diff options
Diffstat (limited to 'drivers/md/raid10.c')
| -rw-r--r-- | drivers/md/raid10.c | 3506 |
1 files changed, 1958 insertions, 1548 deletions
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index cd066b63bdaf..84be4cc7e873 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* * raid10.c : Multiple Devices driver for Linux * @@ -6,16 +7,6 @@ * 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> @@ -25,10 +16,15 @@ #include <linux/seq_file.h> #include <linux/ratelimit.h> #include <linux/kthread.h> +#include <linux/raid/md_p.h> +#include <trace/events/block.h> #include "md.h" + +#define RAID_1_10_NAME "raid10" #include "raid10.h" #include "raid0.h" -#include "bitmap.h" +#include "md-bitmap.h" +#include "md-cluster.h" /* * RAID10 provides a combination of RAID0 and RAID1 functionality. @@ -39,6 +35,7 @@ * 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 @@ -70,62 +67,59 @@ * [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 int enough(struct r10conf *conf, 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, int error); +static void end_reshape_write(struct bio *bio); static void end_reshape(struct r10conf *conf); +#include "raid1-10.c" + +#define NULL_CMD +#define cmd_before(conf, cmd) \ + do { \ + write_sequnlock_irq(&(conf)->resync_lock); \ + cmd; \ + } while (0) +#define cmd_after(conf) write_seqlock_irq(&(conf)->resync_lock) + +#define wait_event_barrier_cmd(conf, cond, cmd) \ + wait_event_cmd((conf)->wait_barrier, cond, cmd_before(conf, cmd), \ + cmd_after(conf)) + +#define wait_event_barrier(conf, cond) \ + wait_event_barrier_cmd(conf, cond, NULL_CMD) + +/* + * for resync bio, r10bio pointer can be retrieved from the per-bio + * 'struct resync_pages'. + */ +static inline struct r10bio *get_resync_r10bio(struct bio *bio) +{ + return get_resync_pages(bio)->raid_bio; +} + static void * r10bio_pool_alloc(gfp_t gfp_flags, void *data) { struct r10conf *conf = data; - int size = offsetof(struct r10bio, devs[conf->copies]); + int size = offsetof(struct r10bio, devs[conf->geo.raid_disks]); /* 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) +#define RESYNC_SECTORS (RESYNC_BLOCK_SIZE >> 9) /* 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) +#define CLUSTER_RESYNC_WINDOW (32 * RESYNC_WINDOW) +#define CLUSTER_RESYNC_WINDOW_SECTORS (CLUSTER_RESYNC_WINDOW >> 9) /* * When performing a resync, we need to read and compare, so @@ -137,11 +131,11 @@ static void r10bio_pool_free(void *r10_bio, void *data) 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; + int j; + int nalloc, nalloc_rp; + struct resync_pages *rps; r10_bio = r10bio_pool_alloc(gfp_flags, conf); if (!r10_bio) @@ -153,95 +147,120 @@ static void * r10buf_pool_alloc(gfp_t gfp_flags, void *data) else nalloc = 2; /* recovery */ + /* allocate once for all bios */ + if (!conf->have_replacement) + nalloc_rp = nalloc; + else + nalloc_rp = nalloc * 2; + rps = kmalloc_array(nalloc_rp, sizeof(struct resync_pages), gfp_flags); + if (!rps) + goto out_free_r10bio; + /* * Allocate bios. */ for (j = nalloc ; j-- ; ) { - bio = bio_kmalloc(gfp_flags, RESYNC_PAGES); + bio = bio_kmalloc(RESYNC_PAGES, gfp_flags); if (!bio) goto out_free_bio; + bio_init_inline(bio, NULL, RESYNC_PAGES, 0); r10_bio->devs[j].bio = bio; if (!conf->have_replacement) continue; - bio = bio_kmalloc(gfp_flags, RESYNC_PAGES); + bio = bio_kmalloc(RESYNC_PAGES, gfp_flags); if (!bio) goto out_free_bio; + bio_init_inline(bio, NULL, RESYNC_PAGES, 0); r10_bio->devs[j].repl_bio = bio; } /* * Allocate RESYNC_PAGES data pages and attach them * where needed. */ - for (j = 0 ; j < nalloc; j++) { + for (j = 0; j < nalloc; j++) { struct bio *rbio = r10_bio->devs[j].repl_bio; + struct resync_pages *rp, *rp_repl; + + rp = &rps[j]; + if (rbio) + rp_repl = &rps[nalloc + j]; + 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)) + + if (!j || test_bit(MD_RECOVERY_SYNC, + &conf->mddev->recovery)) { + if (resync_alloc_pages(rp, gfp_flags)) goto out_free_pages; + } else { + memcpy(rp, &rps[0], sizeof(*rp)); + resync_get_all_pages(rp); + } - bio->bi_io_vec[i].bv_page = page; - if (rbio) - rbio->bi_io_vec[i].bv_page = page; + rp->raid_bio = r10_bio; + bio->bi_private = rp; + if (rbio) { + memcpy(rp_repl, rp, sizeof(*rp)); + rbio->bi_private = rp_repl; } } 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); + while (--j >= 0) + resync_free_pages(&rps[j]); + j = 0; out_free_bio: for ( ; j < nalloc; j++) { if (r10_bio->devs[j].bio) - bio_put(r10_bio->devs[j].bio); + bio_uninit(r10_bio->devs[j].bio); + kfree(r10_bio->devs[j].bio); if (r10_bio->devs[j].repl_bio) - bio_put(r10_bio->devs[j].repl_bio); + bio_uninit(r10_bio->devs[j].repl_bio); + kfree(r10_bio->devs[j].repl_bio); } - r10bio_pool_free(r10_bio, conf); + kfree(rps); +out_free_r10bio: + rbio_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; + struct resync_pages *rp = NULL; - for (j=0; j < conf->copies; j++) { + for (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); + rp = get_resync_pages(bio); + resync_free_pages(rp); + bio_uninit(bio); + kfree(bio); } + bio = r10bio->devs[j].repl_bio; - if (bio) - bio_put(bio); + if (bio) { + bio_uninit(bio); + kfree(bio); + } } - r10bio_pool_free(r10bio, conf); + + /* resync pages array stored in the 1st bio's .bi_private */ + kfree(rp); + + rbio_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++) { + for (i = 0; i < conf->geo.raid_disks; i++) { struct bio **bio = & r10_bio->devs[i].bio; if (!BIO_SPECIAL(*bio)) bio_put(*bio); @@ -258,18 +277,24 @@ 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); + 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); + mempool_free(r10_bio, &conf->r10buf_pool); lower_barrier(conf); } +static void wake_up_barrier(struct r10conf *conf) +{ + if (wq_has_sleeper(&conf->wait_barrier)) + wake_up(&conf->wait_barrier); +} + static void reschedule_retry(struct r10bio *r10_bio) { unsigned long flags; @@ -295,27 +320,20 @@ static void reschedule_retry(struct r10bio *r10_bio) 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)) - clear_bit(BIO_UPTODATE, &bio->bi_flags); - if (done) { - bio_endio(bio, 0); - /* - * Wake up any possible resync thread that waits for the device - * to go idle. - */ - allow_barrier(conf); + if (!test_and_set_bit(R10BIO_Returned, &r10_bio->state)) { + if (!test_bit(R10BIO_Uptodate, &r10_bio->state)) + bio->bi_status = BLK_STS_IOERR; + bio_endio(bio); } + + /* + * Wake up any possible resync thread that waits for the device + * to go idle. + */ + allow_barrier(conf); + free_r10bio(r10_bio); } @@ -339,7 +357,7 @@ static int find_bio_disk(struct r10conf *conf, struct r10bio *r10_bio, int slot; int repl = 0; - for (slot = 0; slot < conf->copies; slot++) { + for (slot = 0; slot < conf->geo.raid_disks; slot++) { if (r10_bio->devs[slot].bio == bio) break; if (r10_bio->devs[slot].repl_bio == bio) { @@ -348,7 +366,6 @@ static int find_bio_disk(struct r10conf *conf, struct r10bio *r10_bio, } } - BUG_ON(slot == conf->copies); update_head_pos(slot, r10_bio); if (slotp) @@ -358,17 +375,15 @@ static int find_bio_disk(struct r10conf *conf, struct r10bio *r10_bio, return r10_bio->devs[slot].devnum; } -static void raid10_end_read_request(struct bio *bio, int error) +static void raid10_end_read_request(struct bio *bio) { - int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags); + int uptodate = !bio->bi_status; struct r10bio *r10_bio = bio->bi_private; - int slot, dev; + int slot; 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: @@ -386,6 +401,8 @@ static void raid10_end_read_request(struct bio *bio, int error) * wait for the 'master' bio. */ set_bit(R10BIO_Uptodate, &r10_bio->state); + } else if (!raid1_should_handle_error(bio)) { + uptodate = 1; } else { /* If all other devices that store this block have * failed, we want to return the error upwards rather @@ -403,11 +420,9 @@ static void raid10_end_read_request(struct bio *bio, int error) /* * 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", + pr_err_ratelimited("md/raid10:%s: %pg: rescheduling sector %llu\n", mdname(conf->mddev), - bdevname(rdev->bdev, b), + rdev->bdev, (unsigned long long)r10_bio->sector); set_bit(R10BIO_ReadError, &r10_bio->state); reschedule_retry(r10_bio); @@ -416,12 +431,9 @@ static void raid10_end_read_request(struct bio *bio, int error) 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); + struct mddev *mddev = r10_bio->mddev; + + md_write_end(mddev); } static void one_write_done(struct r10bio *r10_bio) @@ -439,15 +451,17 @@ static void one_write_done(struct r10bio *r10_bio) } } -static void raid10_end_write_request(struct bio *bio, int error) +static void raid10_end_write_request(struct bio *bio) { - int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags); 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; + struct bio *to_put = NULL; + bool ignore_error = !raid1_should_handle_error(bio) || + (bio->bi_status && bio_op(bio) == REQ_OP_DISCARD); dev = find_bio_disk(conf, r10_bio, bio, &slot, &repl); @@ -461,7 +475,7 @@ static void raid10_end_write_request(struct bio *bio, int error) /* * this branch is our 'one mirror IO has finished' event handler: */ - if (!uptodate) { + if (bio->bi_status && !ignore_error) { if (repl) /* Never record new bad blocks to replacement, * just fail it. @@ -472,8 +486,25 @@ static void raid10_end_write_request(struct bio *bio, int error) 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; + if (test_bit(FailFast, &rdev->flags) && + (bio->bi_opf & MD_FAILFAST)) { + md_error(rdev->mddev, rdev); + } + + /* + * When the device is faulty, it is not necessary to + * handle write error. + */ + if (!test_bit(Faulty, &rdev->flags)) + set_bit(R10BIO_WriteError, &r10_bio->state); + else { + /* Fail the request */ + r10_bio->devs[slot].bio = NULL; + to_put = bio; + dec_rdev = 1; + } } } else { /* @@ -484,11 +515,7 @@ static void raid10_end_write_request(struct bio *bio, int error) * 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 @@ -501,10 +528,9 @@ static void raid10_end_write_request(struct bio *bio, int error) 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)) { + if (rdev_has_badblock(rdev, r10_bio->devs[slot].addr, + r10_bio->sectors) && + !ignore_error) { bio_put(bio); if (repl) r10_bio->devs[slot].repl_bio = IO_MADE_GOOD; @@ -523,6 +549,8 @@ static void raid10_end_write_request(struct bio *bio, int error) one_write_done(r10_bio); if (dec_rdev) rdev_dec_pending(rdev, conf->mddev); + if (to_put) + bio_put(to_put); } /* @@ -673,94 +701,6 @@ static sector_t raid10_find_virt(struct r10conf *conf, sector_t sector, int dev) return (vchunk << geo->chunk_shift) + offset; } -/** - * raid10_mergeable_bvec -- tell bio layer if a two requests can be merged - * @q: request queue - * @bvm: properties of new bio - * @biovec: the request that could be merged to it. - * - * Return amount of bytes we can accept at this offset - * This requires checking for end-of-chunk if near_copies != raid_disks, - * and for subordinate merge_bvec_fns if merge_check_needed. - */ -static int raid10_mergeable_bvec(struct request_queue *q, - struct bvec_merge_data *bvm, - struct bio_vec *biovec) -{ - struct mddev *mddev = q->queuedata; - struct r10conf *conf = mddev->private; - sector_t sector = bvm->bi_sector + get_start_sect(bvm->bi_bdev); - int max; - unsigned int chunk_sectors; - unsigned int bio_sectors = bvm->bi_size >> 9; - struct geom *geo = &conf->geo; - - chunk_sectors = (conf->geo.chunk_mask & conf->prev.chunk_mask) + 1; - if (conf->reshape_progress != MaxSector && - ((sector >= conf->reshape_progress) != - conf->mddev->reshape_backwards)) - geo = &conf->prev; - - if (geo->near_copies < geo->raid_disks) { - max = (chunk_sectors - ((sector & (chunk_sectors - 1)) - + bio_sectors)) << 9; - if (max < 0) - /* bio_add cannot handle a negative return */ - max = 0; - if (max <= biovec->bv_len && bio_sectors == 0) - return biovec->bv_len; - } else - max = biovec->bv_len; - - if (mddev->merge_check_needed) { - struct { - struct r10bio r10_bio; - struct r10dev devs[conf->copies]; - } on_stack; - struct r10bio *r10_bio = &on_stack.r10_bio; - int s; - if (conf->reshape_progress != MaxSector) { - /* Cannot give any guidance during reshape */ - if (max <= biovec->bv_len && bio_sectors == 0) - return biovec->bv_len; - return 0; - } - r10_bio->sector = sector; - raid10_find_phys(conf, r10_bio); - rcu_read_lock(); - for (s = 0; s < conf->copies; s++) { - int disk = r10_bio->devs[s].devnum; - struct md_rdev *rdev = rcu_dereference( - conf->mirrors[disk].rdev); - if (rdev && !test_bit(Faulty, &rdev->flags)) { - struct request_queue *q = - bdev_get_queue(rdev->bdev); - if (q->merge_bvec_fn) { - bvm->bi_sector = r10_bio->devs[s].addr - + rdev->data_offset; - bvm->bi_bdev = rdev->bdev; - max = min(max, q->merge_bvec_fn( - q, bvm, biovec)); - } - } - rdev = rcu_dereference(conf->mirrors[disk].replacement); - if (rdev && !test_bit(Faulty, &rdev->flags)) { - struct request_queue *q = - bdev_get_queue(rdev->bdev); - if (q->merge_bvec_fn) { - bvm->bi_sector = r10_bio->devs[s].addr - + rdev->data_offset; - bvm->bi_bdev = rdev->bdev; - max = min(max, q->merge_bvec_fn( - q, bvm, biovec)); - } - } - } - rcu_read_unlock(); - } - return max; -} - /* * This routine returns the disk from which the requested read should * be done. There is a per-array 'next expected sequential IO' sector @@ -789,46 +729,43 @@ static struct md_rdev *read_balance(struct r10conf *conf, int sectors = r10_bio->sectors; int best_good_sectors; sector_t new_distance, best_dist; - struct md_rdev *best_rdev, *rdev = NULL; + struct md_rdev *best_dist_rdev, *best_pending_rdev, *rdev = NULL; int do_balance; - int best_slot; + int best_dist_slot, best_pending_slot; + bool has_nonrot_disk = false; + unsigned int min_pending; 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_slot = -1; + min_pending = UINT_MAX; + best_dist_rdev = NULL; + best_pending_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)) + clear_bit(R10BIO_FailFast, &r10_bio->state); + + if (raid1_should_read_first(conf->mddev, this_sector, sectors)) do_balance = 0; for (slot = 0; slot < conf->copies ; slot++) { sector_t first_bad; - int bad_sectors; + sector_t bad_sectors; sector_t dev_sector; + unsigned int pending; + bool nonrot; if (r10_bio->devs[slot].bio == IO_BLOCKED) continue; disk = r10_bio->devs[slot].devnum; - rdev = rcu_dereference(conf->mirrors[disk].replacement); + rdev = conf->mirrors[disk].replacement; if (rdev == NULL || test_bit(Faulty, &rdev->flags) || - test_bit(Unmerged, &rdev->flags) || - r10_bio->devs[slot].addr + sectors > rdev->recovery_offset) - rdev = rcu_dereference(conf->mirrors[disk].rdev); + r10_bio->devs[slot].addr + sectors > + rdev->recovery_offset) + rdev = conf->mirrors[disk].rdev; if (rdev == NULL || - test_bit(Faulty, &rdev->flags) || - test_bit(Unmerged, &rdev->flags)) + test_bit(Faulty, &rdev->flags)) continue; if (!test_bit(In_sync, &rdev->flags) && r10_bio->devs[slot].addr + sectors > rdev->recovery_offset) @@ -855,8 +792,8 @@ retry: first_bad - dev_sector; if (good_sectors > best_good_sectors) { best_good_sectors = good_sectors; - best_slot = slot; - best_rdev = rdev; + best_dist_slot = slot; + best_dist_rdev = rdev; } if (!do_balance) /* Must read from here */ @@ -869,82 +806,58 @@ retry: if (!do_balance) break; + nonrot = bdev_nonrot(rdev->bdev); + has_nonrot_disk |= nonrot; + pending = atomic_read(&rdev->nr_pending); + if (min_pending > pending && nonrot) { + min_pending = pending; + best_pending_slot = slot; + best_pending_rdev = rdev; + } + + if (best_dist_slot >= 0) + /* At least 2 disks to choose from so failfast is OK */ + set_bit(R10BIO_FailFast, &r10_bio->state); /* 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; + if (geo->near_copies > 1 && !pending) + new_distance = 0; /* for far > 1 always use the lowest address */ - if (geo->far_copies > 1) + else 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; + best_dist_slot = slot; + best_dist_rdev = rdev; } } if (slot >= conf->copies) { - slot = best_slot; - rdev = best_rdev; + if (has_nonrot_disk) { + slot = best_pending_slot; + rdev = best_pending_rdev; + } else { + slot = best_dist_slot; + rdev = best_dist_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; } -int md_raid10_congested(struct mddev *mddev, int bits) -{ - struct r10conf *conf = mddev->private; - int i, ret = 0; - - if ((bits & (1 << BDI_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; -} -EXPORT_SYMBOL_GPL(md_raid10_congested); - -static int raid10_congested(void *data, int bits) -{ - struct mddev *mddev = data; - - return mddev_congested(mddev, bits) || - md_raid10_congested(mddev, bits); -} - static void flush_pending_writes(struct r10conf *conf) { /* Any writes that have been queued but are awaiting @@ -953,26 +866,35 @@ static void flush_pending_writes(struct r10conf *conf) spin_lock_irq(&conf->device_lock); if (conf->pending_bio_list.head) { + struct blk_plug plug; 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); + + /* + * As this is called in a wait_event() loop (see freeze_array), + * current->state might be TASK_UNINTERRUPTIBLE which will + * cause a warning when we prepare to wait again. As it is + * rare that this path is taken, it is perfectly safe to force + * us to go around the wait_event() loop again, so the warning + * is a false-positive. Silence the warning by resetting + * thread state + */ + __set_current_state(TASK_RUNNING); + + blk_start_plug(&plug); + raid1_prepare_flush_writes(conf->mddev); 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, 0); - else - generic_make_request(bio); + + raid1_submit_write(bio); bio = next; + cond_resched(); } + blk_finish_plug(&plug); } else spin_unlock_irq(&conf->device_lock); } @@ -1001,66 +923,118 @@ static void flush_pending_writes(struct r10conf *conf) static void raise_barrier(struct r10conf *conf, int force) { - BUG_ON(force && !conf->barrier); - spin_lock_irq(&conf->resync_lock); + write_seqlock_irq(&conf->resync_lock); + + if (WARN_ON_ONCE(force && !conf->barrier)) + force = false; /* Wait until no block IO is waiting (unless 'force') */ - wait_event_lock_irq(conf->wait_barrier, force || !conf->nr_waiting, - conf->resync_lock); + wait_event_barrier(conf, force || !conf->nr_waiting); /* block any new IO from starting */ - conf->barrier++; + WRITE_ONCE(conf->barrier, conf->barrier + 1); /* 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); + wait_event_barrier(conf, !atomic_read(&conf->nr_pending) && + conf->barrier < RESYNC_DEPTH); - spin_unlock_irq(&conf->resync_lock); + write_sequnlock_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); + + write_seqlock_irqsave(&conf->resync_lock, flags); + WRITE_ONCE(conf->barrier, conf->barrier - 1); + write_sequnlock_irqrestore(&conf->resync_lock, flags); wake_up(&conf->wait_barrier); } -static void wait_barrier(struct r10conf *conf) +static bool stop_waiting_barrier(struct r10conf *conf) +{ + struct bio_list *bio_list = current->bio_list; + struct md_thread *thread; + + /* barrier is dropped */ + if (!conf->barrier) + return true; + + /* + * 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. + */ + if (atomic_read(&conf->nr_pending) && bio_list && + (!bio_list_empty(&bio_list[0]) || !bio_list_empty(&bio_list[1]))) + return true; + + /* daemon thread must exist while handling io */ + thread = rcu_dereference_protected(conf->mddev->thread, true); + /* + * move on if io is issued from raid10d(), nr_pending is not released + * from original io(see handle_read_error()). All raise barrier is + * blocked until this io is done. + */ + if (thread->tsk == current) { + WARN_ON_ONCE(atomic_read(&conf->nr_pending) == 0); + return true; + } + + return false; +} + +static bool wait_barrier_nolock(struct r10conf *conf) +{ + unsigned int seq = read_seqbegin(&conf->resync_lock); + + if (READ_ONCE(conf->barrier)) + return false; + + atomic_inc(&conf->nr_pending); + if (!read_seqretry(&conf->resync_lock, seq)) + return true; + + if (atomic_dec_and_test(&conf->nr_pending)) + wake_up_barrier(conf); + + return false; +} + +static bool wait_barrier(struct r10conf *conf, bool nowait) { - spin_lock_irq(&conf->resync_lock); + bool ret = true; + + if (wait_barrier_nolock(conf)) + return true; + + write_seqlock_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--; + /* Return false when nowait flag is set */ + if (nowait) { + ret = false; + } else { + conf->nr_waiting++; + mddev_add_trace_msg(conf->mddev, "raid10 wait barrier"); + wait_event_barrier(conf, stop_waiting_barrier(conf)); + conf->nr_waiting--; + } + if (!conf->nr_waiting) + wake_up(&conf->wait_barrier); } - conf->nr_pending++; - spin_unlock_irq(&conf->resync_lock); + /* Only increment nr_pending when we wait */ + if (ret) + atomic_inc(&conf->nr_pending); + write_sequnlock_irq(&conf->resync_lock); + return ret; } 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); + if ((atomic_dec_and_test(&conf->nr_pending)) || + (conf->array_freeze_pending)) + wake_up_barrier(conf); } static void freeze_array(struct r10conf *conf, int extra) @@ -1077,25 +1051,24 @@ static void freeze_array(struct r10conf *conf, int extra) * must match the number of pending IOs (nr_pending) before * we continue. */ - spin_lock_irq(&conf->resync_lock); - conf->barrier++; + write_seqlock_irq(&conf->resync_lock); + conf->array_freeze_pending++; + WRITE_ONCE(conf->barrier, conf->barrier + 1); 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); + wait_event_barrier_cmd(conf, atomic_read(&conf->nr_pending) == + conf->nr_queued + extra, flush_pending_writes(conf)); + conf->array_freeze_pending--; + write_sequnlock_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--; + write_seqlock_irq(&conf->resync_lock); + WRITE_ONCE(conf->barrier, conf->barrier - 1); conf->nr_waiting--; wake_up(&conf->wait_barrier); - spin_unlock_irq(&conf->resync_lock); + write_sequnlock_irq(&conf->resync_lock); } static sector_t choose_data_offset(struct r10bio *r10_bio, @@ -1108,26 +1081,18 @@ static sector_t choose_data_offset(struct r10bio *r10_bio, 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 raid1_plug_cb *plug = container_of(cb, struct raid1_plug_cb, cb); struct mddev *mddev = plug->cb.data; struct r10conf *conf = mddev->private; struct bio *bio; - if (from_schedule || current->bio_list) { + if (from_schedule) { 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); + wake_up_barrier(conf); md_wakeup_thread(mddev->thread); kfree(plug); return; @@ -1135,227 +1100,306 @@ static void raid10_unplug(struct blk_plug_cb *cb, bool from_schedule) /* 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); + raid1_prepare_flush_writes(mddev); + wake_up_barrier(conf); 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, 0); - else - generic_make_request(bio); + + raid1_submit_write(bio); bio = next; + cond_resched(); } kfree(plug); } -static void make_request(struct mddev *mddev, struct bio * bio) +/* + * 1. 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. + * 2. If IO spans the reshape position. Need to wait for reshape to pass. + */ +static bool regular_request_wait(struct mddev *mddev, struct r10conf *conf, + struct bio *bio, sector_t sectors) +{ + /* Bail out if REQ_NOWAIT is set for the bio */ + if (!wait_barrier(conf, bio->bi_opf & REQ_NOWAIT)) { + bio_wouldblock_error(bio); + return false; + } + 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) { + allow_barrier(conf); + if (bio->bi_opf & REQ_NOWAIT) { + bio_wouldblock_error(bio); + return false; + } + mddev_add_trace_msg(conf->mddev, "raid10 wait reshape"); + 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, false); + } + return true; +} + +static void raid10_read_request(struct mddev *mddev, struct bio *bio, + struct r10bio *r10_bio, bool io_accounting) { struct r10conf *conf = mddev->private; - struct r10bio *r10_bio; struct bio *read_bio; - int i; - sector_t chunk_mask = (conf->geo.chunk_mask & conf->prev.chunk_mask); - int chunk_sects = chunk_mask + 1; - 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; - - if (unlikely(bio->bi_rw & REQ_FLUSH)) { - md_flush_request(mddev, bio); - return; - } + struct md_rdev *rdev; + char b[BDEVNAME_SIZE]; + int slot = r10_bio->read_slot; + struct md_rdev *err_rdev = NULL; + gfp_t gfp = GFP_NOIO; - /* 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 & chunk_mask) + bio_sectors(bio) - > chunk_sects - && (conf->geo.near_copies < conf->geo.raid_disks - || conf->prev.near_copies < conf->prev.raid_disks))) { - struct bio_pair *bp; - /* Sanity check -- queue functions should prevent this happening */ - if (bio_segments(bio) > 1) - goto bad_map; - /* This is a one page bio that upper layers - * refuse to split for us, so we need to split it. + if (slot >= 0 && r10_bio->devs[slot].rdev) { + /* + * This is an error retry, but we cannot + * safely dereference the rdev in the r10_bio, + * we must use the one in conf. + * If it has already been disconnected (unlikely) + * we lose the device name in error messages. */ - bp = bio_split(bio, - chunk_sects - (bio->bi_sector & (chunk_sects - 1)) ); - - /* Each of these 'make_request' calls will call 'wait_barrier'. - * If the first succeeds but the second blocks due to the resync - * thread raising the barrier, we will deadlock because the - * IO to the underlying device will be queued in generic_make_request - * and will never complete, so will never reduce nr_pending. - * So increment nr_waiting here so no new raise_barriers will - * succeed, and so the second wait_barrier cannot block. + int disk; + /* + * As we are blocking raid10, it is a little safer to + * use __GFP_HIGH. */ - spin_lock_irq(&conf->resync_lock); - conf->nr_waiting++; - spin_unlock_irq(&conf->resync_lock); + gfp = GFP_NOIO | __GFP_HIGH; - make_request(mddev, &bp->bio1); - make_request(mddev, &bp->bio2); - - spin_lock_irq(&conf->resync_lock); - conf->nr_waiting--; - wake_up(&conf->wait_barrier); - spin_unlock_irq(&conf->resync_lock); + disk = r10_bio->devs[slot].devnum; + err_rdev = conf->mirrors[disk].rdev; + if (err_rdev) + snprintf(b, sizeof(b), "%pg", err_rdev->bdev); + else { + strcpy(b, "???"); + /* This never gets dereferenced */ + err_rdev = r10_bio->devs[slot].rdev; + } + } - bio_pair_release(bp); + if (!regular_request_wait(mddev, conf, bio, r10_bio->sectors)) { + raid_end_bio_io(r10_bio); return; - bad_map: - printk("md/raid10:%s: make_request bug: can't convert block across chunks" - " or bigger than %dk %llu %d\n", mdname(mddev), chunk_sects/2, - (unsigned long long)bio->bi_sector, bio_sectors(bio) / 2); + } - bio_io_error(bio); + rdev = read_balance(conf, r10_bio, &max_sectors); + if (!rdev) { + if (err_rdev) { + pr_crit_ratelimited("md/raid10:%s: %s: unrecoverable I/O read error for block %llu\n", + mdname(mddev), b, + (unsigned long long)r10_bio->sector); + } + raid_end_bio_io(r10_bio); return; } - - 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); - - sectors = bio_sectors(bio); - while (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery) && - bio->bi_sector < conf->reshape_progress && - bio->bi_sector + sectors > conf->reshape_progress) { - /* IO spans the reshape position. Need to wait for - * reshape to pass - */ + if (err_rdev) + pr_err_ratelimited("md/raid10:%s: %pg: redirecting sector %llu to another mirror\n", + mdname(mddev), + rdev->bdev, + (unsigned long long)r10_bio->sector); + if (max_sectors < bio_sectors(bio)) { allow_barrier(conf); - wait_event(conf->wait_barrier, - conf->reshape_progress <= bio->bi_sector || - conf->reshape_progress >= bio->bi_sector + sectors); - wait_barrier(conf); + bio = bio_submit_split_bioset(bio, max_sectors, + &conf->bio_split); + wait_barrier(conf, false); + if (!bio) { + set_bit(R10BIO_Returned, &r10_bio->state); + goto err_handle; + } + + r10_bio->master_bio = bio; + r10_bio->sectors = max_sectors; } - if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery) && - bio_data_dir(bio) == WRITE && - (mddev->reshape_backwards - ? (bio->bi_sector < conf->reshape_safe && - bio->bi_sector + sectors > conf->reshape_progress) - : (bio->bi_sector + sectors > conf->reshape_safe && - bio->bi_sector < conf->reshape_progress))) { - /* Need to update reshape_position in metadata */ - mddev->reshape_position = conf->reshape_progress; - set_bit(MD_CHANGE_DEVS, &mddev->flags); - set_bit(MD_CHANGE_PENDING, &mddev->flags); - md_wakeup_thread(mddev->thread); - wait_event(mddev->sb_wait, - !test_bit(MD_CHANGE_PENDING, &mddev->flags)); + slot = r10_bio->read_slot; - conf->reshape_safe = mddev->reshape_position; + if (io_accounting) { + md_account_bio(mddev, &bio); + r10_bio->master_bio = bio; } + read_bio = bio_alloc_clone(rdev->bdev, bio, gfp, &mddev->bio_set); + read_bio->bi_opf &= ~REQ_NOWAIT; - r10_bio = mempool_alloc(conf->r10bio_pool, GFP_NOIO); + r10_bio->devs[slot].bio = read_bio; + r10_bio->devs[slot].rdev = rdev; - r10_bio->master_bio = bio; - r10_bio->sectors = sectors; + read_bio->bi_iter.bi_sector = r10_bio->devs[slot].addr + + choose_data_offset(r10_bio, rdev); + read_bio->bi_end_io = raid10_end_read_request; + if (test_bit(FailFast, &rdev->flags) && + test_bit(R10BIO_FailFast, &r10_bio->state)) + read_bio->bi_opf |= MD_FAILFAST; + read_bio->bi_private = r10_bio; + mddev_trace_remap(mddev, read_bio, r10_bio->sector); + submit_bio_noacct(read_bio); + return; +err_handle: + atomic_dec(&rdev->nr_pending); + raid_end_bio_io(r10_bio); +} - r10_bio->mddev = mddev; - r10_bio->sector = bio->bi_sector; - r10_bio->state = 0; +static void raid10_write_one_disk(struct mddev *mddev, struct r10bio *r10_bio, + struct bio *bio, bool replacement, + int n_copy) +{ + unsigned long flags; + struct r10conf *conf = mddev->private; + struct md_rdev *rdev; + int devnum = r10_bio->devs[n_copy].devnum; + struct bio *mbio; - /* 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; - clear_bit(BIO_SEG_VALID, &bio->bi_flags); + rdev = replacement ? conf->mirrors[devnum].replacement : + conf->mirrors[devnum].rdev; - if (rw == READ) { - /* - * read balancing logic: - */ - struct md_rdev *rdev; - int slot; + mbio = bio_alloc_clone(rdev->bdev, bio, GFP_NOIO, &mddev->bio_set); + mbio->bi_opf &= ~REQ_NOWAIT; + if (replacement) + r10_bio->devs[n_copy].repl_bio = mbio; + else + r10_bio->devs[n_copy].bio = mbio; + + mbio->bi_iter.bi_sector = (r10_bio->devs[n_copy].addr + + choose_data_offset(r10_bio, rdev)); + mbio->bi_end_io = raid10_end_write_request; + if (!replacement && test_bit(FailFast, + &conf->mirrors[devnum].rdev->flags) + && enough(conf, devnum)) + mbio->bi_opf |= MD_FAILFAST; + mbio->bi_private = r10_bio; + mddev_trace_remap(mddev, mbio, r10_bio->sector); + /* flush_pending_writes() needs access to the rdev so...*/ + mbio->bi_bdev = (void *)rdev; -read_again: - rdev = read_balance(conf, r10_bio, &max_sectors); - if (!rdev) { - raid_end_bio_io(r10_bio); - return; - } - slot = r10_bio->read_slot; + atomic_inc(&r10_bio->remaining); + + if (!raid1_add_bio_to_plug(mddev, mbio, raid10_unplug, conf->copies)) { + spin_lock_irqsave(&conf->device_lock, flags); + bio_list_add(&conf->pending_bio_list, mbio); + spin_unlock_irqrestore(&conf->device_lock, flags); + md_wakeup_thread(mddev->thread); + } +} - read_bio = bio_clone_mddev(bio, GFP_NOIO, mddev); - md_trim_bio(read_bio, r10_bio->sector - bio->bi_sector, - max_sectors); +static void wait_blocked_dev(struct mddev *mddev, struct r10bio *r10_bio) +{ + struct r10conf *conf = mddev->private; + struct md_rdev *blocked_rdev; + int i; - r10_bio->devs[slot].bio = read_bio; - r10_bio->devs[slot].rdev = rdev; +retry_wait: + blocked_rdev = NULL; + for (i = 0; i < conf->copies; i++) { + struct md_rdev *rdev, *rrdev; - read_bio->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; + rdev = conf->mirrors[i].rdev; + if (rdev) { + sector_t dev_sector = r10_bio->devs[i].addr; - if (max_sectors < r10_bio->sectors) { - /* Could not read all from this device, so we will - * need another r10_bio. - */ - sectors_handled = (r10_bio->sectors + max_sectors - - bio->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(&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. + /* + * Discard request doesn't care the write result + * so it doesn't need to wait blocked disk here. */ - reschedule_retry(r10_bio); + if (test_bit(WriteErrorSeen, &rdev->flags) && + r10_bio->sectors && + rdev_has_badblock(rdev, dev_sector, + r10_bio->sectors) < 0) + /* + * Mustn't write here until the bad + * block is acknowledged + */ + set_bit(BlockedBadBlocks, &rdev->flags); - r10_bio = mempool_alloc(conf->r10bio_pool, GFP_NOIO); + if (rdev_blocked(rdev)) { + blocked_rdev = rdev; + atomic_inc(&rdev->nr_pending); + break; + } + } - 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_sector + sectors_handled; - goto read_again; - } else - generic_make_request(read_bio); + rrdev = conf->mirrors[i].replacement; + if (rrdev && rdev_blocked(rrdev)) { + atomic_inc(&rrdev->nr_pending); + blocked_rdev = rrdev; + break; + } + } + + if (unlikely(blocked_rdev)) { + /* Have to wait for this device to get unblocked, then retry */ + allow_barrier(conf); + mddev_add_trace_msg(conf->mddev, + "raid10 %s wait rdev %d blocked", + __func__, blocked_rdev->raid_disk); + md_wait_for_blocked_rdev(blocked_rdev, mddev); + wait_barrier(conf, false); + goto retry_wait; + } +} + +static void raid10_write_request(struct mddev *mddev, struct bio *bio, + struct r10bio *r10_bio) +{ + struct r10conf *conf = mddev->private; + int i, k; + sector_t sectors; + int max_sectors; + + if ((mddev_is_clustered(mddev) && + mddev->cluster_ops->area_resyncing(mddev, WRITE, + bio->bi_iter.bi_sector, + bio_end_sector(bio)))) { + DEFINE_WAIT(w); + /* Bail out if REQ_NOWAIT is set for the bio */ + if (bio->bi_opf & REQ_NOWAIT) { + bio_wouldblock_error(bio); + return; + } + for (;;) { + prepare_to_wait(&conf->wait_barrier, + &w, TASK_IDLE); + if (!mddev->cluster_ops->area_resyncing(mddev, WRITE, + bio->bi_iter.bi_sector, bio_end_sector(bio))) + break; + schedule(); + } + finish_wait(&conf->wait_barrier, &w); + } + + sectors = r10_bio->sectors; + if (!regular_request_wait(mddev, conf, bio, sectors)) { + raid_end_bio_io(r10_bio); return; } - /* - * WRITE: - */ - if (conf->pending_count >= max_queued_requests) { + if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery) && + (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->sb_flags, 0, + BIT(MD_SB_CHANGE_DEVS) | BIT(MD_SB_CHANGE_PENDING)); md_wakeup_thread(mddev->thread); - wait_event(conf->wait_barrier, - conf->pending_count < max_queued_requests); + if (bio->bi_opf & REQ_NOWAIT) { + allow_barrier(conf); + bio_wouldblock_error(bio); + return; + } + mddev_add_trace_msg(conf->mddev, + "raid10 wait reshape metadata"); + wait_event(mddev->sb_wait, + !test_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags)); + + conf->reshape_safe = mddev->reshape_position; } + /* first select target devices under rcu_lock and * inc refcount on their rdev. Record them by setting * bios[x] to bio @@ -1363,67 +1407,40 @@ read_again: * 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. + * gets its own r10_bio with a set of bios attached. */ 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(); + + wait_blocked_dev(mddev, r10_bio); + 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) - || test_bit(Unmerged, &rdev->flags))) + struct md_rdev *rdev, *rrdev; + + rdev = conf->mirrors[d].rdev; + rrdev = conf->mirrors[d].replacement; + if (rdev && (test_bit(Faulty, &rdev->flags))) rdev = NULL; - if (rrdev && (test_bit(Faulty, &rrdev->flags) - || test_bit(Unmerged, &rrdev->flags))) + 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); + if (!rdev && !rrdev) continue; - } if (rdev && test_bit(WriteErrorSeen, &rdev->flags)) { sector_t first_bad; sector_t dev_sector = r10_bio->devs[i].addr; - int bad_sectors; + sector_t bad_sectors; int is_bad; - is_bad = is_badblock(rdev, dev_sector, - max_sectors, + 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); @@ -1432,18 +1449,22 @@ retry_write: * 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; + int good_sectors; + + /* + * We cannot atomically write this, so just + * error in that case. It could be possible to + * atomically write other mirrors, but the + * complexity of supporting that is not worth + * the benefit. + */ + if (bio->bi_opf & REQ_ATOMIC) + goto err_handle; + + good_sectors = first_bad - dev_sector; if (good_sectors < max_sectors) max_sectors = good_sectors; } @@ -1457,155 +1478,433 @@ retry_write: 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; + if (max_sectors < r10_bio->sectors) + r10_bio->sectors = max_sectors; - 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); - } - } + if (r10_bio->sectors < bio_sectors(bio)) { allow_barrier(conf); - md_wait_for_blocked_rdev(blocked_rdev, mddev); - wait_barrier(conf); - goto retry_write; - } + bio = bio_submit_split_bioset(bio, r10_bio->sectors, + &conf->bio_split); + wait_barrier(conf, false); + if (!bio) { + set_bit(R10BIO_Returned, &r10_bio->state); + goto err_handle; + } - 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); + r10_bio->master_bio = bio; } - sectors_handled = r10_bio->sector + max_sectors - bio->bi_sector; + md_account_bio(mddev, &bio); + r10_bio->master_bio = bio; 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); - md_trim_bio(mbio, r10_bio->sector - bio->bi_sector, - max_sectors); - r10_bio->devs[i].bio = mbio; - - mbio->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; + if (r10_bio->devs[i].bio) + raid10_write_one_disk(mddev, r10_bio, bio, false, i); + if (r10_bio->devs[i].repl_bio) + raid10_write_one_disk(mddev, r10_bio, bio, true, i); + } + one_write_done(r10_bio); + return; +err_handle: + for (k = 0; k < i; k++) { + int d = r10_bio->devs[k].devnum; + struct md_rdev *rdev = conf->mirrors[d].rdev; + struct md_rdev *rrdev = conf->mirrors[d].replacement; + + if (r10_bio->devs[k].bio) { + rdev_dec_pending(rdev, mddev); + r10_bio->devs[k].bio = NULL; + } + if (r10_bio->devs[k].repl_bio) { + rdev_dec_pending(rrdev, mddev); + r10_bio->devs[k].repl_bio = NULL; + } + } - atomic_inc(&r10_bio->remaining); + raid_end_bio_io(r10_bio); +} - 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); +static void __make_request(struct mddev *mddev, struct bio *bio, int sectors) +{ + struct r10conf *conf = mddev->private; + struct r10bio *r10_bio; + + 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; + r10_bio->read_slot = -1; + memset(r10_bio->devs, 0, sizeof(r10_bio->devs[0]) * + conf->geo.raid_disks); + + if (bio_data_dir(bio) == READ) + raid10_read_request(mddev, bio, r10_bio, true); + else + raid10_write_request(mddev, bio, r10_bio); +} + +static void raid_end_discard_bio(struct r10bio *r10bio) +{ + struct r10conf *conf = r10bio->mddev->private; + struct r10bio *first_r10bio; + + while (atomic_dec_and_test(&r10bio->remaining)) { + + allow_barrier(conf); + + if (!test_bit(R10BIO_Discard, &r10bio->state)) { + first_r10bio = (struct r10bio *)r10bio->master_bio; + free_r10bio(r10bio); + r10bio = first_r10bio; + } else { + md_write_end(r10bio->mddev); + bio_endio(r10bio->master_bio); + free_r10bio(r10bio); + break; } + } +} - 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); - md_trim_bio(mbio, r10_bio->sector - bio->bi_sector, - max_sectors); - r10_bio->devs[i].repl_bio = mbio; - - mbio->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; +static void raid10_end_discard_request(struct bio *bio) +{ + struct r10bio *r10_bio = bio->bi_private; + struct r10conf *conf = r10_bio->mddev->private; + struct md_rdev *rdev = NULL; + int dev; + int slot, repl; - 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); + /* + * We don't care the return value of discard bio + */ + if (!test_bit(R10BIO_Uptodate, &r10_bio->state)) + set_bit(R10BIO_Uptodate, &r10_bio->state); + + dev = find_bio_disk(conf, r10_bio, bio, &slot, &repl); + rdev = repl ? conf->mirrors[dev].replacement : + conf->mirrors[dev].rdev; + + raid_end_discard_bio(r10_bio); + rdev_dec_pending(rdev, conf->mddev); +} + +/* + * There are some limitations to handle discard bio + * 1st, the discard size is bigger than stripe_size*2. + * 2st, if the discard bio spans reshape progress, we use the old way to + * handle discard bio + */ +static int raid10_handle_discard(struct mddev *mddev, struct bio *bio) +{ + struct r10conf *conf = mddev->private; + struct geom *geo = &conf->geo; + int far_copies = geo->far_copies; + bool first_copy = true; + struct r10bio *r10_bio, *first_r10bio; + struct bio *split; + int disk; + sector_t chunk; + unsigned int stripe_size; + unsigned int stripe_data_disks; + sector_t split_size; + sector_t bio_start, bio_end; + sector_t first_stripe_index, last_stripe_index; + sector_t start_disk_offset; + unsigned int start_disk_index; + sector_t end_disk_offset; + unsigned int end_disk_index; + unsigned int remainder; + + if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery)) + return -EAGAIN; + + if (!wait_barrier(conf, bio->bi_opf & REQ_NOWAIT)) { + bio_wouldblock_error(bio); + return 0; + } + + /* + * Check reshape again to avoid reshape happens after checking + * MD_RECOVERY_RESHAPE and before wait_barrier + */ + if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery)) + goto out; + + if (geo->near_copies) + stripe_data_disks = geo->raid_disks / geo->near_copies + + geo->raid_disks % geo->near_copies; + else + stripe_data_disks = geo->raid_disks; + + stripe_size = stripe_data_disks << geo->chunk_shift; + + bio_start = bio->bi_iter.bi_sector; + bio_end = bio_end_sector(bio); + + /* + * Maybe one discard bio is smaller than strip size or across one + * stripe and discard region is larger than one stripe size. For far + * offset layout, if the discard region is not aligned with stripe + * size, there is hole when we submit discard bio to member disk. + * For simplicity, we only handle discard bio which discard region + * is bigger than stripe_size * 2 + */ + if (bio_sectors(bio) < stripe_size*2) + goto out; + + /* + * Keep bio aligned with strip size. + */ + div_u64_rem(bio_start, stripe_size, &remainder); + if (remainder) { + split_size = stripe_size - remainder; + split = bio_split(bio, split_size, GFP_NOIO, &conf->bio_split); + if (IS_ERR(split)) { + bio->bi_status = errno_to_blk_status(PTR_ERR(split)); + bio_endio(bio); + return 0; + } + + bio_chain(split, bio); + trace_block_split(split, bio->bi_iter.bi_sector); + allow_barrier(conf); + /* Resend the fist split part */ + submit_bio_noacct(split); + wait_barrier(conf, false); + } + div_u64_rem(bio_end, stripe_size, &remainder); + if (remainder) { + split_size = bio_sectors(bio) - remainder; + split = bio_split(bio, split_size, GFP_NOIO, &conf->bio_split); + if (IS_ERR(split)) { + bio->bi_status = errno_to_blk_status(PTR_ERR(split)); + bio_endio(bio); + return 0; } + + bio_chain(split, bio); + trace_block_split(split, bio->bi_iter.bi_sector); + allow_barrier(conf); + /* Resend the second split part */ + submit_bio_noacct(bio); + bio = split; + wait_barrier(conf, false); } - /* Don't remove the bias on 'remaining' (one_write_done) until - * after checking if we need to go around again. + bio_start = bio->bi_iter.bi_sector; + bio_end = bio_end_sector(bio); + + /* + * Raid10 uses chunk as the unit to store data. It's similar like raid0. + * One stripe contains the chunks from all member disk (one chunk from + * one disk at the same HBA address). For layout detail, see 'man md 4' + */ + chunk = bio_start >> geo->chunk_shift; + chunk *= geo->near_copies; + first_stripe_index = chunk; + start_disk_index = sector_div(first_stripe_index, geo->raid_disks); + if (geo->far_offset) + first_stripe_index *= geo->far_copies; + start_disk_offset = (bio_start & geo->chunk_mask) + + (first_stripe_index << geo->chunk_shift); + + chunk = bio_end >> geo->chunk_shift; + chunk *= geo->near_copies; + last_stripe_index = chunk; + end_disk_index = sector_div(last_stripe_index, geo->raid_disks); + if (geo->far_offset) + last_stripe_index *= geo->far_copies; + end_disk_offset = (bio_end & geo->chunk_mask) + + (last_stripe_index << geo->chunk_shift); + +retry_discard: + r10_bio = mempool_alloc(&conf->r10bio_pool, GFP_NOIO); + r10_bio->mddev = mddev; + r10_bio->state = 0; + r10_bio->sectors = 0; + memset(r10_bio->devs, 0, sizeof(r10_bio->devs[0]) * geo->raid_disks); + wait_blocked_dev(mddev, r10_bio); + + /* + * For far layout it needs more than one r10bio to cover all regions. + * Inspired by raid10_sync_request, we can use the first r10bio->master_bio + * to record the discard bio. Other r10bio->master_bio record the first + * r10bio. The first r10bio only release after all other r10bios finish. + * The discard bio returns only first r10bio finishes */ + if (first_copy) { + md_account_bio(mddev, &bio); + r10_bio->master_bio = bio; + set_bit(R10BIO_Discard, &r10_bio->state); + first_copy = false; + first_r10bio = r10_bio; + } else + r10_bio->master_bio = (struct bio *)first_r10bio; + + /* + * first select target devices under rcu_lock and + * inc refcount on their rdev. Record them by setting + * bios[x] to bio + */ + for (disk = 0; disk < geo->raid_disks; disk++) { + struct md_rdev *rdev, *rrdev; + + rdev = conf->mirrors[disk].rdev; + rrdev = conf->mirrors[disk].replacement; + r10_bio->devs[disk].bio = NULL; + r10_bio->devs[disk].repl_bio = NULL; + + if (rdev && (test_bit(Faulty, &rdev->flags))) + rdev = NULL; + if (rrdev && (test_bit(Faulty, &rrdev->flags))) + rrdev = NULL; + if (!rdev && !rrdev) + continue; + + if (rdev) { + r10_bio->devs[disk].bio = bio; + atomic_inc(&rdev->nr_pending); + } + if (rrdev) { + r10_bio->devs[disk].repl_bio = bio; + atomic_inc(&rrdev->nr_pending); + } + } + + atomic_set(&r10_bio->remaining, 1); + for (disk = 0; disk < geo->raid_disks; disk++) { + sector_t dev_start, dev_end; + struct bio *mbio, *rbio = NULL; - 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. + /* + * Now start to calculate the start and end address for each disk. + * The space between dev_start and dev_end is the discard region. + * + * For dev_start, it needs to consider three conditions: + * 1st, the disk is before start_disk, you can imagine the disk in + * the next stripe. So the dev_start is the start address of next + * stripe. + * 2st, the disk is after start_disk, it means the disk is at the + * same stripe of first disk + * 3st, the first disk itself, we can use start_disk_offset directly */ - r10_bio = mempool_alloc(conf->r10bio_pool, GFP_NOIO); + if (disk < start_disk_index) + dev_start = (first_stripe_index + 1) * mddev->chunk_sectors; + else if (disk > start_disk_index) + dev_start = first_stripe_index * mddev->chunk_sectors; + else + dev_start = start_disk_offset; - r10_bio->master_bio = bio; - r10_bio->sectors = bio_sectors(bio) - sectors_handled; + if (disk < end_disk_index) + dev_end = (last_stripe_index + 1) * mddev->chunk_sectors; + else if (disk > end_disk_index) + dev_end = last_stripe_index * mddev->chunk_sectors; + else + dev_end = end_disk_offset; - r10_bio->mddev = mddev; - r10_bio->sector = bio->bi_sector + sectors_handled; - r10_bio->state = 0; - goto retry_write; + /* + * It only handles discard bio which size is >= stripe size, so + * dev_end > dev_start all the time. + * It doesn't need to use rcu lock to get rdev here. We already + * add rdev->nr_pending in the first loop. + */ + if (r10_bio->devs[disk].bio) { + struct md_rdev *rdev = conf->mirrors[disk].rdev; + mbio = bio_alloc_clone(bio->bi_bdev, bio, GFP_NOIO, + &mddev->bio_set); + mbio->bi_end_io = raid10_end_discard_request; + mbio->bi_private = r10_bio; + r10_bio->devs[disk].bio = mbio; + r10_bio->devs[disk].devnum = disk; + atomic_inc(&r10_bio->remaining); + md_submit_discard_bio(mddev, rdev, mbio, + dev_start + choose_data_offset(r10_bio, rdev), + dev_end - dev_start); + bio_endio(mbio); + } + if (r10_bio->devs[disk].repl_bio) { + struct md_rdev *rrdev = conf->mirrors[disk].replacement; + rbio = bio_alloc_clone(bio->bi_bdev, bio, GFP_NOIO, + &mddev->bio_set); + rbio->bi_end_io = raid10_end_discard_request; + rbio->bi_private = r10_bio; + r10_bio->devs[disk].repl_bio = rbio; + r10_bio->devs[disk].devnum = disk; + atomic_inc(&r10_bio->remaining); + md_submit_discard_bio(mddev, rrdev, rbio, + dev_start + choose_data_offset(r10_bio, rrdev), + dev_end - dev_start); + bio_endio(rbio); + } } - one_write_done(r10_bio); + + if (!geo->far_offset && --far_copies) { + first_stripe_index += geo->stride >> geo->chunk_shift; + start_disk_offset += geo->stride; + last_stripe_index += geo->stride >> geo->chunk_shift; + end_disk_offset += geo->stride; + atomic_inc(&first_r10bio->remaining); + raid_end_discard_bio(r10_bio); + wait_barrier(conf, false); + goto retry_discard; + } + + raid_end_discard_bio(r10_bio); + + return 0; +out: + allow_barrier(conf); + return -EAGAIN; +} + +static bool 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; + int sectors = bio_sectors(bio); + + if (unlikely(bio->bi_opf & REQ_PREFLUSH) + && md_flush_request(mddev, bio)) + return true; + + md_write_start(mddev, bio); + + if (unlikely(bio_op(bio) == REQ_OP_DISCARD)) + if (!raid10_handle_discard(mddev, bio)) + return true; + + /* + * If this request crosses a chunk boundary, we need to split + * it. + */ + if (unlikely((bio->bi_iter.bi_sector & chunk_mask) + + sectors > chunk_sects + && (conf->geo.near_copies < conf->geo.raid_disks + || conf->prev.near_copies < + conf->prev.raid_disks))) + sectors = chunk_sects - + (bio->bi_iter.bi_sector & + (chunk_sects - 1)); + __make_request(mddev, bio, sectors); /* In case raid10d snuck in to freeze_array */ - wake_up(&conf->wait_barrier); + wake_up_barrier(conf); + return true; } -static void status(struct seq_file *seq, struct mddev *mddev) +static void raid10_status(struct seq_file *seq, struct mddev *mddev) { struct r10conf *conf = mddev->private; int i; + lockdep_assert_held(&mddev->lock); + if (conf->geo.near_copies < conf->geo.raid_disks) seq_printf(seq, " %dK chunks", mddev->chunk_sectors / 2); if (conf->geo.near_copies > 1) @@ -1615,13 +1914,16 @@ static void status(struct seq_file *seq, struct mddev *mddev) 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" : "_"); + for (i = 0; i < conf->geo.raid_disks; i++) { + struct md_rdev *rdev = READ_ONCE(conf->mirrors[i].rdev); + + seq_printf(seq, "%s", rdev && test_bit(In_sync, &rdev->flags) ? "U" : "_"); + } seq_printf(seq, "]"); } @@ -1643,7 +1945,6 @@ static int _enough(struct r10conf *conf, int previous, int ignore) ncopies = conf->geo.near_copies; } - rcu_read_lock(); do { int n = conf->copies; int cnt = 0; @@ -1651,7 +1952,7 @@ static int _enough(struct r10conf *conf, int previous, int ignore) while (n--) { struct md_rdev *rdev; if (this != ignore && - (rdev = rcu_dereference(conf->mirrors[this].rdev)) && + (rdev = conf->mirrors[this].rdev) && test_bit(In_sync, &rdev->flags)) cnt++; this = (this+1) % disks; @@ -1662,7 +1963,6 @@ static int _enough(struct r10conf *conf, int previous, int ignore) } while (first != 0); has_enough = 1; out: - rcu_read_unlock(); return has_enough; } @@ -1677,76 +1977,81 @@ static int enough(struct r10conf *conf, int ignore) _enough(conf, 1, ignore); } -static void error(struct mddev *mddev, struct md_rdev *rdev) +/** + * raid10_error() - RAID10 error handler. + * @mddev: affected md device. + * @rdev: member device to fail. + * + * The routine acknowledges &rdev failure and determines new @mddev state. + * If it failed, then: + * - &MD_BROKEN flag is set in &mddev->flags. + * Otherwise, it must be degraded: + * - recovery is interrupted. + * - &mddev->degraded is bumped. + * + * @rdev is marked as &Faulty excluding case when array is failed and + * &mddev->fail_last_dev is off. + */ +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_bit(In_sync, &rdev->flags) && !enough(conf, rdev->raid_disk)) { + set_bit(MD_BROKEN, &mddev->flags); + + if (!mddev->fail_last_dev) { + spin_unlock_irqrestore(&conf->device_lock, flags); + return; + } } - if (test_and_clear_bit(In_sync, &rdev->flags)) { + 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(MD_RECOVERY_INTR, &mddev->recovery); set_bit(Blocked, &rdev->flags); set_bit(Faulty, &rdev->flags); - set_bit(MD_CHANGE_DEVS, &mddev->flags); + set_mask_bits(&mddev->sb_flags, 0, + BIT(MD_SB_CHANGE_DEVS) | BIT(MD_SB_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); + pr_crit("md/raid10:%s: Disk failure on %pg, disabling device.\n" + "md/raid10:%s: Operation continuing on %d devices.\n", + mdname(mddev), rdev->bdev, + mdname(mddev), conf->geo.raid_disks - mddev->degraded); } static void print_conf(struct r10conf *conf) { int i; - struct raid10_info *tmp; + struct md_rdev *rdev; - printk(KERN_DEBUG "RAID10 conf printout:\n"); + pr_debug("RAID10 conf printout:\n"); if (!conf) { - printk(KERN_DEBUG "(!conf)\n"); + pr_debug("(!conf)\n"); return; } - printk(KERN_DEBUG " --- wd:%d rd:%d\n", conf->geo.raid_disks - conf->mddev->degraded, - conf->geo.raid_disks); + pr_debug(" --- wd:%d rd:%d\n", conf->geo.raid_disks - conf->mddev->degraded, + conf->geo.raid_disks); + lockdep_assert_held(&conf->mddev->reconfig_mutex); 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)); + rdev = conf->mirrors[i].rdev; + if (rdev) + pr_debug(" disk %d, wo:%d, o:%d, dev:%pg\n", + i, !test_bit(In_sync, &rdev->flags), + !test_bit(Faulty, &rdev->flags), + rdev->bdev); } } static void close_sync(struct r10conf *conf) { - wait_barrier(conf); + wait_barrier(conf, false); allow_barrier(conf); - mempool_destroy(conf->r10buf_pool); - conf->r10buf_pool = NULL; + mempool_exit(&conf->r10buf_pool); } static int raid10_spare_active(struct mddev *mddev) @@ -1782,6 +2087,7 @@ static int raid10_spare_active(struct mddev *mddev) } 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++; @@ -1796,17 +2102,16 @@ static int raid10_spare_active(struct mddev *mddev) 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 mirror, repl_slot = -1; int first = 0; int last = conf->geo.raid_disks - 1; - struct request_queue *q = bdev_get_queue(rdev->bdev); + struct raid10_info *p; - if (mddev->recovery_cp < MaxSector) + if (mddev->resync_offset < MaxSector) /* only hot-add to in-sync arrays, as recovery is * very different from resync */ @@ -1817,65 +2122,47 @@ static int raid10_add_disk(struct mddev *mddev, struct md_rdev *rdev) if (rdev->raid_disk >= 0) first = last = rdev->raid_disk; - if (q->merge_bvec_fn) { - set_bit(Unmerged, &rdev->flags); - mddev->merge_check_needed = 1; - } - if (rdev->saved_raid_disk >= first && + rdev->saved_raid_disk < conf->geo.raid_disks && 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]; + 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 (test_bit(WantReplacement, &p->rdev->flags) && + p->replacement == NULL && repl_slot < 0) + repl_slot = mirror; + continue; } - if (mddev->gendisk) - disk_stack_limits(mddev->gendisk, rdev->bdev, - rdev->data_offset << 9); - + err = mddev_stack_new_rdev(mddev, rdev); + if (err) + return err; 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); + WRITE_ONCE(p->rdev, rdev); break; } - if (err == 0 && test_bit(Unmerged, &rdev->flags)) { - /* Some requests might not have seen this new - * merge_bvec_fn. We must wait for them to complete - * before merging the device fully. - * First we make sure any code which has tested - * our function has submitted the request, then - * we wait for all outstanding requests to complete. - */ - synchronize_sched(); - freeze_array(conf, 0); - unfreeze_array(conf); - clear_bit(Unmerged, &rdev->flags); + + if (err && repl_slot >= 0) { + p = &conf->mirrors[repl_slot]; + clear_bit(In_sync, &rdev->flags); + set_bit(Replacement, &rdev->flags); + rdev->raid_disk = repl_slot; + err = mddev_stack_new_rdev(mddev, rdev); + if (err) + return err; + conf->fullsync = 1; + WRITE_ONCE(p->replacement, rdev); } - md_integrity_add_rdev(rdev, mddev); - 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; @@ -1887,9 +2174,12 @@ static int raid10_remove_disk(struct mddev *mddev, struct md_rdev *rdev) int err = 0; int number = rdev->raid_disk; struct md_rdev **rdevp; - struct raid10_info *p = conf->mirrors + number; + struct raid10_info *p; print_conf(conf); + if (unlikely(number >= mddev->raid_disks)) + return 0; + p = conf->mirrors + number; if (rdev == p->rdev) rdevp = &p->rdev; else if (rdev == p->replacement) @@ -1902,7 +2192,7 @@ static int raid10_remove_disk(struct mddev *mddev, struct md_rdev *rdev) err = -EBUSY; goto abort; } - /* Only remove faulty devices if recovery + /* Only remove non-faulty devices if recovery * is not possible. */ if (!test_bit(Faulty, &rdev->flags) && @@ -1913,28 +2203,15 @@ static int raid10_remove_disk(struct mddev *mddev, struct md_rdev *rdev) err = -EBUSY; goto abort; } - *rdevp = NULL; - synchronize_rcu(); - if (atomic_read(&rdev->nr_pending)) { - /* lost the race, try later */ - err = -EBUSY; - *rdevp = rdev; - goto abort; - } else if (p->replacement) { + WRITE_ONCE(*rdevp, NULL); + if (p->replacement) { /* We must have just cleared 'rdev' */ - p->rdev = p->replacement; + WRITE_ONCE(p->rdev, p->replacement); clear_bit(Replacement, &p->replacement->flags); - smp_mb(); /* Make sure other CPUs may see both as identical - * but will never see neither -- if they are careful. - */ - p->replacement = NULL; - clear_bit(WantReplacement, &rdev->flags); - } else - /* We might have just remove the Replacement as faulty - * Clear the flag just in case - */ - clear_bit(WantReplacement, &rdev->flags); + WRITE_ONCE(p->replacement, NULL); + } + clear_bit(WantReplacement, &rdev->flags); err = md_integrity_register(mddev); abort: @@ -1943,20 +2220,11 @@ abort: return err; } - -static void end_sync_read(struct bio *bio, int error) +static void __end_sync_read(struct r10bio *r10_bio, struct bio *bio, int d) { - struct r10bio *r10_bio = bio->bi_private; struct r10conf *conf = r10_bio->mddev->private; - int d; - if (bio == r10_bio->master_bio) { - /* this is a reshape read */ - d = r10_bio->read_slot; /* really the read dev */ - } else - d = find_bio_disk(conf, r10_bio, bio, NULL, NULL); - - if (test_bit(BIO_UPTODATE, &bio->bi_flags)) + if (!bio->bi_status) set_bit(R10BIO_Uptodate, &r10_bio->state); else /* The write handler will notice the lack of @@ -1978,6 +2246,23 @@ static void end_sync_read(struct bio *bio, int error) } } +static void end_sync_read(struct bio *bio) +{ + struct r10bio *r10_bio = get_resync_r10bio(bio); + struct r10conf *conf = r10_bio->mddev->private; + int d = find_bio_disk(conf, r10_bio, bio, NULL, NULL); + + __end_sync_read(r10_bio, bio, d); +} + +static void end_reshape_read(struct bio *bio) +{ + /* reshape read bio isn't allocated from r10buf_pool */ + struct r10bio *r10_bio = bio->bi_private; + + __end_sync_read(r10_bio, bio, r10_bio->read_slot); +} + static void end_sync_request(struct r10bio *r10_bio) { struct mddev *mddev = r10_bio->mddev; @@ -2005,15 +2290,12 @@ static void end_sync_request(struct r10bio *r10_bio) } } -static void end_sync_write(struct bio *bio, int error) +static void end_sync_write(struct bio *bio) { - int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags); - struct r10bio *r10_bio = bio->bi_private; + struct r10bio *r10_bio = get_resync_r10bio(bio); struct mddev *mddev = r10_bio->mddev; struct r10conf *conf = mddev->private; int d; - sector_t first_bad; - int bad_sectors; int slot; int repl; struct md_rdev *rdev = NULL; @@ -2024,7 +2306,7 @@ static void end_sync_write(struct bio *bio, int error) else rdev = conf->mirrors[d].rdev; - if (!uptodate) { + if (bio->bi_status) { if (repl) md_error(mddev, rdev); else { @@ -2034,11 +2316,10 @@ static void end_sync_write(struct bio *bio, int error) &rdev->mddev->recovery); set_bit(R10BIO_WriteError, &r10_bio->state); } - } else if (is_badblock(rdev, - r10_bio->devs[slot].addr, - r10_bio->sectors, - &first_bad, &bad_sectors)) + } else if (rdev_has_badblock(rdev, r10_bio->devs[slot].addr, + r10_bio->sectors)) { set_bit(R10BIO_MadeGood, &r10_bio->state); + } rdev_dec_pending(rdev, mddev); @@ -2067,12 +2348,13 @@ static void sync_request_write(struct mddev *mddev, struct r10bio *r10_bio) int i, first; struct bio *tbio, *fbio; int vcnt; + struct page **tpages, **fpages; atomic_set(&r10_bio->remaining, 1); /* find the first device with a block */ for (i=0; i<conf->copies; i++) - if (test_bit(BIO_UPTODATE, &r10_bio->devs[i].bio->bi_flags)) + if (!r10_bio->devs[i].bio->bi_status) break; if (i == conf->copies) @@ -2080,11 +2362,16 @@ static void sync_request_write(struct mddev *mddev, struct r10bio *r10_bio) first = i; fbio = r10_bio->devs[i].bio; + fbio->bi_iter.bi_size = r10_bio->sectors << 9; + fbio->bi_iter.bi_idx = 0; + fpages = get_resync_pages(fbio)->pages; vcnt = (r10_bio->sectors + (PAGE_SIZE >> 9) - 1) >> (PAGE_SHIFT - 9); /* now find blocks with errors */ for (i=0 ; i < conf->copies ; i++) { int j, d; + struct md_rdev *rdev; + struct resync_pages *rp; tbio = r10_bio->devs[i].bio; @@ -2092,76 +2379,75 @@ static void sync_request_write(struct mddev *mddev, struct r10bio *r10_bio) continue; if (i == first) continue; - if (test_bit(BIO_UPTODATE, &r10_bio->devs[i].bio->bi_flags)) { + + tpages = get_resync_pages(tbio)->pages; + d = r10_bio->devs[i].devnum; + rdev = conf->mirrors[d].rdev; + if (!r10_bio->devs[i].bio->bi_status) { /* We know that the bi_io_vec layout is the same for * both 'first' and 'i', so we just compare them. * All vec entries are PAGE_SIZE; */ - for (j = 0; j < vcnt; j++) - if (memcmp(page_address(fbio->bi_io_vec[j].bv_page), - page_address(tbio->bi_io_vec[j].bv_page), - fbio->bi_io_vec[j].bv_len)) + int sectors = r10_bio->sectors; + for (j = 0; j < vcnt; j++) { + int len = PAGE_SIZE; + if (sectors < (len / 512)) + len = sectors * 512; + if (memcmp(page_address(fpages[j]), + page_address(tpages[j]), + len)) break; + sectors -= len/512; + } if (j == vcnt) continue; atomic64_add(r10_bio->sectors, &mddev->resync_mismatches); if (test_bit(MD_RECOVERY_CHECK, &mddev->recovery)) /* Don't fix anything. */ continue; + } else if (test_bit(FailFast, &rdev->flags)) { + /* Just give up on this device */ + md_error(rdev->mddev, rdev); + continue; } /* Ok, we need to write this bio, either to correct an * inconsistency or to correct an unreadable block. * First we need to fixup bv_offset, bv_len and * bi_vecs, as the read request might have corrupted these */ - bio_reset(tbio); - - tbio->bi_vcnt = vcnt; - tbio->bi_size = r10_bio->sectors << 9; - tbio->bi_rw = WRITE; - tbio->bi_private = r10_bio; - tbio->bi_sector = r10_bio->devs[i].addr; + rp = get_resync_pages(tbio); + bio_reset(tbio, conf->mirrors[d].rdev->bdev, REQ_OP_WRITE); - for (j=0; j < vcnt ; j++) { - tbio->bi_io_vec[j].bv_offset = 0; - tbio->bi_io_vec[j].bv_len = PAGE_SIZE; + md_bio_reset_resync_pages(tbio, rp, fbio->bi_iter.bi_size); - memcpy(page_address(tbio->bi_io_vec[j].bv_page), - page_address(fbio->bi_io_vec[j].bv_page), - PAGE_SIZE); - } + rp->raid_bio = r10_bio; + tbio->bi_private = rp; + tbio->bi_iter.bi_sector = r10_bio->devs[i].addr; tbio->bi_end_io = end_sync_write; - d = r10_bio->devs[i].devnum; + bio_copy_data(tbio, fbio); + atomic_inc(&conf->mirrors[d].rdev->nr_pending); atomic_inc(&r10_bio->remaining); - md_sync_acct(conf->mirrors[d].rdev->bdev, bio_sectors(tbio)); - tbio->bi_sector += conf->mirrors[d].rdev->data_offset; - tbio->bi_bdev = conf->mirrors[d].rdev->bdev; - generic_make_request(tbio); + if (test_bit(FailFast, &conf->mirrors[d].rdev->flags)) + tbio->bi_opf |= MD_FAILFAST; + tbio->bi_iter.bi_sector += conf->mirrors[d].rdev->data_offset; + submit_bio_noacct(tbio); } /* Now write out to any replacement devices * that are active */ for (i = 0; i < conf->copies; i++) { - int j, d; - tbio = r10_bio->devs[i].repl_bio; if (!tbio || !tbio->bi_end_io) continue; if (r10_bio->devs[i].bio->bi_end_io != end_sync_write && r10_bio->devs[i].bio != fbio) - for (j = 0; j < vcnt; j++) - memcpy(page_address(tbio->bi_io_vec[j].bv_page), - page_address(fbio->bi_io_vec[j].bv_page), - PAGE_SIZE); - d = r10_bio->devs[i].devnum; + bio_copy_data(tbio, fbio); atomic_inc(&r10_bio->remaining); - md_sync_acct(conf->mirrors[d].replacement->bdev, - bio_sectors(tbio)); - generic_make_request(tbio); + submit_bio_noacct(tbio); } done: @@ -2198,6 +2484,7 @@ static void fix_recovery_read_error(struct r10bio *r10_bio) int idx = 0; int dr = r10_bio->devs[0].devnum; int dw = r10_bio->devs[1].devnum; + struct page **pages = get_resync_pages(bio)->pages; while (sectors) { int s = sectors; @@ -2209,20 +2496,20 @@ static void fix_recovery_read_error(struct r10bio *r10_bio) s = PAGE_SIZE >> 9; rdev = conf->mirrors[dr].rdev; - addr = r10_bio->devs[0].addr + sect, + addr = r10_bio->devs[0].addr + sect; ok = sync_page_io(rdev, addr, s << 9, - bio->bi_io_vec[idx].bv_page, - READ, false); + pages[idx], + REQ_OP_READ, false); if (ok) { rdev = conf->mirrors[dw].rdev; addr = r10_bio->devs[1].addr + sect; ok = sync_page_io(rdev, addr, s << 9, - bio->bi_io_vec[idx].bv_page, - WRITE, false); + pages[idx], + REQ_OP_WRITE, false); if (!ok) { set_bit(WriteErrorSeen, &rdev->flags); if (!test_and_set_bit(WantReplacement, @@ -2245,10 +2532,8 @@ static void fix_recovery_read_error(struct r10bio *r10_bio) ok = rdev_set_badblocks(rdev2, addr, s, 0); if (!ok) { /* just abort the recovery */ - printk(KERN_NOTICE - "md/raid10:%s: recovery aborted" - " due to read error\n", - mdname(mddev)); + pr_notice("md/raid10:%s: recovery aborted due to read error\n", + mdname(mddev)); conf->mirrors[dw].recovery_disabled = mddev->recovery_disabled; @@ -2269,11 +2554,22 @@ static void recovery_request_write(struct mddev *mddev, struct r10bio *r10_bio) { struct r10conf *conf = mddev->private; int d; - struct bio *wbio, *wbio2; + struct bio *wbio = r10_bio->devs[1].bio; + struct bio *wbio2 = r10_bio->devs[1].repl_bio; + + /* Need to test wbio2->bi_end_io before we call + * submit_bio_noacct as if the former is NULL, + * the latter is free to free wbio2. + */ + if (wbio2 && !wbio2->bi_end_io) + wbio2 = NULL; if (!test_bit(R10BIO_Uptodate, &r10_bio->state)) { fix_recovery_read_error(r10_bio); - end_sync_request(r10_bio); + if (wbio->bi_end_io) + end_sync_request(r10_bio); + if (wbio2) + end_sync_request(r10_bio); return; } @@ -2282,72 +2578,26 @@ static void recovery_request_write(struct mddev *mddev, struct r10bio *r10_bio) * and submit the write request */ d = r10_bio->devs[1].devnum; - wbio = r10_bio->devs[1].bio; - wbio2 = r10_bio->devs[1].repl_bio; if (wbio->bi_end_io) { atomic_inc(&conf->mirrors[d].rdev->nr_pending); - md_sync_acct(conf->mirrors[d].rdev->bdev, bio_sectors(wbio)); - generic_make_request(wbio); + submit_bio_noacct(wbio); } - if (wbio2 && wbio2->bi_end_io) { + if (wbio2) { atomic_inc(&conf->mirrors[d].replacement->nr_pending); - md_sync_acct(conf->mirrors[d].replacement->bdev, - bio_sectors(wbio2)); - generic_make_request(wbio2); + submit_bio_noacct(wbio2); } } - -/* - * Used by fix_read_error() to decay the per rdev read_errors. - * We halve the read error count for every hour that has elapsed - * since the last recorded read error. - * - */ -static void check_decay_read_errors(struct mddev *mddev, struct md_rdev *rdev) -{ - struct timespec cur_time_mon; - unsigned long hours_since_last; - unsigned int read_errors = atomic_read(&rdev->read_errors); - - ktime_get_ts(&cur_time_mon); - - if (rdev->last_read_error.tv_sec == 0 && - rdev->last_read_error.tv_nsec == 0) { - /* first time we've seen a read error */ - rdev->last_read_error = cur_time_mon; - return; - } - - hours_since_last = (cur_time_mon.tv_sec - - rdev->last_read_error.tv_sec) / 3600; - - rdev->last_read_error = cur_time_mon; - - /* - * if hours_since_last is > the number of bits in read_errors - * just set read errors to 0. We do this to avoid - * overflowing the shift of read_errors by hours_since_last. - */ - if (hours_since_last >= 8 * sizeof(read_errors)) - atomic_set(&rdev->read_errors, 0); - else - atomic_set(&rdev->read_errors, read_errors >> hours_since_last); -} - static int r10_sync_page_io(struct md_rdev *rdev, sector_t sector, - int sectors, struct page *page, int rw) + int sectors, struct page *page, enum req_op op) { - sector_t first_bad; - int bad_sectors; - - if (is_badblock(rdev, sector, sectors, &first_bad, &bad_sectors) - && (rw == READ || test_bit(WriteErrorSeen, &rdev->flags))) + if (rdev_has_badblock(rdev, sector, sectors) && + (op == REQ_OP_READ || test_bit(WriteErrorSeen, &rdev->flags))) return -1; - if (sync_page_io(rdev, sector, sectors << 9, page, rw, false)) + if (sync_page_io(rdev, sector, sectors << 9, page, op, false)) /* success */ return 1; - if (rw == WRITE) { + if (op == REQ_OP_WRITE) { set_bit(WriteErrorSeen, &rdev->flags); if (!test_and_set_bit(WantReplacement, &rdev->flags)) set_bit(MD_RECOVERY_NEEDED, @@ -2370,10 +2620,9 @@ static int r10_sync_page_io(struct md_rdev *rdev, sector_t sector, static void fix_read_error(struct r10conf *conf, struct mddev *mddev, struct r10bio *r10_bio) { int sect = 0; /* Offset from r10_bio->sector */ - int sectors = r10_bio->sectors; - struct md_rdev*rdev; - int max_read_errors = atomic_read(&mddev->max_corr_read_errors); - int d = r10_bio->devs[r10_bio->read_slot].devnum; + int sectors = r10_bio->sectors, slot = r10_bio->read_slot; + struct md_rdev *rdev; + int d = r10_bio->devs[slot].devnum; /* still own a reference to this rdev, so it cannot * have been cleared recently. @@ -2385,79 +2634,60 @@ static void fix_read_error(struct r10conf *conf, struct mddev *mddev, struct r10 more fix_read_error() attempts */ return; - check_decay_read_errors(mddev, rdev); - atomic_inc(&rdev->read_errors); - if (atomic_read(&rdev->read_errors) > max_read_errors) { - char b[BDEVNAME_SIZE]; - bdevname(rdev->bdev, b); - - printk(KERN_NOTICE - "md/raid10:%s: %s: Raid device exceeded " - "read_error threshold [cur %d:max %d]\n", - mdname(mddev), b, - atomic_read(&rdev->read_errors), max_read_errors); - printk(KERN_NOTICE - "md/raid10:%s: %s: Failing raid device\n", - mdname(mddev), b); - md_error(mddev, conf->mirrors[d].rdev); - r10_bio->devs[r10_bio->read_slot].bio = IO_BLOCKED; + if (exceed_read_errors(mddev, rdev)) { + r10_bio->devs[slot].bio = IO_BLOCKED; return; } while(sectors) { int s = sectors; - int sl = r10_bio->read_slot; + int sl = slot; int success = 0; int start; if (s > (PAGE_SIZE>>9)) s = PAGE_SIZE >> 9; - rcu_read_lock(); do { - sector_t first_bad; - int bad_sectors; - d = r10_bio->devs[sl].devnum; - rdev = rcu_dereference(conf->mirrors[d].rdev); + rdev = conf->mirrors[d].rdev; if (rdev && - !test_bit(Unmerged, &rdev->flags) && test_bit(In_sync, &rdev->flags) && - is_badblock(rdev, r10_bio->devs[sl].addr + sect, s, - &first_bad, &bad_sectors) == 0) { + !test_bit(Faulty, &rdev->flags) && + rdev_has_badblock(rdev, + r10_bio->devs[sl].addr + sect, + s) == 0) { atomic_inc(&rdev->nr_pending); - rcu_read_unlock(); success = sync_page_io(rdev, r10_bio->devs[sl].addr + sect, s<<9, - conf->tmppage, READ, false); + conf->tmppage, + REQ_OP_READ, false); rdev_dec_pending(rdev, mddev); - rcu_read_lock(); if (success) break; } sl++; if (sl == conf->copies) sl = 0; - } while (!success && sl != r10_bio->read_slot); - rcu_read_unlock(); + } while (sl != slot); if (!success) { /* Cannot read from anywhere, just mark the block * as bad on the first device to discourage future * reads. */ - int dn = r10_bio->devs[r10_bio->read_slot].devnum; + int dn = r10_bio->devs[slot].devnum; rdev = conf->mirrors[dn].rdev; if (!rdev_set_badblocks( rdev, - r10_bio->devs[r10_bio->read_slot].addr + r10_bio->devs[slot].addr + sect, s, 0)) { md_error(mddev, rdev); - r10_bio->devs[r10_bio->read_slot].bio + r10_bio->devs[slot].bio = IO_BLOCKED; } break; @@ -2465,105 +2695,85 @@ static void fix_read_error(struct r10conf *conf, struct mddev *mddev, struct r10 start = sl; /* write it back and re-read */ - rcu_read_lock(); - while (sl != r10_bio->read_slot) { - char b[BDEVNAME_SIZE]; - + while (sl != slot) { if (sl==0) sl = conf->copies; sl--; d = r10_bio->devs[sl].devnum; - rdev = rcu_dereference(conf->mirrors[d].rdev); + rdev = conf->mirrors[d].rdev; if (!rdev || - test_bit(Unmerged, &rdev->flags) || + test_bit(Faulty, &rdev->flags) || !test_bit(In_sync, &rdev->flags)) continue; atomic_inc(&rdev->nr_pending); - rcu_read_unlock(); if (r10_sync_page_io(rdev, r10_bio->devs[sl].addr + sect, - s, conf->tmppage, WRITE) + s, conf->tmppage, REQ_OP_WRITE) == 0) { /* Well, this device is dead */ - printk(KERN_NOTICE - "md/raid10:%s: read correction " - "write failed" - " (%d sectors at %llu on %s)\n", - mdname(mddev), s, - (unsigned long long)( - sect + - choose_data_offset(r10_bio, - rdev)), - bdevname(rdev->bdev, b)); - printk(KERN_NOTICE "md/raid10:%s: %s: failing " - "drive\n", - mdname(mddev), - bdevname(rdev->bdev, b)); + pr_notice("md/raid10:%s: read correction write failed (%d sectors at %llu on %pg)\n", + mdname(mddev), s, + (unsigned long long)( + sect + + choose_data_offset(r10_bio, + rdev)), + rdev->bdev); + pr_notice("md/raid10:%s: %pg: failing drive\n", + mdname(mddev), + rdev->bdev); } rdev_dec_pending(rdev, mddev); - rcu_read_lock(); } sl = start; - while (sl != r10_bio->read_slot) { - char b[BDEVNAME_SIZE]; - + while (sl != slot) { if (sl==0) sl = conf->copies; sl--; d = r10_bio->devs[sl].devnum; - rdev = rcu_dereference(conf->mirrors[d].rdev); + rdev = conf->mirrors[d].rdev; if (!rdev || + test_bit(Faulty, &rdev->flags) || !test_bit(In_sync, &rdev->flags)) continue; atomic_inc(&rdev->nr_pending); - rcu_read_unlock(); switch (r10_sync_page_io(rdev, r10_bio->devs[sl].addr + sect, - s, conf->tmppage, - READ)) { + s, conf->tmppage, REQ_OP_READ)) { case 0: /* Well, this device is dead */ - printk(KERN_NOTICE - "md/raid10:%s: unable to read back " - "corrected sectors" - " (%d sectors at %llu on %s)\n", + pr_notice("md/raid10:%s: unable to read back corrected sectors (%d sectors at %llu on %pg)\n", mdname(mddev), s, (unsigned long long)( sect + choose_data_offset(r10_bio, rdev)), - bdevname(rdev->bdev, b)); - printk(KERN_NOTICE "md/raid10:%s: %s: failing " - "drive\n", + rdev->bdev); + pr_notice("md/raid10:%s: %pg: failing drive\n", mdname(mddev), - bdevname(rdev->bdev, b)); + rdev->bdev); break; case 1: - printk(KERN_INFO - "md/raid10:%s: read error corrected" - " (%d sectors at %llu on %s)\n", + pr_info("md/raid10:%s: read error corrected (%d sectors at %llu on %pg)\n", mdname(mddev), s, (unsigned long long)( sect + choose_data_offset(r10_bio, rdev)), - bdevname(rdev->bdev, b)); + rdev->bdev); atomic_add(s, &rdev->corrected_errors); } rdev_dec_pending(rdev, mddev); - rcu_read_lock(); } - rcu_read_unlock(); sectors -= s; sect += s; } } -static int narrow_write_error(struct r10bio *r10_bio, int i) +static bool narrow_write_error(struct r10bio *r10_bio, int i) { struct bio *bio = r10_bio->master_bio; struct mddev *mddev = r10_bio->mddev; @@ -2584,12 +2794,13 @@ static int narrow_write_error(struct r10bio *r10_bio, int i) sector_t sector; int sectors; int sect_to_write = r10_bio->sectors; - int ok = 1; + bool ok = true; if (rdev->badblocks.shift < 0) - return 0; + return false; - block_sectors = 1 << rdev->badblocks.shift; + block_sectors = roundup(1 << rdev->badblocks.shift, + bdev_logical_block_size(rdev->bdev) >> 9); sector = r10_bio->sector; sectors = ((r10_bio->sector + block_sectors) & ~(sector_t)(block_sectors - 1)) @@ -2597,18 +2808,21 @@ static int narrow_write_error(struct r10bio *r10_bio, int i) while (sect_to_write) { struct bio *wbio; + sector_t wsector; if (sectors > sect_to_write) sectors = sect_to_write; /* Write at 'sector' for 'sectors' */ - wbio = bio_clone_mddev(bio, GFP_NOIO, mddev); - md_trim_bio(wbio, sector - bio->bi_sector, sectors); - wbio->bi_sector = (r10_bio->devs[i].addr+ - choose_data_offset(r10_bio, rdev) + - (sector - r10_bio->sector)); - wbio->bi_bdev = rdev->bdev; - if (submit_bio_wait(WRITE, wbio) == 0) + wbio = bio_alloc_clone(rdev->bdev, bio, GFP_NOIO, + &mddev->bio_set); + bio_trim(wbio, sector - bio->bi_iter.bi_sector, sectors); + wsector = r10_bio->devs[i].addr + (sector - r10_bio->sector); + wbio->bi_iter.bi_sector = wsector + + choose_data_offset(r10_bio, rdev); + wbio->bi_opf = REQ_OP_WRITE; + + if (submit_bio_wait(wbio) < 0) /* Failure! */ - ok = rdev_set_badblocks(rdev, sector, + ok = rdev_set_badblocks(rdev, wsector, sectors, 0) && ok; @@ -2626,9 +2840,6 @@ static void handle_read_error(struct mddev *mddev, struct r10bio *r10_bio) struct bio *bio; struct r10conf *conf = mddev->private; struct md_rdev *rdev = r10_bio->devs[slot].rdev; - char b[BDEVNAME_SIZE]; - unsigned long do_sync; - int max_sectors; /* we got a read error. Maybe the drive is bad. Maybe just * the block and we can fix it. @@ -2639,81 +2850,26 @@ static void handle_read_error(struct mddev *mddev, struct r10bio *r10_bio) * frozen. */ bio = r10_bio->devs[slot].bio; - bdevname(bio->bi_bdev, b); bio_put(bio); r10_bio->devs[slot].bio = NULL; - if (mddev->ro == 0) { + if (mddev->ro) + r10_bio->devs[slot].bio = IO_BLOCKED; + else if (!test_bit(FailFast, &rdev->flags)) { freeze_array(conf, 1); fix_read_error(conf, mddev, r10_bio); unfreeze_array(conf); } else - r10_bio->devs[slot].bio = IO_BLOCKED; + md_error(mddev, rdev); rdev_dec_pending(rdev, mddev); - -read_more: - rdev = read_balance(conf, r10_bio, &max_sectors); - if (rdev == NULL) { - printk(KERN_ALERT "md/raid10:%s: %s: unrecoverable I/O" - " read error for block %llu\n", - mdname(mddev), b, - (unsigned long long)r10_bio->sector); - raid_end_bio_io(r10_bio); - return; - } - - do_sync = (r10_bio->master_bio->bi_rw & REQ_SYNC); - slot = r10_bio->read_slot; - printk_ratelimited( - KERN_ERR - "md/raid10:%s: %s: redirecting " - "sector %llu to another mirror\n", - mdname(mddev), - bdevname(rdev->bdev, b), - (unsigned long long)r10_bio->sector); - bio = bio_clone_mddev(r10_bio->master_bio, - GFP_NOIO, mddev); - md_trim_bio(bio, - r10_bio->sector - bio->bi_sector, - max_sectors); - r10_bio->devs[slot].bio = bio; - r10_bio->devs[slot].rdev = rdev; - bio->bi_sector = r10_bio->devs[slot].addr - + choose_data_offset(r10_bio, rdev); - bio->bi_bdev = rdev->bdev; - bio->bi_rw = READ | do_sync; - bio->bi_private = r10_bio; - bio->bi_end_io = raid10_end_read_request; - if (max_sectors < r10_bio->sectors) { - /* Drat - have to split this up more */ - struct bio *mbio = r10_bio->master_bio; - int sectors_handled = - r10_bio->sector + max_sectors - - mbio->bi_sector; - r10_bio->sectors = max_sectors; - spin_lock_irq(&conf->device_lock); - if (mbio->bi_phys_segments == 0) - mbio->bi_phys_segments = 2; - else - mbio->bi_phys_segments++; - spin_unlock_irq(&conf->device_lock); - generic_make_request(bio); - - r10_bio = mempool_alloc(conf->r10bio_pool, - GFP_NOIO); - r10_bio->master_bio = mbio; - r10_bio->sectors = bio_sectors(mbio) - sectors_handled; - r10_bio->state = 0; - set_bit(R10BIO_ReadError, - &r10_bio->state); - r10_bio->mddev = mddev; - r10_bio->sector = mbio->bi_sector - + sectors_handled; - - goto read_more; - } else - generic_make_request(bio); + r10_bio->state = 0; + raid10_read_request(mddev, r10_bio->master_bio, r10_bio, false); + /* + * allow_barrier after re-submit to ensure no sync io + * can be issued while regular io pending. + */ + allow_barrier(conf); } static void handle_write_completed(struct r10conf *conf, struct r10bio *r10_bio) @@ -2732,10 +2888,10 @@ static void handle_write_completed(struct r10conf *conf, struct r10bio *r10_bio) for (m = 0; m < conf->copies; m++) { int dev = r10_bio->devs[m].devnum; rdev = conf->mirrors[dev].rdev; - if (r10_bio->devs[m].bio == NULL) + if (r10_bio->devs[m].bio == NULL || + r10_bio->devs[m].bio->bi_end_io == NULL) continue; - if (test_bit(BIO_UPTODATE, - &r10_bio->devs[m].bio->bi_flags)) { + if (!r10_bio->devs[m].bio->bi_status) { rdev_clear_badblocks( rdev, r10_bio->devs[m].addr, @@ -2748,10 +2904,11 @@ static void handle_write_completed(struct r10conf *conf, struct r10bio *r10_bio) md_error(conf->mddev, rdev); } rdev = conf->mirrors[dev].replacement; - if (r10_bio->devs[m].repl_bio == NULL) + if (r10_bio->devs[m].repl_bio == NULL || + r10_bio->devs[m].repl_bio->bi_end_io == NULL) continue; - if (test_bit(BIO_UPTODATE, - &r10_bio->devs[m].repl_bio->bi_flags)) { + + if (!r10_bio->devs[m].repl_bio->bi_status) { rdev_clear_badblocks( rdev, r10_bio->devs[m].addr, @@ -2766,6 +2923,7 @@ static void handle_write_completed(struct r10conf *conf, struct r10bio *r10_bio) } put_buf(r10_bio); } else { + bool fail = false; for (m = 0; m < conf->copies; m++) { int dev = r10_bio->devs[m].devnum; struct bio *bio = r10_bio->devs[m].bio; @@ -2776,13 +2934,10 @@ static void handle_write_completed(struct r10conf *conf, struct r10bio *r10_bio) r10_bio->devs[m].addr, r10_bio->sectors, 0); rdev_dec_pending(rdev, conf->mddev); - } else if (bio != NULL && - !test_bit(BIO_UPTODATE, &bio->bi_flags)) { - if (!narrow_write_error(r10_bio, m)) { + } else if (bio != NULL && bio->bi_status) { + fail = true; + if (!narrow_write_error(r10_bio, m)) md_error(conf->mddev, rdev); - set_bit(R10BIO_Degraded, - &r10_bio->state); - } rdev_dec_pending(rdev, conf->mddev); } bio = r10_bio->devs[m].repl_bio; @@ -2795,10 +2950,23 @@ static void handle_write_completed(struct r10conf *conf, struct r10bio *r10_bio) rdev_dec_pending(rdev, conf->mddev); } } - if (test_bit(R10BIO_WriteError, - &r10_bio->state)) - close_write(r10_bio); - raid_end_bio_io(r10_bio); + if (fail) { + spin_lock_irq(&conf->device_lock); + list_add(&r10_bio->retry_list, &conf->bio_end_io_list); + conf->nr_queued++; + spin_unlock_irq(&conf->device_lock); + /* + * In case freeze_array() is waiting for condition + * nr_pending == nr_queued + extra to be true. + */ + wake_up(&conf->wait_barrier); + md_wakeup_thread(conf->mddev->thread); + } else { + if (test_bit(R10BIO_WriteError, + &r10_bio->state)) + close_write(r10_bio); + raid_end_bio_io(r10_bio); + } } } @@ -2813,6 +2981,29 @@ static void raid10d(struct md_thread *thread) md_check_recovery(mddev); + if (!list_empty_careful(&conf->bio_end_io_list) && + !test_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags)) { + LIST_HEAD(tmp); + spin_lock_irqsave(&conf->device_lock, flags); + if (!test_bit(MD_SB_CHANGE_PENDING, &mddev->sb_flags)) { + while (!list_empty(&conf->bio_end_io_list)) { + list_move(conf->bio_end_io_list.prev, &tmp); + conf->nr_queued--; + } + } + spin_unlock_irqrestore(&conf->device_lock, flags); + while (!list_empty(&tmp)) { + r10_bio = list_first_entry(&tmp, struct r10bio, + retry_list); + list_del(&r10_bio->retry_list); + + if (test_bit(R10BIO_WriteError, + &r10_bio->state)) + close_write(r10_bio); + raid_end_bio_io(r10_bio); + } + } + blk_start_plug(&plug); for (;;) { @@ -2841,40 +3032,100 @@ static void raid10d(struct md_thread *thread) recovery_request_write(mddev, r10_bio); else if (test_bit(R10BIO_ReadError, &r10_bio->state)) handle_read_error(mddev, r10_bio); - else { - /* just a partial read to be scheduled from a - * separate context - */ - int slot = r10_bio->read_slot; - generic_make_request(r10_bio->devs[slot].bio); - } + else + WARN_ON_ONCE(1); cond_resched(); - if (mddev->flags & ~(1<<MD_CHANGE_PENDING)) + if (mddev->sb_flags & ~(1<<MD_SB_CHANGE_PENDING)) md_check_recovery(mddev); } blk_finish_plug(&plug); } - static int init_resync(struct r10conf *conf) { - int buffs; - int i; + int ret, buffs, i; buffs = RESYNC_WINDOW / RESYNC_BLOCK_SIZE; - BUG_ON(conf->r10buf_pool); + BUG_ON(mempool_initialized(&conf->r10buf_pool)); conf->have_replacement = 0; for (i = 0; i < conf->geo.raid_disks; i++) if (conf->mirrors[i].replacement) conf->have_replacement = 1; - conf->r10buf_pool = mempool_create(buffs, r10buf_pool_alloc, r10buf_pool_free, conf); - if (!conf->r10buf_pool) - return -ENOMEM; + ret = mempool_init(&conf->r10buf_pool, buffs, + r10buf_pool_alloc, r10buf_pool_free, conf); + if (ret) + return ret; conf->next_resync = 0; return 0; } +static struct r10bio *raid10_alloc_init_r10buf(struct r10conf *conf) +{ + struct r10bio *r10bio = mempool_alloc(&conf->r10buf_pool, GFP_NOIO); + struct rsync_pages *rp; + struct bio *bio; + int nalloc; + int i; + + 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 */ + + for (i = 0; i < nalloc; i++) { + bio = r10bio->devs[i].bio; + rp = bio->bi_private; + bio_reset(bio, NULL, 0); + bio->bi_private = rp; + bio = r10bio->devs[i].repl_bio; + if (bio) { + rp = bio->bi_private; + bio_reset(bio, NULL, 0); + bio->bi_private = rp; + } + } + return r10bio; +} + +/* + * Set cluster_sync_high since we need other nodes to add the + * range [cluster_sync_low, cluster_sync_high] to suspend list. + */ +static void raid10_set_cluster_sync_high(struct r10conf *conf) +{ + sector_t window_size; + int extra_chunk, chunks; + + /* + * First, here we define "stripe" as a unit which across + * all member devices one time, so we get chunks by use + * raid_disks / near_copies. Otherwise, if near_copies is + * close to raid_disks, then resync window could increases + * linearly with the increase of raid_disks, which means + * we will suspend a really large IO window while it is not + * necessary. If raid_disks is not divisible by near_copies, + * an extra chunk is needed to ensure the whole "stripe" is + * covered. + */ + + chunks = conf->geo.raid_disks / conf->geo.near_copies; + if (conf->geo.raid_disks % conf->geo.near_copies == 0) + extra_chunk = 0; + else + extra_chunk = 1; + window_size = (chunks + extra_chunk) * conf->mddev->chunk_sectors; + + /* + * At least use a 32M window to align with raid1's resync window + */ + window_size = (CLUSTER_RESYNC_WINDOW_SECTORS > window_size) ? + CLUSTER_RESYNC_WINDOW_SECTORS : window_size; + + conf->cluster_sync_high = conf->cluster_sync_low + window_size; +} + /* * perform a "sync" on one "block" * @@ -2898,7 +3149,7 @@ static int init_resync(struct r10conf *conf) * a number of r10_bio structures, one for each out-of-sync device. * As we setup these structures, we collect all bio's together into a list * which we then process collectively to add pages, and then process again - * to pass to generic_make_request. + * to pass to submit_bio_noacct. * * The r10_bio structures are linked using a borrowed master_bio pointer. * This link is counted in ->remaining. When the r10_bio that points to NULL @@ -2907,30 +3158,28 @@ static int init_resync(struct r10conf *conf) * */ -static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, - int *skipped, int go_faster) +static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr, + sector_t max_sector, int *skipped) { struct r10conf *conf = mddev->private; struct r10bio *r10_bio; struct bio *biolist = NULL, *bio; - sector_t max_sector, nr_sectors; + sector_t nr_sectors; int i; int max_sync; sector_t sync_blocks; sector_t sectors_skipped = 0; int chunks_skipped = 0; sector_t chunk_mask = conf->geo.chunk_mask; - - if (!conf->r10buf_pool) - if (init_resync(conf)) - return 0; + int page_idx = 0; + int error_disk = -1; /* * Allow skipping a full rebuild for incremental assembly * of a clean array, like RAID1 does. */ if (mddev->bitmap == NULL && - mddev->recovery_cp == MaxSector && + mddev->resync_offset == MaxSector && mddev->reshape_position == MaxSector && !test_bit(MD_RECOVERY_SYNC, &mddev->recovery) && !test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery) && @@ -2940,12 +3189,15 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, return mddev->dev_sectors - sector_nr; } + if (!mempool_initialized(&conf->r10buf_pool)) + if (init_resync(conf)) + return 0; + skipped: - max_sector = mddev->dev_sectors; - if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery) || - test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery)) - max_sector = mddev->resync_max_sectors; if (sector_nr >= max_sector) { + conf->cluster_sync_low = 0; + conf->cluster_sync_high = 0; + /* If we aborted, we need to abort the * sync on the 'current' bitmap chucks (there can * be several when recovering multiple devices). @@ -2957,18 +3209,19 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, */ if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery)) { end_reshape(conf); + close_sync(conf); return 0; } if (mddev->curr_resync < max_sector) { /* aborted */ if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) - bitmap_end_sync(mddev->bitmap, mddev->curr_resync, - &sync_blocks, 1); + md_bitmap_end_sync(mddev, mddev->curr_resync, + &sync_blocks); else for (i = 0; i < conf->geo.raid_disks; i++) { sector_t sect = raid10_find_virt(conf, mddev->curr_resync, i); - bitmap_end_sync(mddev->bitmap, sect, - &sync_blocks, 1); + + md_bitmap_end_sync(mddev, sect, &sync_blocks); } } else { /* completed sync */ @@ -2978,15 +3231,18 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, /* Completed a full sync so the replacements * are now fully recovered. */ - for (i = 0; i < conf->geo.raid_disks; i++) - if (conf->mirrors[i].replacement) - conf->mirrors[i].replacement - ->recovery_offset - = MaxSector; + for (i = 0; i < conf->geo.raid_disks; i++) { + struct md_rdev *rdev = + conf->mirrors[i].replacement; + + if (rdev) + rdev->recovery_offset = MaxSector; + } } conf->fullsync = 0; } - bitmap_close_sync(mddev->bitmap); + if (md_bitmap_enabled(mddev, false)) + mddev->bitmap_ops->close_sync(mddev); close_sync(conf); *skipped = 1; return sectors_skipped; @@ -2996,8 +3252,21 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, return reshape_request(mddev, sector_nr, skipped); if (chunks_skipped >= conf->geo.raid_disks) { - /* if there has been nothing to do on any drive, - * then there is nothing to do at all.. + pr_err("md/raid10:%s: %s fails\n", mdname(mddev), + test_bit(MD_RECOVERY_SYNC, &mddev->recovery) ? "resync" : "recovery"); + if (error_disk >= 0 && + !test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) { + /* + * recovery fails, set mirrors.recovery_disabled, + * device shouldn't be added to there. + */ + conf->mirrors[error_disk].recovery_disabled = + mddev->recovery_disabled; + return 0; + } + /* + * if there has been nothing to do on any drive, + * then there is nothing to do at all. */ *skipped = 1; return (max_sector - sector_nr) + sectors_skipped; @@ -3012,12 +3281,13 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, if (conf->geo.near_copies < conf->geo.raid_disks && max_sector > (sector_nr | chunk_mask)) max_sector = (sector_nr | chunk_mask) + 1; + /* - * If there is non-resync activity waiting for us then - * put in a delay to throttle resync. + * If there is non-resync activity waiting for a turn, then let it + * though before starting on this new sync request. */ - if (!go_faster && conf->nr_waiting) - msleep_interruptible(1000); + if (conf->nr_waiting) + schedule_timeout_uninterruptible(1); /* Again, very different code for resync and recovery. * Both must result in an r10bio with a list of bios that @@ -3041,41 +3311,45 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, r10_bio = NULL; for (i = 0 ; i < conf->geo.raid_disks; i++) { - int still_degraded; + bool still_degraded; struct r10bio *rb2; sector_t sect; - int must_sync; + bool must_sync; int any_working; struct raid10_info *mirror = &conf->mirrors[i]; + struct md_rdev *mrdev, *mreplace; + + mrdev = mirror->rdev; + mreplace = mirror->replacement; - if ((mirror->rdev == NULL || - test_bit(In_sync, &mirror->rdev->flags)) - && - (mirror->replacement == NULL || - test_bit(Faulty, - &mirror->replacement->flags))) + if (mrdev && (test_bit(Faulty, &mrdev->flags) || + test_bit(In_sync, &mrdev->flags))) + mrdev = NULL; + if (mreplace && test_bit(Faulty, &mreplace->flags)) + mreplace = NULL; + + if (!mrdev && !mreplace) continue; - still_degraded = 0; + still_degraded = false; /* want to reconstruct this device */ rb2 = r10_bio; sect = raid10_find_virt(conf, sector_nr, i); - if (sect >= mddev->resync_max_sectors) { + if (sect >= mddev->resync_max_sectors) /* last stripe is not complete - don't * try to recover this sector. */ continue; - } /* Unless we are doing a full sync, or a replacement * we only need to recover the block if it is set in * the bitmap */ - must_sync = bitmap_start_sync(mddev->bitmap, sect, - &sync_blocks, 1); + must_sync = md_bitmap_start_sync(mddev, sect, + &sync_blocks, true); if (sync_blocks < max_sync) max_sync = sync_blocks; if (!must_sync && - mirror->replacement == NULL && + mreplace == NULL && !conf->fullsync) { /* yep, skip the sync_blocks here, but don't assume * that there will never be anything to do here @@ -3083,8 +3357,13 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, chunks_skipped = -1; continue; } + if (mrdev) + atomic_inc(&mrdev->nr_pending); + if (mreplace) + atomic_inc(&mreplace->nr_pending); - r10_bio = mempool_alloc(conf->r10buf_pool, GFP_NOIO); + r10_bio = raid10_alloc_init_r10buf(conf); + r10_bio->state = 0; raise_barrier(conf, rb2 != NULL); atomic_set(&r10_bio->remaining, 0); @@ -3100,30 +3379,30 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, /* Need to check if the array will still be * degraded */ - for (j = 0; j < conf->geo.raid_disks; j++) - if (conf->mirrors[j].rdev == NULL || - test_bit(Faulty, &conf->mirrors[j].rdev->flags)) { - still_degraded = 1; + for (j = 0; j < conf->geo.raid_disks; j++) { + struct md_rdev *rdev = conf->mirrors[j].rdev; + + if (rdev == NULL || test_bit(Faulty, &rdev->flags)) { + still_degraded = false; break; } + } - must_sync = bitmap_start_sync(mddev->bitmap, sect, - &sync_blocks, still_degraded); - + md_bitmap_start_sync(mddev, sect, &sync_blocks, + still_degraded); any_working = 0; for (j=0; j<conf->copies;j++) { int k; int d = r10_bio->devs[j].devnum; sector_t from_addr, to_addr; - struct md_rdev *rdev; + struct md_rdev *rdev = conf->mirrors[d].rdev; sector_t sector, first_bad; - int bad_sectors; - if (!conf->mirrors[d].rdev || - !test_bit(In_sync, &conf->mirrors[d].rdev->flags)) + sector_t bad_sectors; + if (!rdev || + !test_bit(In_sync, &rdev->flags)) continue; /* This is where we read from */ any_working = 1; - rdev = conf->mirrors[d].rdev; sector = r10_bio->devs[j].addr; if (is_badblock(rdev, sector, max_sync, @@ -3139,15 +3418,16 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, } } bio = r10_bio->devs[0].bio; - bio_reset(bio); bio->bi_next = biolist; biolist = bio; - bio->bi_private = r10_bio; bio->bi_end_io = end_sync_read; - bio->bi_rw = READ; + bio->bi_opf = REQ_OP_READ; + if (test_bit(FailFast, &rdev->flags)) + bio->bi_opf |= MD_FAILFAST; from_addr = r10_bio->devs[j].addr; - bio->bi_sector = from_addr + rdev->data_offset; - bio->bi_bdev = rdev->bdev; + bio->bi_iter.bi_sector = from_addr + + rdev->data_offset; + bio_set_dev(bio, rdev->bdev); atomic_inc(&rdev->nr_pending); /* and we write to 'i' (if not in_sync) */ @@ -3161,18 +3441,15 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, r10_bio->devs[1].devnum = i; r10_bio->devs[1].addr = to_addr; - rdev = mirror->rdev; - if (!test_bit(In_sync, &rdev->flags)) { + if (mrdev) { bio = r10_bio->devs[1].bio; - bio_reset(bio); bio->bi_next = biolist; biolist = bio; - bio->bi_private = r10_bio; bio->bi_end_io = end_sync_write; - bio->bi_rw = WRITE; - bio->bi_sector = to_addr - + rdev->data_offset; - bio->bi_bdev = rdev->bdev; + bio->bi_opf = REQ_OP_WRITE; + bio->bi_iter.bi_sector = to_addr + + mrdev->data_offset; + bio_set_dev(bio, mrdev->bdev); atomic_inc(&r10_bio->remaining); } else r10_bio->devs[1].bio->bi_end_io = NULL; @@ -3181,36 +3458,25 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, bio = r10_bio->devs[1].repl_bio; if (bio) bio->bi_end_io = NULL; - rdev = mirror->replacement; - /* Note: if rdev != NULL, then bio + /* Note: if replace is not NULL, then bio * cannot be NULL as r10buf_pool_alloc will * have allocated it. - * So the second test here is pointless. - * But it keeps semantic-checkers happy, and - * this comment keeps human reviewers - * happy. */ - if (rdev == NULL || bio == NULL || - test_bit(Faulty, &rdev->flags)) + if (!mreplace) break; - bio_reset(bio); bio->bi_next = biolist; biolist = bio; - bio->bi_private = r10_bio; bio->bi_end_io = end_sync_write; - bio->bi_rw = WRITE; - bio->bi_sector = to_addr + rdev->data_offset; - bio->bi_bdev = rdev->bdev; + bio->bi_opf = REQ_OP_WRITE; + bio->bi_iter.bi_sector = to_addr + + mreplace->data_offset; + bio_set_dev(bio, mreplace->bdev); atomic_inc(&r10_bio->remaining); break; } if (j == conf->copies) { /* Cannot recover, so abort the recovery or * record a bad block */ - put_buf(r10_bio); - if (rb2) - atomic_dec(&rb2->remaining); - r10_bio = rb2; if (any_working) { /* problem is that there are bad blocks * on other device(s) @@ -3219,16 +3485,16 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, for (k = 0; k < conf->copies; k++) if (r10_bio->devs[k].devnum == i) break; - if (!test_bit(In_sync, - &mirror->rdev->flags) + if (mrdev && !test_bit(In_sync, + &mrdev->flags) && !rdev_set_badblocks( - mirror->rdev, + mrdev, r10_bio->devs[k].addr, max_sync, 0)) any_working = 0; - if (mirror->replacement && + if (mreplace && !rdev_set_badblocks( - mirror->replacement, + mreplace, r10_bio->devs[k].addr, max_sync, 0)) any_working = 0; @@ -3236,14 +3502,44 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, if (!any_working) { if (!test_and_set_bit(MD_RECOVERY_INTR, &mddev->recovery)) - printk(KERN_INFO "md/raid10:%s: insufficient " - "working devices for recovery.\n", + pr_warn("md/raid10:%s: insufficient working devices for recovery.\n", mdname(mddev)); mirror->recovery_disabled = mddev->recovery_disabled; + } else { + error_disk = i; } + put_buf(r10_bio); + if (rb2) + atomic_dec(&rb2->remaining); + r10_bio = rb2; + if (mrdev) + rdev_dec_pending(mrdev, mddev); + if (mreplace) + rdev_dec_pending(mreplace, mddev); break; } + if (mrdev) + rdev_dec_pending(mrdev, mddev); + if (mreplace) + rdev_dec_pending(mreplace, mddev); + if (r10_bio->devs[0].bio->bi_opf & MD_FAILFAST) { + /* Only want this if there is elsewhere to + * read from. 'j' is currently the first + * readable copy. + */ + int targets = 1; + for (; j < conf->copies; j++) { + int d = r10_bio->devs[j].devnum; + if (conf->mirrors[d].rdev && + test_bit(In_sync, + &conf->mirrors[d].rdev->flags)) + targets++; + } + if (targets == 1) + r10_bio->devs[0].bio->bi_opf + &= ~MD_FAILFAST; + } } if (biolist == NULL) { while (r10_bio) { @@ -3258,10 +3554,20 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, /* resync. Schedule a read for every block at this virt offset */ int count = 0; - bitmap_cond_end_sync(mddev->bitmap, sector_nr); + /* + * Since curr_resync_completed could probably not update in + * time, and we will set cluster_sync_low based on it. + * Let's check against "sector_nr + 2 * RESYNC_SECTORS" for + * safety reason, which ensures curr_resync_completed is + * updated in bitmap_cond_end_sync. + */ + if (md_bitmap_enabled(mddev, false)) + mddev->bitmap_ops->cond_end_sync(mddev, sector_nr, + mddev_is_clustered(mddev) && + (sector_nr + 2 * RESYNC_SECTORS > conf->cluster_sync_high)); - if (!bitmap_start_sync(mddev->bitmap, sector_nr, - &sync_blocks, mddev->degraded) && + if (!md_bitmap_start_sync(mddev, sector_nr, &sync_blocks, + mddev->degraded) && !conf->fullsync && !test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery)) { /* We can skip this block */ @@ -3270,7 +3576,8 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, } if (sync_blocks < max_sync) max_sync = sync_blocks; - r10_bio = mempool_alloc(conf->r10buf_pool, GFP_NOIO); + r10_bio = raid10_alloc_init_r10buf(conf); + r10_bio->state = 0; r10_bio->mddev = mddev; atomic_set(&r10_bio->remaining, 0); @@ -3286,20 +3593,20 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, for (i = 0; i < conf->copies; i++) { int d = r10_bio->devs[i].devnum; sector_t first_bad, sector; - int bad_sectors; + sector_t bad_sectors; + struct md_rdev *rdev; if (r10_bio->devs[i].repl_bio) r10_bio->devs[i].repl_bio->bi_end_io = NULL; bio = r10_bio->devs[i].bio; - bio_reset(bio); - clear_bit(BIO_UPTODATE, &bio->bi_flags); - if (conf->mirrors[d].rdev == NULL || - test_bit(Faulty, &conf->mirrors[d].rdev->flags)) + bio->bi_status = BLK_STS_IOERR; + rdev = conf->mirrors[d].rdev; + if (rdev == NULL || test_bit(Faulty, &rdev->flags)) continue; + sector = r10_bio->devs[i].addr; - if (is_badblock(conf->mirrors[d].rdev, - sector, max_sync, + if (is_badblock(rdev, sector, max_sync, &first_bad, &bad_sectors)) { if (first_bad > sector) max_sync = first_bad - sector; @@ -3310,38 +3617,37 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, continue; } } - atomic_inc(&conf->mirrors[d].rdev->nr_pending); + atomic_inc(&rdev->nr_pending); atomic_inc(&r10_bio->remaining); bio->bi_next = biolist; biolist = bio; - bio->bi_private = r10_bio; bio->bi_end_io = end_sync_read; - bio->bi_rw = READ; - bio->bi_sector = sector + - conf->mirrors[d].rdev->data_offset; - bio->bi_bdev = conf->mirrors[d].rdev->bdev; + bio->bi_opf = REQ_OP_READ; + if (test_bit(FailFast, &rdev->flags)) + bio->bi_opf |= MD_FAILFAST; + bio->bi_iter.bi_sector = sector + rdev->data_offset; + bio_set_dev(bio, rdev->bdev); count++; - if (conf->mirrors[d].replacement == NULL || - test_bit(Faulty, - &conf->mirrors[d].replacement->flags)) + rdev = conf->mirrors[d].replacement; + if (rdev == NULL || test_bit(Faulty, &rdev->flags)) continue; + atomic_inc(&rdev->nr_pending); + /* Need to set up for writing to the replacement */ bio = r10_bio->devs[i].repl_bio; - bio_reset(bio); - clear_bit(BIO_UPTODATE, &bio->bi_flags); + bio->bi_status = BLK_STS_IOERR; sector = r10_bio->devs[i].addr; - atomic_inc(&conf->mirrors[d].rdev->nr_pending); bio->bi_next = biolist; biolist = bio; - bio->bi_private = r10_bio; bio->bi_end_io = end_sync_write; - bio->bi_rw = WRITE; - bio->bi_sector = sector + - conf->mirrors[d].replacement->data_offset; - bio->bi_bdev = conf->mirrors[d].replacement->bdev; + bio->bi_opf = REQ_OP_WRITE; + if (test_bit(FailFast, &rdev->flags)) + bio->bi_opf |= MD_FAILFAST; + bio->bi_iter.bi_sector = sector + rdev->data_offset; + bio_set_dev(bio, rdev->bdev); count++; } @@ -3374,40 +3680,76 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, if (len == 0) break; for (bio= biolist ; bio ; bio=bio->bi_next) { - struct bio *bio2; - page = bio->bi_io_vec[bio->bi_vcnt].bv_page; - if (bio_add_page(bio, page, len, 0)) - continue; - - /* stop here */ - bio->bi_io_vec[bio->bi_vcnt].bv_page = page; - for (bio2 = biolist; - bio2 && bio2 != bio; - bio2 = bio2->bi_next) { - /* remove last page from this bio */ - bio2->bi_vcnt--; - bio2->bi_size -= len; - bio2->bi_flags &= ~(1<< BIO_SEG_VALID); + struct resync_pages *rp = get_resync_pages(bio); + page = resync_fetch_page(rp, page_idx); + if (WARN_ON(!bio_add_page(bio, page, len, 0))) { + bio->bi_status = BLK_STS_RESOURCE; + bio_endio(bio); + goto giveup; } - goto bio_full; } nr_sectors += len>>9; sector_nr += len>>9; - } while (biolist->bi_vcnt < RESYNC_PAGES); - bio_full: + } while (++page_idx < RESYNC_PAGES); r10_bio->sectors = nr_sectors; + if (mddev_is_clustered(mddev) && + test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) { + /* It is resync not recovery */ + if (conf->cluster_sync_high < sector_nr + nr_sectors) { + conf->cluster_sync_low = mddev->curr_resync_completed; + raid10_set_cluster_sync_high(conf); + /* Send resync message */ + mddev->cluster_ops->resync_info_update(mddev, + conf->cluster_sync_low, + conf->cluster_sync_high); + } + } else if (mddev_is_clustered(mddev)) { + /* This is recovery not resync */ + sector_t sect_va1, sect_va2; + bool broadcast_msg = false; + + for (i = 0; i < conf->geo.raid_disks; i++) { + /* + * sector_nr is a device address for recovery, so we + * need translate it to array address before compare + * with cluster_sync_high. + */ + sect_va1 = raid10_find_virt(conf, sector_nr, i); + + if (conf->cluster_sync_high < sect_va1 + nr_sectors) { + broadcast_msg = true; + /* + * curr_resync_completed is similar as + * sector_nr, so make the translation too. + */ + sect_va2 = raid10_find_virt(conf, + mddev->curr_resync_completed, i); + + if (conf->cluster_sync_low == 0 || + conf->cluster_sync_low > sect_va2) + conf->cluster_sync_low = sect_va2; + } + } + if (broadcast_msg) { + raid10_set_cluster_sync_high(conf); + mddev->cluster_ops->resync_info_update(mddev, + conf->cluster_sync_low, + conf->cluster_sync_high); + } + } + while (biolist) { bio = biolist; biolist = biolist->bi_next; bio->bi_next = NULL; - r10_bio = bio->bi_private; + r10_bio = get_resync_r10bio(bio); r10_bio->sectors = nr_sectors; if (bio->bi_end_io == end_sync_read) { - md_sync_acct(bio->bi_bdev, nr_sectors); - generic_make_request(bio); + bio->bi_status = 0; + submit_bio_noacct(bio); } } @@ -3506,7 +3848,7 @@ static int setup_geo(struct geom *geo, struct mddev *mddev, enum geo_type new) disks = mddev->raid_disks + mddev->delta_disks; break; } - if (layout >> 18) + if (layout >> 19) return -1; if (chunk < (PAGE_SIZE >> 9) || !is_power_of_2(chunk)) @@ -3518,12 +3860,41 @@ static int setup_geo(struct geom *geo, struct mddev *mddev, enum geo_type new) geo->near_copies = nc; geo->far_copies = fc; geo->far_offset = fo; - geo->far_set_size = (layout & (1<<17)) ? disks / fc : disks; + switch (layout >> 17) { + case 0: /* original layout. simple but not always optimal */ + geo->far_set_size = disks; + break; + case 1: /* "improved" layout which was buggy. Hopefully no-one is + * actually using this, but leave code here just in case.*/ + geo->far_set_size = disks/fc; + WARN(geo->far_set_size < fc, + "This RAID10 layout does not provide data safety - please backup and create new array\n"); + break; + case 2: /* "improved" layout fixed to match documentation */ + geo->far_set_size = fc * nc; + break; + default: /* Not a valid layout */ + return -1; + } geo->chunk_mask = chunk - 1; geo->chunk_shift = ffz(~chunk); return nc*fc; } +static void raid10_free_conf(struct r10conf *conf) +{ + if (!conf) + return; + + mempool_exit(&conf->r10bio_pool); + kfree(conf->mirrors); + kfree(conf->mirrors_old); + kfree(conf->mirrors_new); + safe_put_page(conf->tmppage); + bioset_exit(&conf->bio_split); + kfree(conf); +} + static struct r10conf *setup_conf(struct mddev *mddev) { struct r10conf *conf = NULL; @@ -3534,15 +3905,14 @@ static struct r10conf *setup_conf(struct mddev *mddev) copies = setup_geo(&geo, mddev, geo_new); if (copies == -2) { - printk(KERN_ERR "md/raid10:%s: chunk size must be " - "at least PAGE_SIZE(%ld) and be a power of 2.\n", - mdname(mddev), PAGE_SIZE); + pr_warn("md/raid10:%s: chunk size must be at least PAGE_SIZE(%ld) and be a power of 2.\n", + mdname(mddev), PAGE_SIZE); goto out; } if (copies < 2 || copies > mddev->raid_disks) { - printk(KERN_ERR "md/raid10:%s: unsupported raid10 layout: 0x%8x\n", - mdname(mddev), mddev->new_layout); + pr_warn("md/raid10:%s: unsupported raid10 layout: 0x%8x\n", + mdname(mddev), mddev->new_layout); goto out; } @@ -3552,8 +3922,8 @@ static struct r10conf *setup_conf(struct mddev *mddev) goto out; /* FIXME calc properly */ - conf->mirrors = kzalloc(sizeof(struct raid10_info)*(mddev->raid_disks + - max(0,-mddev->delta_disks)), + conf->mirrors = kcalloc(mddev->raid_disks + max(0, -mddev->delta_disks), + sizeof(struct raid10_info), GFP_KERNEL); if (!conf->mirrors) goto out; @@ -3564,9 +3934,13 @@ static struct r10conf *setup_conf(struct mddev *mddev) conf->geo = geo; conf->copies = copies; - conf->r10bio_pool = mempool_create(NR_RAID10_BIOS, r10bio_pool_alloc, - r10bio_pool_free, conf); - if (!conf->r10bio_pool) + err = mempool_init(&conf->r10bio_pool, NR_RAID_BIOS, r10bio_pool_alloc, + rbio_pool_free, conf); + if (err) + goto out; + + err = bioset_init(&conf->bio_split, BIO_POOL_SIZE, 0, 0); + if (err) goto out; calc_sectors(conf, mddev->dev_sectors); @@ -3585,13 +3959,18 @@ static struct r10conf *setup_conf(struct mddev *mddev) /* far_copies must be 1 */ conf->prev.stride = conf->dev_sectors; } + conf->reshape_safe = conf->reshape_progress; spin_lock_init(&conf->device_lock); INIT_LIST_HEAD(&conf->retry_list); + INIT_LIST_HEAD(&conf->bio_end_io_list); - spin_lock_init(&conf->resync_lock); + seqlock_init(&conf->resync_lock); init_waitqueue_head(&conf->wait_barrier); + atomic_set(&conf->nr_pending, 0); - conf->thread = md_register_thread(raid10d, mddev, "raid10"); + err = -ENOMEM; + rcu_assign_pointer(conf->thread, + md_register_thread(raid10d, mddev, "raid10")); if (!conf->thread) goto out; @@ -3599,29 +3978,49 @@ static struct r10conf *setup_conf(struct mddev *mddev) return conf; out: - if (err == -ENOMEM) - printk(KERN_ERR "md/raid10:%s: couldn't allocate memory.\n", - mdname(mddev)); - if (conf) { - if (conf->r10bio_pool) - mempool_destroy(conf->r10bio_pool); - kfree(conf->mirrors); - safe_put_page(conf->tmppage); - kfree(conf); - } + raid10_free_conf(conf); return ERR_PTR(err); } -static int run(struct mddev *mddev) +static unsigned int raid10_nr_stripes(struct r10conf *conf) +{ + unsigned int raid_disks = conf->geo.raid_disks; + + if (conf->geo.raid_disks % conf->geo.near_copies) + return raid_disks; + return raid_disks / conf->geo.near_copies; +} + +static int raid10_set_queue_limits(struct mddev *mddev) +{ + struct r10conf *conf = mddev->private; + struct queue_limits lim; + int err; + + md_init_stacking_limits(&lim); + lim.max_write_zeroes_sectors = 0; + lim.max_hw_wzeroes_unmap_sectors = 0; + lim.logical_block_size = mddev->logical_block_size; + lim.io_min = mddev->chunk_sectors << 9; + lim.chunk_sectors = mddev->chunk_sectors; + lim.io_opt = lim.io_min * raid10_nr_stripes(conf); + lim.features |= BLK_FEAT_ATOMIC_WRITES; + err = mddev_stack_rdev_limits(mddev, &lim, MDDEV_STACK_INTEGRITY); + if (err) + return err; + return queue_limits_set(mddev->gendisk->queue, &lim); +} + +static int raid10_run(struct mddev *mddev) { struct r10conf *conf; - int i, disk_idx, chunk_size; + int i, disk_idx; struct raid10_info *disk; struct md_rdev *rdev; sector_t size; sector_t min_offset_diff = 0; int first = 1; - bool discard_supported = false; + int ret = -EIO; if (mddev->private == NULL) { conf = setup_conf(mddev); @@ -3633,25 +4032,23 @@ static int run(struct mddev *mddev) if (!conf) goto out; - mddev->thread = conf->thread; - conf->thread = NULL; - - chunk_size = mddev->chunk_sectors << 9; - if (mddev->queue) { - blk_queue_max_discard_sectors(mddev->queue, - mddev->chunk_sectors); - blk_queue_max_write_same_sectors(mddev->queue, 0); - blk_queue_io_min(mddev->queue, chunk_size); - if (conf->geo.raid_disks % conf->geo.near_copies) - blk_queue_io_opt(mddev->queue, chunk_size * conf->geo.raid_disks); - else - blk_queue_io_opt(mddev->queue, chunk_size * - (conf->geo.raid_disks / conf->geo.near_copies)); + rcu_assign_pointer(mddev->thread, conf->thread); + rcu_assign_pointer(conf->thread, NULL); + + if (mddev_is_clustered(conf->mddev)) { + int fc, fo; + + fc = (mddev->layout >> 8) & 255; + fo = mddev->layout & (1<<16); + if (fc > 1 || fo > 0) { + pr_err("only near layout is supported by clustered" + " raid10\n"); + goto out_free_conf; + } } rdev_for_each(rdev, mddev) { long long diff; - struct request_queue *q; disk_idx = rdev->raid_disk; if (disk_idx < 0) @@ -3670,9 +4067,6 @@ static int run(struct mddev *mddev) goto out_free_conf; disk->rdev = rdev; } - q = bdev_get_queue(rdev->bdev); - if (q->merge_bvec_fn) - mddev->merge_check_needed = 1; diff = (rdev->new_data_offset - rdev->data_offset); if (!mddev->reshape_backwards) diff = -diff; @@ -3681,27 +4075,22 @@ static int run(struct mddev *mddev) if (first || diff < min_offset_diff) min_offset_diff = diff; - if (mddev->gendisk) - disk_stack_limits(mddev->gendisk, rdev->bdev, - rdev->data_offset << 9); - disk->head_position = 0; - - if (blk_queue_discard(bdev_get_queue(rdev->bdev))) - discard_supported = true; + first = 0; } - if (mddev->queue) { - if (discard_supported) - queue_flag_set_unlocked(QUEUE_FLAG_DISCARD, - mddev->queue); - else - queue_flag_clear_unlocked(QUEUE_FLAG_DISCARD, - mddev->queue); + if (!mddev_is_dm(conf->mddev)) { + int err = raid10_set_queue_limits(mddev); + + if (err) { + ret = err; + goto out_free_conf; + } } + /* need to check that every block has at least one working mirror */ if (!enough(conf, -1)) { - printk(KERN_ERR "md/raid10:%s: not enough operational mirrors.\n", + pr_err("md/raid10:%s: not enough operational mirrors.\n", mdname(mddev)); goto out_free_conf; } @@ -3735,18 +4124,24 @@ static int run(struct mddev *mddev) !test_bit(In_sync, &disk->rdev->flags)) { disk->head_position = 0; mddev->degraded++; - if (disk->rdev) + if (disk->rdev && + disk->rdev->saved_raid_disk < 0) conf->fullsync = 1; } + + if (disk->replacement && + !test_bit(In_sync, &disk->replacement->flags) && + disk->replacement->saved_raid_disk < 0) { + conf->fullsync = 1; + } + disk->recovery_disabled = mddev->recovery_disabled - 1; } - if (mddev->recovery_cp != MaxSector) - printk(KERN_NOTICE "md/raid10:%s: not clean" - " -- starting background reconstruction\n", - mdname(mddev)); - printk(KERN_INFO - "md/raid10:%s: active with %d out of %d devices\n", + if (mddev->resync_offset != MaxSector) + pr_notice("md/raid10:%s: not clean -- starting background reconstruction\n", + mdname(mddev)); + pr_info("md/raid10:%s: active with %d out of %d devices\n", mdname(mddev), conf->geo.raid_disks - mddev->degraded, conf->geo.raid_disks); /* @@ -3756,23 +4151,7 @@ static int run(struct mddev *mddev) size = raid10_size(mddev, 0, 0); md_set_array_sectors(mddev, size); mddev->resync_max_sectors = size; - - if (mddev->queue) { - int stripe = conf->geo.raid_disks * - ((mddev->chunk_sectors << 9) / PAGE_SIZE); - mddev->queue->backing_dev_info.congested_fn = raid10_congested; - mddev->queue->backing_dev_info.congested_data = mddev; - - /* Calculate max read-ahead size. - * We need to readahead at least twice a whole stripe.... - * maybe... - */ - stripe /= conf->geo.near_copies; - if (mddev->queue->backing_dev_info.ra_pages < 2 * stripe) - mddev->queue->backing_dev_info.ra_pages = 2 * stripe; - blk_queue_merge_bvec(mddev->queue, raid10_mergeable_bvec); - } - + set_bit(MD_FAILFAST_SUPPORTED, &mddev->flags); if (md_integrity_register(mddev)) goto out_free_conf; @@ -3787,67 +4166,40 @@ static int run(struct mddev *mddev) if (max(before_length, after_length) > min_offset_diff) { /* This cannot work */ - printk("md/raid10: offset difference not enough to continue reshape\n"); + pr_warn("md/raid10: offset difference not enough to continue reshape\n"); goto out_free_conf; } conf->offset_diff = min_offset_diff; - conf->reshape_safe = conf->reshape_progress; clear_bit(MD_RECOVERY_SYNC, &mddev->recovery); clear_bit(MD_RECOVERY_CHECK, &mddev->recovery); set_bit(MD_RECOVERY_RESHAPE, &mddev->recovery); - set_bit(MD_RECOVERY_RUNNING, &mddev->recovery); - mddev->sync_thread = md_register_thread(md_do_sync, mddev, - "reshape"); + set_bit(MD_RECOVERY_NEEDED, &mddev->recovery); } return 0; out_free_conf: - md_unregister_thread(&mddev->thread); - if (conf->r10bio_pool) - mempool_destroy(conf->r10bio_pool); - safe_put_page(conf->tmppage); - kfree(conf->mirrors); - kfree(conf); + md_unregister_thread(mddev, &mddev->thread); + raid10_free_conf(conf); mddev->private = NULL; out: - return -EIO; + return ret; } -static int stop(struct mddev *mddev) +static void raid10_free(struct mddev *mddev, void *priv) { - struct r10conf *conf = mddev->private; - - raise_barrier(conf, 0); - lower_barrier(conf); - - md_unregister_thread(&mddev->thread); - if (mddev->queue) - /* the unplug fn references 'conf'*/ - blk_sync_queue(mddev->queue); - - if (conf->r10bio_pool) - mempool_destroy(conf->r10bio_pool); - safe_put_page(conf->tmppage); - kfree(conf->mirrors); - kfree(conf); - mddev->private = NULL; - return 0; + raid10_free_conf(priv); } -static void raid10_quiesce(struct mddev *mddev, int state) +static void raid10_quiesce(struct mddev *mddev, int quiesce) { struct r10conf *conf = mddev->private; - switch(state) { - case 1: + if (quiesce) raise_barrier(conf, 0); - break; - case 0: + else lower_barrier(conf); - break; - } } static int raid10_resize(struct mddev *mddev, sector_t sectors) @@ -3878,17 +4230,18 @@ static int raid10_resize(struct mddev *mddev, sector_t sectors) if (mddev->external_size && mddev->array_sectors > size) return -EINVAL; - if (mddev->bitmap) { - int ret = bitmap_resize(mddev->bitmap, size, 0, 0); + + if (md_bitmap_enabled(mddev, false)) { + int ret = mddev->bitmap_ops->resize(mddev, size, 0); + if (ret) return ret; } + md_set_array_sectors(mddev, size); - set_capacity(mddev->gendisk, mddev->array_sectors); - revalidate_disk(mddev->gendisk); if (sectors > mddev->dev_sectors && - mddev->recovery_cp > oldsize) { - mddev->recovery_cp = oldsize; + mddev->resync_offset > oldsize) { + mddev->resync_offset = oldsize; set_bit(MD_RECOVERY_NEEDED, &mddev->recovery); } calc_sectors(conf, sectors); @@ -3897,16 +4250,17 @@ static int raid10_resize(struct mddev *mddev, sector_t sectors) return 0; } -static void *raid10_takeover_raid0(struct mddev *mddev) +static void *raid10_takeover_raid0(struct mddev *mddev, sector_t size, int devs) { struct md_rdev *rdev; struct r10conf *conf; if (mddev->degraded > 0) { - printk(KERN_ERR "md/raid10:%s: Error: degraded raid0!\n", - mdname(mddev)); + pr_warn("md/raid10:%s: Error: degraded raid0!\n", + mdname(mddev)); return ERR_PTR(-EINVAL); } + sector_div(size, devs); /* Set new parameters */ mddev->new_level = 10; @@ -3916,14 +4270,16 @@ static void *raid10_takeover_raid0(struct mddev *mddev) mddev->delta_disks = mddev->raid_disks; mddev->raid_disks *= 2; /* make sure it will be not marked as dirty */ - mddev->recovery_cp = MaxSector; + mddev->resync_offset = MaxSector; + mddev->dev_sectors = size; conf = setup_conf(mddev); if (!IS_ERR(conf)) { rdev_for_each(rdev, mddev) - if (rdev->raid_disk >= 0) + if (rdev->raid_disk >= 0) { rdev->new_raid_disk = rdev->raid_disk * 2; - conf->barrier = 1; + rdev->sectors = size; + } } return conf; @@ -3940,12 +4296,13 @@ static void *raid10_takeover(struct mddev *mddev) /* for raid0 takeover only one zone is supported */ raid0_conf = mddev->private; if (raid0_conf->nr_strip_zones > 1) { - printk(KERN_ERR "md/raid10:%s: cannot takeover raid 0" - " with more than one zone.\n", - mdname(mddev)); + pr_warn("md/raid10:%s: cannot takeover raid 0 with more than one zone.\n", + mdname(mddev)); return ERR_PTR(-EINVAL); } - return raid10_takeover_raid0(mddev); + return raid10_takeover_raid0(mddev, + raid0_conf->strip_zone->zone_end, + raid0_conf->strip_zone->nb_dev); } return ERR_PTR(-EINVAL); } @@ -3990,11 +4347,10 @@ static int raid10_check_reshape(struct mddev *mddev) conf->mirrors_new = NULL; if (mddev->delta_disks > 0) { /* allocate new 'mirrors' list */ - conf->mirrors_new = kzalloc( - sizeof(struct raid10_info) - *(mddev->raid_disks + - mddev->delta_disks), - GFP_KERNEL); + conf->mirrors_new = + kcalloc(mddev->raid_disks + mddev->delta_disks, + sizeof(struct raid10_info), + GFP_KERNEL); if (!conf->mirrors_new) return -ENOMEM; } @@ -4019,11 +4375,11 @@ static int calc_degraded(struct r10conf *conf) int degraded, degraded2; int i; - rcu_read_lock(); degraded = 0; /* 'prev' section first */ for (i = 0; i < conf->prev.raid_disks; i++) { - struct md_rdev *rdev = rcu_dereference(conf->mirrors[i].rdev); + struct md_rdev *rdev = conf->mirrors[i].rdev; + if (!rdev || test_bit(Faulty, &rdev->flags)) degraded++; else if (!test_bit(In_sync, &rdev->flags)) @@ -4033,13 +4389,12 @@ static int calc_degraded(struct r10conf *conf) */ degraded++; } - rcu_read_unlock(); if (conf->geo.raid_disks == conf->prev.raid_disks) return degraded; - rcu_read_lock(); degraded2 = 0; for (i = 0; i < conf->geo.raid_disks; i++) { - struct md_rdev *rdev = rcu_dereference(conf->mirrors[i].rdev); + struct md_rdev *rdev = conf->mirrors[i].rdev; + if (!rdev || test_bit(Faulty, &rdev->flags)) degraded2++; else if (!test_bit(In_sync, &rdev->flags)) { @@ -4052,7 +4407,6 @@ static int calc_degraded(struct r10conf *conf) degraded2++; } } - rcu_read_unlock(); if (degraded2 > degraded) return degraded2; return degraded; @@ -4103,6 +4457,7 @@ static int raid10_start_reshape(struct mddev *mddev) diff = 0; if (first || diff < min_offset_diff) min_offset_diff = diff; + first = 0; } } @@ -4118,7 +4473,7 @@ static int raid10_start_reshape(struct mddev *mddev) memcpy(conf->mirrors_new, conf->mirrors, sizeof(struct raid10_info)*conf->prev.raid_disks); smp_mb(); - kfree(conf->mirrors_old); /* FIXME and elsewhere */ + kfree(conf->mirrors_old); conf->mirrors_old = conf->mirrors; conf->mirrors = conf->mirrors_new; conf->mirrors_new = NULL; @@ -4129,24 +4484,60 @@ static int raid10_start_reshape(struct mddev *mddev) sector_t size = raid10_size(mddev, 0, 0); if (size < mddev->array_sectors) { spin_unlock_irq(&conf->device_lock); - printk(KERN_ERR "md/raid10:%s: array size must be reduce before number of disks\n", - mdname(mddev)); + pr_warn("md/raid10:%s: array size must be reduce before number of disks\n", + mdname(mddev)); return -EINVAL; } mddev->resync_max_sectors = size; conf->reshape_progress = size; } else conf->reshape_progress = 0; + conf->reshape_safe = conf->reshape_progress; spin_unlock_irq(&conf->device_lock); if (mddev->delta_disks && mddev->bitmap) { - ret = bitmap_resize(mddev->bitmap, - raid10_size(mddev, 0, - conf->geo.raid_disks), - 0, 0); + struct mdp_superblock_1 *sb = NULL; + sector_t oldsize, newsize; + + oldsize = raid10_size(mddev, 0, 0); + newsize = raid10_size(mddev, 0, conf->geo.raid_disks); + + if (!mddev_is_clustered(mddev) && + md_bitmap_enabled(mddev, false)) { + ret = mddev->bitmap_ops->resize(mddev, newsize, 0); + if (ret) + goto abort; + else + goto out; + } + + rdev_for_each(rdev, mddev) { + if (rdev->raid_disk > -1 && + !test_bit(Faulty, &rdev->flags)) + sb = page_address(rdev->sb_page); + } + + /* + * some node is already performing reshape, and no need to + * call bitmap_ops->resize again since it should be called when + * receiving BITMAP_RESIZE msg + */ + if ((sb && (le32_to_cpu(sb->feature_map) & + MD_FEATURE_RESHAPE_ACTIVE)) || (oldsize == newsize)) + goto out; + + /* cluster can't be setup without bitmap */ + ret = mddev->bitmap_ops->resize(mddev, newsize, 0); if (ret) goto abort; + + ret = mddev->cluster_ops->resize_bitmaps(mddev, newsize, oldsize); + if (ret) { + mddev->bitmap_ops->resize(mddev, oldsize, 0); + goto abort; + } } +out: if (mddev->delta_disks > 0) { rdev_for_each(rdev, mddev) if (rdev->raid_disk < 0 && @@ -4158,8 +4549,8 @@ static int raid10_start_reshape(struct mddev *mddev) else rdev->recovery_offset = 0; - if (sysfs_link_rdev(mddev, rdev)) - /* Failure here is OK */; + /* Failure here is OK */ + sysfs_link_rdev(mddev, rdev); } } else if (rdev->raid_disk >= conf->prev.raid_disks && !test_bit(Faulty, &rdev->flags)) { @@ -4176,22 +4567,15 @@ static int raid10_start_reshape(struct mddev *mddev) spin_unlock_irq(&conf->device_lock); mddev->raid_disks = conf->geo.raid_disks; mddev->reshape_position = conf->reshape_progress; - set_bit(MD_CHANGE_DEVS, &mddev->flags); + set_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags); clear_bit(MD_RECOVERY_SYNC, &mddev->recovery); clear_bit(MD_RECOVERY_CHECK, &mddev->recovery); + clear_bit(MD_RECOVERY_DONE, &mddev->recovery); set_bit(MD_RECOVERY_RESHAPE, &mddev->recovery); - set_bit(MD_RECOVERY_RUNNING, &mddev->recovery); - - mddev->sync_thread = md_register_thread(md_do_sync, mddev, - "reshape"); - if (!mddev->sync_thread) { - ret = -EAGAIN; - goto abort; - } + set_bit(MD_RECOVERY_NEEDED, &mddev->recovery); conf->reshape_checkpoint = jiffies; - md_wakeup_thread(mddev->sync_thread); - md_new_event(mddev); + md_new_event(); return 0; abort: @@ -4203,6 +4587,7 @@ abort: rdev->new_data_offset = rdev->data_offset; smp_wmb(); conf->reshape_progress = MaxSector; + conf->reshape_safe = MaxSector; mddev->reshape_position = MaxSector; spin_unlock_irq(&conf->device_lock); return ret; @@ -4246,7 +4631,7 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, * at a time, possibly less if that exceeds RESYNC_PAGES, * or we hit a bad block or something. * This might mean we pause for normal IO in the middle of - * a chunk, but that is not a problem was mddev->reshape_position + * a chunk, but that is not a problem as mddev->reshape_position * can record any location. * * If we will want to write to a location that isn't @@ -4270,7 +4655,7 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, * * In all this the minimum difference in data offsets * (conf->offset_diff - always positive) allows a bit of slack, - * so next can be after 'safe', but not by more than offset_disk + * so next can be after 'safe', but not by more than offset_diff * * We need to prepare all the bios here before we start any IO * to ensure the size we choose is acceptable to all devices. @@ -4290,6 +4675,7 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, struct bio *blist; struct bio *bio, *read_bio; int sectors_done = 0; + struct page **pages; if (sector_nr == 0) { /* If restarting in the middle, skip the initial sectors */ @@ -4302,7 +4688,7 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, sector_nr = conf->reshape_progress; if (sector_nr) { mddev->curr_resync_completed = sector_nr; - sysfs_notify(&mddev->kobj, NULL, "sync_completed"); + sysfs_notify_dirent_safe(mddev->sysfs_completed); *skipped = 1; return sector_nr; } @@ -4331,8 +4717,8 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, last = conf->reshape_progress - 1; sector_nr = last & ~(sector_t)(conf->geo.chunk_mask & conf->prev.chunk_mask); - if (sector_nr + RESYNC_BLOCK_SIZE/512 < last) - sector_nr = last + 1 - RESYNC_BLOCK_SIZE/512; + if (sector_nr + RESYNC_SECTORS < last) + sector_nr = last + 1 - RESYNC_SECTORS; } else { /* 'next' is after the last device address that we * might write to for this chunk in the new layout @@ -4354,14 +4740,14 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, last = sector_nr | (conf->geo.chunk_mask & conf->prev.chunk_mask); - if (sector_nr + RESYNC_BLOCK_SIZE/512 <= last) - last = sector_nr + RESYNC_BLOCK_SIZE/512 - 1; + if (sector_nr + RESYNC_SECTORS <= last) + last = sector_nr + RESYNC_SECTORS - 1; } if (need_flush || time_after(jiffies, conf->reshape_checkpoint + 10*HZ)) { /* Need to update reshape_position in metadata */ - wait_barrier(conf); + wait_barrier(conf, false); mddev->reshape_position = conf->reshape_progress; if (mddev->reshape_backwards) mddev->curr_resync_completed = raid10_size(mddev, 0, 0) @@ -4369,18 +4755,24 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr, else mddev->curr_resync_completed = conf->reshape_progress; conf->reshape_checkpoint = jiffies; - set_bit(MD_CHANGE_DEVS, &mddev->flags); + set_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags); md_wakeup_thread(mddev->thread); - wait_event(mddev->sb_wait, mddev->flags == 0 || - kthread_should_stop()); + wait_event(mddev->sb_wait, mddev->sb_flags == 0 || + test_bit(MD_RECOVERY_INTR, &mddev->recovery)); + if (test_bit(MD_RECOVERY_INTR, &mddev->recovery)) { + allow_barrier(conf); + return sectors_done; + } conf->reshape_safe = mddev->reshape_position; allow_barrier(conf); } + raise_barrier(conf, 0); read_more: /* Now schedule reads for blocks from sector_nr to last */ - r10_bio = mempool_alloc(conf->r10buf_pool, GFP_NOIO); - raise_barrier(conf, sectors_done != 0); + r10_bio = raid10_alloc_init_r10buf(conf); + r10_bio->state = 0; + raise_barrier(conf, 1); atomic_set(&r10_bio->remaining, 0); r10_bio->mddev = mddev; r10_bio->sector = sector_nr; @@ -4394,25 +4786,46 @@ read_more: * on all the target devices. */ // FIXME + mempool_free(r10_bio, &conf->r10buf_pool); set_bit(MD_RECOVERY_INTR, &mddev->recovery); return sectors_done; } - read_bio = bio_alloc_mddev(GFP_KERNEL, RESYNC_PAGES, mddev); - - read_bio->bi_bdev = rdev->bdev; - read_bio->bi_sector = (r10_bio->devs[r10_bio->read_slot].addr + read_bio = bio_alloc_bioset(rdev->bdev, RESYNC_PAGES, REQ_OP_READ, + GFP_KERNEL, &mddev->bio_set); + read_bio->bi_iter.bi_sector = (r10_bio->devs[r10_bio->read_slot].addr + rdev->data_offset); read_bio->bi_private = r10_bio; - read_bio->bi_end_io = end_sync_read; - read_bio->bi_rw = READ; - read_bio->bi_flags &= ~(BIO_POOL_MASK - 1); - read_bio->bi_flags |= 1 << BIO_UPTODATE; - read_bio->bi_vcnt = 0; - read_bio->bi_size = 0; + read_bio->bi_end_io = end_reshape_read; r10_bio->master_bio = read_bio; r10_bio->read_slot = r10_bio->devs[r10_bio->read_slot].devnum; + /* + * Broadcast RESYNC message to other nodes, so all nodes would not + * write to the region to avoid conflict. + */ + if (mddev_is_clustered(mddev) && conf->cluster_sync_high <= sector_nr) { + struct mdp_superblock_1 *sb = NULL; + int sb_reshape_pos = 0; + + conf->cluster_sync_low = sector_nr; + conf->cluster_sync_high = sector_nr + CLUSTER_RESYNC_WINDOW_SECTORS; + sb = page_address(rdev->sb_page); + if (sb) { + sb_reshape_pos = le64_to_cpu(sb->reshape_position); + /* + * Set cluster_sync_low again if next address for array + * reshape is less than cluster_sync_low. Since we can't + * update cluster_sync_low until it has finished reshape. + */ + if (sb_reshape_pos < conf->cluster_sync_low) + conf->cluster_sync_low = sb_reshape_pos; + } + + mddev->cluster_ops->resync_info_update(mddev, conf->cluster_sync_low, + conf->cluster_sync_high); + } + /* Now find the locations in the new layout */ __raid10_find_phys(&conf->geo, r10_bio); @@ -4433,12 +4846,11 @@ read_more: if (!rdev2 || test_bit(Faulty, &rdev2->flags)) continue; - bio_reset(b); - b->bi_bdev = rdev2->bdev; - b->bi_sector = r10_bio->devs[s/2].addr + rdev2->new_data_offset; - b->bi_private = r10_bio; + bio_set_dev(b, rdev2->bdev); + b->bi_iter.bi_sector = r10_bio->devs[s/2].addr + + rdev2->new_data_offset; b->bi_end_io = end_reshape_write; - b->bi_rw = WRITE; + b->bi_opf = REQ_OP_WRITE; b->bi_next = blist; blist = b; } @@ -4446,43 +4858,34 @@ read_more: /* Now add as many pages as possible to all of these bios. */ nr_sectors = 0; + pages = get_resync_pages(r10_bio->devs[0].bio)->pages; for (s = 0 ; s < max_sectors; s += PAGE_SIZE >> 9) { - struct page *page = r10_bio->devs[0].bio->bi_io_vec[s/(PAGE_SIZE>>9)].bv_page; + struct page *page = pages[s / (PAGE_SIZE >> 9)]; int len = (max_sectors - s) << 9; if (len > PAGE_SIZE) len = PAGE_SIZE; for (bio = blist; bio ; bio = bio->bi_next) { - struct bio *bio2; - if (bio_add_page(bio, page, len, 0)) - continue; - - /* Didn't fit, must stop */ - for (bio2 = blist; - bio2 && bio2 != bio; - bio2 = bio2->bi_next) { - /* Remove last page from this bio */ - bio2->bi_vcnt--; - bio2->bi_size -= len; - bio2->bi_flags &= ~(1<<BIO_SEG_VALID); + if (WARN_ON(!bio_add_page(bio, page, len, 0))) { + bio->bi_status = BLK_STS_RESOURCE; + bio_endio(bio); + return sectors_done; } - goto bio_full; } sector_nr += len >> 9; nr_sectors += len >> 9; } -bio_full: r10_bio->sectors = nr_sectors; /* Now submit the read */ - md_sync_acct(read_bio->bi_bdev, r10_bio->sectors); atomic_inc(&r10_bio->remaining); read_bio->bi_next = NULL; - generic_make_request(read_bio); - sector_nr += nr_sectors; + submit_bio_noacct(read_bio); sectors_done += nr_sectors; if (sector_nr <= last) goto read_more; + lower_barrier(conf); + /* Now that we have done the whole section we can * update reshape_progress */ @@ -4531,11 +4934,11 @@ static void reshape_request_write(struct mddev *mddev, struct r10bio *r10_bio) } if (!rdev || test_bit(Faulty, &rdev->flags)) continue; + atomic_inc(&rdev->nr_pending); - md_sync_acct(b->bi_bdev, r10_bio->sectors); atomic_inc(&r10_bio->remaining); b->bi_next = NULL; - generic_make_request(b); + submit_bio_noacct(b); } end_reshape_request(r10_bio); } @@ -4550,21 +4953,25 @@ static void end_reshape(struct r10conf *conf) md_finish_reshape(conf->mddev); smp_wmb(); conf->reshape_progress = MaxSector; + conf->reshape_safe = MaxSector; spin_unlock_irq(&conf->device_lock); - /* read-ahead size must cover two whole stripes, which is - * 2 * (datadisks) * chunksize where 'n' is the number of raid devices - */ - if (conf->mddev->queue) { - int stripe = conf->geo.raid_disks * - ((conf->mddev->chunk_sectors << 9) / PAGE_SIZE); - stripe /= conf->geo.near_copies; - if (conf->mddev->queue->backing_dev_info.ra_pages < 2 * stripe) - conf->mddev->queue->backing_dev_info.ra_pages = 2 * stripe; - } + mddev_update_io_opt(conf->mddev, raid10_nr_stripes(conf)); conf->fullsync = 0; } +static void raid10_update_reshape_pos(struct mddev *mddev) +{ + struct r10conf *conf = mddev->private; + sector_t lo, hi; + + mddev->cluster_ops->resync_info_get(mddev, &lo, &hi); + if (((mddev->reshape_position <= hi) && (mddev->reshape_position >= lo)) + || mddev->reshape_position == MaxSector) + conf->reshape_progress = mddev->reshape_position; + else + WARN_ON_ONCE(1); +} static int handle_reshape_read_error(struct mddev *mddev, struct r10bio *r10_bio) @@ -4572,14 +4979,19 @@ static int handle_reshape_read_error(struct mddev *mddev, /* Use sync reads to get the blocks from somewhere else */ int sectors = r10_bio->sectors; struct r10conf *conf = mddev->private; - struct { - struct r10bio r10_bio; - struct r10dev devs[conf->copies]; - } on_stack; - struct r10bio *r10b = &on_stack.r10_bio; + struct r10bio *r10b; int slot = 0; int idx = 0; - struct bio_vec *bvec = r10_bio->master_bio->bi_io_vec; + struct page **pages; + + r10b = kmalloc(struct_size(r10b, devs, conf->copies), GFP_NOIO); + if (!r10b) { + set_bit(MD_RECOVERY_INTR, &mddev->recovery); + return -ENOMEM; + } + + /* reshape IOs share pages from .devs[0].bio */ + pages = get_resync_pages(r10_bio->devs[0].bio)->pages; r10b->sector = r10_bio->sector; __raid10_find_phys(&conf->prev, r10b); @@ -4602,11 +5014,13 @@ static int handle_reshape_read_error(struct mddev *mddev, goto failed; addr = r10b->devs[slot].addr + idx * PAGE_SIZE; + atomic_inc(&rdev->nr_pending); success = sync_page_io(rdev, addr, s << 9, - bvec[idx].bv_page, - READ, false); + pages[idx], + REQ_OP_READ, false); + rdev_dec_pending(rdev, mddev); if (success) break; failed: @@ -4620,18 +5034,19 @@ static int handle_reshape_read_error(struct mddev *mddev, /* couldn't read this block, must give up */ set_bit(MD_RECOVERY_INTR, &mddev->recovery); + kfree(r10b); return -EIO; } sectors -= s; idx++; } + kfree(r10b); return 0; } -static void end_reshape_write(struct bio *bio, int error) +static void end_reshape_write(struct bio *bio) { - int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags); - struct r10bio *r10_bio = bio->bi_private; + struct r10bio *r10_bio = get_resync_r10bio(bio); struct mddev *mddev = r10_bio->mddev; struct r10conf *conf = mddev->private; int d; @@ -4640,14 +5055,10 @@ static void end_reshape_write(struct bio *bio, int error) struct md_rdev *rdev = NULL; d = find_bio_disk(conf, r10_bio, bio, &slot, &repl); - if (repl) - rdev = conf->mirrors[d].replacement; - if (!rdev) { - smp_mb(); - rdev = conf->mirrors[d].rdev; - } + rdev = repl ? conf->mirrors[d].replacement : + conf->mirrors[d].rdev; - if (!uptodate) { + if (bio->bi_status) { /* FIXME should record badblock */ md_error(mddev, rdev); } @@ -4673,15 +5084,11 @@ static void raid10_finish_reshape(struct mddev *mddev) return; if (mddev->delta_disks > 0) { - sector_t size = raid10_size(mddev, 0, 0); - md_set_array_sectors(mddev, size); - if (mddev->recovery_cp > mddev->resync_max_sectors) { - mddev->recovery_cp = mddev->resync_max_sectors; + if (mddev->resync_offset > mddev->resync_max_sectors) { + mddev->resync_offset = mddev->resync_max_sectors; set_bit(MD_RECOVERY_NEEDED, &mddev->recovery); } - mddev->resync_max_sectors = size; - set_capacity(mddev->gendisk, mddev->array_sectors); - revalidate_disk(mddev->gendisk); + mddev->resync_max_sectors = mddev->array_sectors; } else { int d; for (d = conf->geo.raid_disks ; @@ -4704,18 +5111,22 @@ static void raid10_finish_reshape(struct mddev *mddev) static struct md_personality raid10_personality = { - .name = "raid10", - .level = 10, - .owner = THIS_MODULE, - .make_request = make_request, - .run = run, - .stop = stop, - .status = status, - .error_handler = error, + .head = { + .type = MD_PERSONALITY, + .id = ID_RAID10, + .name = "raid10", + .owner = THIS_MODULE, + }, + + .make_request = raid10_make_request, + .run = raid10_run, + .free = raid10_free, + .status = raid10_status, + .error_handler = raid10_error, .hot_add_disk = raid10_add_disk, .hot_remove_disk= raid10_remove_disk, .spare_active = raid10_spare_active, - .sync_request = sync_request, + .sync_request = raid10_sync_request, .quiesce = raid10_quiesce, .size = raid10_size, .resize = raid10_resize, @@ -4723,24 +5134,23 @@ static struct md_personality raid10_personality = .check_reshape = raid10_check_reshape, .start_reshape = raid10_start_reshape, .finish_reshape = raid10_finish_reshape, + .update_reshape_pos = raid10_update_reshape_pos, }; -static int __init raid_init(void) +static int __init raid10_init(void) { - return register_md_personality(&raid10_personality); + return register_md_submodule(&raid10_personality.head); } -static void raid_exit(void) +static void __exit raid10_exit(void) { - unregister_md_personality(&raid10_personality); + unregister_md_submodule(&raid10_personality.head); } -module_init(raid_init); -module_exit(raid_exit); +module_init(raid10_init); +module_exit(raid10_exit); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("RAID10 (striped mirror) personality for MD"); MODULE_ALIAS("md-personality-9"); /* RAID10 */ MODULE_ALIAS("md-raid10"); MODULE_ALIAS("md-level-10"); - -module_param(max_queued_requests, int, S_IRUGO|S_IWUSR); |