raid5.c

linux和2410结合开发 用他可以生成2410所需的zImage文件

/*
 * raid5.c : Multiple Devices driver for Linux
 *	   Copyright (C) 1996, 1997 Ingo Molnar, Miguel de Icaza, Gadi Oxman
 *	   Copyright (C) 1999, 2000 Ingo Molnar
 *
 * RAID-5 management functions.
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2, or (at your option)
 * any later version.
 *
 * You should have received a copy of the GNU General Public License
 * (for example /usr/src/linux/COPYING); if not, write to the Free
 * Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
 */


#include <linux/config.h>
#include <linux/module.h>
#include <linux/locks.h>
#include <linux/slab.h>
#include <linux/raid/raid5.h>
#include <asm/bitops.h>
#include <asm/atomic.h>

static mdk_personality_t raid5_personality;

/*
 * Stripe cache
 */

#define NR_STRIPES		256
#define	IO_THRESHOLD		1
#define HASH_PAGES		1
#define HASH_PAGES_ORDER	0
#define NR_HASH			(HASH_PAGES * PAGE_SIZE / sizeof(struct stripe_head *))
#define HASH_MASK		(NR_HASH - 1)
#define stripe_hash(conf, sect)	((conf)->stripe_hashtbl[((sect) / ((conf)->buffer_size >> 9)) & HASH_MASK])

/*
 * The following can be used to debug the driver
 */
#define RAID5_DEBUG	0
#define RAID5_PARANOIA	1
#if RAID5_PARANOIA && CONFIG_SMP
# define CHECK_DEVLOCK() if (!spin_is_locked(&conf->device_lock)) BUG()
#else
# define CHECK_DEVLOCK()
#endif

#if RAID5_DEBUG
#define PRINTK(x...) printk(x)
#define inline
#define __inline__
#else
#define PRINTK(x...) do { } while (0)
#endif

static void print_raid5_conf (raid5_conf_t *conf);

static inline void __release_stripe(raid5_conf_t *conf, struct stripe_head *sh)
{
	if (atomic_dec_and_test(&sh->count)) {
		if (!list_empty(&sh->lru))
			BUG();
		if (atomic_read(&conf->active_stripes)==0)
			BUG();
		if (test_bit(STRIPE_HANDLE, &sh->state)) {
			if (test_bit(STRIPE_DELAYED, &sh->state))
				list_add_tail(&sh->lru, &conf->delayed_list);
			else
				list_add_tail(&sh->lru, &conf->handle_list);
			md_wakeup_thread(conf->thread);
		} else {
			if (test_and_clear_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) {
				atomic_dec(&conf->preread_active_stripes);
				if (atomic_read(&conf->preread_active_stripes) < IO_THRESHOLD)
					md_wakeup_thread(conf->thread);
			}
			list_add_tail(&sh->lru, &conf->inactive_list);
			atomic_dec(&conf->active_stripes);
			if (!conf->inactive_blocked ||
			    atomic_read(&conf->active_stripes) < (NR_STRIPES*3/4))
				wake_up(&conf->wait_for_stripe);
		}
	}
}
static void release_stripe(struct stripe_head *sh)
{
	raid5_conf_t *conf = sh->raid_conf;
	unsigned long flags;
	
	spin_lock_irqsave(&conf->device_lock, flags);
	__release_stripe(conf, sh);
	spin_unlock_irqrestore(&conf->device_lock, flags);
}

static void remove_hash(struct stripe_head *sh)
{
	PRINTK("remove_hash(), stripe %lu\n", sh->sector);

	if (sh->hash_pprev) {
		if (sh->hash_next)
			sh->hash_next->hash_pprev = sh->hash_pprev;
		*sh->hash_pprev = sh->hash_next;
		sh->hash_pprev = NULL;
	}
}

static __inline__ void insert_hash(raid5_conf_t *conf, struct stripe_head *sh)
{
	struct stripe_head **shp = &stripe_hash(conf, sh->sector);

	PRINTK("insert_hash(), stripe %lu\n",sh->sector);

	CHECK_DEVLOCK();
	if ((sh->hash_next = *shp) != NULL)
		(*shp)->hash_pprev = &sh->hash_next;
	*shp = sh;
	sh->hash_pprev = shp;
}


/* find an idle stripe, make sure it is unhashed, and return it. */
static struct stripe_head *get_free_stripe(raid5_conf_t *conf)
{
	struct stripe_head *sh = NULL;
	struct list_head *first;

	CHECK_DEVLOCK();
	if (list_empty(&conf->inactive_list))
		goto out;
	first = conf->inactive_list.next;
	sh = list_entry(first, struct stripe_head, lru);
	list_del_init(first);
	remove_hash(sh);
	atomic_inc(&conf->active_stripes);
out:
	return sh;
}

static void shrink_buffers(struct stripe_head *sh, int num)
{
	struct buffer_head *bh;
	int i;

	for (i=0; i<num ; i++) {
		bh = sh->bh_cache[i];
		if (!bh)
			return;
		sh->bh_cache[i] = NULL;
		free_page((unsigned long) bh->b_data);
		kfree(bh);
	}
}

static int grow_buffers(struct stripe_head *sh, int num, int b_size, int priority)
{
	struct buffer_head *bh;
	int i;

	for (i=0; i<num; i++) {
		struct page *page;
		bh = kmalloc(sizeof(struct buffer_head), priority);
		if (!bh)
			return 1;
		memset(bh, 0, sizeof (struct buffer_head));
		init_waitqueue_head(&bh->b_wait);
		if ((page = alloc_page(priority)))
			bh->b_data = page_address(page);
		else {
			kfree(bh);
			return 1;
		}
		atomic_set(&bh->b_count, 0);
		bh->b_page = page;
		sh->bh_cache[i] = bh;

	}
	return 0;
}

static struct buffer_head *raid5_build_block (struct stripe_head *sh, int i);

static inline void init_stripe(struct stripe_head *sh, unsigned long sector)
{
	raid5_conf_t *conf = sh->raid_conf;
	int disks = conf->raid_disks, i;

	if (atomic_read(&sh->count) != 0)
		BUG();
	if (test_bit(STRIPE_HANDLE, &sh->state))
		BUG();
	
	CHECK_DEVLOCK();
	PRINTK("init_stripe called, stripe %lu\n", sh->sector);

	remove_hash(sh);
	
	sh->sector = sector;
	sh->size = conf->buffer_size;
	sh->state = 0;

	for (i=disks; i--; ) {
		if (sh->bh_read[i] || sh->bh_write[i] || sh->bh_written[i] ||
		    buffer_locked(sh->bh_cache[i])) {
			printk("sector=%lx i=%d %p %p %p %d\n",
			       sh->sector, i, sh->bh_read[i],
			       sh->bh_write[i], sh->bh_written[i],
			       buffer_locked(sh->bh_cache[i]));
			BUG();
		}
		clear_bit(BH_Uptodate, &sh->bh_cache[i]->b_state);
		raid5_build_block(sh, i);
	}
	insert_hash(conf, sh);
}

/* the buffer size has changed, so unhash all stripes
 * as active stripes complete, they will go onto inactive list
 */
static void shrink_stripe_cache(raid5_conf_t *conf)
{
	int i;
	CHECK_DEVLOCK();
	if (atomic_read(&conf->active_stripes))
		BUG();
	for (i=0; i < NR_HASH; i++) {
		struct stripe_head *sh;
		while ((sh = conf->stripe_hashtbl[i])) 
			remove_hash(sh);
	}
}

static struct stripe_head *__find_stripe(raid5_conf_t *conf, unsigned long sector)
{
	struct stripe_head *sh;

	CHECK_DEVLOCK();
	PRINTK("__find_stripe, sector %lu\n", sector);
	for (sh = stripe_hash(conf, sector); sh; sh = sh->hash_next)
		if (sh->sector == sector)
			return sh;
	PRINTK("__stripe %lu not in cache\n", sector);
	return NULL;
}

static struct stripe_head *get_active_stripe(raid5_conf_t *conf, unsigned long sector, int size, int noblock) 
{
	struct stripe_head *sh;

	PRINTK("get_stripe, sector %lu\n", sector);

	md_spin_lock_irq(&conf->device_lock);

	do {
		if (conf->buffer_size == 0 ||
		    (size && size != conf->buffer_size)) {
			/* either the size is being changed (buffer_size==0) or
			 * we need to change it.
			 * If size==0, we can proceed as soon as buffer_size gets set.
			 * If size>0, we can proceed when active_stripes reaches 0, or
			 * when someone else sets the buffer_size to size.
			 * If someone sets the buffer size to something else, we will need to
			 * assert that we want to change it again
			 */
			int oldsize = conf->buffer_size;
			PRINTK("get_stripe %ld/%d buffer_size is %d, %d active\n", sector, size, conf->buffer_size, atomic_read(&conf->active_stripes));
			if (size==0)
				wait_event_lock_irq(conf->wait_for_stripe,
						    conf->buffer_size,
						    conf->device_lock);
			else {
				while (conf->buffer_size != size && atomic_read(&conf->active_stripes)) {
					conf->buffer_size = 0;
					wait_event_lock_irq(conf->wait_for_stripe,
							    atomic_read(&conf->active_stripes)==0 || conf->buffer_size,
							    conf->device_lock);
					PRINTK("waited and now  %ld/%d buffer_size is %d - %d active\n", sector, size,
					       conf->buffer_size, atomic_read(&conf->active_stripes));
				}

				if (conf->buffer_size != size) {
					printk("raid5: switching cache buffer size, %d --> %d\n", oldsize, size);
					shrink_stripe_cache(conf);
					if (size==0) BUG();
					conf->buffer_size = size;
					PRINTK("size now %d\n", conf->buffer_size);
				}
			}
		}
		if (size == 0)
			sector -= sector & ((conf->buffer_size>>9)-1);

		sh = __find_stripe(conf, sector);
		if (!sh) {
			if (!conf->inactive_blocked)
				sh = get_free_stripe(conf);
			if (noblock && sh == NULL)
				break;
			if (!sh) {
				conf->inactive_blocked = 1;
				wait_event_lock_irq(conf->wait_for_stripe,
						    !list_empty(&conf->inactive_list) &&
						    (atomic_read(&conf->active_stripes) < (NR_STRIPES *3/4)
						     || !conf->inactive_blocked),
						    conf->device_lock);
				conf->inactive_blocked = 0;
			} else
				init_stripe(sh, sector);
		} else {
			if (atomic_read(&sh->count)) {
				if (!list_empty(&sh->lru))
					BUG();
			} else {
				if (!test_bit(STRIPE_HANDLE, &sh->state))
					atomic_inc(&conf->active_stripes);
				if (list_empty(&sh->lru))
					BUG();
				list_del_init(&sh->lru);
			}
		}
	} while (sh == NULL);

	if (sh)
		atomic_inc(&sh->count);

	md_spin_unlock_irq(&conf->device_lock);
	return sh;
}

static int grow_stripes(raid5_conf_t *conf, int num, int priority)
{
	struct stripe_head *sh;

	while (num--) {
		sh = kmalloc(sizeof(struct stripe_head), priority);
		if (!sh)
			return 1;
		memset(sh, 0, sizeof(*sh));
		sh->raid_conf = conf;
		sh->lock = SPIN_LOCK_UNLOCKED;

		if (grow_buffers(sh, conf->raid_disks, PAGE_SIZE, priority)) {
			shrink_buffers(sh, conf->raid_disks);
			kfree(sh);
			return 1;
		}
		/* we just created an active stripe so... */
		atomic_set(&sh->count, 1);
		atomic_inc(&conf->active_stripes);
		INIT_LIST_HEAD(&sh->lru);
		release_stripe(sh);
	}
	return 0;
}

static void shrink_stripes(raid5_conf_t *conf, int num)
{
	struct stripe_head *sh;

	while (num--) {
		spin_lock_irq(&conf->device_lock);
		sh = get_free_stripe(conf);
		spin_unlock_irq(&conf->device_lock);
		if (!sh)
			break;
		if (atomic_read(&sh->count))
			BUG();
		shrink_buffers(sh, conf->raid_disks);
		kfree(sh);
		atomic_dec(&conf->active_stripes);
	}
}


static void raid5_end_read_request (struct buffer_head * bh, int uptodate)
{
 	struct stripe_head *sh = bh->b_private;
	raid5_conf_t *conf = sh->raid_conf;
	int disks = conf->raid_disks, i;
	unsigned long flags;

	for (i=0 ; i<disks; i++)
		if (bh == sh->bh_cache[i])
			break;

	PRINTK("end_read_request %lu/%d, count: %d, uptodate %d.\n", sh->sector, i, atomic_read(&sh->count), uptodate);
	if (i == disks) {
		BUG();
		return;
	}

	if (uptodate) {
		struct buffer_head *buffer;
		spin_lock_irqsave(&conf->device_lock, flags);
		/* we can return a buffer if we bypassed the cache or
		 * if the top buffer is not in highmem.  If there are
		 * multiple buffers, leave the extra work to
		 * handle_stripe
		 */
		buffer = sh->bh_read[i];
		if (buffer &&
		    (!PageHighMem(buffer->b_page)
		     || buffer->b_page == bh->b_page )
			) {
			sh->bh_read[i] = buffer->b_reqnext;
			buffer->b_reqnext = NULL;
		} else
			buffer = NULL;
		spin_unlock_irqrestore(&conf->device_lock, flags);
		if (sh->bh_page[i]==NULL)
			set_bit(BH_Uptodate, &bh->b_state);
		if (buffer) {
			if (buffer->b_page != bh->b_page)
				memcpy(buffer->b_data, bh->b_data, bh->b_size);
			buffer->b_end_io(buffer, 1);
		}
	} else {
		md_error(conf->mddev, bh->b_dev);
		clear_bit(BH_Uptodate, &bh->b_state);
	}
	/* must restore b_page before unlocking buffer... */
	if (sh->bh_page[i]) {
		bh->b_page = sh->bh_page[i];
		bh->b_data = page_address(bh->b_page);
		sh->bh_page[i] = NULL;
		clear_bit(BH_Uptodate, &bh->b_state);
	}
	clear_bit(BH_Lock, &bh->b_state);
	set_bit(STRIPE_HANDLE, &sh->state);
	release_stripe(sh);
}

static void raid5_end_write_request (struct buffer_head *bh, int uptodate)
{
 	struct stripe_head *sh = bh->b_private;
	raid5_conf_t *conf = sh->raid_conf;
	int disks = conf->raid_disks, i;
	unsigned long flags;

	for (i=0 ; i<disks; i++)
		if (bh == sh->bh_cache[i])
			break;

	PRINTK("end_write_request %lu/%d, count %d, uptodate: %d.\n", sh->sector, i, atomic_read(&sh->count), uptodate);
	if (i == disks) {
		BUG();
		return;
	}

	md_spin_lock_irqsave(&conf->device_lock, flags);
	if (!uptodate)
		md_error(conf->mddev, bh->b_dev);
	clear_bit(BH_Lock, &bh->b_state);
	set_bit(STRIPE_HANDLE, &sh->state);
	__release_stripe(conf, sh);
	md_spin_unlock_irqrestore(&conf->device_lock, flags);
}
	


static struct buffer_head *raid5_build_block (struct stripe_head *sh, int i)
{
	raid5_conf_t *conf = sh->raid_conf;
	struct buffer_head *bh = sh->bh_cache[i];
	unsigned long block = sh->sector / (sh->size >> 9);

	init_buffer(bh, raid5_end_read_request, sh);
	bh->b_dev       = conf->disks[i].dev;
	bh->b_blocknr   = block;

	bh->b_state	= (1 << BH_Req) | (1 << BH_Mapped);
	bh->b_size	= sh->size;
	bh->b_list	= BUF_LOCKED;
	return bh;
}

static int raid5_error (mddev_t *mddev, kdev_t dev)
{
	raid5_conf_t *conf = (raid5_conf_t *) mddev->private;
	mdp_super_t *sb = mddev->sb;
	struct disk_info *disk;
	int i;

	PRINTK("raid5_error called\n");

	for (i = 0, disk = conf->disks; i < conf->raid_disks; i++, disk++) {
		if (disk->dev == dev) {
			if (disk->operational) {
				disk->operational = 0;
				mark_disk_faulty(sb->disks+disk->number);
				mark_disk_nonsync(sb->disks+disk->number);
				mark_disk_inactive(sb->disks+disk->number);
				sb->active_disks--;
				sb->working_disks--;
				sb->failed_disks++;
				mddev->sb_dirty = 1;
				conf->working_disks--;
				conf->failed_disks++;
				md_wakeup_thread(conf->thread);
				printk (KERN_ALERT
					"raid5: Disk failure on %s, disabling device."
					" Operation continuing on %d devices\n",
					partition_name (dev), conf->working_disks);
			}
			return 0;