Linux内核4.14版本——mtd子系统(4)——块mtd设备缓冲操作（mtdblock.c）

目录

1. 简介

2. 缓冲数据结构(struct mtd_blktrans_dev)

3. mtdblock_open

4. mtdblock_release

5. mtdblock_writesect

6. mtdblock_readsect

7. mtdblock_flush

8. mtdblock_add_mtd->add_mtd_blktrans_dev

---

1. 简介

      从这架构图中，我们今天分析的是mtdblock.c文件。

      这个文件中并没有增加mtd块设备，mtd块设备的代码文件是Mtd_blkdevs.c。它的功能是为mtd块设备读写提供缓冲操作。
      另外还有一个文件mtdblock_ro.c，它定义的是mtd块设备缓冲的只读操作。

static struct mtd_blktrans_ops mtdblock_tr = {
	.name		= "mtdblock",
	.major		= MTD_BLOCK_MAJOR,
	.part_bits	= 0,
	.blksize 	= 512,
	.open		= mtdblock_open,
	.flush		= mtdblock_flush,
	.release	= mtdblock_release,
	.readsect	= mtdblock_readsect,
	.writesect	= mtdblock_writesect,
	.add_mtd	= mtdblock_add_mtd,         //增加一个mtd_blktrans_dev
	.remove_dev	= mtdblock_remove_dev,
	.owner		= THIS_MODULE,
};

static int __init init_mtdblock(void)
{
	return register_mtd_blktrans(&mtdblock_tr);
}

static void __exit cleanup_mtdblock(void)
{
	deregister_mtd_blktrans(&mtdblock_tr);
}

2. 缓冲数据结构(struct mtd_blktrans_dev)

struct mtd_blktrans_dev {
	struct mtd_blktrans_ops *tr;
	struct list_head list;
	struct mtd_info *mtd;
	struct mutex lock;
	int devnum;           
	bool bg_stop;
	unsigned long size;
	int readonly;
	int open;
	struct kref ref;
	struct gendisk *disk;
	struct attribute_group *disk_attributes;
	struct workqueue_struct *wq;
	struct work_struct work;
	struct request_queue *rq;
	spinlock_t queue_lock;
	void *priv;
	fmode_t file_mode;
};

devnum: 分区序号，等于mtd->index
size:   分区大小，单位是512字节
readonly: 是否是只读的
priv:    指向gen_disk结构

      mtd_blktrans_dev对应于mtd_info，每个分区有一个mtd_blktrans_dev实体。

3. mtdblock_open

static int mtdblock_open(struct mtd_blktrans_dev *mbd)
{
	struct mtdblk_dev *mtdblk = container_of(mbd, struct mtdblk_dev, mbd);

	pr_debug("mtdblock_open\n");

	if (mtdblk->count) {
		mtdblk->count++;
		return 0;
	}

	/* OK, it's not open. Create cache info for it */
	mtdblk->count = 1;
	mutex_init(&mtdblk->cache_mutex);
	mtdblk->cache_state = STATE_EMPTY;
	if (!(mbd->mtd->flags & MTD_NO_ERASE) && mbd->mtd->erasesize) {
		mtdblk->cache_size = mbd->mtd->erasesize;
		mtdblk->cache_data = NULL;
	}

	pr_debug("ok\n");

	return 0;
}

4. mtdblock_release

static void mtdblock_release(struct mtd_blktrans_dev *mbd)
{
	struct mtdblk_dev *mtdblk = container_of(mbd, struct mtdblk_dev, mbd);

	pr_debug("mtdblock_release\n");

	mutex_lock(&mtdblk->cache_mutex);
	write_cached_data(mtdblk);
	mutex_unlock(&mtdblk->cache_mutex);

	if (!--mtdblk->count) {
		/*
		 * It was the last usage. Free the cache, but only sync if
		 * opened for writing.
		 */
		if (mbd->file_mode & FMODE_WRITE)
			mtd_sync(mbd->mtd);
		vfree(mtdblk->cache_data);
	}

	pr_debug("ok\n");
}

      它检查检查mtdblk_dev的使用计数，若为0，则释放内存。

5. mtdblock_writesect

static int mtdblock_writesect(struct mtd_blktrans_dev *dev,
			      unsigned long block, char *buf)
{
	struct mtdblk_dev *mtdblk = container_of(dev, struct mtdblk_dev, mbd);
	if (unlikely(!mtdblk->cache_data && mtdblk->cache_size)) {
		mtdblk->cache_data = vmalloc(mtdblk->mbd.mtd->erasesize);
		if (!mtdblk->cache_data)
			return -EINTR;
		/* -EINTR is not really correct, but it is the best match
		 * documented in man 2 write for all cases.  We could also
		 * return -EAGAIN sometimes, but why bother?
		 */
	}
	return do_cached_write(mtdblk, block<<9, 512, buf);
}

      如果mtdblk->cache_data为空，则分配内存。最终调用do_cached_write把512字节数据写入到block<<9的地址处。

static int do_cached_write (struct mtdblk_dev *mtdblk, unsigned long pos,
			    int len, const char *buf)
{
	struct mtd_info *mtd = mtdblk->mbd.mtd;
	unsigned int sect_size = mtdblk->cache_size;
	size_t retlen;
	int ret;

	pr_debug("mtdblock: write on \"%s\" at 0x%lx, size 0x%x\n",
		mtd->name, pos, len);

	if (!sect_size)      //直接写入
		return mtd_write(mtd, pos, len, &retlen, buf);
	//单次写入的数据大小不能超过sect_size，写入时如果小于sect_size,则先写入到cache
	while (len > 0) {
		unsigned long sect_start = (pos/sect_size)*sect_size;
		unsigned int offset = pos - sect_start;
		unsigned int size = sect_size - offset;
		if( size > len )
			size = len;

		if (size == sect_size) {        // 如果刚好是一个sect的大小，则不经cache直接写入
			/*
			 * We are covering a whole sector.  Thus there is no
			 * need to bother with the cache while it may still be
			 * useful for other partial writes.
			 */
			ret = erase_write (mtd, pos, size, buf);
			if (ret)
				return ret;
		} else {                        // 使用cache
			/* Partial sector: need to use the cache */
			//如果cache有数据且cache的数据和写入的数据不在同一个sect中，则先把cache中的数据写入
			if (mtdblk->cache_state == STATE_DIRTY &&
			    mtdblk->cache_offset != sect_start) {
				ret = write_cached_data(mtdblk);
				if (ret)
					return ret;
			}

			//如果cache中没有数据，则先把FLASH中对应的数据读出来放到cache
			if (mtdblk->cache_state == STATE_EMPTY ||
			    mtdblk->cache_offset != sect_start) {
				/* fill the cache with the current sector */
				mtdblk->cache_state = STATE_EMPTY;
				ret = mtd_read(mtd, sect_start, sect_size,
					       &retlen, mtdblk->cache_data);
				if (ret)
					return ret;
				if (retlen != sect_size)
					return -EIO;

				mtdblk->cache_offset = sect_start;
				mtdblk->cache_size = sect_size;
				mtdblk->cache_state = STATE_CLEAN;
			}

			/* write data to our local cache */
			memcpy (mtdblk->cache_data + offset, buf, size);
			mtdblk->cache_state = STATE_DIRTY;
		}

		buf += size;
		pos += size;
		len -= size;
	}

	return 0;
}

static int write_cached_data (struct mtdblk_dev *mtdblk)
{
	struct mtd_info *mtd = mtdblk->mbd.mtd;
	int ret;

	if (mtdblk->cache_state != STATE_DIRTY)
		return 0;

	pr_debug("mtdblock: writing cached data for \"%s\" "
			"at 0x%lx, size 0x%x\n", mtd->name,
			mtdblk->cache_offset, mtdblk->cache_size);
	// erase_write调用mtd_info的相应操作
	ret = erase_write (mtd, mtdblk->cache_offset,
			   mtdblk->cache_size, mtdblk->cache_data);
	if (ret)
		return ret;

	/*
	 * Here we could arguably set the cache state to STATE_CLEAN.
	 * However this could lead to inconsistency since we will not
	 * be notified if this content is altered on the flash by other
	 * means.  Let's declare it empty and leave buffering tasks to
	 * the buffer cache instead.
	 */
	mtdblk->cache_state = STATE_EMPTY;
	return 0;
}

6. mtdblock_readsect

static int mtdblock_readsect(struct mtd_blktrans_dev *dev,
			      unsigned long block, char *buf)
{
	struct mtdblk_dev *mtdblk = container_of(dev, struct mtdblk_dev, mbd);
	return do_cached_read(mtdblk, block<<9, 512, buf);
}

static int do_cached_read (struct mtdblk_dev *mtdblk, unsigned long pos,
			   int len, char *buf)
{
	struct mtd_info *mtd = mtdblk->mbd.mtd;
	unsigned int sect_size = mtdblk->cache_size;
	size_t retlen;
	int ret;

	pr_debug("mtdblock: read on \"%s\" at 0x%lx, size 0x%x\n",
			mtd->name, pos, len);

	if (!sect_size)
		return mtd_read(mtd, pos, len, &retlen, buf);

	while (len > 0) {
		unsigned long sect_start = (pos/sect_size)*sect_size;
		unsigned int offset = pos - sect_start;
		unsigned int size = sect_size - offset;
		if (size > len)
			size = len;

		/*
		 * Check if the requested data is already cached
		 * Read the requested amount of data from our internal cache if it
		 * contains what we want, otherwise we read the data directly
		 * from flash.
		 */
		 // 检查读取的数据是否已经在cache中
		if (mtdblk->cache_state != STATE_EMPTY &&
		    mtdblk->cache_offset == sect_start) {
			memcpy (buf, mtdblk->cache_data + offset, size);
		} else {
			ret = mtd_read(mtd, pos, size, &retlen, buf);
			if (ret)
				return ret;
			if (retlen != size)
				return -EIO;
		}

		buf += size;
		pos += size;
		len -= size;
	}

	return 0;
}

7. mtdblock_flush

它将cache中数据写入到存储

static int write_cached_data (struct mtdblk_dev *mtdblk)
{
	struct mtd_info *mtd = mtdblk->mbd.mtd;
	int ret;

	if (mtdblk->cache_state != STATE_DIRTY)
		return 0;

	pr_debug("mtdblock: writing cached data for \"%s\" "
			"at 0x%lx, size 0x%x\n", mtd->name,
			mtdblk->cache_offset, mtdblk->cache_size);

	ret = erase_write (mtd, mtdblk->cache_offset,
			   mtdblk->cache_size, mtdblk->cache_data);
	if (ret)
		return ret;

	/*
	 * Here we could arguably set the cache state to STATE_CLEAN.
	 * However this could lead to inconsistency since we will not
	 * be notified if this content is altered on the flash by other
	 * means.  Let's declare it empty and leave buffering tasks to
	 * the buffer cache instead.
	 */
	mtdblk->cache_state = STATE_EMPTY;
	return 0;
}

8. mtdblock_add_mtd->add_mtd_blktrans_dev

static void mtdblock_add_mtd(struct mtd_blktrans_ops *tr, struct mtd_info *mtd)
{
	struct mtdblk_dev *dev = kzalloc(sizeof(*dev), GFP_KERNEL);

	if (!dev)
		return;

	dev->mbd.mtd = mtd;
	dev->mbd.devnum = mtd->index;

	dev->mbd.size = mtd->size >> 9;
	dev->mbd.tr = tr;

	if (!(mtd->flags & MTD_WRITEABLE))
		dev->mbd.readonly = 1;

	if (add_mtd_blktrans_dev(&dev->mbd))
		kfree(dev);
}

      mtdblock_add_mtd这个函数，后面会经常用到，它是把mtd设备真正加入到mtd块设备中。