Ramdisk源代码分析

 /*
* ramdisk.c - Multiple RAM disk driver - gzip-loading version - v. 0.8 beta.
*
* (C) Chad Page, Theodore Ts'o, et. al, 1995.
*
* This RAM disk is designed to have filesystems created on it and mounted
* just like a regular floppy disk.
*
* It also does something suggested by Linus: use the buffer cache as the
* RAM disk data. This makes it possible to dynamically allocate the RAM disk
* buffer - with some consequences I have to deal with as I write this.
*
* This code is based on the original ramdisk.c, written mostly by
* Theodore Ts'o (TYT) in 1991. The code was largely rewritten by
* Chad Page to use the buffer cache to store the RAM disk data in
* 1995; Theodore then took over the driver again, and cleaned it up
* for inclusion in the mainline kernel.
*
* The original CRAMDISK code was written by Richard Lyons, and
* adapted by Chad Page to use the new RAM disk interface. Theodore
* Ts'o rewrote it so that both the compressed RAM disk loader and the
* kernel decompressor uses the same inflate.c codebase. The RAM disk
* loader now also loads into a dynamic (buffer cache based) RAM disk,
* not the old static RAM disk. Support for the old static RAM disk has
* been completely removed.
*
* Loadable module support added by Tom Dyas.
*
*/

#include <linux/string.h>
#include <linux/slab.h>
#include <asm/atomic.h>
#include <linux/bio.h>
#include <linux/module.h>
#include <linux/moduleparam.h>
#include <linux/init.h>
#include <linux/pagemap.h>
#include <linux/blkdev.h>
#include <linux/genhd.h>
#include <linux/buffer_head.h>       /* for invalidate_bdev() */
#include <linux/backing-dev.h>
#include <linux/blkpg.h>
#include <linux/writeback.h>

#include <asm/uaccess.h>

/* Various static variables go here. Most are used only in the RAM disk code.
*/

static struct gendisk *rd_disks[ CONFIG_BLK_DEV_RAM_COUNT];    //the default value is 16
static struct block_device *rd_bdev[CONFIG_BLK_DEV_RAM_COUNT];/* Protected device data */
static struct request_queue *rd_queue[CONFIG_BLK_DEV_RAM_COUNT];

/*
* Parameters for the boot-loading of the RAM disk. These are set by
* init/main.c (from arguments to the kernel command line) or from the
* architecture-specific setup routine (from the stored boot sector
* information).
*/
int rd_size = CONFIG_BLK_DEV_RAM_SIZE; // 4M     /* Size of the RAM disks */
/*
* It would be very desirable to have a soft-blocksize (that in the case
* of the ramdisk driver is also the hardblocksize ;) of PAGE_SIZE because
* doing that we'll achieve a far better MM footprint. Using a rd_blocksize of
* BLOCK_SIZE in the worst case we'll make PAGE_SIZE/BLOCK_SIZE buffer-pages
* unfreeable. With a rd_blocksize of PAGE_SIZE instead we are sure that only
* 1 page will be protected. Depending on the size of the ramdisk you
* may want to change the ramdisk blocksize to achieve a better or worse MM
* behaviour. The default is still BLOCK_SIZE (needed by rd_load_image that
* supposes the filesystem in the image uses a BLOCK_SIZE blocksize).
*/
static int rd_blocksize = CONFIG_BLK_DEV_RAM_BLOCKSIZE; //the block size

/*
* Copyright (C) 2000 Linus Torvalds.
*               2000 Transmeta Corp.
* aops copied from ramfs.
*/

/*
* If a ramdisk page has buffers, some may be uptodate and some may be not.
* To bring the page uptodate we zero out the non-uptodate buffers. The
* page must be locked.
*/
static void make_page_uptodate(struct page *page)
{
   if (page_has_buffers(page)) {
       struct buffer_head *bh = page_buffers(page);
       struct buffer_head *head = bh;

       do {
           if (!buffer_uptodate(bh)) {
               memset(bh->b_data, 0, bh->b_size);
               /*
               * akpm: I'm totally undecided about this. The
               * buffer has just been magically brought "up to
               * date", but nobody should want to be reading
               * it anyway, because it hasn't been used for
               * anything yet. It is still in a "not read
               * from disk yet" state.
               *
               * But non-uptodate buffers against an uptodate
               * page are against the rules. So do it anyway.
               */
               set_buffer_uptodate(bh);
           }
       } while ((bh = bh->b_this_page) != head);
   } else {
       memset(page_address(page), 0, PAGE_CACHE_SIZE);
   }
   flush_dcache_page(page);
   SetPageUptodate(page);
}

static int ramdisk_readpage(struct file *file, struct page *page)
{
   if (!PageUptodate(page))
       make_page_uptodate(page);
   unlock_page(page);
   return 0;
}

static int ramdisk_prepare_write(struct file *file, struct page *page,
               unsigned offset, unsigned to)
{
   if (!PageUptodate(page))
       make_page_uptodate(page);
   return 0;
}

static int ramdisk_commit_write(struct file *file, struct page *page,
               unsigned offset, unsigned to)
{
   set_page_dirty(page);
   return 0;
}

/*
* ->writepage to the the blockdev's mapping has to redirty the page so that the
* VM doesn't go and steal it. We return AOP_WRITEPAGE_ACTIVATE so that the VM
* won't try to (pointlessly) write the page again for a while.
*
* Really, these pages should not be on the LRU at all.
*/
static int ramdisk_writepage(struct page *page, struct writeback_control *wbc)
{
   if (!PageUptodate(page))
       make_page_uptodate(page);
   SetPageDirty(page);
   if (wbc->for_reclaim)
       return AOP_WRITEPAGE_ACTIVATE;
   unlock_page(page);
   return 0;
}

/*
* This is a little speedup thing: short-circuit attempts to write back the
* ramdisk blockdev inode to its non-existent backing store.
*/
static int ramdisk_writepages(struct address_space *mapping,
               struct writeback_control *wbc)
{
   return 0;
}

/*
* ramdisk blockdev pages have their own ->set_page_dirty() because we don't
* want them to contribute to dirty memory accounting.
*/
static int ramdisk_set_page_dirty(struct page *page)
{
   if (!TestSetPageDirty(page))
       return 1;
   return 0;
}


//块设备的address_space_operations结构，注意与文件系统的相应结果的区别。
static const struct address_space_operations ramdisk_aops = {
   .readpage   = ramdisk_readpage,
   .prepare_write   = ramdisk_prepare_write,
   .commit_write   = ramdisk_commit_write,
   .writepage   = ramdisk_writepage,
   .set_page_dirty   = ramdisk_set_page_dirty,
   .writepages   = ramdisk_writepages,
};

static int rd_blkdev_pagecache_IO(int rw, struct bio_vec *vec, sector_t sector,
               struct address_space *mapping)
{
   pgoff_t index = sector >> (PAGE_CACHE_SHIFT - 9);
   unsigned int vec_offset = vec->bv_offset;
   int offset = (sector << 9) & ~PAGE_CACHE_MASK;
   int size = vec->bv_len;
   int err = 0;

   do {
       int count;
       struct page *page;
       char *src;
       char *dst;

       count = PAGE_CACHE_SIZE - offset;
       if (count > size)
           count = size;
       size -= count;

       //从页面缓存中读，如果没有，则出错，而不像对于ide-disk的处理，它出错会导致从磁盘中读入。
       page = grab_cache_page(mapping, index);
       if (!page) {
           err = -ENOMEM;
           goto out;
       }

       if (!PageUptodate(page))
           make_page_uptodate(page);

       index++;

       //建立映射
       if (rw == READ) {
           src = kmap_atomic(page, KM_USER0) + offset;
           dst = kmap_atomic(vec->bv_page, KM_USER1) + vec_offset;
       } else {
           src = kmap_atomic(vec->bv_page, KM_USER0) + vec_offset;
           dst = kmap_atomic(page, KM_USER1) + offset;
       }
       offset = 0;
       vec_offset += count;

       memcpy(dst, src, count); //拷贝

       //临时映射必须在使用完后立即解除映射
       kunmap_atomic(src, KM_USER0);
       kunmap_atomic(dst, KM_USER1);

       if (rw == READ)
           flush_dcache_page(vec->bv_page);
       else
           set_page_dirty(page);   //使页面不可被释放
       unlock_page(page);
       put_page(page);
   } while (size);

out:
   return err;
}

/*
* Basically, my strategy here is to set up a buffer-head which can't be
* deleted, and make that my Ramdisk. If the request is outside of the
* allocated size, we must get rid of it...
*
* 19-JAN-1998 Richard Gooch <rgooch@atnf.csiro.au> Added devfs support
*
*/
//这是个处理BIO请求的函数，对于IDE-DISK相应的函数是通过IO-scheduler将待处理的BIO请求加入到请求队列中，或将这个BIO合并到合适的请求(request)中，在做下一步的处理。而现在对于ramdisk这个随机访问设备，IO-scheduler则不必要，所以这个函数直接处理的BIO请求。

static int rd_make_request(request_queue_t *q, struct bio *bio)
{
   struct block_device *bdev = bio->bi_bdev;
   struct address_space * mapping = bdev->bd_inode->i_mapping;
   sector_t sector = bio->bi_sector;
   unsigned long len = bio->bi_size >> 9;
   int rw = bio_data_dir(bio);
   struct bio_vec *bvec;
   int ret = 0, i;

   if (sector + len > get_capacity(bdev->bd_disk))
       goto fail;

   if (rw==READA) //预备读处理
       rw=READ;

   bio_for_each_segment(bvec, bio, i) { //处理BIO中的每一个segment。
       ret |= rd_blkdev_pagecache_IO(rw, bvec, sector, mapping);
       sector += bvec->bv_len >> 9;
   }
   if (ret)
       goto fail;

   bio_endio(bio, bio->bi_size, 0);
   return 0;
fail:
   bio_io_error(bio, bio->bi_size);
   return 0;
}

static int rd_ioctl(struct inode *inode, struct file *file,
           unsigned int cmd, unsigned long arg)
{
   int error;
   struct block_device *bdev = inode->i_bdev;

   if (cmd != BLKFLSBUF)
       return -ENOTTY;

   /*
   * special: we want to release the ramdisk memory, it's not like with
   * the other blockdevices where this ioctl only flushes away the buffer
   * cache
   */
   error = -EBUSY;
   mutex_lock(&bdev->bd_mutex);
   if (bdev->bd_openers <= 2) {
       truncate_inode_pages(bdev->bd_inode->i_mapping, 0);
       error = 0;
   }
   mutex_unlock(&bdev->bd_mutex);
   return error;
}

/*
* This is the backing_dev_info for the blockdev inode itself. It doesn't need
* writeback and it does not contribute to dirty memory accounting.
*/
static struct backing_dev_info rd_backing_dev_info = {
   .ra_pages   = 0,   /* No readahead */
   .capabilities   = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK | BDI_CAP_MAP_COPY,
   .unplug_io_fn   = default_unplug_io_fn,
};

/*
* This is the backing_dev_info for the files which live atop the ramdisk
* "device". These files do need writeback and they do contribute to dirty
* memory accounting.
*/
static struct backing_dev_info rd_file_backing_dev_info = {
   .ra_pages   = 0,   /* No readahead */
   .capabilities   = BDI_CAP_MAP_COPY,   /* Does contribute to dirty memory */
   .unplug_io_fn   = default_unplug_io_fn,
};

static int rd_open(struct inode *inode, struct file *filp)
{
   unsigned unit = iminor(inode);

   if (rd_bdev[unit] == NULL) {
       struct block_device *bdev = inode->i_bdev;
       struct address_space *mapping;
       unsigned bsize;
       gfp_t gfp_mask;

       inode = igrab(bdev->bd_inode);
       rd_bdev[unit] = bdev;
       bdev->bd_openers++;
       bsize = bdev_hardsect_size(bdev);
       bdev->bd_block_size = bsize;
       inode->i_blkbits = blksize_bits(bsize);
       inode->i_size = get_capacity(bdev->bd_disk)<<9;

       mapping = inode->i_mapping;
       mapping->a_ops = &ramdisk_aops;
       mapping->backing_dev_info = &rd_backing_dev_info;
       bdev->bd_inode_backing_dev_info = &rd_file_backing_dev_info;

       /*
       * Deep badness. rd_blkdev_pagecache_IO() needs to allocate
       * pagecache pages within a request_fn. We cannot recur back
       * into the filesytem which is mounted atop the ramdisk, because
       * that would deadlock on fs locks. And we really don't want
       * to reenter rd_blkdev_pagecache_IO when we're already within
       * that function.
       *
       * So we turn off __GFP_FS and __GFP_IO.
       *
       * And to give this thing a hope of working, turn on __GFP_HIGH.
       * Hopefully, there's enough regular memory allocation going on
       * for the page allocator emergency pools to keep the ramdisk
       * driver happy.
       */
       gfp_mask = mapping_gfp_mask(mapping);
       gfp_mask &= ~(__GFP_FS|__GFP_IO);
       gfp_mask |= __GFP_HIGH;
       mapping_set_gfp_mask(mapping, gfp_mask);
   }

   return 0;
}

static struct block_device_operations rd_bd_op = {
   .owner =   THIS_MODULE,
   .open =       rd_open,
   .ioctl =   rd_ioctl,
};

/*
* Before freeing the module, invalidate all of the protected buffers!
*/
static void __exit rd_cleanup(void)
{
   int i;

   for (i = 0; i < CONFIG_BLK_DEV_RAM_COUNT; i++) {
       struct block_device *bdev = rd_bdev[i];
       rd_bdev[i] = NULL;
       if (bdev) {
           invalidate_bdev(bdev, 1);
           blkdev_put(bdev);
       }
       del_gendisk(rd_disks[i]);
       put_disk(rd_disks[i]);
       blk_cleanup_queue(rd_queue[i]);
   }
   unregister_blkdev(RAMDISK_MAJOR, "ramdisk");
}

/*
* This is the registration and initialization section of the RAM disk driver
*/
static int __init rd_init(void)
{
   int i;
   int err = -ENOMEM;

   //检查块大小是否合适，它必须小于一个内存页面的大小，且要大于512字节，还要是2的N次幂。
   if (rd_blocksize > PAGE_SIZE || rd_blocksize < 512 ||
           (rd_blocksize & (rd_blocksize-1))) {
       printk("RAMDISK: wrong blocksize %d, reverting to defaults\n",
               rd_blocksize);
       rd_blocksize = BLOCK_SIZE; //如果不是则使用默认值
   }

   for (i = 0; i < CONFIG_BLK_DEV_RAM_COUNT; i++) {
       rd_disks[i] = alloc_disk(1);   //分配硬盘描述符
       if (!rd_disks[i])
           goto out;

       rd_queue[i] = blk_alloc_queue(GFP_KERNEL); //分配请求队列描述符
       if (!rd_queue[i]) {
           put_disk(rd_disks[i]);
           goto out;
       }
   }

   //注册进内核
   if (register_blkdev(RAMDISK_MAJOR, "ramdisk")) {
       err = -EIO;
       goto out;
   }

   //初始化请求队列，并将磁盘加入到队列中去
   for (i = 0; i < CONFIG_BLK_DEV_RAM_COUNT; i++) {
       struct gendisk *disk = rd_disks[i];

   //注册请求队列的BIO请求函数
   blk_queue_make_request(rd_queue[i], &rd_make_request);
       //盘块大小
blk_queue_hardsect_size(rd_queue[i], rd_blocksize);

       /* rd_size is given in kB */
       //以下两行注册设备的主次号
       disk->major = RAMDISK_MAJOR; //对于RAMDISK，主设备号都是这个
       disk->first_minor = i;    //次设备号标志是第几个盘
       disk->fops = &rd_bd_op;
       disk->queue = rd_queue[i]; //每个盘一个请求队列。
       disk->flags |= GENHD_FL_SUPPRESS_PARTITION_INFO;
       sprintf(disk->disk_name, "ram%d", i);
       set_capacity(disk, rd_size * 2);
       add_disk(rd_disks[i]); //加入，至此，该盘可以被访问
   }

   /* rd_size is given in kB */
   printk("RAMDISK driver initialized: "
       "%d RAM disks of %dK size %d blocksize\n",
       CONFIG_BLK_DEV_RAM_COUNT, rd_size, rd_blocksize);

   return 0;
out:
   while (i--) {
       put_disk(rd_disks[i]);
       blk_cleanup_queue(rd_queue[i]);
   }
   return err;
}

module_init(rd_init);
module_exit(rd_cleanup);

/* options - nonmodular */
#ifndef MODULE
static int __init ramdisk_size(char *str)
{
   rd_size = simple_strtol(str,NULL,0);
   return 1;
}
static int __init ramdisk_size2(char *str)   /* kludge */
{
   return ramdisk_size(str);
}
static int __init ramdisk_blocksize(char *str)
{
   rd_blocksize = simple_strtol(str,NULL,0);
   return 1;
}
__setup("ramdisk=", ramdisk_size); //为向前兼容而存在，等于ramdisk_size
__setup("ramdisk_size=", ramdisk_size2);
__setup("ramdisk_blocksize=", ramdisk_blocksize);
#endif

/* options - modular */
module_param(rd_size, int, 0);
MODULE_PARM_DESC(rd_size, "Size of each RAM disk in kbytes.");
module_param(rd_blocksize, int, 0);
MODULE_PARM_DESC(rd_blocksize, "Blocksize of each RAM disk in bytes.");
MODULE_ALIAS_BLOCKDEV_MAJOR(RAMDISK_MAJOR);

MODULE_LICENSE("GPL");