patch日期
收发流程的重大修改,来源于2012年的如下补丁
内核提交收发流程的patch
spi: create a message queueing infrastructure - kernel/git/stable/linux.git - Linux kernel stable tree
源代码路径及功能
| 源码 | 作用 | 
| \drivers\spi\spi.c | spi 通用接口,包括发送,队列管理等。核心文件 | 
| \drivers\mtd\devices\m25p80.c | 具体flash驱动,由此驱动进行SPI协议的组装,并通过spi.c的接口将数据下发 | 
| drivers\spi\spidev.c | 通用spi设备驱动,即直接将用户发送的数据通过spi.c中的接口发送。具体发送的命令字构成由用户填写。 | 
| \drivers\spi\spi-dw.c | 具体控制器的驱动,实现对控制器硬件的访问控制 | 
kthread_queue_work
kthread_worker 和 kthread_work
    
两者关系:kworker 类似任务容器;而kwork是具体的任务。
kthread_worker 的初始化
代码来源: spi_init_queue(spi.c)
kthread_init_worker(&ctlr->kworker);
	ctlr->kworker_task = kthread_run(kthread_worker_fn, &ctlr->kworker,
					 "%s", dev_name(&ctlr->dev));可以看出worker本质上即为一个内核线程,线程采用的参数为kworker这个结构体,即容器。那就可以猜测,这个线程worker_fn的功能,就是检测kworker这个容器中是否由work,如果有,则进行处理。
问题: 这个线程的调度策略及优先级是怎么样的?
kthread_worker 的优先级
代码来源: spi_init_queue(spi.c)
struct sched_param param = { .sched_priority = MAX_RT_PRIO - 1 };
	/*
	 * Controller config will indicate if this controller should run the
	 * message pump with high (realtime) priority to reduce the transfer
	 * latency on the bus by minimising the delay between a transfer
	 * request and the scheduling of the message pump thread. Without this
	 * setting the message pump thread will remain at default priority.
	 */
	if (ctlr->rt) {
		dev_info(&ctlr->dev,
			"will run message pump with realtime priority\n");
		sched_setscheduler(ctlr->kworker_task, SCHED_FIFO, ¶m);
	}
这里设置的调度策略为FIFO。
kthread_work的初始化
代码来源: spi_init_queue(spi.c)
	kthread_init_work(&ctlr->pump_messages, spi_pump_messages);
struct kthread_worker		kworker;
	struct task_struct		*kworker_task;
	struct kthread_work		pump_messages;即将 work的功能函数设置为 spi_pump_messages.
kthread_work的执行
kthread_queue_work(&ctlr->kworker, &ctlr->pump_messages);if (!worker->current_work && likely(worker->task))
		wake_up_process(worker->task)将work加入到 worker中,并唤醒worker这个线程。当worker为空,即没有work要处理时,则worker休眠。
总体上讲,从字面意思,worker是干活的人,work是活,有活就唤醒干活的人,没活干活的人就休息。
消息队列
发送流程
数据的传输载体
struct spi_message 与 struct spi_transfer
fmsh\include\linux\spi\spi.h
核心数据结构,用于表示一次数据的传输
/**
 * struct spi_message - one multi-segment SPI transaction
 * @transfers: list of transfer segments in this transaction
 * @spi: SPI device to which the transaction is queued
 * @is_dma_mapped: if true, the caller provided both dma and cpu virtual
 *	addresses for each transfer buffer
 * @complete: called to report transaction completions
 * @context: the argument to complete() when it's called
 * @frame_length: the total number of bytes in the message
 * @actual_length: the total number of bytes that were transferred in all
 *	successful segments
 * @status: zero for success, else negative errno
 * @queue: for use by whichever driver currently owns the message
 * @state: for use by whichever driver currently owns the message
 * @resources: for resource management when the spi message is processed
 *
 * A @spi_message is used to execute an atomic sequence of data transfers,
 * each represented by a struct spi_transfer.  The sequence is "atomic"
 * in the sense that no other spi_message may use that SPI bus until that
 * sequence completes.  On some systems, many such sequences can execute as
 * as single programmed DMA transfer.  On all systems, these messages are
 * queued, and might complete after transactions to other devices.  Messages
 * sent to a given spi_device are always executed in FIFO order.
 *
 * The code that submits an spi_message (and its spi_transfers)
 * to the lower layers is responsible for managing its memory.
 * Zero-initialize every field you don't set up explicitly, to
 * insulate against future API updates.  After you submit a message
 * and its transfers, ignore them until its completion callback.
 */
struct spi_message {
	struct list_head	transfers;
	struct spi_device	*spi;
	unsigned		is_dma_mapped:1;
	/* REVISIT:  we might want a flag affecting the behavior of the
	 * last transfer ... allowing things like "read 16 bit length L"
	 * immediately followed by "read L bytes".  Basically imposing
	 * a specific message scheduling algorithm.
	 *
	 * Some controller drivers (message-at-a-time queue processing)
	 * could provide that as their default scheduling algorithm.  But
	 * others (with multi-message pipelines) could need a flag to
	 * tell them about such special cases.
	 */
	/* completion is reported through a callback */
	void			(*complete)(void *context);
	void			*context;
	unsigned		frame_length;
	unsigned		actual_length;
	int			status;
	/* for optional use by whatever driver currently owns the
	 * spi_message ...  between calls to spi_async and then later
	 * complete(), that's the spi_controller controller driver.
	 */
	struct list_head	queue;
	void			*state;
	/* list of spi_res reources when the spi message is processed */
	struct list_head        resources;
};
一个transfer即一次硬件的命令字及数据的传输过程;
一个message则包含多个transfer
将数据发送到控制器的队列--spi_sync
当flash驱动准备好数据,完成message 数据部分的初始化后,就调用此接口将数据传递给控制器的队列。
__spi_sync
static int __spi_sync(struct spi_device *spi, struct spi_message *message)
{
	DECLARE_COMPLETION_ONSTACK(done);
	int status;
	struct spi_controller *ctlr = spi->controller;
	unsigned long flags;
	status = __spi_validate(spi, message);
	if (status != 0)
		return status;
	message->complete = spi_complete;
	message->context = &done;
	message->spi = spi;
	SPI_STATISTICS_INCREMENT_FIELD(&ctlr->statistics, spi_sync);
	SPI_STATISTICS_INCREMENT_FIELD(&spi->statistics, spi_sync);
	/* If we're not using the legacy transfer method then we will
	 * try to transfer in the calling context so special case.
	 * This code would be less tricky if we could remove the
	 * support for driver implemented message queues.
	 */
	if (ctlr->transfer == spi_queued_transfer) {
		spin_lock_irqsave(&ctlr->bus_lock_spinlock, flags);
		trace_spi_message_submit(message);
		status = __spi_queued_transfer(spi, message, false);
		spin_unlock_irqrestore(&ctlr->bus_lock_spinlock, flags);
	} else {
		status = spi_async_locked(spi, message);
	}
	if (status == 0) {
		/* Push out the messages in the calling context if we
		 * can.
		 */
		if (ctlr->transfer == spi_queued_transfer) {
			SPI_STATISTICS_INCREMENT_FIELD(&ctlr->statistics,
						       spi_sync_immediate);
			SPI_STATISTICS_INCREMENT_FIELD(&spi->statistics,
						       spi_sync_immediate);
			__spi_pump_messages(ctlr, false);
		}
		wait_for_completion(&done);
		status = message->status;
	}
	message->context = NULL;
	return status;
}
这里主要注意3点:
1) status = __spi_queued_transfer(spi, message, false);时,第二个参数为flase,也就是说,此处仅仅是将message 放入到ctrl的queue中,并不会唤醒 kthead worker开始正式的下发数据。此接口在用户进程上下文中。最终变成如下图,一个ctrl queue里面有了多个message,处于待发送状态。

2) __spi_pump_messages(ctlr, false); ,第二个参数也为false,在此时才会唤醒kthread worker干活。
3) 将message放入队列后,接口调用并没有立即返回,而是调用如下接口,等待完成。
wait_for_completion(&done);往硬件发送数据__spi_pump_messages
此函数的调用处有两个:
1) 上一节描述的调用进程的上下文。
2) work的工作函数。
/**
 * __spi_pump_messages - function which processes spi message queue
 * @ctlr: controller to process queue for
 * @in_kthread: true if we are in the context of the message pump thread
 *
 * This function checks if there is any spi message in the queue that
 * needs processing and if so call out to the driver to initialize hardware
 * and transfer each message.
 *
 * Note that it is called both from the kthread itself and also from
 * inside spi_sync(); the queue extraction handling at the top of the
 * function should deal with this safely.
 */
static void __spi_pump_messages(struct spi_controller *ctlr, bool in_kthread)
{
	unsigned long flags;
	bool was_busy = false;
	int ret;
	/* Lock queue */
	spin_lock_irqsave(&ctlr->queue_lock, flags);
	/* Make sure we are not already running a message */
	if (ctlr->cur_msg) {
		spin_unlock_irqrestore(&ctlr->queue_lock, flags);
		return;
	}
	/* If another context is idling the device then defer */
	if (ctlr->idling) {
		kthread_queue_work(&ctlr->kworker, &ctlr->pump_messages);
		spin_unlock_irqrestore(&ctlr->queue_lock, flags);
		return;
	}
	/* Check if the queue is idle */
	if (list_empty(&ctlr->queue) || !ctlr->running) {
		if (!ctlr->busy) {
			spin_unlock_irqrestore(&ctlr->queue_lock, flags);
			return;
		}
		/* Only do teardown in the thread */
		if (!in_kthread) {
			kthread_queue_work(&ctlr->kworker,
					   &ctlr->pump_messages);
			spin_unlock_irqrestore(&ctlr->queue_lock, flags);
			return;
		}
		ctlr->busy = false;
		ctlr->idling = true;
		spin_unlock_irqrestore(&ctlr->queue_lock, flags);
		kfree(ctlr->dummy_rx);
		ctlr->dummy_rx = NULL;
		kfree(ctlr->dummy_tx);
		ctlr->dummy_tx = NULL;
		if (ctlr->unprepare_transfer_hardware &&
		    ctlr->unprepare_transfer_hardware(ctlr))
			dev_err(&ctlr->dev,
				"failed to unprepare transfer hardware\n");
		if (ctlr->auto_runtime_pm) {
			pm_runtime_mark_last_busy(ctlr->dev.parent);
			pm_runtime_put_autosuspend(ctlr->dev.parent);
		}
		trace_spi_controller_idle(ctlr);
		spin_lock_irqsave(&ctlr->queue_lock, flags);
		ctlr->idling = false;
		spin_unlock_irqrestore(&ctlr->queue_lock, flags);
		return;
	}
	/* Extract head of queue */
	ctlr->cur_msg =
		list_first_entry(&ctlr->queue, struct spi_message, queue);
	list_del_init(&ctlr->cur_msg->queue);
	if (ctlr->busy)
		was_busy = true;
	else
		ctlr->busy = true;
	spin_unlock_irqrestore(&ctlr->queue_lock, flags);
	mutex_lock(&ctlr->io_mutex);
	if (!was_busy && ctlr->auto_runtime_pm) {
		ret = pm_runtime_get_sync(ctlr->dev.parent);
		if (ret < 0) {
			dev_err(&ctlr->dev, "Failed to power device: %d\n",
				ret);
			mutex_unlock(&ctlr->io_mutex);
			return;
		}
	}
	if (!was_busy)
		trace_spi_controller_busy(ctlr);
	if (!was_busy && ctlr->prepare_transfer_hardware) {
		ret = ctlr->prepare_transfer_hardware(ctlr);
		if (ret) {
			dev_err(&ctlr->dev,
				"failed to prepare transfer hardware\n");
			if (ctlr->auto_runtime_pm)
				pm_runtime_put(ctlr->dev.parent);
			mutex_unlock(&ctlr->io_mutex);
			return;
		}
	}
	trace_spi_message_start(ctlr->cur_msg);
	if (ctlr->prepare_message) {
		ret = ctlr->prepare_message(ctlr, ctlr->cur_msg);
		if (ret) {
			dev_err(&ctlr->dev, "failed to prepare message: %d\n",
				ret);
			ctlr->cur_msg->status = ret;
			spi_finalize_current_message(ctlr);
			goto out;
		}
		ctlr->cur_msg_prepared = true;
	}
	ret = spi_map_msg(ctlr, ctlr->cur_msg);
	if (ret) {
		ctlr->cur_msg->status = ret;
		spi_finalize_current_message(ctlr);
		goto out;
	}
	ret = ctlr->transfer_one_message(ctlr, ctlr->cur_msg);
	if (ret) {
		dev_err(&ctlr->dev,
			"failed to transfer one message from queue\n");
		goto out;
	}
out:
	mutex_unlock(&ctlr->io_mutex);
	/* Prod the scheduler in case transfer_one() was busy waiting */
	if (!ret)
		cond_resched();
}
发送messages
spi_transfer_one_message
\drivers\spi\spi.c
1) 设置片选
spi_set_cs(msg->spi, true);2) 遍历message每个transfer,调用具体控制器接口发送数据
3) 等待transfer的传输完成。
if (ret > 0) {
				ret = 0;
				ms = 8LL * 1000LL * xfer->len;
				do_div(ms, xfer->speed_hz);
				ms += ms + 200; /* some tolerance */
				if (ms > UINT_MAX)
					ms = UINT_MAX;
				ms = wait_for_completion_timeout(&ctlr->xfer_completion,
								 msecs_to_jiffies(ms));
			}4)整个message的传输完成
if (mesg->complete)
		mesg->complete(mesg->context);设备驱动的发送接口
.55-fmsh\drivers\spi\spi-dw.c
static int dw_spi_transfer_one(struct spi_master *master,
		struct spi_device *spi, struct spi_transfer *transfer)总体流程

注意这里有几个上下文:
1) 用户发送上下文,将message 传递个控制器后,就在此上下文中等待message的处理结果。
2)发送线程上下文。
3)设备驱动发送数据时的中断上下文。
实现细节
锁的使用
1. ctlr->queue_lock
用户进程和实际发送线程之间对queue 链表的互斥。
spinlock_t queue_lock;
使用处:
将message 放入ctrl 队列时
spin_lock_irqsave(&ctlr->queue_lock, flags);以及内核线程将message 从队列取出,下发到硬件时
__spi_pump_messages
2. bus_lock_spinlock
/* If we're not using the legacy transfer method then we will
	 * try to transfer in the calling context so special case.
	 * This code would be less tricky if we could remove the
	 * support for driver implemented message queues.
	 */
	if (ctlr->transfer == spi_queued_transfer) {
		spin_lock_irqsave(&ctlr->bus_lock_spinlock, flags);
		trace_spi_message_submit(message);
		status = __spi_queued_transfer(spi, message, false);
		spin_unlock_irqrestore(&ctlr->bus_lock_spinlock, flags);
	}3. bus_lock_mutex
mutex_lock(&spi->controller->bus_lock_mutex);
	ret = __spi_sync(spi, message);
	mutex_unlock(&spi->controller->bus_lock_mutex);
多个用户或者flash、spi外设驱动之间互斥。
也就是说,一个SPI发送过程,需要先获取 mutex,然后bus_spinlock ,然后 queue spinlock。
这里有了bus mutex,为什么还有一个bus spin lock?
  
总结
为什么针对SPI的驱动,单独设置了FIFO的调度?而实时进程则是采用SCHED_FIFO或SCHED_RR。
参考资料
Kthread worker
linux kthread_worker-CSDN博客
内核 kthread_worker 和 kthread_work 机制_kthread_queue_work-CSDN博客
查看进程调度策略
Linux系统动态查看每个CPU上任务的调度情况_linux cpu 调度 记录-CSDN博客
linux进程的查看和调度 - 知乎 (zhihu.com)
Linux的进程线程调度策略_如何查看线程的policy-CSDN博客
linux内核调度的机制 tasklet/workqueue/kthread_worker/kthreadx详解及示例【转】 - Sky&Zhang - 博客园 (cnblogs.com)
混乱的Linux内核实时线程优先级 - 知乎 (zhihu.com) 5: linux内核调度的机制 tasklet/workqueue/kthread_worker/kthreadx详解及示例_kthread_worker和workqueue-CSDN博客
【驱动】SPI驱动分析(四)-关键API解析 - 学习,积累,成长 - 博客园 (cnblogs.com)




















