed_work { struct work_struct work; struct timer_list timer; /* target workqueue and CPU ->timer uses to queue ->work */ struct workqueue_struct *wq; int cpu; };
2.在工作队列中被调用的函数原形如下
typedef void (*work_func_t)(struct work_struct *work);
3. 初始化数据结构
INIT_DELAYED_WORK(struct delayed_work *work, work_func_t func)
4. 提交延时任务到工作队列
int schedule_delayed_work(struct delayed_work *work, unsigned long delay);
5.删除提交到工作队列的任务
int cancel_delayed_work(strcut delayed_work *work);
6. 刷新默认工作队列(常跟cancle_delayed_work一起使用)
void flush_schedlue_work(void); 或者 int cancel_delayed_work_sync(strcut delayed_work *work);
使用延时工作队列处理中断底半部的设备驱动程序模板代码如下:
struct xxx_struct {
int xxx_irq; struct delayed_work xxx_dwork; /* 定义延时工作队列 */ ... ... }; /* 中断处理底半部 */ void xxx_do_work(struct work_struct *work) { struct xxx_struct *pdata = container_of(work, struct xxx_struct, xxx_dwork.work); ... } /* 中断处理顶半部 */ irqreturn_t xxx_interrupt(int irq, void *dev_id) { struct xxx_struct *pdata = (struct xxx_struct *)dev_id; ...
/* * delay 2000ms */ schedule_delayed_work(&pdata->xxx_dwork, msesc_to_jiffies(2000)); ...
return IRQ_HANDLED; } /* 设备驱动模块探测函数 */ int xxx_probe(struct platform_device *pdev) { ... struct xxx_struct *pdata = devm_kzalloc(&pdev->dev, sizeof(struct xxx_struct ), GFP_KERNEL); /*申请中断*/ result = request_irq(pdata->xxx_irq, xxx_interrupt, 0, "xxx", pdata); ... /* 初始化工作队列 */ INIT_DELAYED_WORK(&pdata->xxx_dwork, xxx_do_work); platform_set_drvdata(pdev, pdata); ... } /* 设备驱动模块remove函数 */ int xxx_remove(struct platform_device *pdev) { struct xxx_struct *pdata = platform_get_drvdata(pdev); ...
cancel_delayed_work(&pdata->xxx_dwork);
cancel_delayed_work_sync(&pdata->xxx_dwork);
/* 释放中断 */ free_irq(pdata->xxx_irq, NULL); ... }
3.软中断
软中断(Softirq)也是一种传统的底半部处理机制,它的执行时机通常是顶半部返回的时候,tasklet是基于软中断实现的,因此也运行于软中断上下文。
在Linux内核中,用softirq_action结构体表征一个软中断,这个结构体包含软中断处理函数指针和传递给该函数的参数。使用open_softirq()函数可以注册软中断对应的处理函数,而raise_softirq()函数可以触发一个软中断。
软中断和tasklet运行于软中断上下文,仍然属于原子上下文的一种,而工作队列则运行于进程上下文。因此,在软中断和tasklet处理函数中不允许睡眠,而在工作队列处理函数中允许睡眠。
local_bh_disable() 和 local_bh_enable() 是内核中用于禁止和使能软中断及tasklet底半部机制的函数。
内核中采用softirq的地方包括 HI_SOFTIRQ、TIMER_SOFTIRQ、NET_TX_SOFTIRQ、NET_RX_SOFTIRQ、SCSI_SOFTIRQ、TASKLET_SOFTIRQ等,一般来说,驱动的编写者不会也不宜直接使用softirq。
硬中断、软中断和信号的区别:
硬中断是外部设备对CPU的中断,软中断是中断底半部的一种处理机制,而信号则是由内核(或其他进程)对某个进程的中断。在设计系统调用的场合,人们也常说通过软中断(例ARM为swi)陷入内核,此时软中断的概念是指由软件指令引发的中断,和我们这个地方所说的softirq是两个完全不同的概念,一个是software,一个是soft。
需要特别说明的是,软中断以及基于软中断的tasklet如果在某段时间内大量出现的话,内核会把后续软中断放入 ksoftirqd 内核线程中执行。总的来说,中断优先级高于软中断,软中断优先级又高于任何一个线程。软中断适度线程化,可以缓解高负载情况下系统的响应。
4.threaded_irq
在内核中除了可以通过request_irq()、devm_request_irq()申请中断以外,还可以通过request_threaded_irq() 和 devm_request_threaded_irq() 申请。这两个函数的原型为:
/** * request_threaded_irq - allocate an interrupt line * @irq: Interrupt line to allocate * @handler: Function to be called when the IRQ occurs. * Primary handler for threaded interrupts. * If handler is NULL and thread_fn != NULL * the default primary handler is installed. * @thread_fn: Function called from the irq handler thread * If NULL, no irq thread is created * @irqflags: Interrupt type flags * @devname: An ascii name for the claiming device * @dev_id: A cookie passed back to the handler function * * This call allocates interrupt resources and enables the * interrupt line and IRQ handling. From the point this * call is made your handler function may be invoked. Since * your handler function must clear any interrupt the board * raises, you must take care both to initialise your hardware * and to set up the interrupt handler in the right order. * * If you want to set up a threaded irq handler for your device * then you need to supply @handler and @thread_fn. @handler is * still called in hard interrupt context and has to check * whether the interrupt originates from the device. If yes it * needs to disable the interrupt on the device and return * IRQ_WAKE_THREAD which will wake up the handler thread and run * @thread_fn. This split handler design is necessary to support * shared interrupts. * * Dev_id must be globally unique. Normally the address of the * device data structure is used as the cookie. Since the handler * receives this value it makes sense to use it. * * If your interrupt is shared you must pass a non NULL dev_id * as this is required when freeing the interrupt. * * Flags: * * IRQF_SHARED Interrupt is shared * IRQF_TRIGGER_* Specify active edge(s) or level * IRQF_ONESHOT Run thread_fn with interrupt line masked */ int request_threaded_irq(unsigned int irq, irq_handler_t handler, irq_handler_t thread_fn, unsigned long irqflags, const char *devname, void *dev_id); /** * devm_request_threaded_irq - allocate an interrupt line for a managed device * @dev: device to request interrupt for * @irq: Interrupt line to allocate * @handler: Function to be called when the IRQ occurs * @thread_fn: function to be called in a threaded interrupt context. NULL * for devices which handle everything in @handler * @irqflags: Interrupt type flags * @devname: An ascii name for the claiming device, dev_name(dev) if NULL * @dev_id: A cookie passed back to the handler function * * Except for the extra @dev argument, this function takes the * same arguments and performs the same function as * request_threaded_irq(). IRQs requested with this function will be * automatically freed on driver detach. * * If an IRQ allocated with this function needs to be freed * separately, devm_free_irq() must be used. */ int devm_request_threaded_irq(struct device *dev, unsigned int irq, irq_handler_t handler, irq_handler_t thread_fn, unsigned long irqflags, const char *devname, void *dev_id);
由此可见,它们比request_irq()、devm_request_irq()多了一个参数 thread_fn。用这两个API申请中断的时候,内核会为相应的中断号分配一个对应的内核线程。注意这个线程只针对这个中断号,如果其他中断也通过request_threaded_irq()申请,自然会得到新的内核线程。
参数handler对应的函数执行于中断上下文,thread_fn参数对应的函数则执行于内核线程。如果handler结束的时候,返回值是 IRQ_WAKE_THREAD,内核会调度对应线程执行 thread_fn 对应的函数。
request_threaded_irq() 和 devm_request_threaded_irq() 支持在 irqflags 中设置 IRQF_ONESHOT标记,这样内核会自动帮助我们在中断上下文中屏蔽对应的中断号,而在内核调度 thread_fn 执行后,重新使能该中断号。对于我们无法在上半部清除中断的情况, IRQ_ONESHOT 特别有用,避免了中断服务程序一退出,中断就洪泛的情况。
handler 参数可以设置为NULL,这种情况下,内核会用默认的 irq_default_primary_handler() 代替 handler,并会使用 IRQ_ONESHOT标记。 irq_default_primary_handler() 定义为:
/* * Default primary interrupt handler for threaded interrupts. Is * assigned as primary handler when request_threaded_irq is called * with handler == NULL. Useful for oneshot interrupts. */ static irqreturn_t irq_default_primary_handler(int irq, void *dev_id) { return IRQ_WAKE_THREAD; }