/drivers/md/raid10.c

/*
 * raid10.c : Multiple Devices driver for Linux
 *
 * Copyright (C) 2000-2004 Neil Brown
 *
 * RAID-10 support for md.
 *
 * Base on code in raid1.c.  See raid1.c for further copyright information.
 *
 *
 * 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/slab.h>
#include <linux/delay.h>
#include <linux/blkdev.h>
#include <linux/module.h>
#include <linux/seq_file.h>
#include <linux/ratelimit.h>
#include <linux/kthread.h>
#include "md.h"
#include "raid10.h"
#include "raid0.h"
#include "bitmap.h"

/*
 * RAID10 provides a combination of RAID0 and RAID1 functionality.
 * The layout of data is defined by
 *    chunk_size
 *    raid_disks
 *    near_copies (stored in low byte of layout)
 *    far_copies (stored in second byte of layout)
 *    far_offset (stored in bit 16 of layout )
 *    use_far_sets (stored in bit 17 of layout )
 *    use_far_sets_bugfixed (stored in bit 18 of layout )
 *
 * The data to be stored is divided into chunks using chunksize.  Each device
 * is divided into far_copies sections.   In each section, chunks are laid out
 * in a style similar to raid0, but near_copies copies of each chunk is stored
 * (each on a different drive).  The starting device for each section is offset
 * near_copies from the starting device of the previous section.  Thus there
 * are (near_copies * far_copies) of each chunk, and each is on a different
 * drive.  near_copies and far_copies must be at least one, and their product
 * is at most raid_disks.
 *
 * If far_offset is true, then the far_copies are handled a bit differently.
 * The copies are still in different stripes, but instead of being very far
 * apart on disk, there are adjacent stripes.
 *
 * The far and offset algorithms are handled slightly differently if
 * 'use_far_sets' is true.  In this case, the array's devices are grouped into
 * sets that are (near_copies * far_copies) in size.  The far copied stripes
 * are still shifted by 'near_copies' devices, but this shifting stays confined
 * to the set rather than the entire array.  This is done to improve the number
 * of device combinations that can fail without causing the array to fail.
 * Example 'far' algorithm w/o 'use_far_sets' (each letter represents a chunk
 * on a device):
 *    A B C D    A B C D E
 *      ...         ...
 *    D A B C    E A B C D
 * Example 'far' algorithm w/ 'use_far_sets' enabled (sets illustrated w/ []'s):
 *    [A B] [C D]    [A B] [C D E]
 *    |...| |...|    |...| | ... |
 *    [B A] [D C]    [B A] [E C D]
 */

/*
 * Number of guaranteed r10bios in case of extreme VM load:
 */
#define	NR_RAID10_BIOS 256

/* when we get a read error on a read-only array, we redirect to another
 * device without failing the first device, or trying to over-write to
 * correct the read error.  To keep track of bad blocks on a per-bio
 * level, we store IO_BLOCKED in the appropriate 'bios' pointer
 */
#define IO_BLOCKED ((struct bio *)1)
/* When we successfully write to a known bad-block, we need to remove the
 * bad-block marking which must be done from process context.  So we record
 * the success by setting devs[n].bio to IO_MADE_GOOD
 */
#define IO_MADE_GOOD ((struct bio *)2)

#define BIO_SPECIAL(bio) ((unsigned long)bio <= 2)

/* When there are this many requests queued to be written by
 * the raid10 thread, we become 'congested' to provide back-pressure
 * for writeback.
 */
static int max_queued_requests = 1024;

static void allow_barrier(struct r10conf *conf);
static void lower_barrier(struct r10conf *conf);
static int _enough(struct r10conf *conf, int previous, int ignore);
static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr,
				int *skipped);
static void reshape_request_write(struct mddev *mddev, struct r10bio *r10_bio);
static void end_reshape_write(struct bio *bio);
static void end_reshape(struct r10conf *conf);

static void * r10bio_pool_alloc(gfp_t gfp_flags, void *data)
{
	struct r10conf *conf = data;
	int size = offsetof(struct r10bio, devs[conf->copies]);

	/* allocate a r10bio with room for raid_disks entries in the
	 * bios array */
	return kzalloc(size, gfp_flags);
}

static void r10bio_pool_free(void *r10_bio, void *data)
{
	kfree(r10_bio);
}

/* Maximum size of each resync request */
#define RESYNC_BLOCK_SIZE (64*1024)
#define RESYNC_PAGES ((RESYNC_BLOCK_SIZE + PAGE_SIZE-1) / PAGE_SIZE)
/* amount of memory to reserve for resync requests */
#define RESYNC_WINDOW (1024*1024)
/* maximum number of concurrent requests, memory permitting */
#define RESYNC_DEPTH (32*1024*1024/RESYNC_BLOCK_SIZE)

/*
 * When performing a resync, we need to read and compare, so
 * we need as many pages are there are copies.
 * When performing a recovery, we need 2 bios, one for read,
 * one for write (we recover only one drive per r10buf)
 *
 */
static void * r10buf_pool_alloc(gfp_t gfp_flags, void *data)
{
	struct r10conf *conf = data;
	struct page *page;
	struct r10bio *r10_bio;
	struct bio *bio;
	int i, j;
	int nalloc;

	r10_bio = r10bio_pool_alloc(gfp_flags, conf);
	if (!r10_bio)
		return NULL;

	if (test_bit(MD_RECOVERY_SYNC, &conf->mddev->recovery) ||
	    test_bit(MD_RECOVERY_RESHAPE, &conf->mddev->recovery))
		nalloc = conf->copies; /* resync */
	else
		nalloc = 2; /* recovery */

	/*
	 * Allocate bios.
	 */
	for (j = nalloc ; j-- ; ) {
		bio = bio_kmalloc(gfp_flags, RESYNC_PAGES);
		if (!bio)
			goto out_free_bio;
		r10_bio->devs[j].bio = bio;
		if (!conf->have_replacement)
			continue;
		bio = bio_kmalloc(gfp_flags, RESYNC_PAGES);
		if (!bio)
			goto out_free_bio;
		r10_bio->devs[j].repl_bio = bio;
	}
	/*
	 * Allocate RESYNC_PAGES data pages and attach them
	 * where needed.
	 */
	for (j = 0 ; j < nalloc; j++) {
		struct bio *rbio = r10_bio->devs[j].repl_bio;
		bio = r10_bio->devs[j].bio;
		for (i = 0; i < RESYNC_PAGES; i++) {
			if (j > 0 && !test_bit(MD_RECOVERY_SYNC,
					       &conf->mddev->recovery)) {
				/* we can share bv_page's during recovery
				 * and reshape */
				struct bio *rbio = r10_bio->devs[0].bio;
				page = rbio->bi_io_vec[i].bv_page;
				get_page(page);
			} else
				page = alloc_page(gfp_flags);
			if (unlikely(!page))
				goto out_free_pages;

			bio->bi_io_vec[i].bv_page = page;
			if (rbio)
				rbio->bi_io_vec[i].bv_page = page;
		}
	}

	return r10_bio;

out_free_pages:
	for ( ; i > 0 ; i--)
		safe_put_page(bio->bi_io_vec[i-1].bv_page);
	while (j--)
		for (i = 0; i < RESYNC_PAGES ; i++)
			safe_put_page(r10_bio->devs[j].bio->bi_io_vec[i].bv_page);
	j = 0;
out_free_bio:
	for ( ; j < nalloc; j++) {
		if (r10_bio->devs[j].bio)
			bio_put(r10_bio->devs[j].bio);
		if (r10_bio->devs[j].repl_bio)
			bio_put(r10_bio->devs[j].repl_bio);
	}
	r10bio_pool_free(r10_bio, conf);
	return NULL;
}

static void r10buf_pool_free(void *__r10_bio, void *data)
{
	int i;
	struct r10conf *conf = data;
	struct r10bio *r10bio = __r10_bio;
	int j;

	for (j=0; j < conf->copies; j++) {
		struct bio *bio = r10bio->devs[j].bio;
		if (bio) {
			for (i = 0; i < RESYNC_PAGES; i++) {
				safe_put_page(bio->bi_io_vec[i].bv_page);
				bio->bi_io_vec[i].bv_page = NULL;
			}
			bio_put(bio);
		}
		bio = r10bio->devs[j].repl_bio;
		if (bio)
			bio_put(bio);
	}
	r10bio_pool_free(r10bio, conf);
}

static void put_all_bios(struct r10conf *conf, struct r10bio *r10_bio)
{
	int i;

	for (i = 0; i < conf->copies; i++) {
		struct bio **bio = & r10_bio->devs[i].bio;
		if (!BIO_SPECIAL(*bio))
			bio_put(*bio);
		*bio = NULL;
		bio = &r10_bio->devs[i].repl_bio;
		if (r10_bio->read_slot < 0 && !BIO_SPECIAL(*bio))
			bio_put(*bio);
		*bio = NULL;
	}
}

static void free_r10bio(struct r10bio *r10_bio)
{
	struct r10conf *conf = r10_bio->mddev->private;

	put_all_bios(conf, r10_bio);
	mempool_free(r10_bio, conf->r10bio_pool);
}

static void put_buf(struct r10bio *r10_bio)
{
	struct r10conf *conf = r10_bio->mddev->private;

	mempool_free(r10_bio, conf->r10buf_pool);

	lower_barrier(conf);
}

static void reschedule_retry(struct r10bio *r10_bio)
{
	unsigned long flags;
	struct mddev *mddev = r10_bio->mddev;
	struct r10conf *conf = mddev->private;

	spin_lock_irqsave(&conf->device_lock, flags);
	list_add(&r10_bio->retry_list, &conf->retry_list);
	conf->nr_queued ++;
	spin_unlock_irqrestore(&conf->device_lock, flags);

	/* wake up frozen array... */
	wake_up(&conf->wait_barrier);

	md_wakeup_thread(mddev->thread);
}

/*
 * raid_end_bio_io() is called when we have finished servicing a mirrored
 * operation and are ready to return a success/failure code to the buffer
 * cache layer.
 */
static void raid_end_bio_io(struct r10bio *r10_bio)
{
	struct bio *bio = r10_bio->master_bio;
	int done;
	struct r10conf *conf = r10_bio->mddev->private;

	if (bio->bi_phys_segments) {
		unsigned long flags;
		spin_lock_irqsave(&conf->device_lock, flags);
		bio->bi_phys_segments--;
		done = (bio->bi_phys_segments == 0);
		spin_unlock_irqrestore(&conf->device_lock, flags);
	} else
		done = 1;
	if (!test_bit(R10BIO_Uptodate, &r10_bio->state))
		bio->bi_error = -EIO;
	if (done) {
		bio_endio(bio);
		/*
		 * Wake up any possible resync thread that waits for the device
		 * to go idle.
		 */
		allow_barrier(conf);
	}
	free_r10bio(r10_bio);
}

/*
 * Update disk head position estimator based on IRQ completion info.
 */
static inline void update_head_pos(int slot, struct r10bio *r10_bio)
{
	struct r10conf *conf = r10_bio->mddev->private;

	conf->mirrors[r10_bio->devs[slot].devnum].head_position =
		r10_bio->devs[slot].addr + (r10_bio->sectors);
}

/*
 * Find the disk number which triggered given bio
 */
static int find_bio_disk(struct r10conf *conf, struct r10bio *r10_bio,
			 struct bio *bio, int *slotp, int *replp)
{
	int slot;
	int repl = 0;

	for (slot = 0; slot < conf->copies; slot++) {
		if (r10_bio->devs[slot].bio == bio)
			break;
		if (r10_bio->devs[slot].repl_bio == bio) {
			repl = 1;
			break;
		}
	}

	BUG_ON(slot == conf->copies);
	update_head_pos(slot, r10_bio);

	if (slotp)
		*slotp = slot;
	if (replp)
		*replp = repl;
	return r10_bio->devs[slot].devnum;
}

static void raid10_end_read_request(struct bio *bio)
{
	int uptodate = !bio->bi_error;
	struct r10bio *r10_bio = bio->bi_private;
	int slot, dev;
	struct md_rdev *rdev;
	struct r10conf *conf = r10_bio->mddev->private;

	slot = r10_bio->read_slot;
	dev = r10_bio->devs[slot].devnum;
	rdev = r10_bio->devs[slot].rdev;
	/*
	 * this branch is our 'one mirror IO has finished' event handler:
	 */
	update_head_pos(slot, r10_bio);

	if (uptodate) {
		/*
		 * Set R10BIO_Uptodate in our master bio, so that
		 * we will return a good error code to the higher
		 * levels even if IO on some other mirrored buffer fails.
		 *
		 * The 'master' represents the composite IO operation to
		 * user-side. So if something waits for IO, then it will
		 * wait for the 'master' bio.
		 */
		set_bit(R10BIO_Uptodate, &r10_bio->state);
	} else {
		/* If all other devices that store this block have
		 * failed, we want to return the error upwards rather
		 * than fail the last device.  Here we redefine
		 * "uptodate" to mean "Don't want to retry"
		 */
		if (!_enough(conf, test_bit(R10BIO_Previous, &r10_bio->state),
			     rdev->raid_disk))
			uptodate = 1;
	}
	if (uptodate) {
		raid_end_bio_io(r10_bio);
		rdev_dec_pending(rdev, conf->mddev);
	} else {
		/*
		 * oops, read error - keep the refcount on the rdev
		 */
		char b[BDEVNAME_SIZE];
		printk_ratelimited(KERN_ERR
				   "md/raid10:%s: %s: rescheduling sector %llu\n",
				   mdname(conf->mddev),
				   bdevname(rdev->bdev, b),
				   (unsigned long long)r10_bio->sector);
		set_bit(R10BIO_ReadError, &r10_bio->state);
		reschedule_retry(r10_bio);
	}
}

static void close_write(struct r10bio *r10_bio)
{
	/* clear the bitmap if all writes complete successfully */
	bitmap_endwrite(r10_bio->mddev->bitmap, r10_bio->sector,
			r10_bio->sectors,
			!test_bit(R10BIO_Degraded, &r10_bio->state),
			0);
	md_write_end(r10_bio->mddev);
}

static void one_write_done(struct r10bio *r10_bio)
{
	if (atomic_dec_and_test(&r10_bio->remaining)) {
		if (test_bit(R10BIO_WriteError, &r10_bio->state))
			reschedule_retry(r10_bio);
		else {
			close_write(r10_bio);
			if (test_bit(R10BIO_MadeGood, &r10_bio->state))
				reschedule_retry(r10_bio);
			else
				raid_end_bio_io(r10_bio);
		}
	}
}

static void raid10_end_write_request(struct bio *bio)
{
	struct r10bio *r10_bio = bio->bi_private;
	int dev;
	int dec_rdev = 1;
	struct r10conf *conf = r10_bio->mddev->private;
	int slot, repl;
	struct md_rdev *rdev = NULL;

	dev = find_bio_disk(conf, r10_bio, bio, &slot, &repl);

	if (repl)
		rdev = conf->mirrors[dev].replacement;
	if (!rdev) {
		smp_rmb();
		repl = 0;
		rdev = conf->mirrors[dev].rdev;
	}
	/*
	 * this branch is our 'one mirror IO has finished' event handler:
	 */
	if (bio->bi_error) {
		if (repl)
			/* Never record new bad blocks to replacement,
			 * just fail it.
			 */
			md_error(rdev->mddev, rdev);
		else {
			set_bit(WriteErrorSeen,	&rdev->flags);
			if (!test_and_set_bit(WantReplacement, &rdev->flags))
				set_bit(MD_RECOVERY_NEEDED,
					&rdev->mddev->recovery);
			set_bit(R10BIO_WriteError, &r10_bio->state);
			dec_rdev = 0;
		}
	} else {
		/*
		 * Set R10BIO_Uptodate in our master bio, so that
		 * we will return a good error code for to the higher
		 * levels even if IO on some other mirrored buffer fails.
		 *
		 * The 'master' represents the composite IO operation to
		 * user-side. So if something waits for IO, then it will
		 * wait for the 'master' bio.
		 */
		sector_t first_bad;
		int bad_sectors;

		/*
		 * Do not set R10BIO_Uptodate if the current device is
		 * rebuilding or Faulty. This is because we cannot use
		 * such device for properly reading the data back (we could
		 * potentially use it, if the current write would have felt
		 * before rdev->recovery_offset, but for simplicity we don't
		 * check this here.
		 */
		if (test_bit(In_sync, &rdev->flags) &&
		    !test_bit(Faulty, &rdev->flags))
			set_bit(R10BIO_Uptodate, &r10_bio->state);

		/* Maybe we can clear some bad blocks. */
		if (is_badblock(rdev,
				r10_bio->devs[slot].addr,
				r10_bio->sectors,
				&first_bad, &bad_sectors)) {
			bio_put(bio);
			if (repl)
				r10_bio->devs[slot].repl_bio = IO_MADE_GOOD;
			else
				r10_bio->devs[slot].bio = IO_MADE_GOOD;
			dec_rdev = 0;
			set_bit(R10BIO_MadeGood, &r10_bio->state);
		}
	}

	/*
	 *
	 * Let's see if all mirrored write operations have finished
	 * already.
	 */
	one_write_done(r10_bio);
	if (dec_rdev)
		rdev_dec_pending(rdev, conf->mddev);
}

/*
 * RAID10 layout manager
 * As well as the chunksize and raid_disks count, there are two
 * parameters: near_copies and far_copies.
 * near_copies * far_copies must be <= raid_disks.
 * Normally one of these will be 1.
 * If both are 1, we get raid0.
 * If near_copies == raid_disks, we get raid1.
 *
 * Chunks are laid out in raid0 style with near_copies copies of the
 * first chunk, followed by near_copies copies of the next chunk and
 * so on.
 * If far_copies > 1, then after 1/far_copies of the array has been assigned
 * as described above, we start again with a device offset of near_copies.
 * So we effectively have another copy of the whole array further down all
 * the drives, but with blocks on different drives.
 * With this layout, and block is never stored twice on the one device.
 *
 * raid10_find_phys finds the sector offset of a given virtual sector
 * on each device that it is on.
 *
 * raid10_find_virt does the reverse mapping, from a device and a
 * sector offset to a virtual address
 */

static void __raid10_find_phys(struct geom *geo, struct r10bio *r10bio)
{
	int n,f;
	sector_t sector;
	sector_t chunk;
	sector_t stripe;
	int dev;
	int slot = 0;
	int last_far_set_start, last_far_set_size;

	last_far_set_start = (geo->raid_disks / geo->far_set_size) - 1;
	last_far_set_start *= geo->far_set_size;

	last_far_set_size = geo->far_set_size;
	last_far_set_size += (geo->raid_disks % geo->far_set_size);

	/* now calculate first sector/dev */
	chunk = r10bio->sector >> geo->chunk_shift;
	sector = r10bio->sector & geo->chunk_mask;

	chunk *= geo->near_copies;
	stripe = chunk;
	dev = sector_div(stripe, geo->raid_disks);
	if (geo->far_offset)
		stripe *= geo->far_copies;

	sector += stripe << geo->chunk_shift;

	/* and calculate all the others */
	for (n = 0; n < geo->near_copies; n++) {
		int d = dev;
		int set;
		sector_t s = sector;
		r10bio->devs[slot].devnum = d;
		r10bio->devs[slot].addr = s;
		slot++;

		for (f = 1; f < geo->far_copies; f++) {
			set = d / geo->far_set_size;
			d += geo->near_copies;

			if ((geo->raid_disks % geo->far_set_size) &&
			    (d > last_far_set_start)) {
				d -= last_far_set_start;
				d %= last_far_set_size;
				d += last_far_set_start;
			} else {
				d %= geo->far_set_size;
				d += geo->far_set_size * set;
			}
			s += geo->stride;
			r10bio->devs[slot].devnum = d;
			r10bio->devs[slot].addr = s;
			slot++;
		}
		dev++;
		if (dev >= geo->raid_disks) {
			dev = 0;
			sector += (geo->chunk_mask + 1);
		}
	}
}

static void raid10_find_phys(struct r10conf *conf, struct r10bio *r10bio)
{
	struct geom *geo = &conf->geo;

	if (conf->reshape_progress != MaxSector &&
	    ((r10bio->sector >= conf->reshape_progress) !=
	     conf->mddev->reshape_backwards)) {
		set_bit(R10BIO_Previous, &r10bio->state);
		geo = &conf->prev;
	} else
		clear_bit(R10BIO_Previous, &r10bio->state);

	__raid10_find_phys(geo, r10bio);
}

static sector_t raid10_find_virt(struct r10conf *conf, sector_t sector, int dev)
{
	sector_t offset, chunk, vchunk;
	/* Never use conf->prev as this is only called during resync
	 * or recovery, so reshape isn't happening
	 */
	struct geom *geo = &conf->geo;
	int far_set_start = (dev / geo->far_set_size) * geo->far_set_size;
	int far_set_size = geo->far_set_size;
	int last_far_set_start;

	if (geo->raid_disks % geo->far_set_size) {
		last_far_set_start = (geo->raid_disks / geo->far_set_size) - 1;
		last_far_set_start *= geo->far_set_size;

		if (dev >= last_far_set_start) {
			far_set_size = geo->far_set_size;
			far_set_size += (geo->raid_disks % geo->far_set_size);
			far_set_start = last_far_set_start;
		}
	}

	offset = sector & geo->chunk_mask;
	if (geo->far_offset) {
		int fc;
		chunk = sector >> geo->chunk_shift;
		fc = sector_div(chunk, geo->far_copies);
		dev -= fc * geo->near_copies;
		if (dev < far_set_start)
			dev += far_set_size;
	} else {
		while (sector >= geo->stride) {
			sector -= geo->stride;
			if (dev < (geo->near_copies + far_set_start))
				dev += far_set_size - geo->near_copies;
			else
				dev -= geo->near_copies;
		}
		chunk = sector >> geo->chunk_shift;
	}
	vchunk = chunk * geo->raid_disks + dev;
	sector_div(vchunk, geo->near_copies);
	return (vchunk << geo->chunk_shift) + offset;
}

/*
 * This routine returns the disk from which the requested read should
 * be done. There is a per-array 'next expected sequential IO' sector
 * number - if this matches on the next IO then we use the last disk.
 * There is also a per-disk 'last know head position' sector that is
 * maintained from IRQ contexts, both the normal and the resync IO
 * completion handlers update this position correctly. If there is no
 * perfect sequential match then we pick the disk whose head is closest.
 *
 * If there are 2 mirrors in the same 2 devices, performance degrades
 * because position is mirror, not device based.
 *
 * The rdev for the device selected will have nr_pending incremented.
 */

/*
 * FIXME: possibly should rethink readbalancing and do it differently
 * depending on near_copies / far_copies geometry.
 */
static struct md_rdev *read_balance(struct r10conf *conf,
				    struct r10bio *r10_bio,
				    int *max_sectors)
{
	const sector_t this_sector = r10_bio->sector;
	int disk, slot;
	int sectors = r10_bio->sectors;
	int best_good_sectors;
	sector_t new_distance, best_dist;
	struct md_rdev *best_rdev, *rdev = NULL;
	int do_balance;
	int best_slot;
	struct geom *geo = &conf->geo;

	raid10_find_phys(conf, r10_bio);
	rcu_read_lock();
retry:
	sectors = r10_bio->sectors;
	best_slot = -1;
	best_rdev = NULL;
	best_dist = MaxSector;
	best_good_sectors = 0;
	do_balance = 1;
	/*
	 * Check if we can balance. We can balance on the whole
	 * device if no resync is going on (recovery is ok), or below
	 * the resync window. We take the first readable disk when
	 * above the resync window.
	 */
	if (conf->mddev->recovery_cp < MaxSector
	    && (this_sector + sectors >= conf->next_resync))
		do_balance = 0;

	for (slot = 0; slot < conf->copies ; slot++) {
		sector_t first_bad;
		int bad_sectors;
		sector_t dev_sector;

		if (r10_bio->devs[slot].bio == IO_BLOCKED)
			continue;
		disk = r10_bio->devs[slot].devnum;
		rdev = rcu_dereference(conf->mirrors[disk].replacement);
		if (rdev == NULL || test_bit(Faulty, &rdev->flags) ||
		    r10_bio->devs[slot].addr + sectors > rdev->recovery_offset)
			rdev = rcu_dereference(conf->mirrors[disk].rdev);
		if (rdev == NULL ||
		    test_bit(Faulty, &rdev->flags))
			continue;
		if (!test_bit(In_sync, &rdev->flags) &&
		    r10_bio->devs[slot].addr + sectors > rdev->recovery_offset)
			continue;

		dev_sector = r10_bio->devs[slot].addr;
		if (is_badblock(rdev, dev_sector, sectors,
				&first_bad, &bad_sectors)) {
			if (best_dist < MaxSector)
				/* Already have a better slot */
				continue;
			if (first_bad <= dev_sector) {
				/* Cannot read here.  If this is the
				 * 'primary' device, then we must not read
				 * beyond 'bad_sectors' from another device.
				 */
				bad_sectors -= (dev_sector - first_bad);
				if (!do_balance && sectors > bad_sectors)
					sectors = bad_sectors;
				if (best_good_sectors > sectors)
					best_good_sectors = sectors;
			} else {
				sector_t good_sectors =
					first_bad - dev_sector;
				if (good_sectors > best_good_sectors) {
					best_good_sectors = good_sectors;
					best_slot = slot;
					best_rdev = rdev;
				}
				if (!do_balance)
					/* Must read from here */
					break;
			}
			continue;
		} else
			best_good_sectors = sectors;

		if (!do_balance)
			break;

		/* 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 (geo->near_copies > 1 && !atomic_read(&rdev->nr_pending))
			break;

		/* for far > 1 always use the lowest address */
		if (geo->far_copies > 1)
			new_distance = r10_bio->devs[slot].addr;
		else
			new_distance = abs(r10_bio->devs[slot].addr -
					   conf->mirrors[disk].head_position);
		if (new_distance < best_dist) {
			best_dist = new_distance;
			best_slot = slot;
			best_rdev = rdev;
		}
	}
	if (slot >= conf->copies) {
		slot = best_slot;
		rdev = best_rdev;
	}

	if (slot >= 0) {
		atomic_inc(&rdev->nr_pending);
		if (test_bit(Faulty, &rdev->flags)) {
			/* Cannot risk returning a device that failed
			 * before we inc'ed nr_pending
			 */
			rdev_dec_pending(rdev, conf->mddev);
			goto retry;
		}
		r10_bio->read_slot = slot;
	} else
		rdev = NULL;
	rcu_read_unlock();
	*max_sectors = best_good_sectors;

	return rdev;
}

static int raid10_congested(struct mddev *mddev, int bits)
{
	struct r10conf *conf = mddev->private;
	int i, ret = 0;

	if ((bits & (1 << WB_async_congested)) &&
	    conf->pending_count >= max_queued_requests)
		return 1;

	rcu_read_lock();
	for (i = 0;
	     (i < conf->geo.raid_disks || i < conf->prev.raid_disks)
		     && ret == 0;
	     i++) {
		struct md_rdev *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;
}

static void flush_pending_writes(struct r10conf *conf)
{
	/* Any writes that have been queued but are awaiting
	 * bitmap updates get flushed here.
	 */
	spin_lock_irq(&conf->device_lock);

	if (conf->pending_bio_list.head) {
		struct bio *bio;
		bio = bio_list_get(&conf->pending_bio_list);
		conf->pending_count = 0;
		spin_unlock_irq(&conf->device_lock);
		/* flush any pending bitmap writes to disk
		 * before proceeding w/ I/O */
		bitmap_unplug(conf->mddev->bitmap);
		wake_up(&conf->wait_barrier);

		while (bio) { /* submit pending writes */
			struct bio *next = bio->bi_next;
			bio->bi_next = NULL;
			if (unlikely((bio->bi_rw & REQ_DISCARD) &&
			    !blk_queue_discard(bdev_get_queue(bio->bi_bdev))))
				/* Just ignore it */
				bio_endio(bio);
			else
				generic_make_request(bio);
			bio = next;
		}
	} else
		spin_unlock_irq(&conf->device_lock);
}

/* 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.
 */

static void raise_barrier(struct r10conf *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);

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

	/* Now wait for all pending IO to complete */
	wait_event_lock_irq(conf->wait_barrier,
			    !conf->nr_pending && conf->barrier < RESYNC_DEPTH,
			    conf->resync_lock);

	spin_unlock_irq(&conf->resync_lock);
}

static void lower_barrier(struct r10conf *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(struct r10conf *conf)
{
	spin_lock_irq(&conf->resync_lock);
	if (conf->barrier) {
		conf->nr_waiting++;
		/* Wait for the barrier to drop.
		 * However if there are already pending
		 * requests (preventing the barrier from
		 * rising completely), and the
		 * pre-process bio queue isn't empty,
		 * then don't wait, as we need to empty
		 * that queue to get the nr_pending
		 * count down.
		 */
		wait_event_lock_irq(conf->wait_barrier,
				    !conf->barrier ||
				    (conf->nr_pending &&
				     current->bio_list &&
				     !bio_list_empty(current->bio_list)),
				    conf->resync_lock);
		conf->nr_waiting--;
	}
	conf->nr_pending++;
	spin_unlock_irq(&conf->resync_lock);
}

static void allow_barrier(struct r10conf *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(struct r10conf *conf, int extra)
{
	/* stop syncio and normal IO and wait for everything to
	 * go quiet.
	 * We increment barrier and nr_waiting, and then
	 * wait until nr_pending match nr_queued+extra
	 * This is called in the context of one normal IO request
	 * that has failed. Thus any sync request that might be pending
	 * will be blocked by nr_pending, and we need to wait for
	 * pending IO requests to complete or be queued for re-try.
	 * Thus the number queued (nr_queued) plus this request (extra)
	 * must match the number of pending IOs (nr_pending) before
	 * we continue.
	 */
	spin_lock_irq(&conf->resync_lock);
	conf->barrier++;
	conf->nr_waiting++;
	wait_event_lock_irq_cmd(conf->wait_barrier,
				conf->nr_pending == conf->nr_queued+extra,
				conf->resync_lock,
				flush_pending_writes(conf));

	spin_unlock_irq(&conf->resync_lock);
}

static void unfreeze_array(struct r10conf *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 sector_t choose_data_offset(struct r10bio *r10_bio,
				   struct md_rdev *rdev)
{
	if (!test_bit(MD_RECOVERY_RESHAPE, &rdev->mddev->recovery) ||
	    test_bit(R10BIO_Previous, &r10_bio->state))
		return rdev->data_offset;
	else
		return rdev->new_data_offset;
}

struct raid10_plug_cb {
	struct blk_plug_cb	cb;
	struct bio_list		pending;
	int			pending_cnt;
};

static void raid10_unplug(struct blk_plug_cb *cb, bool from_schedule)
{
	struct raid10_plug_cb *plug = container_of(cb, struct raid10_plug_cb,
						   cb);
	struct mddev *mddev = plug->cb.data;
	struct r10conf *conf = mddev->private;
	struct bio *bio;

	if (from_schedule || current->bio_list) {
		spin_lock_irq(&conf->device_lock);
		bio_list_merge(&conf->pending_bio_list, &plug->pending);
		conf->pending_count += plug->pending_cnt;
		spin_unlock_irq(&conf->device_lock);
		wake_up(&conf->wait_barrier);
		md_wakeup_thread(mddev->thread);
		kfree(plug);
		return;
	}

	/* we aren't scheduling, so we can do the write-out directly. */
	bio = bio_list_get(&plug->pending);
	bitmap_unplug(mddev->bitmap);
	wake_up(&conf->wait_barrier);

	while (bio) { /* submit pending writes */
		struct bio *next = bio->bi_next;
		bio->bi_next = NULL;
		if (unlikely((bio->bi_rw & REQ_DISCARD) &&
		    !blk_queue_discard(bdev_get_queue(bio->bi_bdev))))
			/* Just ignore it */
			bio_endio(bio);
		else
			generic_make_request(bio);
		bio = next;
	}
	kfree(plug);
}

static void __make_request(struct mddev *mddev, struct bio *bio)
{
	struct r10conf *conf = mddev->private;
	struct r10bio *r10_bio;
	struct bio *read_bio;
	int i;
	const int rw = bio_data_dir(bio);
	const unsigned long do_sync = (bio->bi_rw & REQ_SYNC);
	const unsigned long do_fua = (bio->bi_rw & REQ_FUA);
	const unsigned long do_discard = (bio->bi_rw
					  & (REQ_DISCARD | REQ_SECURE));
	const unsigned long do_same = (bio->bi_rw & REQ_WRITE_SAME);
	unsigned long flags;
	struct md_rdev *blocked_rdev;
	struct blk_plug_cb *cb;
	struct raid10_plug_cb *plug = NULL;
	int sectors_handled;
	int max_sectors;
	int sectors;

	/*
	 * 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);

	sectors = bio_sectors(bio);
	while (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery) &&
	    bio->bi_iter.bi_sector < conf->reshape_progress &&
	    bio->bi_iter.bi_sector + sectors > conf->reshape_progress) {
		/* IO spans the reshape position.  Need to wait for
		 * reshape to pass
		 */
		allow_barrier(conf);
		wait_event(conf->wait_barrier,
			   conf->reshape_progress <= bio->bi_iter.bi_sector ||
			   conf->reshape_progress >= bio->bi_iter.bi_sector +
			   sectors);
		wait_barrier(conf);
	}
	if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery) &&
	    bio_data_dir(bio) == WRITE &&
	    (mddev->reshape_backwards
	     ? (bio->bi_iter.bi_sector < conf->reshape_safe &&
		bio->bi_iter.bi_sector + sectors > conf->reshape_progress)
	     : (bio->bi_iter.bi_sector + sectors > conf->reshape_safe &&
		bio->bi_iter.bi_sector < conf->reshape_progress))) {
		/* Need to update reshape_position in metadata */
		mddev->reshape_position = conf->reshape_progress;
		set_mask_bits(&mddev->flags, 0,
			      BIT(MD_CHANGE_DEVS) | BIT(MD_CHANGE_PENDING));
		md_wakeup_thread(mddev->thread);
		wait_event(mddev->sb_wait,
			   !test_bit(MD_CHANGE_PENDING, &mddev->flags));

		conf->reshape_safe = mddev->reshape_position;
	}

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

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

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

	/* We might need to issue multiple reads to different
	 * devices if there are bad blocks around, so we keep
	 * track of the number of reads in bio->bi_phys_segments.
	 * If this is 0, there is only one r10_bio and no locking
	 * will be needed when the request completes.  If it is
	 * non-zero, then it is the number of not-completed requests.
	 */
	bio->bi_phys_segments = 0;
	bio_clear_flag(bio, BIO_SEG_VALID);

	if (rw == READ) {
		/*
		 * read balancing logic:
		 */
		struct md_rdev *rdev;
		int slot;

read_again:
		rdev = read_balance(conf, r10_bio, &max_sectors);
		if (!rdev) {
			raid_end_bio_io(r10_bio);
			return;
		}
		slot = r10_bio->read_slot;

		read_bio = bio_clone_mddev(bio, GFP_NOIO, mddev);
		bio_trim(read_bio, r10_bio->sector - bio->bi_iter.bi_sector,
			 max_sectors);

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

		read_bio->bi_iter.bi_sector = r10_bio->devs[slot].addr +
			choose_data_offset(r10_bio, rdev);
		read_bio->bi_bdev = rdev->bdev;
		read_bio->bi_end_io = raid10_end_read_request;
		read_bio->bi_rw = READ | do_sync;
		read_bio->bi_private = r10_bio;

		if (max_sectors < r10_bio->sectors) {
			/* Could not read all from this device, so we will
			 * need another r10_bio.
			 */
			sectors_handled = (r10_bio->sector + max_sectors
					   - bio->bi_iter.bi_sector);
			r10_bio->sectors = max_sectors;
			spin_lock_irq(&conf->device_lock);
			if (bio->bi_phys_segments == 0)
				bio->bi_phys_segments = 2;
			else
				bio->bi_phys_segments++;
			spin_unlock_irq(&conf->device_lock);
			/* Cannot call generic_make_request directly
			 * as that will be queued in __generic_make_request
			 * and subsequent mempool_alloc might block
			 * waiting for it.  so hand bio over to raid10d.
			 */
			reschedule_retry(r10_bio);

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

			r10_bio->master_bio = bio;
			r10_bio->sectors = bio_sectors(bio) - sectors_handled;
			r10_bio->state = 0;
			r10_bio->mddev = mddev;
			r10_bio->sector = bio->bi_iter.bi_sector +
				sectors_handled;
			goto read_again;
		} else
			generic_make_request(read_bio);
		return;
	}

	/*
	 * WRITE:
	 */
	if (conf->pending_count >= max_queued_requests) {
		md_wakeup_thread(mddev->thread);
		wait_event(conf->wait_barrier,
			   conf->pending_count < max_queued_requests);
	}
	/* first select target devices under rcu_lock and
	 * inc refcount on their rdev.  Record them by setting
	 * bios[x] to bio
	 * If there are known/acknowledged bad blocks on any device
	 * on which we have seen a write error, we want to avoid
	 * writing to those blocks.  This potentially requires several
	 * writes to write around the bad blocks.  Each set of writes
	 * gets its own r10_bio with a set of bios attached.  The number
	 * of r10_bios is recored in bio->bi_phys_segments just as with
	 * the read case.
	 */

	r10_bio->read_slot = -1; /* make sure repl_bio gets freed */
	raid10_find_phys(conf, r10_bio);
retry_write:
	blocked_rdev = NULL;
	rcu_read_lock();
	max_sectors = r10_bio->sectors;

	for (i = 0;  i < conf->copies; i++) {
		int d = r10_bio->devs[i].devnum;
		struct md_rdev *rdev = rcu_dereference(conf->mirrors[d].rdev);
		struct md_rdev *rrdev = rcu_dereference(
			conf->mirrors[d].replacement);
		if (rdev == rrdev)
			rrdev = NULL;
		if (rdev && unlikely(test_bit(Blocked, &rdev->flags))) {
			atomic_inc(&rdev->nr_pending);
			blocked_rdev = rdev;
			break;
		}
		if (rrdev && unlikely(test_bit(Blocked, &rrdev->flags))) {
			atomic_inc(&rrdev->nr_pending);
			blocked_rdev = rrdev;
			break;
		}
		if (rdev && (test_bit(Faulty, &rdev->flags)))
			rdev = NULL;
		if (rrdev && (test_bit(Faulty, &rrdev->flags)))
			rrdev = NULL;

		r10_bio->devs[i].bio = NULL;
		r10_bio->devs[i].repl_bio = NULL;

		if (!rdev && !rrdev) {
			set_bit(R10BIO_Degraded, &r10_bio->state);
			continue;
		}
		if (rdev && test_bit(WriteErrorSeen, &rdev->flags)) {
			sector_t first_bad;
			sector_t dev_sector = r10_bio->devs[i].addr;
			int bad_sectors;
			int is_bad;

			is_bad = is_badblock(rdev, dev_sector,
					     max_sectors,
					     &first_bad, &bad_sectors);
			if (is_bad < 0) {
				/* Mustn't write here until the bad block
				 * is acknowledged
				 */
				atomic_inc(&rdev->nr_pending);
				set_bit(BlockedBadBlocks, &rdev->flags);
				blocked_rdev = rdev;
				break;
			}
			if (is_bad && first_bad <= dev_sector) {
				/* Cannot write here at all */
				bad_sectors -= (dev_sector - first_bad);
				if (bad_sectors < max_sectors)
					/* Mustn't write more than bad_sectors
					 * to other devices yet
					 */
					max_sectors = bad_sectors;
				/* We don't set R10BIO_Degraded as that
				 * only applies if the disk is missing,
				 * so it might be re-added, and we want to
				 * know to recover this chunk.
				 * In this case the device is here, and the
				 * fact that this chunk is not in-sync is
				 * recorded in the bad block log.
				 */
				continue;
			}
			if (is_bad) {
				int good_sectors = first_bad - dev_sector;
				if (good_sectors < max_sectors)
					max_sectors = good_sectors;
			}
		}
		if (rdev) {
			r10_bio->devs[i].bio = bio;
			atomic_inc(&rdev->nr_pending);
		}
		if (rrdev) {
			r10_bio->devs[i].repl_bio = bio;
			atomic_inc(&rrdev->nr_pending);
		}
	}
	rcu_read_unlock();

	if (unlikely(blocked_rdev)) {
		/* Have to wait for this device to get unblocked, then retry */
		int j;
		int d;

		for (j = 0; j < i; j++) {
			if (r10_bio->devs[j].bio) {
				d = r10_bio->devs[j].devnum;
				rdev_dec_pending(conf->mirrors[d].rdev, mddev);
			}
			if (r10_bio->devs[j].repl_bio) {
				struct md_rdev *rdev;
				d = r10_bio->devs[j].devnum;
				rdev = conf->mirrors[d].replacement;
				if (!rdev) {
					/* Race with remove_disk */
					smp_mb();
					rdev = conf->mirrors[d].rdev;
				}
				rdev_dec_pending(rdev, mddev);
			}
		}
		allow_barrier(conf);
		md_wait_for_blocked_rdev(blocked_rdev, mddev);
		wait_barrier(conf);
		goto retry_write;
	}

	if (max_sectors < r10_bio->sectors) {
		/* We are splitting this into multiple parts, so
		 * we need to prepare for allocating another r10_bio.
		 */
		r10_bio->sectors = max_sectors;
		spin_lock_irq(&conf->device_lock);
		if (bio->bi_phys_segments == 0)
			bio->bi_phys_segments = 2;
		else
			bio->bi_phys_segments++;
		spin_unlock_irq(&conf->device_lock);
	}
	sectors_handled = r10_bio->sector + max_sectors -
		bio->bi_iter.bi_sector;

	atomic_set(&r10_bio->remaining, 1);
	bitmap_startwrite(mddev->bitmap, r10_bio->sector, r10_bio->sectors, 0);

	for (i = 0; i < conf->copies; i++) {
		struct bio *mbio;
		int d = r10_bio->devs[i].devnum;
		if (r10_bio->devs[i].bio) {
			struct md_rdev *rdev = conf->mirrors[d].rdev;
			mbio = bio_clone_mddev(bio, GFP_NOIO, mddev);
			bio_trim(mbio, r10_bio->sector - bio->bi_iter.bi_sector,
				 max_sectors);
			r10_bio->devs[i].bio = mbio;

			mbio->bi_iter.bi_sector	= (r10_bio->devs[i].addr+
					   choose_data_offset(r10_bio,
							      rdev));
			mbio->bi_bdev = rdev->bdev;
			mbio->bi_end_io	= raid10_end_write_request;
			mbio->bi_rw =
				WRITE | do_sync | do_fua | do_discard | do_same;
			mbio->bi_private = r10_bio;

			atomic_inc(&r10_bio->remaining);

			cb = blk_check_plugged(raid10_unplug, mddev,
					       sizeof(*plug));
			if (cb)
				plug = container_of(cb, struct raid10_plug_cb,
						    cb);
			else
				plug = NULL;
			spin_lock_irqsave(&conf->device_lock, flags);
			if (plug) {
				bio_list_add(&plug->pending, mbio);
				plug->pending_cnt++;
			} else {
				bio_list_add(&conf->pending_bio_list, mbio);
				conf->pending_count++;
			}
			spin_unlock_irqrestore(&conf->device_lock, flags);
			if (!plug)
				md_wakeup_thread(mddev->thread);
		}

		if (r10_bio->devs[i].repl_bio) {
			struct md_rdev *rdev = conf->mirrors[d].replacement;
			if (rdev == NULL) {
				/* Replacement just got moved to main 'rdev' */
				smp_mb();
				rdev = conf->mirrors[d].rdev;
			}
			mbio = bio_clone_mddev(bio, GFP_NOIO, mddev);
			bio_trim(mbio, r10_bio->sector - bio->bi_iter.bi_sector,
				 max_sectors);
			r10_bio->devs[i].repl_bio = mbio;

			mbio->bi_iter.bi_sector	= (r10_bio->devs[i].addr +
					   choose_data_offset(
						   r10_bio, rdev));
			mbio->bi_bdev = rdev->bdev;
			mbio->bi_end_io	= raid10_end_write_request;
			mbio->bi_rw =
				WRITE | do_sync | do_fua | do_discard | do_same;
			mbio->bi_private = r10_bio;

			atomic_inc(&r10_bio->remaining);
			spin_lock_irqsave(&conf->device_lock, flags);
			bio_list_add(&conf->pending_bio_list, mbio);
			conf->pending_count++;
			spin_unlock_irqrestore(&conf->device_lock, flags);
			if (!mddev_check_plugged(mddev))
				md_wakeup_thread(mddev->thread);
		}
	}

	/* Don't remove the bias on 'remaining' (one_write_done) until
	 * after checking if we need to go around again.
	 */

	if (sectors_handled < bio_sectors(bio)) {
		one_write_done(r10_bio);
		/* We need another r10_bio.  It has already been counted
		 * in bio->bi_phys_segments.
		 */
		r10_bio = mempool_alloc(conf->r10bio_pool, GFP_NOIO);

		r10_bio->master_bio = bio;
		r10_bio->sectors = bio_sectors(bio) - sectors_handled;

		r10_bio->mddev = mddev;
		r10_bio->sector = bio->bi_iter.bi_sector + sectors_handled;
		r10_bio->state = 0;
		goto retry_write;
	}
	one_write_done(r10_bio);
}

static void raid10_make_request(struct mddev *mddev, struct bio *bio)
{
	struct r10conf *conf = mddev->private;
	sector_t chunk_mask = (conf->geo.chunk_mask & conf->prev.chunk_mask);
	int chunk_sects = chunk_mask + 1;

	struct bio *split;

	if (unlikely(bio->bi_rw & REQ_FLUSH)) {
		md_flush_request(mddev, bio);
		return;
	}

	md_write_start(mddev, bio);

	do {

		/*
		 * If this request crosses a chunk boundary, we need to split
		 * it.
		 */
		if (unlikely((bio->bi_iter.bi_sector & chunk_mask) +
			     bio_sectors(bio) > chunk_sects
			     && (conf->geo.near_copies < conf->geo.raid_disks
				 || conf->prev.near_copies <
				 conf->prev.raid_disks))) {
			split = bio_split(bio, chunk_sects -
					  (bio->bi_iter.bi_sector &
					   (chunk_sects - 1)),
					  GFP_NOIO, fs_bio_set);
			bio_chain(split, bio);
		} else {
			split = bio;
		}

		__make_request(mddev, split);
	} while (split != bio);

	/* In case raid10d snuck in to freeze_array */
	wake_up(&conf->wait_barrier);
}

static void raid10_status(struct seq_file *seq, struct mddev *mddev)
{
	struct r10conf *conf = mddev->private;
	int i;

	if (conf->geo.near_copies < conf->geo.raid_disks)
		seq_printf(seq, " %dK chunks", mddev->chunk_sectors / 2);
	if (conf->geo.near_copies > 1)
		seq_printf(seq, " %d near-copies", conf->geo.near_copies);
	if (conf->geo.far_copies > 1) {
		if (conf->geo.far_offset)
			seq_printf(seq, " %d offset-copies", conf->geo.far_copies);
		else
			seq_printf(seq, " %d far-copies", conf->geo.far_copies);
		if (conf->geo.far_set_size != conf->geo.raid_disks)
			seq_printf(seq, " %d devices per set", conf->geo.far_set_size);
	}
	seq_printf(seq, " [%d/%d] [", conf->geo.raid_disks,
					conf->geo.raid_disks - mddev->degraded);
	for (i = 0; i < conf->geo.raid_disks; i++)
		seq_printf(seq, "%s",
			      conf->mirrors[i].rdev &&
			      test_bit(In_sync, &conf->mirrors[i].rdev->flags) ? "U" : "_");
	seq_printf(seq, "]");
}

/* check if there are enough drives for
 * every block to appear on atleast one.
 * Don't consider the device numbered 'ignore'
 * as we might be about to remove it.
 */
static int _enough(struct r10conf *conf, int previous, int ignore)
{
	int first = 0;
	int has_enough = 0;
	int disks, ncopies;
	if (previous) {
		disks = conf->prev.raid_disks;
		ncopies = conf->prev.near_copies;
	} else {
		disks = conf->geo.raid_disks;
		ncopies = conf->geo.near_copies;
	}

	rcu_read_lock();
	do {
		int n = conf->copies;
		int cnt = 0;
		int this = first;
		while (n--) {
			struct md_rdev *rdev;
			if (this != ignore &&
			    (rdev = rcu_dereference(conf->mirrors[this].rdev)) &&
			    test_bit(In_sync, &rdev->flags))
				cnt++;
			this = (this+1) % disks;
		}
		if (cnt == 0)
			goto out;
		first = (first + ncopies) % disks;
	} while (first != 0);
	has_enough = 1;
out:
	rcu_read_unlock();
	return has_enough;
}

static int enough(struct r10conf *conf, int ignore)
{
	/* when calling 'enough', both 'prev' and 'geo' must
	 * be stable.
	 * This is ensured if ->reconfig_mutex or ->device_lock
	 * is held.
	 */
	return _enough(conf, 0, ignore) &&
		_enough(conf, 1, ignore);
}

static void raid10_error(struct mddev *mddev, struct md_rdev *rdev)
{
	char b[BDEVNAME_SIZE];
	struct r10conf *conf = mddev->private;
	unsigned long flags;

	/*
	 * 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
	 */
	spin_lock_irqsave(&conf->device_lock, flags);
	if (test_bit(In_sync, &rdev->flags)
	    && !enough(conf, rdev->raid_disk)) {
		/*
		 * Don't fail the drive, just return an IO error.
		 */
		spin_unlock_irqrestore(&conf->device_lock, flags);
		return;
	}
	if (test_and_clear_bit(In_sync, &rdev->flags))
		mddev->degraded++;
	/*
	 * If recovery is running, make sure it aborts.
	 */
	set_bit(MD_RECOVERY_INTR, &mddev->recovery);
	set_bit(Blocked, &rdev->flags);
	set_bit(Faulty, &rdev->flags);
	set_mask_bits(&mddev->flags, 0,
		      BIT(MD_CHANGE_DEVS) | BIT(MD_CHANGE_PENDING));
	spin_unlock_irqrestore(&conf->device_lock, flags);
	printk(KERN_ALERT
	       "md/raid10:%s: Disk failure on %s, disabling device.\n"
	       "md/raid10:%s: Operation continuing on %d devices.\n",
	       mdname(mddev), bdevname(rdev->bdev, b),
	       mdname(mddev), conf->geo.raid_disks - mddev->degraded);
}

static void print_conf(struct r10conf *conf)
{
	int i;
	struct raid10_info *tmp;

	printk(KERN_DEBUG "RAID10 conf printout:\n");
	if (!conf) {
		printk(KERN_DEBUG "(!conf)\n");
		return;
	}
	printk(KERN_DEBUG " --- wd:%d rd:%d\n", conf->geo.raid_disks - conf->mddev->degraded,
		conf->geo.raid_disks);

	for (i = 0; i < conf->geo.raid_disks; i++) {
		char b[BDEVNAME_SIZE];
		tmp = conf->mirrors + i;
		if (tmp->rdev)
			printk(KERN_DEBUG " 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(struct r10conf *conf)
{
	wait_barrier(conf);
	allow_barrier(conf);

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

static int raid10_spare_active(struct mddev *mddev)
{
	int i;
	struct r10conf *conf = mddev->private;
	struct raid10_info *tmp;
	int count = 0;
	unsigned long flags;

	/*
	 * Find all non-in_sync disks within the RAID10 configuration
	 * and mark them in_sync
	 */
	for (i = 0; i < conf->geo.raid_disks; i++) {
		tmp = conf->mirrors + i;
		if (tmp->replacement
		    && tmp->replacement->recovery_offset == MaxSector
		    && !test_bit(Faulty, &tmp->replacement->flags)
		    && !test_and_set_bit(In_sync, &tmp->replacement->flags)) {
			/* Replacement has just become active */
			if (!tmp->rdev
			    || !test_and_clear_bit(In_sync, &tmp->rdev->flags))
				count++;
			if (tmp->rdev) {
				/* Replaced device not technically faulty,
				 * but we need to be sure it gets removed
				 * and never re-added.
				 */
				set_bit(Faulty, &tmp->rdev->flags);
				sysfs_notify_dirent_safe(
					tmp->rdev->sysfs_state);
			}
			sysfs_notify_dirent_safe(tmp->replacement->sysfs_state);
		} else if (tmp->rdev
			   && tmp->rdev->recovery_offset == MaxSector
			   && !test_bit(Faulty, &tmp->rdev->flags)
			   && !test_and_set_bit(In_sync, &tmp->rdev->flags)) {
			count++;
			sysfs_notify_dirent_safe(tmp->rdev->sysfs_state);
		}
	}
	spin_lock_irqsave(&conf->device_lock, flags);
	mddev->degraded -= count;
	spin_unlock_irqrestore(&conf->device_lock, flags);

	print_conf(conf);
	return count;
}

static int raid10_add_disk(struct mddev *mddev, struct md_rdev *rdev)
{
	struct r10conf *conf = mddev->private;
	int err = -EEXIST;
	int mirror;
	int first = 0;
	int last = conf->geo.raid_disks - 1;

	if (mddev->recovery_cp < MaxSector)
		/* only hot-add to in-sync arrays, as recovery is
		 * very different from resync
		 */
		return -EBUSY;
	if (rdev->saved_raid_disk < 0 && !_enough(conf, 1, -1))
		return -EINVAL;

	if (md_integrity_add_rdev(rdev, mddev))
		return -ENXIO;

	if (rdev->raid_disk >= 0)
		first = last = rdev->raid_disk;

	if (rdev->saved_raid_disk >= first &&
	    conf->mirrors[rdev->saved_raid_disk].rdev == NULL)
		mirror = rdev->saved_raid_disk;
	else
		mirror = first;
	for ( ; mirror <= last ; mirror++) {
		struct raid10_info *p = &conf->mirrors[mirror];
		if (p->recovery_disabled == mddev->recovery_disabled)
			continue;
		if (p->rdev) {
			if (!test_bit(WantReplacement, &p->rdev->flags) ||
			    p->replacement != NULL)
				continue;
			clear_bit(In_sync, &rdev->flags);
			set_bit(Replacement, &rdev->flags);
			rdev->raid_disk = mirror;
			err = 0;
			if (mddev->gendisk)
				disk_stack_limits(mddev->gendisk, rdev->bdev,
						  rdev->data_offset << 9);
			conf->fullsync = 1;
			rcu_assign_pointer(p->replacement, rdev);
			break;
		}

		if (mddev->gendisk)
			disk_stack_limits(mddev->gendisk, rdev->bdev,
					  rdev->data_offset << 9);

		p->head_position = 0;
		p->recovery_disabled = mddev->recovery_disabled - 1;
		rdev->raid_disk = mirror;
		err = 0;
		if (rdev->saved_raid_disk != mirror)
			conf->fullsync = 1;
		rcu_assign_pointer(p->rdev, rdev);
		break;
	}
	if (mddev->queue && blk_queue_discard(bdev_get_queue(rdev->bdev)))
		queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, mddev->queue);

	print_conf(conf);
	return err;
}

static int raid10_remove_disk(struct mddev *mddev, struct md_rdev