raid10.c

来自「linux 内核源代码」· C语言 代码 · 共 2,188 行 · 第 1/4 页

		    !test_bit(In_sync, &rdev->flags))
			continue;

		/* This optimisation is debatable, and completely destroys
		 * sequential read speed for 'far copies' arrays.  So only
		 * keep it for 'near' arrays, and review those later.
		 */
		if (conf->near_copies > 1 && !atomic_read(&rdev->nr_pending)) {
			disk = ndisk;
			slot = nslot;
			break;
		}
		new_distance = abs(r10_bio->devs[nslot].addr -
				   conf->mirrors[ndisk].head_position);
		if (new_distance < current_distance) {
			current_distance = new_distance;
			disk = ndisk;
			slot = nslot;
		}
	}

rb_out:
	r10_bio->read_slot = slot;
/*	conf->next_seq_sect = this_sector + sectors;*/

	if (disk >= 0 && (rdev=rcu_dereference(conf->mirrors[disk].rdev))!= NULL)
		atomic_inc(&conf->mirrors[disk].rdev->nr_pending);
	else
		disk = -1;
	rcu_read_unlock();

	return disk;
}

static void unplug_slaves(mddev_t *mddev)
{
	conf_t *conf = mddev_to_conf(mddev);
	int i;

	rcu_read_lock();
	for (i=0; i<mddev->raid_disks; i++) {
		mdk_rdev_t *rdev = rcu_dereference(conf->mirrors[i].rdev);
		if (rdev && !test_bit(Faulty, &rdev->flags) && atomic_read(&rdev->nr_pending)) {
			struct request_queue *r_queue = bdev_get_queue(rdev->bdev);

			atomic_inc(&rdev->nr_pending);
			rcu_read_unlock();

			blk_unplug(r_queue);

			rdev_dec_pending(rdev, mddev);
			rcu_read_lock();
		}
	}
	rcu_read_unlock();
}

static void raid10_unplug(struct request_queue *q)
{
	mddev_t *mddev = q->queuedata;

	unplug_slaves(q->queuedata);
	md_wakeup_thread(mddev->thread);
}

static int raid10_congested(void *data, int bits)
{
	mddev_t *mddev = data;
	conf_t *conf = mddev_to_conf(mddev);
	int i, ret = 0;

	rcu_read_lock();
	for (i = 0; i < mddev->raid_disks && ret == 0; i++) {
		mdk_rdev_t *rdev = rcu_dereference(conf->mirrors[i].rdev);
		if (rdev && !test_bit(Faulty, &rdev->flags)) {
			struct request_queue *q = bdev_get_queue(rdev->bdev);

			ret |= bdi_congested(&q->backing_dev_info, bits);
		}
	}
	rcu_read_unlock();
	return ret;
}


/* Barriers....
 * Sometimes we need to suspend IO while we do something else,
 * either some resync/recovery, or reconfigure the array.
 * To do this we raise a 'barrier'.
 * The 'barrier' is a counter that can be raised multiple times
 * to count how many activities are happening which preclude
 * normal IO.
 * We can only raise the barrier if there is no pending IO.
 * i.e. if nr_pending == 0.
 * We choose only to raise the barrier if no-one is waiting for the
 * barrier to go down.  This means that as soon as an IO request
 * is ready, no other operations which require a barrier will start
 * until the IO request has had a chance.
 *
 * So: regular IO calls 'wait_barrier'.  When that returns there
 *    is no backgroup IO happening,  It must arrange to call
 *    allow_barrier when it has finished its IO.
 * backgroup IO calls must call raise_barrier.  Once that returns
 *    there is no normal IO happeing.  It must arrange to call
 *    lower_barrier when the particular background IO completes.
 */
#define RESYNC_DEPTH 32

static void raise_barrier(conf_t *conf, int force)
{
	BUG_ON(force && !conf->barrier);
	spin_lock_irq(&conf->resync_lock);

	/* Wait until no block IO is waiting (unless 'force') */
	wait_event_lock_irq(conf->wait_barrier, force || !conf->nr_waiting,
			    conf->resync_lock,
			    raid10_unplug(conf->mddev->queue));

	/* block any new IO from starting */
	conf->barrier++;

	/* No wait for all pending IO to complete */
	wait_event_lock_irq(conf->wait_barrier,
			    !conf->nr_pending && conf->barrier < RESYNC_DEPTH,
			    conf->resync_lock,
			    raid10_unplug(conf->mddev->queue));

	spin_unlock_irq(&conf->resync_lock);
}

static void lower_barrier(conf_t *conf)
{
	unsigned long flags;
	spin_lock_irqsave(&conf->resync_lock, flags);
	conf->barrier--;
	spin_unlock_irqrestore(&conf->resync_lock, flags);
	wake_up(&conf->wait_barrier);
}

static void wait_barrier(conf_t *conf)
{
	spin_lock_irq(&conf->resync_lock);
	if (conf->barrier) {
		conf->nr_waiting++;
		wait_event_lock_irq(conf->wait_barrier, !conf->barrier,
				    conf->resync_lock,
				    raid10_unplug(conf->mddev->queue));
		conf->nr_waiting--;
	}
	conf->nr_pending++;
	spin_unlock_irq(&conf->resync_lock);
}

static void allow_barrier(conf_t *conf)
{
	unsigned long flags;
	spin_lock_irqsave(&conf->resync_lock, flags);
	conf->nr_pending--;
	spin_unlock_irqrestore(&conf->resync_lock, flags);
	wake_up(&conf->wait_barrier);
}

static void freeze_array(conf_t *conf)
{
	/* stop syncio and normal IO and wait for everything to
	 * go quiet.
	 * We increment barrier and nr_waiting, and then
	 * wait until barrier+nr_pending match nr_queued+2
	 */
	spin_lock_irq(&conf->resync_lock);
	conf->barrier++;
	conf->nr_waiting++;
	wait_event_lock_irq(conf->wait_barrier,
			    conf->barrier+conf->nr_pending == conf->nr_queued+2,
			    conf->resync_lock,
			    raid10_unplug(conf->mddev->queue));
	spin_unlock_irq(&conf->resync_lock);
}

static void unfreeze_array(conf_t *conf)
{
	/* reverse the effect of the freeze */
	spin_lock_irq(&conf->resync_lock);
	conf->barrier--;
	conf->nr_waiting--;
	wake_up(&conf->wait_barrier);
	spin_unlock_irq(&conf->resync_lock);
}

static int make_request(struct request_queue *q, struct bio * bio)
{
	mddev_t *mddev = q->queuedata;
	conf_t *conf = mddev_to_conf(mddev);
	mirror_info_t *mirror;
	r10bio_t *r10_bio;
	struct bio *read_bio;
	int i;
	int chunk_sects = conf->chunk_mask + 1;
	const int rw = bio_data_dir(bio);
	const int do_sync = bio_sync(bio);
	struct bio_list bl;
	unsigned long flags;

	if (unlikely(bio_barrier(bio))) {
		bio_endio(bio, -EOPNOTSUPP);
		return 0;
	}

	/* If this request crosses a chunk boundary, we need to
	 * split it.  This will only happen for 1 PAGE (or less) requests.
	 */
	if (unlikely( (bio->bi_sector & conf->chunk_mask) + (bio->bi_size >> 9)
		      > chunk_sects &&
		    conf->near_copies < conf->raid_disks)) {
		struct bio_pair *bp;
		/* Sanity check -- queue functions should prevent this happening */
		if (bio->bi_vcnt != 1 ||
		    bio->bi_idx != 0)
			goto bad_map;
		/* This is a one page bio that upper layers
		 * refuse to split for us, so we need to split it.
		 */
		bp = bio_split(bio, bio_split_pool,
			       chunk_sects - (bio->bi_sector & (chunk_sects - 1)) );
		if (make_request(q, &bp->bio1))
			generic_make_request(&bp->bio1);
		if (make_request(q, &bp->bio2))
			generic_make_request(&bp->bio2);

		bio_pair_release(bp);
		return 0;
	bad_map:
		printk("raid10_make_request bug: can't convert block across chunks"
		       " or bigger than %dk %llu %d\n", chunk_sects/2,
		       (unsigned long long)bio->bi_sector, bio->bi_size >> 10);

		bio_io_error(bio);
		return 0;
	}

	md_write_start(mddev, bio);

	/*
	 * 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.
	 */
	wait_barrier(conf);

	disk_stat_inc(mddev->gendisk, ios[rw]);
	disk_stat_add(mddev->gendisk, sectors[rw], bio_sectors(bio));

	r10_bio = mempool_alloc(conf->r10bio_pool, GFP_NOIO);

	r10_bio->master_bio = bio;
	r10_bio->sectors = bio->bi_size >> 9;

	r10_bio->mddev = mddev;
	r10_bio->sector = bio->bi_sector;
	r10_bio->state = 0;

	if (rw == READ) {
		/*
		 * read balancing logic:
		 */
		int disk = read_balance(conf, r10_bio);
		int slot = r10_bio->read_slot;
		if (disk < 0) {
			raid_end_bio_io(r10_bio);
			return 0;
		}
		mirror = conf->mirrors + disk;

		read_bio = bio_clone(bio, GFP_NOIO);

		r10_bio->devs[slot].bio = read_bio;

		read_bio->bi_sector = r10_bio->devs[slot].addr +
			mirror->rdev->data_offset;
		read_bio->bi_bdev = mirror->rdev->bdev;
		read_bio->bi_end_io = raid10_end_read_request;
		read_bio->bi_rw = READ | do_sync;
		read_bio->bi_private = r10_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
	 */
	raid10_find_phys(conf, r10_bio);
	rcu_read_lock();
	for (i = 0;  i < conf->copies; i++) {
		int d = r10_bio->devs[i].devnum;
		mdk_rdev_t *rdev = rcu_dereference(conf->mirrors[d].rdev);
		if (rdev &&
		    !test_bit(Faulty, &rdev->flags)) {
			atomic_inc(&rdev->nr_pending);
			r10_bio->devs[i].bio = bio;
		} else {
			r10_bio->devs[i].bio = NULL;
			set_bit(R10BIO_Degraded, &r10_bio->state);
		}
	}
	rcu_read_unlock();

	atomic_set(&r10_bio->remaining, 0);

	bio_list_init(&bl);
	for (i = 0; i < conf->copies; i++) {
		struct bio *mbio;
		int d = r10_bio->devs[i].devnum;
		if (!r10_bio->devs[i].bio)
			continue;

		mbio = bio_clone(bio, GFP_NOIO);
		r10_bio->devs[i].bio = mbio;

		mbio->bi_sector	= r10_bio->devs[i].addr+
			conf->mirrors[d].rdev->data_offset;
		mbio->bi_bdev = conf->mirrors[d].rdev->bdev;
		mbio->bi_end_io	= raid10_end_write_request;
		mbio->bi_rw = WRITE | do_sync;
		mbio->bi_private = r10_bio;

		atomic_inc(&r10_bio->remaining);
		bio_list_add(&bl, mbio);
	}

	if (unlikely(!atomic_read(&r10_bio->remaining))) {
		/* the array is dead */
		md_write_end(mddev);
		raid_end_bio_io(r10_bio);
		return 0;
	}

	bitmap_startwrite(mddev->bitmap, bio->bi_sector, r10_bio->sectors, 0);
	spin_lock_irqsave(&conf->device_lock, flags);
	bio_list_merge(&conf->pending_bio_list, &bl);
	blk_plug_device(mddev->queue);
	spin_unlock_irqrestore(&conf->device_lock, flags);

	if (do_sync)
		md_wakeup_thread(mddev->thread);

	return 0;
}

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

	if (conf->near_copies < conf->raid_disks)
		seq_printf(seq, " %dK chunks", mddev->chunk_size/1024);
	if (conf->near_copies > 1)
		seq_printf(seq, " %d near-copies", conf->near_copies);
	if (conf->far_copies > 1) {
		if (conf->far_offset)
			seq_printf(seq, " %d offset-copies", conf->far_copies);
		else
			seq_printf(seq, " %d far-copies", conf->far_copies);
	}
	seq_printf(seq, " [%d/%d] [", conf->raid_disks,
					conf->raid_disks - mddev->degraded);
	for (i = 0; i < conf->raid_disks; i++)
		seq_printf(seq, "%s",
			      conf->mirrors[i].rdev &&
			      test_bit(In_sync, &conf->mirrors[i].rdev->flags) ? "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 (test_bit(In_sync, &rdev->flags)
	    && conf->raid_disks-mddev->degraded == 1)
		/*
		 * Don't fail the drive, just return an IO error.
		 * The test should really be more sophisticated than
		 * "working_disks == 1", but it isn't critical, and
		 * can wait until we do more sophisticated "is the drive
		 * really dead" tests...
		 */
		return;
	if (test_and_clear_bit(In_sync, &rdev->flags)) {
		unsigned long flags;
		spin_lock_irqsave(&conf->device_lock, flags);
		mddev->degraded++;
		spin_unlock_irqrestore(&conf->device_lock, flags);
		/*
		 * if recovery is running, make sure it aborts.
		 */
		set_bit(MD_RECOVERY_ERR, &mddev->recovery);
	}
	set_bit(Faulty, &rdev->flags);
	set_bit(MD_CHANGE_DEVS, &mddev->flags);
	printk(KERN_ALERT "raid10: Disk failure on %s, disabling device. \n"
		"	Operation continuing on %d devices\n",
		bdevname(rdev->bdev,b), conf->raid_disks - mddev->degraded);
}

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

	printk("RAID10 conf printout:\n");
	if (!conf) {
		printk("(!conf)\n");
		return;
	}
	printk(" --- wd:%d rd:%d\n", conf->raid_disks - conf->mddev->degraded,
		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, !test_bit(In_sync, &tmp->rdev->flags),
			        !test_bit(Faulty, &tmp->rdev->flags),
				bdevname(tmp->rdev->bdev,b));
	}
}

static void close_sync(conf_t *conf)
{
	wait_barrier(conf);
	allow_barrier(conf);

	mempool_destroy(conf->r10buf_pool);
	conf->r10buf_pool = NULL;
}

/* check if there are enough drives for
 * every block to appear on atleast one
 */
static int enough(conf_t *conf)
{
	int first = 0;

	do {
		int n = conf->copies;
		int cnt = 0;
		while (n--) {
			if (conf->mirrors[first].rdev)
				cnt++;
			first = (first+1) % conf->raid_disks;
		}
		if (cnt == 0)
			return 0;
	} while (first != 0);
	return 1;
}

static int raid10_spare_active(mddev_t *mddev)
{
	int i;
	conf_t *conf = mddev->private;
	mirror_info_t *tmp;

	/*
	 * Find all non-in_sync disks within the RAID10 configuration
	 * and mark them in_sync
	 */
	for (i = 0; i < conf->raid_disks; i++) {
		tmp = conf->mirrors + i;
		if (tmp->rdev
		    && !test_bit(Faulty, &tmp->rdev->flags)
		    && !test_and_set_bit(In_sync, &tmp->rdev->flags)) {
			unsigned long flags;
			spin_lock_irqsave(&conf->device_lock, flags);
			mddev->degraded--;
			spin_unlock_irqrestore(&conf->device_lock, flags);
		}
	}

	print_conf(conf);
	return 0;
}


static int raid10_add_disk(mddev_t *mddev, mdk_rdev_t *rdev)
{
	conf_t *conf = mddev->private;
	int found = 0;
	int mirror;
	mirror_info_t *p;

	if (mddev->recovery_cp < MaxSector)
		/* only hot-add to in-sync arrays, as recovery is
		 * very different from resync
		 */
		return 0;
	if (!enough(conf))
		return 0;

	if (rdev->saved_raid_disk >= 0 &&
	    conf->mirrors[rdev->saved_raid_disk].rdev == NULL)
		mirror = rdev->saved_raid_disk;
	else
		mirror = 0;
	for ( ; mirror < mddev->raid_disks; mirror++)
		if ( !(p=conf->mirrors+mirror)->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))
				mddev->queue->max_sectors = (PAGE_SIZE>>9);

			p->head_position = 0;
			rdev->raid_disk = mirror;
			found = 1;
			if (rdev->saved_raid_disk != mirror)
				conf->fullsync = 1;
			rcu_assign_pointer(p->rdev, rdev);
			break;
		}

	print_conf(conf);
	return found;
}

static int raid10_remove_disk(mddev_t *mddev, int number)
{
	conf_t *conf = mddev->private;
	int err = 0;
	mdk_rdev_t *rdev;