Linux内核延迟写机制学习

Linux内核延迟写机制

Linux内核延迟写的特点，是指在Linux通过write的场景下写入数据之后，会将数据直接标记为dirty，然后通过延迟读写的方式最后将数据回写到磁盘上。在本文的Linux-3.10的源码基础上简单学习一下相关机制。

Linux-bdi机制

Linux内核当前都是通过bdi_writeback函数来进行数据回写，该初始化方式如下。

static int __init default_bdi_init(void)
{
	int err;

	bdi_wq = alloc_workqueue("writeback", WQ_MEM_RECLAIM | WQ_FREEZABLE |
					      WQ_UNBOUND | WQ_SYSFS, 0);   // 通过workqueue机制来进行的函数执行
	if (!bdi_wq)
		return -ENOMEM;

	err = bdi_init(&default_backing_dev_info);   // 初始化bdi相关信息
	if (!err)
		bdi_register(&default_backing_dev_info, NULL, "default");
	err = bdi_init(&noop_backing_dev_info);

	return err;
}
subsys_initcall(default_bdi_init);



int bdi_init(struct backing_dev_info *bdi)
{
	int i, err;

	bdi->dev = NULL;

	bdi->min_ratio = 0;
	bdi->max_ratio = 100;
	bdi->max_prop_frac = FPROP_FRAC_BASE;
	spin_lock_init(&bdi->wb_lock);
	INIT_LIST_HEAD(&bdi->bdi_list);
	INIT_LIST_HEAD(&bdi->work_list);

	bdi_wb_init(&bdi->wb, bdi);   // 注册回调执行的函数类型

	for (i = 0; i < NR_BDI_STAT_ITEMS; i++) {
		err = percpu_counter_init(&bdi->bdi_stat[i], 0);
		if (err)
			goto err;
	}

	bdi->dirty_exceeded = 0;

	bdi->bw_time_stamp = jiffies;
	bdi->written_stamp = 0;

	bdi->balanced_dirty_ratelimit = INIT_BW;
	bdi->dirty_ratelimit = INIT_BW;
	bdi->write_bandwidth = INIT_BW;
	bdi->avg_write_bandwidth = INIT_BW;

	err = fprop_local_init_percpu(&bdi->completions);

	if (err) {
err:
		while (i--)
			percpu_counter_destroy(&bdi->bdi_stat[i]);
	}

	return err;
}
EXPORT_SYMBOL(bdi_init);


static void bdi_wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi)
{
	memset(wb, 0, sizeof(*wb));

	wb->bdi = bdi;
	wb->last_old_flush = jiffies;
	INIT_LIST_HEAD(&wb->b_dirty);
	INIT_LIST_HEAD(&wb->b_io);
	INIT_LIST_HEAD(&wb->b_more_io);
	spin_lock_init(&wb->list_lock);
	INIT_DELAYED_WORK(&wb->dwork, bdi_writeback_workfn);  // 注册bdi_writeback_workfn为延迟写入的回调处理函数
}

其中subsys_initcall函数会在start_kernel --> rest_init中通过kernel_init最终调用。

fs_initcall、early_initcall、__init宏_LPSTC123的博客-CSDN博客_early_initcall

主动调用sync触发回写操作

SYSCALL_DEFINE0(sync)
{
	int nowait = 0, wait = 1;

	wakeup_flusher_threads(0, WB_REASON_SYNC);
	iterate_supers(sync_inodes_one_sb, NULL);
	iterate_supers(sync_fs_one_sb, &nowait);
	iterate_supers(sync_fs_one_sb, &wait);
	iterate_bdevs(fdatawrite_one_bdev, NULL);
	iterate_bdevs(fdatawait_one_bdev, NULL);
	if (unlikely(laptop_mode))
		laptop_sync_completion();
	return 0;
}

主动唤醒了wakeup_flusher_threads线程来进行数据回写的操作。

/*
 * Start writeback of `nr_pages' pages.  If `nr_pages' is zero, write back
 * the whole world.
 */
void wakeup_flusher_threads(long nr_pages, enum wb_reason reason)
{
	struct backing_dev_info *bdi;

	if (!nr_pages) {
		nr_pages = global_page_state(NR_FILE_DIRTY) +
				global_page_state(NR_UNSTABLE_NFS);
	}

	rcu_read_lock();
	list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) {
		if (!bdi_has_dirty_io(bdi))
			continue;
		__bdi_start_writeback(bdi, nr_pages, false, reason); // 开启回写的page的任务
	}
	rcu_read_unlock();
}

如果需要回写的数据为dirty状态，则主动调用__bdi_start_writeback来回写。

static void
__bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages,
		      bool range_cyclic, enum wb_reason reason)
{
	struct wb_writeback_work *work;

	/*
	 * This is WB_SYNC_NONE writeback, so if allocation fails just
	 * wakeup the thread for old dirty data writeback
	 */
	work = kzalloc(sizeof(*work), GFP_ATOMIC);
	if (!work) {
		trace_writeback_nowork(bdi);
		mod_delayed_work(bdi_wq, &bdi->wb.dwork, 0);
		return;
	}

	work->sync_mode	= WB_SYNC_NONE;
	work->nr_pages	= nr_pages;
	work->range_cyclic = range_cyclic;
	work->reason	= reason;

	bdi_queue_work(bdi, work);  // 将回写的work放入到任务队列中进行处理
}


static void bdi_queue_work(struct backing_dev_info *bdi,
			   struct wb_writeback_work *work)
{
	trace_writeback_queue(bdi, work);

	spin_lock_bh(&bdi->wb_lock);
	list_add_tail(&work->list, &bdi->work_list); 
	spin_unlock_bh(&bdi->wb_lock);

	mod_delayed_work(bdi_wq, &bdi->wb.dwork, 0);  // 放入队列中
}

通过mod_delayed_work函数放入到队列中进行延迟处理。

/**
 * mod_delayed_work - modify delay of or queue a delayed work
 * @wq: workqueue to use
 * @dwork: work to queue
 * @delay: number of jiffies to wait before queueing
 *
 * mod_delayed_work_on() on local CPU.
 */
static inline bool mod_delayed_work(struct workqueue_struct *wq,
				    struct delayed_work *dwork,
				    unsigned long delay)
{
	return mod_delayed_work_on(WORK_CPU_UNBOUND, wq, dwork, delay);
}


bool mod_delayed_work_on(int cpu, struct workqueue_struct *wq,
			 struct delayed_work *dwork, unsigned long delay)
{
	unsigned long flags;
	int ret;

	do {
		ret = try_to_grab_pending(&dwork->work, true, &flags);
	} while (unlikely(ret == -EAGAIN));

	if (likely(ret >= 0)) {
		__queue_delayed_work(cpu, wq, dwork, delay);  // 放入队列中进行处理
		local_irq_restore(flags);
	}

	/* -ENOENT from try_to_grab_pending() becomes %true */
	return ret;
}
EXPORT_SYMBOL_GPL(mod_delayed_work_on);

此时就会调用在work_queue中注册的回调函数。

bdi_writeback_workfn回写函数

/*
 * Handle writeback of dirty data for the device backed by this bdi. Also
 * reschedules periodically and does kupdated style flushing.
 */
void bdi_writeback_workfn(struct work_struct *work)
{
	struct bdi_writeback *wb = container_of(to_delayed_work(work),
						struct bdi_writeback, dwork);
	struct backing_dev_info *bdi = wb->bdi;
	long pages_written;

	set_worker_desc("flush-%s", dev_name(bdi->dev));
	current->flags |= PF_SWAPWRITE;

	if (likely(!current_is_workqueue_rescuer() ||
		   list_empty(&bdi->bdi_list))) {
		/*
		 * The normal path.  Keep writing back @bdi until its
		 * work_list is empty.  Note that this path is also taken
		 * if @bdi is shutting down even when we're running off the
		 * rescuer as work_list needs to be drained.
		 */
		do {
			pages_written = wb_do_writeback(wb, 0);   // 处理回调page
			trace_writeback_pages_written(pages_written);
		} while (!list_empty(&bdi->work_list));
	} else {
		/*
		 * bdi_wq can't get enough workers and we're running off
		 * the emergency worker.  Don't hog it.  Hopefully, 1024 is
		 * enough for efficient IO.
		 */
		pages_written = writeback_inodes_wb(&bdi->wb, 1024,
						    WB_REASON_FORKER_THREAD);
		trace_writeback_pages_written(pages_written);
	}

	if (!list_empty(&bdi->work_list) ||
	    (wb_has_dirty_io(wb) && dirty_writeback_interval))
		queue_delayed_work(bdi_wq, &wb->dwork,
			msecs_to_jiffies(dirty_writeback_interval * 10)); // 如果未处理完则放入队列继续处理

	current->flags &= ~PF_SWAPWRITE;
}


long wb_do_writeback(struct bdi_writeback *wb, int force_wait)
{
	struct backing_dev_info *bdi = wb->bdi;
	struct wb_writeback_work *work;
	long wrote = 0;

	set_bit(BDI_writeback_running, &wb->bdi->state);
	while ((work = get_next_work_item(bdi)) != NULL) {  // 获取队列
		/*
		 * Override sync mode, in case we must wait for completion
		 * because this thread is exiting now.
		 */
		if (force_wait)
			work->sync_mode = WB_SYNC_ALL;

		trace_writeback_exec(bdi, work);

		wrote += wb_writeback(wb, work);  // 处理work

		/*
		 * Notify the caller of completion if this is a synchronous
		 * work item, otherwise just free it.
		 */
		if (work->done)
			complete(work->done);
		else
			kfree(work);
	}

	/*
	 * Check for periodic writeback, kupdated() style
	 */
	wrote += wb_check_old_data_flush(wb);
	wrote += wb_check_background_flush(wb);
	clear_bit(BDI_writeback_running, &wb->bdi->state);

	return wrote;
}


static long wb_writeback(struct bdi_writeback *wb,
			 struct wb_writeback_work *work)
{
	unsigned long wb_start = jiffies;
	long nr_pages = work->nr_pages;
	unsigned long oldest_jif;
	struct inode *inode;
	long progress;

	oldest_jif = jiffies;
	work->older_than_this = &oldest_jif;

	spin_lock(&wb->list_lock);
	for (;;) {
		/*
		 * Stop writeback when nr_pages has been consumed
		 */
		if (work->nr_pages <= 0)
			break;

		/*
		 * Background writeout and kupdate-style writeback may
		 * run forever. Stop them if there is other work to do
		 * so that e.g. sync can proceed. They'll be restarted
		 * after the other works are all done.
		 */
		if ((work->for_background || work->for_kupdate) &&
		    !list_empty(&wb->bdi->work_list))
			break;

		/*
		 * For background writeout, stop when we are below the
		 * background dirty threshold
		 */
		if (work->for_background && !over_bground_thresh(wb->bdi))
			break;

		/*
		 * Kupdate and background works are special and we want to
		 * include all inodes that need writing. Livelock avoidance is
		 * handled by these works yielding to any other work so we are
		 * safe.
		 */
		if (work->for_kupdate) {
			oldest_jif = jiffies -
				msecs_to_jiffies(dirty_expire_interval * 10);
		} else if (work->for_background)
			oldest_jif = jiffies;

		trace_writeback_start(wb->bdi, work);
		if (list_empty(&wb->b_io))
			queue_io(wb, work);
		if (work->sb)
			progress = writeback_sb_inodes(work->sb, wb, work);  // 写入
		else
			progress = __writeback_inodes_wb(wb, work);
		trace_writeback_written(wb->bdi, work);

		wb_update_bandwidth(wb, wb_start);

		/*
		 * Did we write something? Try for more
		 *
		 * Dirty inodes are moved to b_io for writeback in batches.
		 * The completion of the current batch does not necessarily
		 * mean the overall work is done. So we keep looping as long
		 * as made some progress on cleaning pages or inodes.
		 */
		if (progress)
			continue;
		/*
		 * No more inodes for IO, bail
		 */
		if (list_empty(&wb->b_more_io))
			break;
		/*
		 * Nothing written. Wait for some inode to
		 * become available for writeback. Otherwise
		 * we'll just busyloop.
		 */
		if (!list_empty(&wb->b_more_io))  {
			trace_writeback_wait(wb->bdi, work);
			inode = wb_inode(wb->b_more_io.prev);
			spin_lock(&inode->i_lock);
			spin_unlock(&wb->list_lock);
			/* This function drops i_lock... */
			inode_sleep_on_writeback(inode);
			spin_lock(&wb->list_lock);
		}
	}
	spin_unlock(&wb->list_lock);

	return nr_pages - work->nr_pages;
}

在进入wb_writeback函数之后就是一个死循环进行处理。接着就调用writeback_sb_inodes进一步写入。

/*
 * Write a portion of b_io inodes which belong to @sb.
 *
 * Return the number of pages and/or inodes written.
 */
static long writeback_sb_inodes(struct super_block *sb,
				struct bdi_writeback *wb,
				struct wb_writeback_work *work)
{
	struct writeback_control wbc = {
		.sync_mode		= work->sync_mode,
		.tagged_writepages	= work->tagged_writepages,
		.for_kupdate		= work->for_kupdate,
		.for_background		= work->for_background,
		.range_cyclic		= work->range_cyclic,
		.range_start		= 0,
		.range_end		= LLONG_MAX,
	};
	unsigned long start_time = jiffies;
	long write_chunk;
	long wrote = 0;  /* count both pages and inodes */

	while (!list_empty(&wb->b_io)) {   // 检查队列是否为空
		struct inode *inode = wb_inode(wb->b_io.prev);

		if (inode->i_sb != sb) {
			if (work->sb) {
				/*
				 * We only want to write back data for this
				 * superblock, move all inodes not belonging
				 * to it back onto the dirty list.
				 */
				redirty_tail(inode, wb);
				continue;
			}

			/*
			 * The inode belongs to a different superblock.
			 * Bounce back to the caller to unpin this and
			 * pin the next superblock.
			 */
			break;
		}

		/*
		 * Don't bother with new inodes or inodes being freed, first
		 * kind does not need periodic writeout yet, and for the latter
		 * kind writeout is handled by the freer.
		 */
		spin_lock(&inode->i_lock);
		if (inode->i_state & (I_NEW | I_FREEING | I_WILL_FREE)) {
			spin_unlock(&inode->i_lock);
			redirty_tail(inode, wb);
			continue;
		}
		if ((inode->i_state & I_SYNC) && wbc.sync_mode != WB_SYNC_ALL) {
			/*
			 * If this inode is locked for writeback and we are not
			 * doing writeback-for-data-integrity, move it to
			 * b_more_io so that writeback can proceed with the
			 * other inodes on s_io.
			 *
			 * We'll have another go at writing back this inode
			 * when we completed a full scan of b_io.
			 */
			spin_unlock(&inode->i_lock);
			requeue_io(inode, wb);
			trace_writeback_sb_inodes_requeue(inode);
			continue;
		}
		spin_unlock(&wb->list_lock);

		/*
		 * We already requeued the inode if it had I_SYNC set and we
		 * are doing WB_SYNC_NONE writeback. So this catches only the
		 * WB_SYNC_ALL case.
		 */
		if (inode->i_state & I_SYNC) {
			/* Wait for I_SYNC. This function drops i_lock... */
			inode_sleep_on_writeback(inode);
			/* Inode may be gone, start again */
			spin_lock(&wb->list_lock);
			continue;
		}
		inode->i_state |= I_SYNC;   // 设置同步状态
		spin_unlock(&inode->i_lock);

		write_chunk = writeback_chunk_size(wb->bdi, work);
		wbc.nr_to_write = write_chunk;
		wbc.pages_skipped = 0;

		/*
		 * We use I_SYNC to pin the inode in memory. While it is set
		 * evict_inode() will wait so the inode cannot be freed.
		 */
		__writeback_single_inode(inode, &wbc);   // 调用single_inode函数进行处理

		work->nr_pages -= write_chunk - wbc.nr_to_write;
		wrote += write_chunk - wbc.nr_to_write;
		spin_lock(&wb->list_lock);
		spin_lock(&inode->i_lock);
		if (!(inode->i_state & I_DIRTY))
			wrote++;
		requeue_inode(inode, wb, &wbc);
		inode_sync_complete(inode);
		spin_unlock(&inode->i_lock);
		cond_resched_lock(&wb->list_lock);
		/*
		 * bail out to wb_writeback() often enough to check
		 * background threshold and other termination conditions.
		 */
		if (wrote) {
			if (time_is_before_jiffies(start_time + HZ / 10UL))
				break;
			if (work->nr_pages <= 0)
				break;
		}
	}
	return wrote;
}


static int
__writeback_single_inode(struct inode *inode, struct writeback_control *wbc)
{
	struct address_space *mapping = inode->i_mapping;
	long nr_to_write = wbc->nr_to_write;
	unsigned dirty;
	int ret;

	WARN_ON(!(inode->i_state & I_SYNC));

	trace_writeback_single_inode_start(inode, wbc, nr_to_write);

	ret = do_writepages(mapping, wbc);

	/*
	 * Make sure to wait on the data before writing out the metadata.
	 * This is important for filesystems that modify metadata on data
	 * I/O completion.
	 */
	if (wbc->sync_mode == WB_SYNC_ALL) {
		int err = filemap_fdatawait(mapping);
		if (ret == 0)
			ret = err;
	}

	/*
	 * Some filesystems may redirty the inode during the writeback
	 * due to delalloc, clear dirty metadata flags right before
	 * write_inode()
	 */
	spin_lock(&inode->i_lock);
	/* Clear I_DIRTY_PAGES if we've written out all dirty pages */
	if (!mapping_tagged(mapping, PAGECACHE_TAG_DIRTY))
		inode->i_state &= ~I_DIRTY_PAGES;
	dirty = inode->i_state & I_DIRTY;
	inode->i_state &= ~(I_DIRTY_SYNC | I_DIRTY_DATASYNC);
	spin_unlock(&inode->i_lock);
	/* Don't write the inode if only I_DIRTY_PAGES was set */
	if (dirty & (I_DIRTY_SYNC | I_DIRTY_DATASYNC)) {
		int err = write_inode(inode, wbc);
		if (ret == 0)
			ret = err;
	}
	trace_writeback_single_inode(inode, wbc, nr_to_write);
	return ret;
}


int do_writepages(struct address_space *mapping, struct writeback_control *wbc)
{
	int ret;

	if (wbc->nr_to_write <= 0)
		return 0;
	if (mapping->a_ops->writepages)
		ret = mapping->a_ops->writepages(mapping, wbc); // 调用文件系统注册的回调的函数
	else
		ret = generic_writepages(mapping, wbc);
	return ret;
}

至此，就将回写的操作分配到了底层的文件系统来进行处理。如果是ext4的文件系统则调用的为ext4_writepage函数进行下一步数据的写入操作。

总结

本文主要是通过资料跟着源码学习了一下数据回写的机制，主要是通过bdi机制，通过workqueue来进行数据的写入，并且在数据写入的时候其实就写到了page层就返回，然后再延迟写入机制将page中的数据落入到底层的存储介质中。触发的时机也引用到下文中的内容，即可以通过系统调用sync主动调用，也可以等待系统配置的定时来执行，在内存分配数据不足的时候也会触发数据回写的流程。由于本人才疏学浅，如有错误请批评指正。

page cache回写的几种触发方式_程序猿Ricky的日常干货的博客-CSDN博客_pagecache 如何回写

https://zhuanlan.zhihu.com/p/532262364

writeback bdi脏页回写原理linux内核源码解析_dongzhiyan_hjp的博客-CSDN博客_脏页回写

vm内核参数之内存脏页dirty_writeback_centisecs和dirty_expire_centisecs_Blue summer的博客-CSDN博客_dirty_writeback_centisecs

https://www.leviathan.vip/2019/06/01/Linux%E5%86%85%E6%A0%B8%E6%BA%90%E7%A0%81%E5%88%86%E6%9E%90-Page-Cache%E5%8E%9F%E7%90%86%E5%88%86%E6%9E%90/#%E5%90%8E%E5%8F%B0%E7%BA%BF%E7%A8%8B%E5%91%A8%E6%9C%9F%E5%9B%9E%E5%86%99

【linux内核源码】io操作之page cache - 简书

浅谈Linux dirty data配置 - 腾讯云开发者社区-腾讯云
Linux缓存回写——基于linux-4.15_SweeNeil的博客-CSDN博客
https://ywang-wnlo.github.io/posts/646202b9.html