raid1.c

来自「Linux Kernel 2.6.9 for OMAP1710」· C语言 代码 · 共 1,426 行 · 第 1/3 页

	return ret;
}

/*
 * Throttle resync depth, so that we can both get proper overlapping of
 * requests, but are still able to handle normal requests quickly.
 */
#define RESYNC_DEPTH 32

static void device_barrier(conf_t *conf, sector_t sect)
{
	spin_lock_irq(&conf->resync_lock);
	wait_event_lock_irq(conf->wait_idle, !waitqueue_active(&conf->wait_resume),
			    conf->resync_lock, unplug_slaves(conf->mddev));
	
	if (!conf->barrier++) {
		wait_event_lock_irq(conf->wait_idle, !conf->nr_pending,
				    conf->resync_lock, unplug_slaves(conf->mddev));
		if (conf->nr_pending)
			BUG();
	}
	wait_event_lock_irq(conf->wait_resume, conf->barrier < RESYNC_DEPTH,
			    conf->resync_lock, unplug_slaves(conf->mddev));
	conf->next_resync = sect;
	spin_unlock_irq(&conf->resync_lock);
}

static int make_request(request_queue_t *q, struct bio * bio)
{
	mddev_t *mddev = q->queuedata;
	conf_t *conf = mddev_to_conf(mddev);
	mirror_info_t *mirror;
	r1bio_t *r1_bio;
	struct bio *read_bio;
	int i, disks;

	/*
	 * Register the new request and wait if the reconstruction
	 * thread has put up a bar for new requests.
	 * Continue immediately if no resync is active currently.
	 */
	spin_lock_irq(&conf->resync_lock);
	wait_event_lock_irq(conf->wait_resume, !conf->barrier, conf->resync_lock, );
	conf->nr_pending++;
	spin_unlock_irq(&conf->resync_lock);

	if (bio_data_dir(bio)==WRITE) {
		disk_stat_inc(mddev->gendisk, writes);
		disk_stat_add(mddev->gendisk, write_sectors, bio_sectors(bio));
	} else {
		disk_stat_inc(mddev->gendisk, reads);
		disk_stat_add(mddev->gendisk, read_sectors, bio_sectors(bio));
	}

	/*
	 * make_request() can abort the operation when READA is being
	 * used and no empty request is available.
	 *
	 */
	r1_bio = mempool_alloc(conf->r1bio_pool, GFP_NOIO);

	r1_bio->master_bio = bio;
	r1_bio->sectors = bio->bi_size >> 9;

	r1_bio->mddev = mddev;
	r1_bio->sector = bio->bi_sector;

	r1_bio->state = 0;

	if (bio_data_dir(bio) == READ) {
		/*
		 * read balancing logic:
		 */
		int rdisk = read_balance(conf, r1_bio);

		if (rdisk < 0) {
			/* couldn't find anywhere to read from */
			raid_end_bio_io(r1_bio);
			return 0;
		}
		mirror = conf->mirrors + rdisk;

		r1_bio->read_disk = rdisk;

		read_bio = bio_clone(bio, GFP_NOIO);

		r1_bio->bios[rdisk] = read_bio;

		read_bio->bi_sector = r1_bio->sector + mirror->rdev->data_offset;
		read_bio->bi_bdev = mirror->rdev->bdev;
		read_bio->bi_end_io = raid1_end_read_request;
		read_bio->bi_rw = READ;
		read_bio->bi_private = r1_bio;

		generic_make_request(read_bio);
		return 0;
	}

	/*
	 * WRITE:
	 */
	/* first select target devices under spinlock and
	 * inc refcount on their rdev.  Record them by setting
	 * bios[x] to bio
	 */
	disks = conf->raid_disks;
	spin_lock_irq(&conf->device_lock);
	for (i = 0;  i < disks; i++) {
		if (conf->mirrors[i].rdev &&
		    !conf->mirrors[i].rdev->faulty) {
			atomic_inc(&conf->mirrors[i].rdev->nr_pending);
			r1_bio->bios[i] = bio;
		} else
			r1_bio->bios[i] = NULL;
	}
	spin_unlock_irq(&conf->device_lock);

	atomic_set(&r1_bio->remaining, 1);
	md_write_start(mddev);
	for (i = 0; i < disks; i++) {
		struct bio *mbio;
		if (!r1_bio->bios[i])
			continue;

		mbio = bio_clone(bio, GFP_NOIO);
		r1_bio->bios[i] = mbio;

		mbio->bi_sector	= r1_bio->sector + conf->mirrors[i].rdev->data_offset;
		mbio->bi_bdev = conf->mirrors[i].rdev->bdev;
		mbio->bi_end_io	= raid1_end_write_request;
		mbio->bi_rw = WRITE;
		mbio->bi_private = r1_bio;

		atomic_inc(&r1_bio->remaining);
		generic_make_request(mbio);
	}

	if (atomic_dec_and_test(&r1_bio->remaining)) {
		md_write_end(mddev);
		raid_end_bio_io(r1_bio);
	}

	return 0;
}

static void status(struct seq_file *seq, mddev_t *mddev)
{
	conf_t *conf = mddev_to_conf(mddev);
	int i;

	seq_printf(seq, " [%d/%d] [", conf->raid_disks,
						conf->working_disks);
	for (i = 0; i < conf->raid_disks; i++)
		seq_printf(seq, "%s",
			      conf->mirrors[i].rdev &&
			      conf->mirrors[i].rdev->in_sync ? "U" : "_");
	seq_printf(seq, "]");
}


static void error(mddev_t *mddev, mdk_rdev_t *rdev)
{
	char b[BDEVNAME_SIZE];
	conf_t *conf = mddev_to_conf(mddev);

	/*
	 * If it is not operational, then we have already marked it as dead
	 * else if it is the last working disks, ignore the error, let the
	 * next level up know.
	 * else mark the drive as failed
	 */
	if (rdev->in_sync
	    && conf->working_disks == 1)
		/*
		 * Don't fail the drive, act as though we were just a
		 * normal single drive
		 */
		return;
	if (rdev->in_sync) {
		mddev->degraded++;
		conf->working_disks--;
		/*
		 * if recovery is running, make sure it aborts.
		 */
		set_bit(MD_RECOVERY_ERR, &mddev->recovery);
	}
	rdev->in_sync = 0;
	rdev->faulty = 1;
	mddev->sb_dirty = 1;
	printk(KERN_ALERT "raid1: Disk failure on %s, disabling device. \n"
		"	Operation continuing on %d devices\n",
		bdevname(rdev->bdev,b), conf->working_disks);
}

static void print_conf(conf_t *conf)
{
	int i;
	mirror_info_t *tmp;

	printk("RAID1 conf printout:\n");
	if (!conf) {
		printk("(!conf)\n");
		return;
	}
	printk(" --- wd:%d rd:%d\n", conf->working_disks,
		conf->raid_disks);

	for (i = 0; i < conf->raid_disks; i++) {
		char b[BDEVNAME_SIZE];
		tmp = conf->mirrors + i;
		if (tmp->rdev)
			printk(" disk %d, wo:%d, o:%d, dev:%s\n",
				i, !tmp->rdev->in_sync, !tmp->rdev->faulty,
				bdevname(tmp->rdev->bdev,b));
	}
}

static void close_sync(conf_t *conf)
{
	spin_lock_irq(&conf->resync_lock);
	wait_event_lock_irq(conf->wait_resume, !conf->barrier,
			    conf->resync_lock, 	unplug_slaves(conf->mddev));
	spin_unlock_irq(&conf->resync_lock);

	if (conf->barrier) BUG();
	if (waitqueue_active(&conf->wait_idle)) BUG();

	mempool_destroy(conf->r1buf_pool);
	conf->r1buf_pool = NULL;
}

static int raid1_spare_active(mddev_t *mddev)
{
	int i;
	conf_t *conf = mddev->private;
	mirror_info_t *tmp;

	spin_lock_irq(&conf->device_lock);
	/*
	 * Find all failed disks within the RAID1 configuration 
	 * and mark them readable
	 */
	for (i = 0; i < conf->raid_disks; i++) {
		tmp = conf->mirrors + i;
		if (tmp->rdev 
		    && !tmp->rdev->faulty
		    && !tmp->rdev->in_sync) {
			conf->working_disks++;
			mddev->degraded--;
			tmp->rdev->in_sync = 1;
		}
	}
	spin_unlock_irq(&conf->device_lock);

	print_conf(conf);
	return 0;
}


static int raid1_add_disk(mddev_t *mddev, mdk_rdev_t *rdev)
{
	conf_t *conf = mddev->private;
	int found = 0;
	int mirror;
	mirror_info_t *p;

	spin_lock_irq(&conf->device_lock);
	for (mirror=0; mirror < mddev->raid_disks; mirror++)
		if ( !(p=conf->mirrors+mirror)->rdev) {
			p->rdev = rdev;

			blk_queue_stack_limits(mddev->queue,
					       rdev->bdev->bd_disk->queue);
			/* as we don't honour merge_bvec_fn, we must never risk
			 * violating it, so limit ->max_sector to one PAGE, as
			 * a one page request is never in violation.
			 */
			if (rdev->bdev->bd_disk->queue->merge_bvec_fn &&
			    mddev->queue->max_sectors > (PAGE_SIZE>>9))
				blk_queue_max_sectors(mddev->queue, PAGE_SIZE>>9);

			p->head_position = 0;
			rdev->raid_disk = mirror;
			found = 1;
			break;
		}
	spin_unlock_irq(&conf->device_lock);

	print_conf(conf);
	return found;
}

static int raid1_remove_disk(mddev_t *mddev, int number)
{
	conf_t *conf = mddev->private;
	int err = 1;
	mirror_info_t *p = conf->mirrors+ number;

	print_conf(conf);
	spin_lock_irq(&conf->device_lock);
	if (p->rdev) {
		if (p->rdev->in_sync ||
		    atomic_read(&p->rdev->nr_pending)) {
			err = -EBUSY;
			goto abort;
		}
		p->rdev = NULL;
		err = 0;
	}
	if (err)
		MD_BUG();
abort:
	spin_unlock_irq(&conf->device_lock);

	print_conf(conf);
	return err;
}


static int end_sync_read(struct bio *bio, unsigned int bytes_done, int error)
{
	int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
	r1bio_t * r1_bio = (r1bio_t *)(bio->bi_private);
	conf_t *conf = mddev_to_conf(r1_bio->mddev);

	if (bio->bi_size)
		return 1;

	if (r1_bio->bios[r1_bio->read_disk] != bio)
		BUG();
	update_head_pos(r1_bio->read_disk, r1_bio);
	/*
	 * we have read a block, now it needs to be re-written,
	 * or re-read if the read failed.
	 * We don't do much here, just schedule handling by raid1d
	 */
	if (!uptodate)
		md_error(r1_bio->mddev,
			 conf->mirrors[r1_bio->read_disk].rdev);
	else
		set_bit(R1BIO_Uptodate, &r1_bio->state);
	rdev_dec_pending(conf->mirrors[r1_bio->read_disk].rdev, conf->mddev);
	reschedule_retry(r1_bio);
	return 0;
}

static int end_sync_write(struct bio *bio, unsigned int bytes_done, int error)
{
	int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
	r1bio_t * r1_bio = (r1bio_t *)(bio->bi_private);
	mddev_t *mddev = r1_bio->mddev;
	conf_t *conf = mddev_to_conf(mddev);
	int i;
	int mirror=0;

	if (bio->bi_size)
		return 1;

	for (i = 0; i < conf->raid_disks; i++)
		if (r1_bio->bios[i] == bio) {
			mirror = i;
			break;
		}
	if (!uptodate)
		md_error(mddev, conf->mirrors[mirror].rdev);
	update_head_pos(mirror, r1_bio);

	if (atomic_dec_and_test(&r1_bio->remaining)) {
		md_done_sync(mddev, r1_bio->sectors, uptodate);
		put_buf(r1_bio);
	}
	rdev_dec_pending(conf->mirrors[mirror].rdev, mddev);
	return 0;
}

static void sync_request_write(mddev_t *mddev, r1bio_t *r1_bio)
{
	conf_t *conf = mddev_to_conf(mddev);
	int i;
	int disks = conf->raid_disks;
	struct bio *bio, *wbio;

	bio = r1_bio->bios[r1_bio->read_disk];

	/*
	 * schedule writes
	 */
	if (!test_bit(R1BIO_Uptodate, &r1_bio->state)) {
		/*
		 * There is no point trying a read-for-reconstruct as
		 * reconstruct is about to be aborted
		 */
		char b[BDEVNAME_SIZE];
		printk(KERN_ALERT "raid1: %s: unrecoverable I/O read error"
			" for block %llu\n",
			bdevname(bio->bi_bdev,b), 
			(unsigned long long)r1_bio->sector);
		md_done_sync(mddev, r1_bio->sectors, 0);
		put_buf(r1_bio);
		return;
	}

	atomic_set(&r1_bio->remaining, 1);
	for (i = 0; i < disks ; i++) {
		wbio = r1_bio->bios[i];
		if (wbio->bi_end_io != end_sync_write)
			continue;

		atomic_inc(&conf->mirrors[i].rdev->nr_pending);
		atomic_inc(&r1_bio->remaining);
		md_sync_acct(conf->mirrors[i].rdev->bdev, wbio->bi_size >> 9);
		generic_make_request(wbio);
	}

	if (atomic_dec_and_test(&r1_bio->remaining)) {
		md_done_sync(mddev, r1_bio->sectors, 1);
		put_buf(r1_bio);
	}
}

/*
 * This is a kernel thread which:
 *
 *	1.	Retries failed read operations on working mirrors.
 *	2.	Updates the raid superblock when problems encounter.
 *	3.	Performs writes following reads for array syncronising.
 */

static void raid1d(mddev_t *mddev)
{
	struct list_head *head = &retry_list_head;
	r1bio_t *r1_bio;
	struct bio *bio;
	unsigned long flags;
	conf_t *conf = mddev_to_conf(mddev);
	int unplug=0;
	mdk_rdev_t *rdev;

	md_check_recovery(mddev);
	md_handle_safemode(mddev);
	
	for (;;) {
		char b[BDEVNAME_SIZE];
		spin_lock_irqsave(&retry_list_lock, flags);
		if (list_empty(head))
			break;
		r1_bio = list_entry(head->prev, r1bio_t, retry_list);
		list_del(head->prev);
		spin_unlock_irqrestore(&retry_list_lock, flags);

		mddev = r1_bio->mddev;
		conf = mddev_to_conf(mddev);
		if (test_bit(R1BIO_IsSync, &r1_bio->state)) {
			sync_request_write(mddev, r1_bio);
			unplug = 1;
		} else {
			int disk;
			bio = r1_bio->bios[r1_bio->read_disk];
			if ((disk=read_balance(conf, r1_bio)) == -1) {
				printk(KERN_ALERT "raid1: %s: unrecoverable I/O"
				       " read error for block %llu\n",
				       bdevname(bio->bi_bdev,b),
				       (unsigned long long)r1_bio->sector);
				raid_end_bio_io(r1_bio);
			} else {
				r1_bio->bios[r1_bio->read_disk] = NULL;
				r1_bio->read_disk = disk;
				r1_bio->bios[r1_bio->read_disk] = bio;
				rdev = conf->mirrors[disk].rdev;
				if (printk_ratelimit())
					printk(KERN_ERR "raid1: %s: redirecting sector %llu to"
					       " another mirror\n",
					       bdevname(rdev->bdev,b),
					       (unsigned long long)r1_bio->sector);
				bio->bi_bdev = rdev->bdev;
				bio->bi_sector = r1_bio->sector + rdev->data_offset;