linux内核之块设备驱动图解

2019独角兽企业重金招聘Python工程师标准>>>

块设备驱动程序的分层结构

                  

      块设备驱动：在Linux中，驱动对块设备的输入或输出(I/O)操作，都会向块设备发出一个请求，在驱动中用request结构体描述。但对于一些磁盘设备而言请求的速度很慢，这时候内核就提供一种队列的机制把这些I/O请求添加到队列中（即：请求队列），在驱动中用request_queue结构体描述。在向块设备提交这些请求前内核会先执行请求的合并和排序预操作，以提高访问的效率，然后再由内核中的I/O调度程序子系统来负责提交  I/O 请求，调度程序将磁盘资源分配给系统中所有挂起的块 I/O  请求，其工作是管理块设备的请求队列，决定队列中的请求的排列顺序以及什么时候派发请求到设备。

      通用块层(Generic Block Layer)：负责维持一个I/O请求在上层文件系统与底层物理磁盘之间的关系。在通用块层中，通常用一个bio结构体来对应一个I/O请求。

      Linux提供了一个gendisk数据结构体，用来表示一个独立的磁盘设备或分区，用于对底层物理磁盘进行访问。在gendisk中有一个类似字符设备中file_operations的硬件操作结构指针，是block_device_operations结构体。

      IO调度层：当多个请求提交给块设备时，执行效率依赖于请求的顺序。如果所有的请求是同一个方向（如：写数据），执行效率是最大的。内核在调用块设备驱动程序例程处理请求之前，先收集I/O请求并将请求排序，然后，将连续扇区操作的多个请求进行合并以提高执行效率（内核算法会自己做，不用你管），对I/O请求排序的算法称为电梯算法（elevator algorithm）。电梯算法在I/O调度层完成。内核提供了不同类型的电梯算法，电梯算法有

• noop（实现简单的FIFO，基本的直接合并与排序）
• anticipatory（延迟I/O请求，进行临界区的优化排序）
• Deadline（针对anticipatory缺点进行改善，降低延迟时间）
• Cfq（均匀分配I/O带宽，公平机制）

     映射层(Mapping Layer)：起映射作用，将文件访问映射为设备的访问。

     VFS：对各种文件系统进行统一封装，为用户程序访问文件提供统一的接口，包含ext2，FAT,NFS,设备文件。

     磁盘缓存(Caches):将访问频率很高的文件放入其中。

块设备驱动内核数据结构关系图

                

块设备驱动实例

/*
 * Sample disk driver, from the beginning.
 */

#include <linux/module.h>
#include <linux/moduleparam.h>
#include <linux/init.h>

#include <linux/sched.h>
#include <linux/kernel.h>	/* printk() */
#include <linux/slab.h>		/* kmalloc() */
#include <linux/fs.h>		/* everything... */
#include <linux/errno.h>	/* error codes */
#include <linux/types.h>	/* size_t */
#include <linux/fcntl.h>	/* O_ACCMODE */
#include <linux/hdreg.h>	/* HDIO_GETGEO */
#include <linux/kdev_t.h>
#include <linux/vmalloc.h>
#include <linux/genhd.h>
#include <linux/blk-mq.h>
#include <linux/buffer_head.h>	/* invalidate_bdev */
#include <linux/bio.h>

#ifndef BLK_STS_OK
typedef int blk_status_t;
#define BLK_STS_OK 0
#define OLDER_KERNEL 1
#endif

#ifndef BLK_STS_IOERR
#define BLK_STS_IOERR 10
#endif

#ifndef SECTOR_SHIFT
#define SECTOR_SHIFT 9
#endif

/* FIXME: implement these macros in kernel mainline */
#define size_to_sectors(size) ((size) >> SECTOR_SHIFT)
#define sectors_to_size(size) ((size) << SECTOR_SHIFT)

MODULE_LICENSE("Dual BSD/GPL");

static int sbull_major;
module_param(sbull_major, int, 0);
static int logical_block_size = 512;
module_param(logical_block_size, int, 0);
static char* disk_size = "256M";
module_param(disk_size, charp, 0);
static int ndevices = 1;
module_param(ndevices, int, 0);
static bool debug = false;
module_param(debug, bool, false);

/*
 * The different "request modes" we can use.
 */
enum {
	RM_SIMPLE  = 0,	/* The extra-simple request function */
	RM_FULL    = 1,	/* The full-blown version */
	RM_NOQUEUE = 2,	/* Use make_request */
};

/*
 * Minor number and partition management.
 */
#define SBULL_MINORS	16

/*
 * We can tweak our hardware sector size, but the kernel talks to us
 * in terms of small sectors, always.
 */
#define KERNEL_SECTOR_SIZE	512

/*
 * The internal representation of our device.
 */
struct sbull_dev {
	int size;                       /* Device size in sectors */
	u8 *data;                       /* The data array */
	spinlock_t lock;                /* For mutual exclusion */
	struct request_queue *queue;    /* The device request queue */
	struct gendisk *gd;             /* The gendisk structure */
	struct blk_mq_tag_set tag_set;
};

static struct sbull_dev *Devices;

/* Handle an I/O request */
static blk_status_t sbull_transfer(struct sbull_dev *dev, unsigned long sector,
			unsigned long nsect, char *buffer, int op)

{
	unsigned long offset = sectors_to_size(sector);
	unsigned long nbytes = sectors_to_size(nsect);

	if ((offset + nbytes) > dev->size) {
		pr_notice("Beyond-end write (%ld %ld)\n", offset, nbytes);
		return BLK_STS_IOERR;
	}

	if  (debug)
		pr_info("%s: %s, sector: %ld, nsectors: %ld, offset: %ld,"
			       " nbytes: %ld",
			dev->gd->disk_name,
			op == REQ_OP_WRITE ? "WRITE" : "READ", sector, nsect,
			offset, nbytes);

	/* will be only REQ_OP_READ or REQ_OP_WRITE */
	if (op == REQ_OP_WRITE)
		memcpy(dev->data + offset, buffer, nbytes);
	else
		memcpy(buffer, dev->data + offset, nbytes);

	return BLK_STS_OK;
}

static blk_status_t sbull_queue_rq(struct blk_mq_hw_ctx *hctx,
		const struct blk_mq_queue_data *bd)
{
	struct request *req = bd->rq;
	struct sbull_dev *dev = req->rq_disk->private_data;
	int op = req_op(req);
	blk_status_t ret;

	blk_mq_start_request(req);
	spin_lock(&dev->lock);

	if (op != REQ_OP_READ && op != REQ_OP_WRITE) {
		pr_notice("Skip non-fs request\n");
		blk_mq_end_request(req, BLK_STS_IOERR);
		spin_unlock(&dev->lock);
		return BLK_STS_IOERR;
	}

	ret = sbull_transfer(dev, blk_rq_pos(req),
			blk_rq_cur_sectors(req),
			bio_data(req->bio), op);

	blk_mq_end_request(req, ret);
	spin_unlock(&dev->lock);
	return ret;
}

/*
 * The device operations structure.
 */
static const struct block_device_operations sbull_ops = {
	.owner		= THIS_MODULE,
};

static const struct blk_mq_ops sbull_mq_ops = {
	.queue_rq = sbull_queue_rq,
};

static struct request_queue *create_req_queue(struct blk_mq_tag_set *set)
{
	struct request_queue *q;

#ifndef OLDER_KERNEL
	q = blk_mq_init_sq_queue(set, &sbull_mq_ops,
			2, BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_BLOCKING);
#else
	int ret;

	memset(set, 0, sizeof(*set));
	set->ops = &sbull_mq_ops;
	set->nr_hw_queues = 1;
	/*set->nr_maps = 1;*/
	set->queue_depth = 2;
	set->numa_node = NUMA_NO_NODE;
	set->flags = BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_BLOCKING;

	ret = blk_mq_alloc_tag_set(set);
	if (ret)
		return ERR_PTR(ret);

	q = blk_mq_init_queue(set);
	if (IS_ERR(q)) {
		blk_mq_free_tag_set(set);
		return q;
	}
#endif

	return q;
}

/*
 * Set up our internal device.
 */
static void setup_device(struct sbull_dev *dev, int which)
{
	long long sbull_size = memparse(disk_size, NULL);

	memset(dev, 0, sizeof(struct sbull_dev));
	dev->size = sbull_size;
	dev->data = vzalloc(dev->size);
	if (dev->data == NULL) {
		pr_notice("vmalloc failure.\n");
		return;
	}
	spin_lock_init(&dev->lock);

	dev->queue = create_req_queue(&dev->tag_set);
	if (IS_ERR(dev->queue))
		goto out_vfree;

	blk_queue_logical_block_size(dev->queue, logical_block_size);
	dev->queue->queuedata = dev;
	/*
	 * And the gendisk structure.
	 */
	dev->gd = alloc_disk(SBULL_MINORS);
	if (!dev->gd) {
		pr_notice("alloc_disk failure\n");
		goto out_vfree;
	}
	dev->gd->major = sbull_major;
	dev->gd->first_minor = which*SBULL_MINORS;
	dev->gd->fops = &sbull_ops;
	dev->gd->queue = dev->queue;
	dev->gd->private_data = dev;
	snprintf(dev->gd->disk_name, 32, "sbull%c", which + 'a');
	set_capacity(dev->gd, size_to_sectors(sbull_size));
	add_disk(dev->gd);
	return;

out_vfree:
	if (dev->data)
		vfree(dev->data);
}



static int __init sbull_init(void)
{
	int i;
	/*
	 * Get registered.
	 */
	sbull_major = register_blkdev(sbull_major, "sbull");
	if (sbull_major <= 0) {
		pr_warn("sbull: unable to get major number\n");
		return -EBUSY;
	}

	/*
	 * Allocate the device array, and initialize each one.
	 */
	Devices = kmalloc(ndevices * sizeof(struct sbull_dev), GFP_KERNEL);
	if (Devices == NULL)
		goto out_unregister;
	for (i = 0; i < ndevices; i++)
		setup_device(Devices + i, i);

	return 0;

out_unregister:
	unregister_blkdev(sbull_major, "sbull");
	return -ENOMEM;
}

static void sbull_exit(void)
{
	int i;

	for (i = 0; i < ndevices; i++) {
		struct sbull_dev *dev = Devices + i;

		if (dev->gd) {
			del_gendisk(dev->gd);
			put_disk(dev->gd);
		}
		if (dev->queue)
			blk_cleanup_queue(dev->queue);

		if (dev->data)
			vfree(dev->data);
	}
	unregister_blkdev(sbull_major, "sbull");
	kfree(Devices);
}

module_init(sbull_init);
module_exit(sbull_exit);

     可以编写Makefile：

obj-m += sbull.o

CURRENT_PATH:=$(shell pwd)
LINUX_KERNEL:=$(shell uname -r)
LINUX_KERNEL_PATH:=/usr/src/kernels/$(LINUX_KERNEL)

all:
    make -C $(LINUX_KERNEL_PATH) M=$(CURRENT_PATH) modules
clean:
    make -C $(LINUX_KERNEL_PATH) M=$(CURRENT_PATH) clean

 

或者直接使用linux内核中提供的/drivers/block/brd.c

/*
 * Ram backed block device driver.
 *
 * Copyright (C) 2007 Nick Piggin
 * Copyright (C) 2007 Novell Inc.
 *
 * Parts derived from drivers/block/rd.c, and drivers/block/loop.c, copyright
 * of their respective owners.
 */

#include <linux/init.h>
#include <linux/initrd.h>
#include <linux/module.h>
#include <linux/moduleparam.h>
#include <linux/major.h>
#include <linux/blkdev.h>
#include <linux/bio.h>
#include <linux/highmem.h>
#include <linux/mutex.h>
#include <linux/radix-tree.h>
#include <linux/fs.h>
#include <linux/slab.h>
#include <linux/backing-dev.h>

#include <linux/uaccess.h>

#define PAGE_SECTORS_SHIFT	(PAGE_SHIFT - SECTOR_SHIFT)
#define PAGE_SECTORS		(1 << PAGE_SECTORS_SHIFT)

static int ramdisk_major;

/*
 * Each block ramdisk device has a radix_tree brd_pages of pages that stores
 * the pages containing the block device's contents. A brd page's ->index is
 * its offset in PAGE_SIZE units. This is similar to, but in no way connected
 * with, the kernel's pagecache or buffer cache (which sit above our block
 * device).
 */
struct brd_device {
	int		brd_number;

	struct request_queue	*brd_queue;
	struct gendisk		*brd_disk;
	struct list_head	brd_list;

	/*
	 * Backing store of pages and lock to protect it. This is the contents
	 * of the block device.
	 */
	spinlock_t		brd_lock;
	struct radix_tree_root	brd_pages;
};

/*
 * Look up and return a brd's page for a given sector.
 */
static struct page *brd_lookup_page(struct brd_device *brd, sector_t sector)
{
	pgoff_t idx;
	struct page *page;

	/*
	 * The page lifetime is protected by the fact that we have opened the
	 * device node -- brd pages will never be deleted under us, so we
	 * don't need any further locking or refcounting.
	 *
	 * This is strictly true for the radix-tree nodes as well (ie. we
	 * don't actually need the rcu_read_lock()), however that is not a
	 * documented feature of the radix-tree API so it is better to be
	 * safe here (we don't have total exclusion from radix tree updates
	 * here, only deletes).
	 */
	rcu_read_lock();
	idx = sector >> PAGE_SECTORS_SHIFT; /* sector to page index */
	page = radix_tree_lookup(&brd->brd_pages, idx);
	rcu_read_unlock();

	BUG_ON(page && page->index != idx);

	return page;
}

/*
 * Look up and return a brd's page for a given sector.
 * If one does not exist, allocate an empty page, and insert that. Then
 * return it.
 */
static struct page *brd_insert_page(struct brd_device *brd, sector_t sector)
{
	pgoff_t idx;
	struct page *page;
	gfp_t gfp_flags;

	page = brd_lookup_page(brd, sector);
	if (page)
		return page;

	/*
	 * Must use NOIO because we don't want to recurse back into the
	 * block or filesystem layers from page reclaim.
	 */
	gfp_flags = GFP_NOIO | __GFP_ZERO | __GFP_HIGHMEM;
	page = alloc_page(gfp_flags);
	if (!page)
		return NULL;

	if (radix_tree_preload(GFP_NOIO)) {
		__free_page(page);
		return NULL;
	}

	spin_lock(&brd->brd_lock);
	idx = sector >> PAGE_SECTORS_SHIFT;
	page->index = idx;
	if (radix_tree_insert(&brd->brd_pages, idx, page)) {
		__free_page(page);
		page = radix_tree_lookup(&brd->brd_pages, idx);
		BUG_ON(!page);
		BUG_ON(page->index != idx);
	}
	spin_unlock(&brd->brd_lock);

	radix_tree_preload_end();

	return page;
}

/*
 * Free all backing store pages and radix tree. This must only be called when
 * there are no other users of the device.
 */
#define FREE_BATCH 16
static void brd_free_pages(struct brd_device *brd)
{
	unsigned long pos = 0;
	struct page *pages[FREE_BATCH];
	int nr_pages;

	do {
		int i;

		nr_pages = radix_tree_gang_lookup(&brd->brd_pages,
				(void **)pages, pos, FREE_BATCH);

		for (i = 0; i < nr_pages; i++) {
			void *ret;

			BUG_ON(pages[i]->index < pos);
			pos = pages[i]->index;
			ret = radix_tree_delete(&brd->brd_pages, pos);
			BUG_ON(!ret || ret != pages[i]);
			__free_page(pages[i]);
		}

		pos++;

		/*
		 * This assumes radix_tree_gang_lookup always returns as
		 * many pages as possible. If the radix-tree code changes,
		 * so will this have to.
		 */
	} while (nr_pages == FREE_BATCH);
}

/*
 * copy_to_brd_setup must be called before copy_to_brd. It may sleep.
 */
static int copy_to_brd_setup(struct brd_device *brd, sector_t sector, size_t n)
{
	unsigned int offset = (sector & (PAGE_SECTORS-1)) << SECTOR_SHIFT;
	size_t copy;

	copy = min_t(size_t, n, PAGE_SIZE - offset);
	if (!brd_insert_page(brd, sector))
		return -ENOSPC;
	if (copy < n) {
		sector += copy >> SECTOR_SHIFT;
		if (!brd_insert_page(brd, sector))
			return -ENOSPC;
	}
	return 0;
}

/*
 * Copy n bytes from src to the brd starting at sector. Does not sleep.
 */
static void copy_to_brd(struct brd_device *brd, const void *src,
			sector_t sector, size_t n)
{
	struct page *page;
	void *dst;
	unsigned int offset = (sector & (PAGE_SECTORS-1)) << SECTOR_SHIFT;
	size_t copy;

    pr_info("copy_to_ramdisk: sector: %d,size: %d\n",(int)sector,(int)n);

	copy = min_t(size_t, n, PAGE_SIZE - offset);
	page = brd_lookup_page(brd, sector);
	BUG_ON(!page);

	dst = kmap_atomic(page);
	memcpy(dst + offset, src, copy);
	kunmap_atomic(dst);

	if (copy < n) {
		src += copy;
		sector += copy >> SECTOR_SHIFT;
		copy = n - copy;
		page = brd_lookup_page(brd, sector);
		BUG_ON(!page);

		dst = kmap_atomic(page);
		memcpy(dst, src, copy);
		kunmap_atomic(dst);
	}
}

/*
 * Copy n bytes to dst from the brd starting at sector. Does not sleep.
 */
static void copy_from_brd(void *dst, struct brd_device *brd,
			sector_t sector, size_t n)
{
	struct page *page;
	void *src;
	unsigned int offset = (sector & (PAGE_SECTORS-1)) << SECTOR_SHIFT;
	size_t copy;

    pr_info("copy_from_ramdisk: sector: %d,size: %d\n",(int)sector,(int)n);

	copy = min_t(size_t, n, PAGE_SIZE - offset);
	page = brd_lookup_page(brd, sector);
	if (page) {
		src = kmap_atomic(page);
		memcpy(dst, src + offset, copy);
		kunmap_atomic(src);
	} else
		memset(dst, 0, copy);

	if (copy < n) {
		dst += copy;
		sector += copy >> SECTOR_SHIFT;
		copy = n - copy;
		page = brd_lookup_page(brd, sector);
		if (page) {
			src = kmap_atomic(page);
			memcpy(dst, src, copy);
			kunmap_atomic(src);
		} else
			memset(dst, 0, copy);
	}
}

/*
 * Process a single bvec of a bio.
 */
static int brd_do_bvec(struct brd_device *brd, struct page *page,
			unsigned int len, unsigned int off, unsigned int op,
			sector_t sector)
{
	void *mem;
	int err = 0;

	if (op_is_write(op)) {
		err = copy_to_brd_setup(brd, sector, len);
		if (err)
			goto out;
	}

	mem = kmap_atomic(page);
	if (!op_is_write(op)) {
		copy_from_brd(mem + off, brd, sector, len);
		flush_dcache_page(page);
	} else {
		flush_dcache_page(page);
		copy_to_brd(brd, mem + off, sector, len);
	}
	kunmap_atomic(mem);

out:
	return err;
}

static blk_qc_t brd_make_request(struct request_queue *q, struct bio *bio)
{
	struct brd_device *brd = bio->bi_disk->private_data;
	struct bio_vec bvec;
	sector_t sector;
	struct bvec_iter iter;

	sector = bio->bi_iter.bi_sector;
	if (bio_end_sector(bio) > get_capacity(bio->bi_disk))
		goto io_error;

	bio_for_each_segment(bvec, bio, iter) {
		unsigned int len = bvec.bv_len;
		int err;

		err = brd_do_bvec(brd, bvec.bv_page, len, bvec.bv_offset,
				  bio_op(bio), sector);
		if (err)
			goto io_error;
		sector += len >> SECTOR_SHIFT;
	}

	bio_endio(bio);
	return BLK_QC_T_NONE;
io_error:
	bio_io_error(bio);
	return BLK_QC_T_NONE;
}

static int brd_rw_page(struct block_device *bdev, sector_t sector,
		       struct page *page, unsigned int op)
{
	struct brd_device *brd = bdev->bd_disk->private_data;
	int err;

	if (PageTransHuge(page))
		return -ENOTSUPP;
	err