/* md.c : Multiple Devices driver for Linux Copyright (C) 1998, 1999, 2000 Ingo Molnar completely rewritten, based on the MD driver code from Marc Zyngier Changes: - RAID-1/RAID-5 extensions by Miguel de Icaza, Gadi Oxman, Ingo Molnar - RAID-6 extensions by H. Peter Anvin - boot support for linear and striped mode by Harald Hoyer - kerneld support by Boris Tobotras - kmod support by: Cyrus Durgin - RAID0 bugfixes: Mark Anthony Lisher - Devfs support by Richard Gooch - lots of fixes and improvements to the RAID1/RAID5 and generic RAID code (such as request based resynchronization): Neil Brown . - persistent bitmap code Copyright (C) 2003-2004, Paul Clements, SteelEye Technology, Inc. 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 #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include "md.h" #include "bitmap.h" #ifndef MODULE static void autostart_arrays(int part); #endif /* pers_list is a list of registered personalities protected * by pers_lock. * pers_lock does extra service to protect accesses to * mddev->thread when the mutex cannot be held. */ static LIST_HEAD(pers_list); static DEFINE_SPINLOCK(pers_lock); static void md_print_devices(void); static DECLARE_WAIT_QUEUE_HEAD(resync_wait); static struct workqueue_struct *md_wq; static struct workqueue_struct *md_misc_wq; #define MD_BUG(x...) { printk("md: bug in file %s, line %d\n", __FILE__, __LINE__); md_print_devices(); } /* * Default number of read corrections we'll attempt on an rdev * before ejecting it from the array. We divide the read error * count by 2 for every hour elapsed between read errors. */ #define MD_DEFAULT_MAX_CORRECTED_READ_ERRORS 20 /* * Current RAID-1,4,5 parallel reconstruction 'guaranteed speed limit' * is 1000 KB/sec, so the extra system load does not show up that much. * Increase it if you want to have more _guaranteed_ speed. Note that * the RAID driver will use the maximum available bandwidth if the IO * subsystem is idle. There is also an 'absolute maximum' reconstruction * speed limit - in case reconstruction slows down your system despite * idle IO detection. * * you can change it via /proc/sys/dev/raid/speed_limit_min and _max. * or /sys/block/mdX/md/sync_speed_{min,max} */ static int sysctl_speed_limit_min = 1000; static int sysctl_speed_limit_max = 200000; static inline int speed_min(struct mddev *mddev) { return mddev->sync_speed_min ? mddev->sync_speed_min : sysctl_speed_limit_min; } static inline int speed_max(struct mddev *mddev) { return mddev->sync_speed_max ? mddev->sync_speed_max : sysctl_speed_limit_max; } static struct ctl_table_header *raid_table_header; static ctl_table raid_table[] = { { .procname = "speed_limit_min", .data = &sysctl_speed_limit_min, .maxlen = sizeof(int), .mode = S_IRUGO|S_IWUSR, .proc_handler = proc_dointvec, }, { .procname = "speed_limit_max", .data = &sysctl_speed_limit_max, .maxlen = sizeof(int), .mode = S_IRUGO|S_IWUSR, .proc_handler = proc_dointvec, }, { } }; static ctl_table raid_dir_table[] = { { .procname = "raid", .maxlen = 0, .mode = S_IRUGO|S_IXUGO, .child = raid_table, }, { } }; static ctl_table raid_root_table[] = { { .procname = "dev", .maxlen = 0, .mode = 0555, .child = raid_dir_table, }, { } }; static const struct block_device_operations md_fops; static int start_readonly; /* bio_clone_mddev * like bio_clone, but with a local bio set */ static void mddev_bio_destructor(struct bio *bio) { struct mddev *mddev, **mddevp; mddevp = (void*)bio; mddev = mddevp[-1]; bio_free(bio, mddev->bio_set); } struct bio *bio_alloc_mddev(gfp_t gfp_mask, int nr_iovecs, struct mddev *mddev) { struct bio *b; struct mddev **mddevp; if (!mddev || !mddev->bio_set) return bio_alloc(gfp_mask, nr_iovecs); b = bio_alloc_bioset(gfp_mask, nr_iovecs, mddev->bio_set); if (!b) return NULL; mddevp = (void*)b; mddevp[-1] = mddev; b->bi_destructor = mddev_bio_destructor; return b; } EXPORT_SYMBOL_GPL(bio_alloc_mddev); struct bio *bio_clone_mddev(struct bio *bio, gfp_t gfp_mask, struct mddev *mddev) { struct bio *b; struct mddev **mddevp; if (!mddev || !mddev->bio_set) return bio_clone(bio, gfp_mask); b = bio_alloc_bioset(gfp_mask, bio->bi_max_vecs, mddev->bio_set); if (!b) return NULL; mddevp = (void*)b; mddevp[-1] = mddev; b->bi_destructor = mddev_bio_destructor; __bio_clone(b, bio); if (bio_integrity(bio)) { int ret; ret = bio_integrity_clone(b, bio, gfp_mask, mddev->bio_set); if (ret < 0) { bio_put(b); return NULL; } } return b; } EXPORT_SYMBOL_GPL(bio_clone_mddev); void md_trim_bio(struct bio *bio, int offset, int size) { /* 'bio' is a cloned bio which we need to trim to match * the given offset and size. * This requires adjusting bi_sector, bi_size, and bi_io_vec */ int i; struct bio_vec *bvec; int sofar = 0; size <<= 9; if (offset == 0 && size == bio->bi_size) return; bio->bi_sector += offset; bio->bi_size = size; offset <<= 9; clear_bit(BIO_SEG_VALID, &bio->bi_flags); while (bio->bi_idx < bio->bi_vcnt && bio->bi_io_vec[bio->bi_idx].bv_len <= offset) { /* remove this whole bio_vec */ offset -= bio->bi_io_vec[bio->bi_idx].bv_len; bio->bi_idx++; } if (bio->bi_idx < bio->bi_vcnt) { bio->bi_io_vec[bio->bi_idx].bv_offset += offset; bio->bi_io_vec[bio->bi_idx].bv_len -= offset; } /* avoid any complications with bi_idx being non-zero*/ if (bio->bi_idx) { memmove(bio->bi_io_vec, bio->bi_io_vec+bio->bi_idx, (bio->bi_vcnt - bio->bi_idx) * sizeof(struct bio_vec)); bio->bi_vcnt -= bio->bi_idx; bio->bi_idx = 0; } /* Make sure vcnt and last bv are not too big */ bio_for_each_segment(bvec, bio, i) { if (sofar + bvec->bv_len > size) bvec->bv_len = size - sofar; if (bvec->bv_len == 0) { bio->bi_vcnt = i; break; } sofar += bvec->bv_len; } } EXPORT_SYMBOL_GPL(md_trim_bio); /* * We have a system wide 'event count' that is incremented * on any 'interesting' event, and readers of /proc/mdstat * can use 'poll' or 'select' to find out when the event * count increases. * * Events are: * start array, stop array, error, add device, remove device, * start build, activate spare */ static DECLARE_WAIT_QUEUE_HEAD(md_event_waiters); static atomic_t md_event_count; void md_new_event(struct mddev *mddev) { atomic_inc(&md_event_count); wake_up(&md_event_waiters); } EXPORT_SYMBOL_GPL(md_new_event); /* Alternate version that can be called from interrupts * when calling sysfs_notify isn't needed. */ static void md_new_event_inintr(struct mddev *mddev) { atomic_inc(&md_event_count); wake_up(&md_event_waiters); } /* * Enables to iterate over all existing md arrays * all_mddevs_lock protects this list. */ static LIST_HEAD(all_mddevs); static DEFINE_SPINLOCK(all_mddevs_lock); /* * iterates through all used mddevs in the system. * We take care to grab the all_mddevs_lock whenever navigating * the list, and to always hold a refcount when unlocked. * Any code which breaks out of this loop while own * a reference to the current mddev and must mddev_put it. */ #define for_each_mddev(_mddev,_tmp) \ \ for (({ spin_lock(&all_mddevs_lock); \ _tmp = all_mddevs.next; \ _mddev = NULL;}); \ ({ if (_tmp != &all_mddevs) \ mddev_get(list_entry(_tmp, struct mddev, all_mddevs));\ spin_unlock(&all_mddevs_lock); \ if (_mddev) mddev_put(_mddev); \ _mddev = list_entry(_tmp, struct mddev, all_mddevs); \ _tmp != &all_mddevs;}); \ ({ spin_lock(&all_mddevs_lock); \ _tmp = _tmp->next;}) \ ) /* Rather than calling directly into the personality make_request function, * IO requests come here first so that we can check if the device is * being suspended pending a reconfiguration. * We hold a refcount over the call to ->make_request. By the time that * call has finished, the bio has been linked into some internal structure * and so is visible to ->quiesce(), so we don't need the refcount any more. */ static void md_make_request(struct request_queue *q, struct bio *bio) { const int rw = bio_data_dir(bio); struct mddev *mddev = q->queuedata; int cpu; unsigned int sectors; if (mddev == NULL || mddev->pers == NULL || !mddev->ready) { bio_io_error(bio); return; } smp_rmb(); /* Ensure implications of 'active' are visible */ rcu_read_lock(); if (mddev->suspended) { DEFINE_WAIT(__wait); for (;;) { prepare_to_wait(&mddev->sb_wait, &__wait, TASK_UNINTERRUPTIBLE); if (!mddev->suspended) break; rcu_read_unlock(); schedule(); rcu_read_lock(); } finish_wait(&mddev->sb_wait, &__wait); } atomic_inc(&mddev->active_io); rcu_read_unlock(); /* * save the sectors now since our bio can * go away inside make_request */ sectors = bio_sectors(bio); mddev->pers->make_request(mddev, bio); cpu = part_stat_lock(); part_stat_inc(cpu, &mddev->gendisk->part0, ios[rw]); part_stat_add(cpu, &mddev->gendisk->part0, sectors[rw], sectors); part_stat_unlock(); if (atomic_dec_and_test(&mddev->active_io) && mddev->suspended) wake_up(&mddev->sb_wait); } /* mddev_suspend makes sure no new requests are submitted * to the device, and that any requests that have been submitted * are completely handled. * Once ->stop is called and completes, the module will be completely * unused. */ void mddev_suspend(struct mddev *mddev) { BUG_ON(mddev->suspended); mddev->suspended = 1; synchronize_rcu(); wait_event(mddev->sb_wait, atomic_read(&mddev->active_io) == 0); mddev->pers->quiesce(mddev, 1); del_timer_sync(&mddev->safemode_timer); } EXPORT_SYMBOL_GPL(mddev_suspend); void mddev_resume(struct mddev *mddev) { mddev->suspended = 0; wake_up(&mddev->sb_wait); mddev->pers->quiesce(mddev, 0); set_bit(MD_RECOVERY_NEEDED, &mddev->recovery); md_wakeup_thread(mddev->thread); md_wakeup_thread(mddev->sync_thread); /* possibly kick off a reshape */ } EXPORT_SYMBOL_GPL(mddev_resume); int mddev_congested(struct mddev *mddev, int bits) { return mddev->suspended; } EXPORT_SYMBOL(mddev_congested); /* * Generic flush handling for md */ static void md_end_flush(struct bio *bio, int err) { struct md_rdev *rdev = bio->bi_private; struct mddev *mddev = rdev->mddev; rdev_dec_pending(rdev, mddev); if (atomic_dec_and_test(&mddev->flush_pending)) { /* The pre-request flush has finished */ queue_work(md_wq, &mddev->flush_work); } bio_put(bio); } static void md_submit_flush_data(struct work_struct *ws); static void submit_flushes(struct work_struct *ws) { struct mddev *mddev = container_of(ws, struct mddev, flush_work); struct md_rdev *rdev; INIT_WORK(&mddev->flush_work, md_submit_flush_data); atomic_set(&mddev->flush_pending, 1); rcu_read_lock(); rdev_for_each_rcu(rdev, mddev) if (rdev->raid_disk >= 0 && !test_bit(Faulty, &rdev->flags)) { /* Take two references, one is dropped * when request finishes, one after * we reclaim rcu_read_lock */ struct bio *bi; atomic_inc(&rdev->nr_pending); atomic_inc(&rdev->nr_pending); rcu_read_unlock(); bi = bio_alloc_mddev(GFP_NOIO, 0, mddev); bi->bi_end_io = md_end_flush; bi->bi_private = rdev; bi->bi_bdev = rdev->bdev; atomic_inc(&mddev->flush_pending); submit_bio(WRITE_FLUSH, bi); rcu_read_lock(); rdev_dec_pending(rdev, mddev); } rcu_read_unlock(); if (atomic_dec_and_test(&mddev->flush_pending)) queue_work(md_wq, &mddev->flush_work); } static void md_submit_flush_data(struct work_struct *ws) { struct mddev *mddev = container_of(ws, struct mddev, flush_work); struct bio *bio = mddev->flush_bio; if (bio->bi_size == 0) /* an empty barrier - all done */ bio_endio(bio, 0); else { bio->bi_rw &= ~REQ_FLUSH; mddev->pers->make_request(mddev, bio); } mddev->flush_bio = NULL; wake_up(&mddev->sb_wait); } void md_flush_request(struct mddev *mddev, struct bio *bio) { spin_lock_irq(&mddev->write_lock); wait_event_lock_irq(mddev->sb_wait, !mddev->flush_bio, mddev->write_lock, /*nothing*/); mddev->flush_bio = bio; spin_unlock_irq(&mddev->write_lock); INIT_WORK(&mddev->flush_work, submit_flushes); queue_work(md_wq, &mddev->flush_work); } EXPORT_SYMBOL(md_flush_request); /* Support for plugging. * This mirrors the plugging support in request_queue, but does not * require having a whole queue or request structures. * We allocate an md_plug_cb for each md device and each thread it gets * plugged on. This links tot the private plug_handle structure in the * personality data where we keep a count of the number of outstanding * plugs so other code can see if a plug is active. */ struct md_plug_cb { struct blk_plug_cb cb; struct mddev *mddev; }; static void plugger_unplug(struct blk_plug_cb *cb) { struct md_plug_cb *mdcb = container_of(cb, struct md_plug_cb, cb); if (atomic_dec_and_test(&mdcb->mddev->plug_cnt)) md_wakeup_thread(mdcb->mddev->thread); kfree(mdcb); } /* Check that an unplug wakeup will come shortly. * If not, wakeup the md thread immediately */ int mddev_check_plugged(struct mddev *mddev) { struct blk_plug *plug = current->plug; struct md_plug_cb *mdcb; if (!plug) return 0; list_for_each_entry(mdcb, &plug->cb_list, cb.list) { if (mdcb->cb.callback == plugger_unplug && mdcb->mddev == mddev) { /* Already on the list, move to top */ if (mdcb != list_first_entry(&plug->cb_list, struct md_plug_cb, cb.list)) list_move(&mdcb->cb.list, &plug->cb_list); return 1; } } /* Not currently on the callback list */ mdcb = kmalloc(sizeof(*mdcb), GFP_ATOMIC); if (!mdcb) return 0; mdcb->mddev = mddev; mdcb->cb.callback = plugger_unplug; atomic_inc(&mddev->plug_cnt); list_add(&mdcb->cb.list, &plug->cb_list); return 1; } EXPORT_SYMBOL_GPL(mddev_check_plugged); static inline struct mddev *mddev_get(struct mddev *mddev) { atomic_inc(&mddev->active); return mddev; } static void mddev_delayed_delete(struct work_struct *ws); static void mddev_put(struct mddev *mddev) { struct bio_set *bs = NULL; if (!atomic_dec_and_lock(&mddev->active, &all_mddevs_lock)) return; if (!mddev->raid_disks && list_empty(&mddev->disks) && mddev->ctime == 0 && !mddev->hold_active) { /* Array is not configured at all, and not held active, * so destroy it */ list_del_init(&mddev->all_mddevs); bs = mddev->bio_set; mddev->bio_set = NULL; if (mddev->gendisk) { /* We did a probe so need to clean up. Call * queue_work inside the spinlock so that * flush_workqueue() after mddev_find will * succeed in waiting for the work to be done. */ INIT_WORK(&mddev->del_work, mddev_delayed_delete); queue_work(md_misc_wq, &mddev->del_work); } else kfree(mddev); } spin_unlock(&all_mddevs_lock); if (bs) bioset_free(bs); } void mddev_init(struct mddev *mddev) { mutex_init(&mddev->open_mutex); mutex_init(&mddev->reconfig_mutex); mutex_init(&mddev->bitmap_info.mutex); INIT_LIST_HEAD(&mddev->disks); INIT_LIST_HEAD(&mddev->all_mddevs); init_timer(&mddev->safemode_timer); atomic_set(&mddev->active, 1); atomic_set(&mddev->openers, 0); atomic_set(&mddev->active_io, 0); atomic_set(&mddev->plug_cnt, 0); spin_lock_init(&mddev->write_lock); atomic_set(&mddev->flush_pending, 0); init_waitqueue_head(&mddev->sb_wait); init_waitqueue_head(&mddev->recovery_wait); mddev->reshape_position = MaxSector; mddev->reshape_backwards = 0; mddev->resync_min = 0; mddev->resync_max = MaxSector; mddev->level = LEVEL_NONE; } EXPORT_SYMBOL_GPL(mddev_init); static struct mddev * mddev_find(dev_t unit) { struct mddev *mddev, *new = NULL; if (unit && MAJOR(unit) != MD_MAJOR) unit &= ~((1<unit == unit) { mddev_get(mddev); spin_unlock(&all_mddevs_lock); kfree(new); return mddev; } if (new) { list_add(&new->all_mddevs, &all_mddevs); spin_unlock(&all_mddevs_lock); new->hold_active = UNTIL_IOCTL; return new; } } else if (new) { /* find an unused unit number */ static int next_minor = 512; int start = next_minor; int is_free = 0; int dev = 0; while (!is_free) { dev = MKDEV(MD_MAJOR, next_minor); next_minor++; if (next_minor > MINORMASK) next_minor = 0; if (next_minor == start) { /* Oh dear, all in use. */ spin_unlock(&all_mddevs_lock); kfree(new); return NULL; } is_free = 1; list_for_each_entry(mddev, &all_mddevs, all_mddevs) if (mddev->unit == dev) { is_free = 0; break; } } new->unit = dev; new->md_minor = MINOR(dev); new->hold_active = UNTIL_STOP; list_add(&new->all_mddevs, &all_mddevs); spin_unlock(&all_mddevs_lock); return new; } spin_unlock(&all_mddevs_lock); new = kzalloc(sizeof(*new), GFP_KERNEL); if (!new) return NULL; new->unit = unit; if (MAJOR(unit) == MD_MAJOR) new->md_minor = MINOR(unit); else new->md_minor = MINOR(unit) >> MdpMinorShift; mddev_init(new); goto retry; } static inline int mddev_lock(struct mddev * mddev) { return mutex_lock_interruptible(&mddev->reconfig_mutex); } static inline int mddev_is_locked(struct mddev *mddev) { return mutex_is_locked(&mddev->reconfig_mutex); } static inline int mddev_trylock(struct mddev * mddev) { return mutex_trylock(&mddev->reconfig_mutex); } static struct attribute_group md_redundancy_group; static void mddev_unlock(struct mddev * mddev) { if (mddev->to_remove) { /* These cannot be removed under reconfig_mutex as * an access to the files will try to take reconfig_mutex * while holding the file unremovable, which leads to * a deadlock. * So hold set sysfs_active while the remove in happeing, * and anything else which might set ->to_remove or my * otherwise change the sysfs namespace will fail with * -EBUSY if sysfs_active is still set. * We set sysfs_active under reconfig_mutex and elsewhere * test it under the same mutex to ensure its correct value * is seen. */ struct attribute_group *to_remove = mddev->to_remove; mddev->to_remove = NULL; mddev->sysfs_active = 1; mutex_unlock(&mddev->reconfig_mutex); if (mddev->kobj.sd) { if (to_remove != &md_redundancy_group) sysfs_remove_group(&mddev->kobj, to_remove); if (mddev->pers == NULL || mddev->pers->sync_request == NULL) { sysfs_remove_group(&mddev->kobj, &md_redundancy_group); if (mddev->sysfs_action) sysfs_put(mddev->sysfs_action); mddev->sysfs_action = NULL; } } mddev->sysfs_active = 0; } else mutex_unlock(&mddev->reconfig_mutex); /* As we've dropped the mutex we need a spinlock to * make sure the thread doesn't disappear */ spin_lock(&pers_lock); md_wakeup_thread(mddev->thread); spin_unlock(&pers_lock); } static struct md_rdev * find_rdev_nr(struct mddev *mddev, int nr) { struct md_rdev *rdev; rdev_for_each(rdev, mddev) if (rdev->desc_nr == nr) return rdev; return NULL; } static struct md_rdev * find_rdev(struct mddev * mddev, dev_t dev) { struct md_rdev *rdev; rdev_for_each(rdev, mddev) if (rdev->bdev->bd_dev == dev) return rdev; return NULL; } static struct md_personality *find_pers(int level, char *clevel) { struct md_personality *pers; list_for_each_entry(pers, &pers_list, list) { if (level != LEVEL_NONE && pers->level == level) return pers; if (strcmp(pers->name, clevel)==0) return pers; } return NULL; } /* return the offset of the super block in 512byte sectors */ static inline sector_t calc_dev_sboffset(struct md_rdev *rdev) { sector_t num_sectors = i_size_read(rdev->bdev->bd_inode) / 512; return MD_NEW_SIZE_SECTORS(num_sectors); } static int alloc_disk_sb(struct md_rdev * rdev) { if (rdev->sb_page) MD_BUG(); rdev->sb_page = alloc_page(GFP_KERNEL); if (!rdev->sb_page) { printk(KERN_ALERT "md: out of memory.\n"); return -ENOMEM; } return 0; } void md_rdev_clear(struct md_rdev *rdev) { if (rdev->sb_page) { put_page(rdev->sb_page); rdev->sb_loaded = 0; rdev->sb_page = NULL; rdev->sb_start = 0; rdev->sectors = 0; } if (rdev->bb_page) { put_page(rdev->bb_page); rdev->bb_page = NULL; } kfree(rdev->badblocks.page); rdev->badblocks.page = NULL; } EXPORT_SYMBOL_GPL(md_rdev_clear); static void super_written(struct bio *bio, int error) { struct md_rdev *rdev = bio->bi_private; struct mddev *mddev = rdev->mddev; if (error || !test_bit(BIO_UPTODATE, &bio->bi_flags)) { printk("md: super_written gets error=%d, uptodate=%d\n", error, test_bit(BIO_UPTODATE, &bio->bi_flags)); WARN_ON(test_bit(BIO_UPTODATE, &bio->bi_flags)); md_error(mddev, rdev); } if (atomic_dec_and_test(&mddev->pending_writes)) wake_up(&mddev->sb_wait); bio_put(bio); } void md_super_write(struct mddev *mddev, struct md_rdev *rdev, sector_t sector, int size, struct page *page) { /* write first size bytes of page to sector of rdev * Increment mddev->pending_writes before returning * and decrement it on completion, waking up sb_wait * if zero is reached. * If an error occurred, call md_error */ struct bio *bio = bio_alloc_mddev(GFP_NOIO, 1, mddev); bio->bi_bdev = rdev->meta_bdev ? rdev->meta_bdev : rdev->bdev; bio->bi_sector = sector; bio_add_page(bio, page, size, 0); bio->bi_private = rdev; bio->bi_end_io = super_written; atomic_inc(&mddev->pending_writes); submit_bio(WRITE_FLUSH_FUA, bio); } void md_super_wait(struct mddev *mddev) { /* wait for all superblock writes that were scheduled to complete */ DEFINE_WAIT(wq); for(;;) { prepare_to_wait(&mddev->sb_wait, &wq, TASK_UNINTERRUPTIBLE); if (atomic_read(&mddev->pending_writes)==0) break; schedule(); } finish_wait(&mddev->sb_wait, &wq); } static void bi_complete(struct bio *bio, int error) { complete((struct completion*)bio->bi_private); } int sync_page_io(struct md_rdev *rdev, sector_t sector, int size, struct page *page, int rw, bool metadata_op) { struct bio *bio = bio_alloc_mddev(GFP_NOIO, 1, rdev->mddev); struct completion event; int ret; rw |= REQ_SYNC; bio->bi_bdev = (metadata_op && rdev->meta_bdev) ? rdev->meta_bdev : rdev->bdev; if (metadata_op) bio->bi_sector = sector + rdev->sb_start; else if (rdev->mddev->reshape_position != MaxSector && (rdev->mddev->reshape_backwards == (sector >= rdev->mddev->reshape_position))) bio->bi_sector = sector + rdev->new_data_offset; else bio->bi_sector = sector + rdev->data_offset; bio_add_page(bio, page, size, 0); init_completion(&event); bio->bi_private = &event; bio->bi_end_io = bi_complete; submit_bio(rw, bio); wait_for_completion(&event); ret = test_bit(BIO_UPTODATE, &bio->bi_flags); bio_put(bio); return ret; } EXPORT_SYMBOL_GPL(sync_page_io); static int read_disk_sb(struct md_rdev * rdev, int size) { char b[BDEVNAME_SIZE]; if (!rdev->sb_page) { MD_BUG(); return -EINVAL; } if (rdev->sb_loaded) return 0; if (!sync_page_io(rdev, 0, size, rdev->sb_page, READ, true)) goto fail; rdev->sb_loaded = 1; return 0; fail: printk(KERN_WARNING "md: disabled device %s, could not read superblock.\n", bdevname(rdev->bdev,b)); return -EINVAL; } static int uuid_equal(mdp_super_t *sb1, mdp_super_t *sb2) { return sb1->set_uuid0 == sb2->set_uuid0 && sb1->set_uuid1 == sb2->set_uuid1 && sb1->set_uuid2 == sb2->set_uuid2 && sb1->set_uuid3 == sb2->set_uuid3; } static int sb_equal(mdp_super_t *sb1, mdp_super_t *sb2) { int ret; mdp_super_t *tmp1, *tmp2; tmp1 = kmalloc(sizeof(*tmp1),GFP_KERNEL); tmp2 = kmalloc(sizeof(*tmp2),GFP_KERNEL); if (!tmp1 || !tmp2) { ret = 0; printk(KERN_INFO "md.c sb_equal(): failed to allocate memory!\n"); goto abort; } *tmp1 = *sb1; *tmp2 = *sb2; /* * nr_disks is not constant */ tmp1->nr_disks = 0; tmp2->nr_disks = 0; ret = (memcmp(tmp1, tmp2, MD_SB_GENERIC_CONSTANT_WORDS * 4) == 0); abort: kfree(tmp1); kfree(tmp2); return ret; } static u32 md_csum_fold(u32 csum) { csum = (csum & 0xffff) + (csum >> 16); return (csum & 0xffff) + (csum >> 16); } static unsigned int calc_sb_csum(mdp_super_t * sb) { u64 newcsum = 0; u32 *sb32 = (u32*)sb; int i; unsigned int disk_csum, csum; disk_csum = sb->sb_csum; sb->sb_csum = 0; for (i = 0; i < MD_SB_BYTES/4 ; i++) newcsum += sb32[i]; csum = (newcsum & 0xffffffff) + (newcsum>>32); #ifdef CONFIG_ALPHA /* This used to use csum_partial, which was wrong for several * reasons including that different results are returned on * different architectures. It isn't critical that we get exactly * the same return value as before (we always csum_fold before * testing, and that removes any differences). However as we * know that csum_partial always returned a 16bit value on * alphas, do a fold to maximise conformity to previous behaviour. */ sb->sb_csum = md_csum_fold(disk_csum); #else sb->sb_csum = disk_csum; #endif return csum; } /* * Handle superblock details. * We want to be able to handle multiple superblock formats * so we have a common interface to them all, and an array of * different handlers. * We rely on user-space to write the initial superblock, and support * reading and updating of superblocks. * Interface methods are: * int load_super(struct md_rdev *dev, struct md_rdev *refdev, int minor_version) * loads and validates a superblock on dev. * if refdev != NULL, compare superblocks on both devices * Return: * 0 - dev has a superblock that is compatible with refdev * 1 - dev has a superblock that is compatible and newer than refdev * so dev should be used as the refdev in future * -EINVAL superblock incompatible or invalid * -othererror e.g. -EIO * * int validate_super(struct mddev *mddev, struct md_rdev *dev) * Verify that dev is acceptable into mddev. * The first time, mddev->raid_disks will be 0, and data from * dev should be merged in. Subsequent calls check that dev * is new enough. Return 0 or -EINVAL * * void sync_super(struct mddev *mddev, struct md_rdev *dev) * Update the superblock for rdev with data in mddev * This does not write to disc. * */ struct super_type { char *name; struct module *owner; int (*load_super)(struct md_rdev *rdev, struct md_rdev *refdev, int minor_version); int (*validate_super)(struct mddev *mddev, struct md_rdev *rdev); void (*sync_super)(struct mddev *mddev, struct md_rdev *rdev); unsigned long long (*rdev_size_change)(struct md_rdev *rdev, sector_t num_sectors); int (*allow_new_offset)(struct md_rdev *rdev, unsigned long long new_offset); }; /* * Check that the given mddev has no bitmap. * * This function is called from the run method of all personalities that do not * support bitmaps. It prints an error message and returns non-zero if mddev * has a bitmap. Otherwise, it returns 0. * */ int md_check_no_bitmap(struct mddev *mddev) { if (!mddev->bitmap_info.file && !mddev->bitmap_info.offset) return 0; printk(KERN_ERR "%s: bitmaps are not supported for %s\n", mdname(mddev), mddev->pers->name); return 1; } EXPORT_SYMBOL(md_check_no_bitmap); /* * load_super for 0.90.0 */ static int super_90_load(struct md_rdev *rdev, struct md_rdev *refdev, int minor_version) { char b[BDEVNAME_SIZE], b2[BDEVNAME_SIZE]; mdp_super_t *sb; int ret; /* * Calculate the position of the superblock (512byte sectors), * it's at the end of the disk. * * It also happens to be a multiple of 4Kb. */ rdev->sb_start = calc_dev_sboffset(rdev); ret = read_disk_sb(rdev, MD_SB_BYTES); if (ret) return ret; ret = -EINVAL; bdevname(rdev->bdev, b); sb = page_address(rdev->sb_page); if (sb->md_magic != MD_SB_MAGIC) { printk(KERN_ERR "md: invalid raid superblock magic on %s\n", b); goto abort; } if (sb->major_version != 0 || sb->minor_version < 90 || sb->minor_version > 91) { printk(KERN_WARNING "Bad version number %d.%d on %s\n", sb->major_version, sb->minor_version, b); goto abort; } if (sb->raid_disks <= 0) goto abort; if (md_csum_fold(calc_sb_csum(sb)) != md_csum_fold(sb->sb_csum)) { printk(KERN_WARNING "md: invalid superblock checksum on %s\n", b); goto abort; } rdev->preferred_minor = sb->md_minor; rdev->data_offset = 0; rdev->new_data_offset = 0; rdev->sb_size = MD_SB_BYTES; rdev->badblocks.shift = -1; if (sb->level == LEVEL_MULTIPATH) rdev->desc_nr = -1; else rdev->desc_nr = sb->this_disk.number; if (!refdev) { ret = 1; } else { __u64 ev1, ev2; mdp_super_t *refsb = page_address(refdev->sb_page); if (!uuid_equal(refsb, sb)) { printk(KERN_WARNING "md: %s has different UUID to %s\n", b, bdevname(refdev->bdev,b2)); goto abort; } if (!sb_equal(refsb, sb)) { printk(KERN_WARNING "md: %s has same UUID" " but different superblock to %s\n", b, bdevname(refdev->bdev, b2)); goto abort; } ev1 = md_event(sb); ev2 = md_event(refsb); if (ev1 > ev2) ret = 1; else ret = 0; } rdev->sectors = rdev->sb_start; /* Limit to 4TB as metadata cannot record more than that */ if (rdev->sectors >= (2ULL << 32)) rdev->sectors = (2ULL << 32) - 2; if (rdev->sectors < ((sector_t)sb->size) * 2 && sb->level >= 1) /* "this cannot possibly happen" ... */ ret = -EINVAL; abort: return ret; } /* * validate_super for 0.90.0 */ static int super_90_validate(struct mddev *mddev, struct md_rdev *rdev) { mdp_disk_t *desc; mdp_super_t *sb = page_address(rdev->sb_page); __u64 ev1 = md_event(sb); rdev->raid_disk = -1; clear_bit(Faulty, &rdev->flags); clear_bit(In_sync, &rdev->flags); clear_bit(WriteMostly, &rdev->flags); if (mddev->raid_disks == 0) { mddev->major_version = 0; mddev->minor_version = sb->minor_version; mddev->patch_version = sb->patch_version; mddev->external = 0; mddev->chunk_sectors = sb->chunk_size >> 9; mddev->ctime = sb->ctime; mddev->utime = sb->utime; mddev->level = sb->level; mddev->clevel[0] = 0; mddev->layout = sb->layout; mddev->raid_disks = sb->raid_disks; mddev->dev_sectors = ((sector_t)sb->size) * 2; mddev->events = ev1; mddev->bitmap_info.offset = 0; mddev->bitmap_info.space = 0; /* bitmap can use 60 K after the 4K superblocks */ mddev->bitmap_info.default_offset = MD_SB_BYTES >> 9; mddev->bitmap_info.default_space = 64*2 - (MD_SB_BYTES >> 9); mddev->reshape_backwards = 0; if (mddev->minor_version >= 91) { mddev->reshape_position = sb->reshape_position; mddev->delta_disks = sb->delta_disks; mddev->new_level = sb->new_level; mddev->new_layout = sb->new_layout; mddev->new_chunk_sectors = sb->new_chunk >> 9; if (mddev->delta_disks < 0) mddev->reshape_backwards = 1; } else { mddev->reshape_position = MaxSector; mddev->delta_disks = 0; mddev->new_level = mddev->level; mddev->new_layout = mddev->layout; mddev->new_chunk_sectors = mddev->chunk_sectors; } if (sb->state & (1<recovery_cp = MaxSector; else { if (sb->events_hi == sb->cp_events_hi && sb->events_lo == sb->cp_events_lo) { mddev->recovery_cp = sb->recovery_cp; } else mddev->recovery_cp = 0; } memcpy(mddev->uuid+0, &sb->set_uuid0, 4); memcpy(mddev->uuid+4, &sb->set_uuid1, 4); memcpy(mddev->uuid+8, &sb->set_uuid2, 4); memcpy(mddev->uuid+12,&sb->set_uuid3, 4); mddev->max_disks = MD_SB_DISKS; if (sb->state & (1<bitmap_info.file == NULL) { mddev->bitmap_info.offset = mddev->bitmap_info.default_offset; mddev->bitmap_info.space = mddev->bitmap_info.space; } } else if (mddev->pers == NULL) { /* Insist on good event counter while assembling, except * for spares (which don't need an event count) */ ++ev1; if (sb->disks[rdev->desc_nr].state & ( (1<events) return -EINVAL; } else if (mddev->bitmap) { /* if adding to array with a bitmap, then we can accept an * older device ... but not too old. */ if (ev1 < mddev->bitmap->events_cleared) return 0; } else { if (ev1 < mddev->events) /* just a hot-add of a new device, leave raid_disk at -1 */ return 0; } if (mddev->level != LEVEL_MULTIPATH) { desc = sb->disks + rdev->desc_nr; if (desc->state & (1<flags); else if (desc->state & (1<raid_disk < mddev->raid_disks */) { set_bit(In_sync, &rdev->flags); rdev->raid_disk = desc->raid_disk; } else if (desc->state & (1<minor_version >= 91) { rdev->recovery_offset = 0; rdev->raid_disk = desc->raid_disk; } } if (desc->state & (1<flags); } else /* MULTIPATH are always insync */ set_bit(In_sync, &rdev->flags); return 0; } /* * sync_super for 0.90.0 */ static void super_90_sync(struct mddev *mddev, struct md_rdev *rdev) { mdp_super_t *sb; struct md_rdev *rdev2; int next_spare = mddev->raid_disks; /* make rdev->sb match mddev data.. * * 1/ zero out disks * 2/ Add info for each disk, keeping track of highest desc_nr (next_spare); * 3/ any empty disks < next_spare become removed * * disks[0] gets initialised to REMOVED because * we cannot be sure from other fields if it has * been initialised or not. */ int i; int active=0, working=0,failed=0,spare=0,nr_disks=0; rdev->sb_size = MD_SB_BYTES; sb = page_address(rdev->sb_page); memset(sb, 0, sizeof(*sb)); sb->md_magic = MD_SB_MAGIC; sb->major_version = mddev->major_version; sb->patch_version = mddev->patch_version; sb->gvalid_words = 0; /* ignored */ memcpy(&sb->set_uuid0, mddev->uuid+0, 4); memcpy(&sb->set_uuid1, mddev->uuid+4, 4); memcpy(&sb->set_uuid2, mddev->uuid+8, 4); memcpy(&sb->set_uuid3, mddev->uuid+12,4); sb->ctime = mddev->ctime; sb->level = mddev->level; sb->size = mddev->dev_sectors / 2; sb->raid_disks = mddev->raid_disks; sb->md_minor = mddev->md_minor; sb->not_persistent = 0; sb->utime = mddev->utime; sb->state = 0; sb->events_hi = (mddev->events>>32); sb->events_lo = (u32)mddev->events; if (mddev->reshape_position == MaxSector) sb->minor_version = 90; else { sb->minor_version = 91; sb->reshape_position = mddev->reshape_position; sb->new_level = mddev->new_level; sb->delta_disks = mddev->delta_disks; sb->new_layout = mddev->new_layout; sb->new_chunk = mddev->new_chunk_sectors << 9; } mddev->minor_version = sb->minor_version; if (mddev->in_sync) { sb->recovery_cp = mddev->recovery_cp; sb->cp_events_hi = (mddev->events>>32); sb->cp_events_lo = (u32)mddev->events; if (mddev->recovery_cp == MaxSector) sb->state = (1<< MD_SB_CLEAN); } else sb->recovery_cp = 0; sb->layout = mddev->layout; sb->chunk_size = mddev->chunk_sectors << 9; if (mddev->bitmap && mddev->bitmap_info.file == NULL) sb->state |= (1<disks[0].state = (1<flags); if (rdev2->raid_disk >= 0 && sb->minor_version >= 91) /* we have nowhere to store the recovery_offset, * but if it is not below the reshape_position, * we can piggy-back on that. */ is_active = 1; if (rdev2->raid_disk < 0 || test_bit(Faulty, &rdev2->flags)) is_active = 0; if (is_active) desc_nr = rdev2->raid_disk; else desc_nr = next_spare++; rdev2->desc_nr = desc_nr; d = &sb->disks[rdev2->desc_nr]; nr_disks++; d->number = rdev2->desc_nr; d->major = MAJOR(rdev2->bdev->bd_dev); d->minor = MINOR(rdev2->bdev->bd_dev); if (is_active) d->raid_disk = rdev2->raid_disk; else d->raid_disk = rdev2->desc_nr; /* compatibility */ if (test_bit(Faulty, &rdev2->flags)) d->state = (1<state = (1<flags)) d->state |= (1<state = 0; spare++; working++; } if (test_bit(WriteMostly, &rdev2->flags)) d->state |= (1<raid_disks ; i++) { mdp_disk_t *d = &sb->disks[i]; if (d->state == 0 && d->number == 0) { d->number = i; d->raid_disk = i; d->state = (1<state |= (1<nr_disks = nr_disks; sb->active_disks = active; sb->working_disks = working; sb->failed_disks = failed; sb->spare_disks = spare; sb->this_disk = sb->disks[rdev->desc_nr]; sb->sb_csum = calc_sb_csum(sb); } /* * rdev_size_change for 0.90.0 */ static unsigned long long super_90_rdev_size_change(struct md_rdev *rdev, sector_t num_sectors) { if (num_sectors && num_sectors < rdev->mddev->dev_sectors) return 0; /* component must fit device */ if (rdev->mddev->bitmap_info.offset) return 0; /* can't move bitmap */ rdev->sb_start = calc_dev_sboffset(rdev); if (!num_sectors || num_sectors > rdev->sb_start) num_sectors = rdev->sb_start; /* Limit to 4TB as metadata cannot record more than that. * 4TB == 2^32 KB, or 2*2^32 sectors. */ if (num_sectors >= (2ULL << 32)) num_sectors = (2ULL << 32) - 2; md_super_write(rdev->mddev, rdev, rdev->sb_start, rdev->sb_size, rdev->sb_page); md_super_wait(rdev->mddev); return num_sectors; } static int super_90_allow_new_offset(struct md_rdev *rdev, unsigned long long new_offset) { /* non-zero offset changes not possible with v0.90 */ return new_offset == 0; } /* * version 1 superblock */ static __le32 calc_sb_1_csum(struct mdp_superblock_1 * sb) { __le32 disk_csum; u32 csum; unsigned long long newcsum; int size = 256 + le32_to_cpu(sb->max_dev)*2; __le32 *isuper = (__le32*)sb; int i; disk_csum = sb->sb_csum; sb->sb_csum = 0; newcsum = 0; for (i=0; size>=4; size -= 4 ) newcsum += le32_to_cpu(*isuper++); if (size == 2) newcsum += le16_to_cpu(*(__le16*) isuper); csum = (newcsum & 0xffffffff) + (newcsum >> 32); sb->sb_csum = disk_csum; return cpu_to_le32(csum); } static int md_set_badblocks(struct badblocks *bb, sector_t s, int sectors, int acknowledged); static int super_1_load(struct md_rdev *rdev, struct md_rdev *refdev, int minor_version) { struct mdp_superblock_1 *sb; int ret; sector_t sb_start; sector_t sectors; char b[BDEVNAME_SIZE], b2[BDEVNAME_SIZE]; int bmask; /* * Calculate the position of the superblock in 512byte sectors. * It is always aligned to a 4K boundary and * depeding on minor_version, it can be: * 0: At least 8K, but less than 12K, from end of device * 1: At start of device * 2: 4K from start of device. */ switch(minor_version) { case 0: sb_start = i_size_read(rdev->bdev->bd_inode) >> 9; sb_start -= 8*2; sb_start &= ~(sector_t)(4*2-1); break; case 1: sb_start = 0; break; case 2: sb_start = 8; break; default: return -EINVAL; } rdev->sb_start = sb_start; /* superblock is rarely larger than 1K, but it can be larger, * and it is safe to read 4k, so we do that */ ret = read_disk_sb(rdev, 4096); if (ret) return ret; sb = page_address(rdev->sb_page); if (sb->magic != cpu_to_le32(MD_SB_MAGIC) || sb->major_version != cpu_to_le32(1) || le32_to_cpu(sb->max_dev) > (4096-256)/2 || le64_to_cpu(sb->super_offset) != rdev->sb_start || (le32_to_cpu(sb->feature_map) & ~MD_FEATURE_ALL) != 0) return -EINVAL; if (calc_sb_1_csum(sb) != sb->sb_csum) { printk("md: invalid superblock checksum on %s\n", bdevname(rdev->bdev,b)); return -EINVAL; } if (le64_to_cpu(sb->data_size) < 10) { printk("md: data_size too small on %s\n", bdevname(rdev->bdev,b)); return -EINVAL; } if (sb->pad0 || sb->pad3[0] || memcmp(sb->pad3, sb->pad3+1, sizeof(sb->pad3) - sizeof(sb->pad3[1]))) /* Some padding is non-zero, might be a new feature */ return -EINVAL; rdev->preferred_minor = 0xffff; rdev->data_offset = le64_to_cpu(sb->data_offset); rdev->new_data_offset = rdev->data_offset; if ((le32_to_cpu(sb->feature_map) & MD_FEATURE_RESHAPE_ACTIVE) && (le32_to_cpu(sb->feature_map) & MD_FEATURE_NEW_OFFSET)) rdev->new_data_offset += (s32)le32_to_cpu(sb->new_offset); atomic_set(&rdev->corrected_errors, le32_to_cpu(sb->cnt_corrected_read)); rdev->sb_size = le32_to_cpu(sb->max_dev) * 2 + 256; bmask = queue_logical_block_size(rdev->bdev->bd_disk->queue)-1; if (rdev->sb_size & bmask) rdev->sb_size = (rdev->sb_size | bmask) + 1; if (minor_version && rdev->data_offset < sb_start + (rdev->sb_size/512)) return -EINVAL; if (minor_version && rdev->new_data_offset < sb_start + (rdev->sb_size/512)) return -EINVAL; if (sb->level == cpu_to_le32(LEVEL_MULTIPATH)) rdev->desc_nr = -1; else rdev->desc_nr = le32_to_cpu(sb->dev_number); if (!rdev->bb_page) { rdev->bb_page = alloc_page(GFP_KERNEL); if (!rdev->bb_page) return -ENOMEM; } if ((le32_to_cpu(sb->feature_map) & MD_FEATURE_BAD_BLOCKS) && rdev->badblocks.count == 0) { /* need to load the bad block list. * Currently we limit it to one page. */ s32 offset; sector_t bb_sector; u64 *bbp; int i; int sectors = le16_to_cpu(sb->bblog_size); if (sectors > (PAGE_SIZE / 512)) return -EINVAL; offset = le32_to_cpu(sb->bblog_offset); if (offset == 0) return -EINVAL; bb_sector = (long long)offset; if (!sync_page_io(rdev, bb_sector, sectors << 9, rdev->bb_page, READ, true)) return -EIO; bbp = (u64 *)page_address(rdev->bb_page); rdev->badblocks.shift = sb->bblog_shift; for (i = 0 ; i < (sectors << (9-3)) ; i++, bbp++) { u64 bb = le64_to_cpu(*bbp); int count = bb & (0x3ff); u64 sector = bb >> 10; sector <<= sb->bblog_shift; count <<= sb->bblog_shift; if (bb + 1 == 0) break; if (md_set_badblocks(&rdev->badblocks, sector, count, 1) == 0) return -EINVAL; } } else if (sb->bblog_offset == 0) rdev->badblocks.shift = -1; if (!refdev) { ret = 1; } else { __u64 ev1, ev2; struct mdp_superblock_1 *refsb = page_address(refdev->sb_page); if (memcmp(sb->set_uuid, refsb->set_uuid, 16) != 0 || sb->level != refsb->level || sb->layout != refsb->layout || sb->chunksize != refsb->chunksize) { printk(KERN_WARNING "md: %s has strangely different" " superblock to %s\n", bdevname(rdev->bdev,b), bdevname(refdev->bdev,b2)); return -EINVAL; } ev1 = le64_to_cpu(sb->events); ev2 = le64_to_cpu(refsb->events); if (ev1 > ev2) ret = 1; else ret = 0; } if (minor_version) { sectors = (i_size_read(rdev->bdev->bd_inode) >> 9); sectors -= rdev->data_offset; } else sectors = rdev->sb_start; if (sectors < le64_to_cpu(sb->data_size)) return -EINVAL; rdev->sectors = le64_to_cpu(sb->data_size); return ret; } static int super_1_validate(struct mddev *mddev, struct md_rdev *rdev) { struct mdp_superblock_1 *sb = page_address(rdev->sb_page); __u64 ev1 = le64_to_cpu(sb->events); rdev->raid_disk = -1; clear_bit(Faulty, &rdev->flags); clear_bit(In_sync, &rdev->flags); clear_bit(WriteMostly, &rdev->flags); if (mddev->raid_disks == 0) { mddev->major_version = 1; mddev->patch_version = 0; mddev->external = 0; mddev->chunk_sectors = le32_to_cpu(sb->chunksize); mddev->ctime = le64_to_cpu(sb->ctime) & ((1ULL << 32)-1); mddev->utime = le64_to_cpu(sb->utime) & ((1ULL << 32)-1); mddev->level = le32_to_cpu(sb->level); mddev->clevel[0] = 0; mddev->layout = le32_to_cpu(sb->layout); mddev->raid_disks = le32_to_cpu(sb->raid_disks); mddev->dev_sectors = le64_to_cpu(sb->size); mddev->events = ev1; mddev->bitmap_info.offset = 0; mddev->bitmap_info.space = 0; /* Default location for bitmap is 1K after superblock * using 3K - total of 4K */ mddev->bitmap_info.default_offset = 1024 >> 9; mddev->bitmap_info.default_space = (4096-1024) >> 9; mddev->reshape_backwards = 0; mddev->recovery_cp = le64_to_cpu(sb->resync_offset); memcpy(mddev->uuid, sb->set_uuid, 16); mddev->max_disks = (4096-256)/2; if ((le32_to_cpu(sb->feature_map) & MD_FEATURE_BITMAP_OFFSET) && mddev->bitmap_info.file == NULL) { mddev->bitmap_info.offset = (__s32)le32_to_cpu(sb->bitmap_offset); /* Metadata doesn't record how much space is available. * For 1.0, we assume we can use up to the superblock * if before, else to 4K beyond superblock. * For others, assume no change is possible. */ if (mddev->minor_version > 0) mddev->bitmap_info.space = 0; else if (mddev->bitmap_info.offset > 0) mddev->bitmap_info.space = 8 - mddev->bitmap_info.offset; else mddev->bitmap_info.space = -mddev->bitmap_info.offset; } if ((le32_to_cpu(sb->feature_map) & MD_FEATURE_RESHAPE_ACTIVE)) { mddev->reshape_position = le64_to_cpu(sb->reshape_position); mddev->delta_disks = le32_to_cpu(sb->delta_disks); mddev->new_level = le32_to_cpu(sb->new_level); mddev->new_layout = le32_to_cpu(sb->new_layout); mddev->new_chunk_sectors = le32_to_cpu(sb->new_chunk); if (mddev->delta_disks < 0 || (mddev->delta_disks == 0 && (le32_to_cpu(sb->feature_map) & MD_FEATURE_RESHAPE_BACKWARDS))) mddev->reshape_backwards = 1; } else { mddev->reshape_position = MaxSector; mddev->delta_disks = 0; mddev->new_level = mddev->level; mddev->new_layout = mddev->layout; mddev->new_chunk_sectors = mddev->chunk_sectors; } } else if (mddev->pers == NULL) { /* Insist of good event counter while assembling, except for * spares (which don't need an event count) */ ++ev1; if (rdev->desc_nr >= 0 && rdev->desc_nr < le32_to_cpu(sb->max_dev) && le16_to_cpu(sb->dev_roles[rdev->desc_nr]) < 0xfffe) if (ev1 < mddev->events) return -EINVAL; } else if (mddev->bitmap) { /* If adding to array with a bitmap, then we can accept an * older device, but not too old. */ if (ev1 < mddev->bitmap->events_cleared) return 0; } else { if (ev1 < mddev->events) /* just a hot-add of a new device, leave raid_disk at -1 */ return 0; } if (mddev->level != LEVEL_MULTIPATH) { int role; if (rdev->desc_nr < 0 || rdev->desc_nr >= le32_to_cpu(sb->max_dev)) { role = 0xffff; rdev->desc_nr = -1; } else role = le16_to_cpu(sb->dev_roles[rdev->desc_nr]); switch(role) { case 0xffff: /* spare */ break; case 0xfffe: /* faulty */ set_bit(Faulty, &rdev->flags); break; default: if ((le32_to_cpu(sb->feature_map) & MD_FEATURE_RECOVERY_OFFSET)) rdev->recovery_offset = le64_to_cpu(sb->recovery_offset); else set_bit(In_sync, &rdev->flags); rdev->raid_disk = role; break; } if (sb->devflags & WriteMostly1) set_bit(WriteMostly, &rdev->flags); if (le32_to_cpu(sb->feature_map) & MD_FEATURE_REPLACEMENT) set_bit(Replacement, &rdev->flags); } else /* MULTIPATH are always insync */ set_bit(In_sync, &rdev->flags); return 0; } static void super_1_sync(struct mddev *mddev, struct md_rdev *rdev) { struct mdp_superblock_1 *sb; struct md_rdev *rdev2; int max_dev, i; /* make rdev->sb match mddev and rdev data. */ sb = page_address(rdev->sb_page); sb->feature_map = 0; sb->pad0 = 0; sb->recovery_offset = cpu_to_le64(0); memset(sb->pad3, 0, sizeof(sb->pad3)); sb->utime = cpu_to_le64((__u64)mddev->utime); sb->events = cpu_to_le64(mddev->events); if (mddev->in_sync) sb->resync_offset = cpu_to_le64(mddev->recovery_cp); else sb->resync_offset = cpu_to_le64(0); sb->cnt_corrected_read = cpu_to_le32(atomic_read(&rdev->corrected_errors)); sb->raid_disks = cpu_to_le32(mddev->raid_disks); sb->size = cpu_to_le64(mddev->dev_sectors); sb->chunksize = cpu_to_le32(mddev->chunk_sectors); sb->level = cpu_to_le32(mddev->level); sb->layout = cpu_to_le32(mddev->layout); if (test_bit(WriteMostly, &rdev->flags)) sb->devflags |= WriteMostly1; else sb->devflags &= ~WriteMostly1; sb->data_offset = cpu_to_le64(rdev->data_offset); sb->data_size = cpu_to_le64(rdev->sectors); if (mddev->bitmap && mddev->bitmap_info.file == NULL) { sb->bitmap_offset = cpu_to_le32((__u32)mddev->bitmap_info.offset); sb->feature_map = cpu_to_le32(MD_FEATURE_BITMAP_OFFSET); } if (rdev->raid_disk >= 0 && !test_bit(In_sync, &rdev->flags)) { sb->feature_map |= cpu_to_le32(MD_FEATURE_RECOVERY_OFFSET); sb->recovery_offset = cpu_to_le64(rdev->recovery_offset); } if (test_bit(Replacement, &rdev->flags)) sb->feature_map |= cpu_to_le32(MD_FEATURE_REPLACEMENT); if (mddev->reshape_position != MaxSector) { sb->feature_map |= cpu_to_le32(MD_FEATURE_RESHAPE_ACTIVE); sb->reshape_position = cpu_to_le64(mddev->reshape_position); sb->new_layout = cpu_to_le32(mddev->new_layout); sb->delta_disks = cpu_to_le32(mddev->delta_disks); sb->new_level = cpu_to_le32(mddev->new_level); sb->new_chunk = cpu_to_le32(mddev->new_chunk_sectors); if (mddev->delta_disks == 0 && mddev->reshape_backwards) sb->feature_map |= cpu_to_le32(MD_FEATURE_RESHAPE_BACKWARDS); if (rdev->new_data_offset != rdev->data_offset) { sb->feature_map |= cpu_to_le32(MD_FEATURE_NEW_OFFSET); sb->new_offset = cpu_to_le32((__u32)(rdev->new_data_offset - rdev->data_offset)); } } if (rdev->badblocks.count == 0) /* Nothing to do for bad blocks*/ ; else if (sb->bblog_offset == 0) /* Cannot record bad blocks on this device */ md_error(mddev, rdev); else { struct badblocks *bb = &rdev->badblocks; u64 *bbp = (u64 *)page_address(rdev->bb_page); u64 *p = bb->page; sb->feature_map |= cpu_to_le32(MD_FEATURE_BAD_BLOCKS); if (bb->changed) { unsigned seq; retry: seq = read_seqbegin(&bb->lock); memset(bbp, 0xff, PAGE_SIZE); for (i = 0 ; i < bb->count ; i++) { u64 internal_bb = *p++; u64 store_bb = ((BB_OFFSET(internal_bb) << 10) | BB_LEN(internal_bb)); *bbp++ = cpu_to_le64(store_bb); } bb->changed = 0; if (read_seqretry(&bb->lock, seq)) goto retry; bb->sector = (rdev->sb_start + (int)le32_to_cpu(sb->bblog_offset)); bb->size = le16_to_cpu(sb->bblog_size); } } max_dev = 0; rdev_for_each(rdev2, mddev) if (rdev2->desc_nr+1 > max_dev) max_dev = rdev2->desc_nr+1; if (max_dev > le32_to_cpu(sb->max_dev)) { int bmask; sb->max_dev = cpu_to_le32(max_dev); rdev->sb_size = max_dev * 2 + 256; bmask = queue_logical_block_size(rdev->bdev->bd_disk->queue)-1; if (rdev->sb_size & bmask) rdev->sb_size = (rdev->sb_size | bmask) + 1; } else max_dev = le32_to_cpu(sb->max_dev); for (i=0; idev_roles[i] = cpu_to_le16(0xfffe); rdev_for_each(rdev2, mddev) { i = rdev2->desc_nr; if (test_bit(Faulty, &rdev2->flags)) sb->dev_roles[i] = cpu_to_le16(0xfffe); else if (test_bit(In_sync, &rdev2->flags)) sb->dev_roles[i] = cpu_to_le16(rdev2->raid_disk); else if (rdev2->raid_disk >= 0) sb->dev_roles[i] = cpu_to_le16(rdev2->raid_disk); else sb->dev_roles[i] = cpu_to_le16(0xffff); } sb->sb_csum = calc_sb_1_csum(sb); } static unsigned long long super_1_rdev_size_change(struct md_rdev *rdev, sector_t num_sectors) { struct mdp_superblock_1 *sb; sector_t max_sectors; if (num_sectors && num_sectors < rdev->mddev->dev_sectors) return 0; /* component must fit device */ if (rdev->data_offset != rdev->new_data_offset) return 0; /* too confusing */ if (rdev->sb_start < rdev->data_offset) { /* minor versions 1 and 2; superblock before data */ max_sectors = i_size_read(rdev->bdev->bd_inode) >> 9; max_sectors -= rdev->data_offset; if (!num_sectors || num_sectors > max_sectors) num_sectors = max_sectors; } else if (rdev->mddev->bitmap_info.offset) { /* minor version 0 with bitmap we can't move */ return 0; } else { /* minor version 0; superblock after data */ sector_t sb_start; sb_start = (i_size_read(rdev->bdev->bd_inode) >> 9) - 8*2; sb_start &= ~(sector_t)(4*2 - 1); max_sectors = rdev->sectors + sb_start - rdev->sb_start; if (!num_sectors || num_sectors > max_sectors) num_sectors = max_sectors; rdev->sb_start = sb_start; } sb = page_address(rdev->sb_page); sb->data_size = cpu_to_le64(num_sectors); sb->super_offset = rdev->sb_start; sb->sb_csum = calc_sb_1_csum(sb); md_super_write(rdev->mddev, rdev, rdev->sb_start, rdev->sb_size, rdev->sb_page); md_super_wait(rdev->mddev); return num_sectors; } static int super_1_allow_new_offset(struct md_rdev *rdev, unsigned long long new_offset) { /* All necessary checks on new >= old have been done */ struct bitmap *bitmap; if (new_offset >= rdev->data_offset) return 1; /* with 1.0 metadata, there is no metadata to tread on * so we can always move back */ if (rdev->mddev->minor_version == 0) return 1; /* otherwise we must be sure not to step on * any metadata, so stay: * 36K beyond start of superblock * beyond end of badblocks * beyond write-intent bitmap */ if (rdev->sb_start + (32+4)*2 > new_offset) return 0; bitmap = rdev->mddev->bitmap; if (bitmap && !rdev->mddev->bitmap_info.file && rdev->sb_start + rdev->mddev->bitmap_info.offset + bitmap->storage.file_pages * (PAGE_SIZE>>9) > new_offset) return 0; if (rdev->badblocks.sector + rdev->badblocks.size > new_offset) return 0; return 1; } static struct super_type super_types[] = { [0] = { .name = "0.90.0", .owner = THIS_MODULE, .load_super = super_90_load, .validate_super = super_90_validate, .sync_super = super_90_sync, .rdev_size_change = super_90_rdev_size_change, .allow_new_offset = super_90_allow_new_offset, }, [1] = { .name = "md-1", .owner = THIS_MODULE, .load_super = super_1_load, .validate_super = super_1_validate, .sync_super = super_1_sync, .rdev_size_change = super_1_rdev_size_change, .allow_new_offset = super_1_allow_new_offset, }, }; static void sync_super(struct mddev *mddev, struct md_rdev *rdev) { if (mddev->sync_super) { mddev->sync_super(mddev, rdev); return; } BUG_ON(mddev->major_version >= ARRAY_SIZE(super_types)); super_types[mddev->major_version].sync_super(mddev, rdev); } static int match_mddev_units(struct mddev *mddev1, struct mddev *mddev2) { struct md_rdev *rdev, *rdev2; rcu_read_lock(); rdev_for_each_rcu(rdev, mddev1) rdev_for_each_rcu(rdev2, mddev2) if (rdev->bdev->bd_contains == rdev2->bdev->bd_contains) { rcu_read_unlock(); return 1; } rcu_read_unlock(); return 0; } static LIST_HEAD(pending_raid_disks); /* * Try to register data integrity profile for an mddev * * This is called when an array is started and after a disk has been kicked * from the array. It only succeeds if all working and active component devices * are integrity capable with matching profiles. */ int md_integrity_register(struct mddev *mddev) { struct md_rdev *rdev, *reference = NULL; if (list_empty(&mddev->disks)) return 0; /* nothing to do */ if (!mddev->gendisk || blk_get_integrity(mddev->gendisk)) return 0; /* shouldn't register, or already is */ rdev_for_each(rdev, mddev) { /* skip spares and non-functional disks */ if (test_bit(Faulty, &rdev->flags)) continue; if (rdev->raid_disk < 0) continue; if (!reference) { /* Use the first rdev as the reference */ reference = rdev; continue; } /* does this rdev's profile match the reference profile? */ if (blk_integrity_compare(reference->bdev->bd_disk, rdev->bdev->bd_disk) < 0) return -EINVAL; } if (!reference || !bdev_get_integrity(reference->bdev)) return 0; /* * All component devices are integrity capable and have matching * profiles, register the common profile for the md device. */ if (blk_integrity_register(mddev->gendisk, bdev_get_integrity(reference->bdev)) != 0) { printk(KERN_ERR "md: failed to register integrity for %s\n", mdname(mddev)); return -EINVAL; } printk(KERN_NOTICE "md: data integrity enabled on %s\n", mdname(mddev)); if (bioset_integrity_create(mddev->bio_set, BIO_POOL_SIZE)) { printk(KERN_ERR "md: failed to create integrity pool for %s\n", mdname(mddev)); return -EINVAL; } return 0; } EXPORT_SYMBOL(md_integrity_register); /* Disable data integrity if non-capable/non-matching disk is being added */ void md_integrity_add_rdev(struct md_rdev *rdev, struct mddev *mddev) { struct blk_integrity *bi_rdev = bdev_get_integrity(rdev->bdev); struct blk_integrity *bi_mddev = blk_get_integrity(mddev->gendisk); if (!bi_mddev) /* nothing to do */ return; if (rdev->raid_disk < 0) /* skip spares */ return; if (bi_rdev && blk_integrity_compare(mddev->gendisk, rdev->bdev->bd_disk) >= 0) return; printk(KERN_NOTICE "disabling data integrity on %s\n", mdname(mddev)); blk_integrity_unregister(mddev->gendisk); } EXPORT_SYMBOL(md_integrity_add_rdev); static int bind_rdev_to_array(struct md_rdev * rdev, struct mddev * mddev) { char b[BDEVNAME_SIZE]; struct kobject *ko; char *s; int err; if (rdev->mddev) { MD_BUG(); return -EINVAL; } /* prevent duplicates */ if (find_rdev(mddev, rdev->bdev->bd_dev)) return -EEXIST; /* make sure rdev->sectors exceeds mddev->dev_sectors */ if (rdev->sectors && (mddev->dev_sectors == 0 || rdev->sectors < mddev->dev_sectors)) { if (mddev->pers) { /* Cannot change size, so fail * If mddev->level <= 0, then we don't care * about aligning sizes (e.g. linear) */ if (mddev->level > 0) return -ENOSPC; } else mddev->dev_sectors = rdev->sectors; } /* Verify rdev->desc_nr is unique. * If it is -1, assign a free number, else * check number is not in use */ if (rdev->desc_nr < 0) { int choice = 0; if (mddev->pers) choice = mddev->raid_disks; while (find_rdev_nr(mddev, choice)) choice++; rdev->desc_nr = choice; } else { if (find_rdev_nr(mddev, rdev->desc_nr)) return -EBUSY; } if (mddev->max_disks && rdev->desc_nr >= mddev->max_disks) { printk(KERN_WARNING "md: %s: array is limited to %d devices\n", mdname(mddev), mddev->max_disks); return -EBUSY; } bdevname(rdev->bdev,b); while ( (s=strchr(b, '/')) != NULL) *s = '!'; rdev->mddev = mddev; printk(KERN_INFO "md: bind<%s>\n", b); if ((err = kobject_add(&rdev->kobj, &mddev->kobj, "dev-%s", b))) goto fail; ko = &part_to_dev(rdev->bdev->bd_part)->kobj; if (sysfs_create_link(&rdev->kobj, ko, "block")) /* failure here is OK */; rdev->sysfs_state = sysfs_get_dirent_safe(rdev->kobj.sd, "state"); list_add_rcu(&rdev->same_set, &mddev->disks); bd_link_disk_holder(rdev->bdev, mddev->gendisk); /* May as well allow recovery to be retried once */ mddev->recovery_disabled++; return 0; fail: printk(KERN_WARNING "md: failed to register dev-%s for %s\n", b, mdname(mddev)); return err; } static void md_delayed_delete(struct work_struct *ws) { struct md_rdev *rdev = container_of(ws, struct md_rdev, del_work); kobject_del(&rdev->kobj); kobject_put(&rdev->kobj); } static void unbind_rdev_from_array(struct md_rdev * rdev) { char b[BDEVNAME_SIZE]; if (!rdev->mddev) { MD_BUG(); return; } bd_unlink_disk_holder(rdev->bdev, rdev->mddev->gendisk); list_del_rcu(&rdev->same_set); printk(KERN_INFO "md: unbind<%s>\n", bdevname(rdev->bdev,b)); rdev->mddev = NULL; sysfs_remove_link(&rdev->kobj, "block"); sysfs_put(rdev->sysfs_state); rdev->sysfs_state = NULL; rdev->badblocks.count = 0; /* We need to delay this, otherwise we can deadlock when * writing to 'remove' to "dev/state". We also need * to delay it due to rcu usage. */ synchronize_rcu(); INIT_WORK(&rdev->del_work, md_delayed_delete); kobject_get(&rdev->kobj); queue_work(md_misc_wq, &rdev->del_work); } /* * prevent the device from being mounted, repartitioned or * otherwise reused by a RAID array (or any other kernel * subsystem), by bd_claiming the device. */ static int lock_rdev(struct md_rdev *rdev, dev_t dev, int shared) { int err = 0; struct block_device *bdev; char b[BDEVNAME_SIZE]; bdev = blkdev_get_by_dev(dev, FMODE_READ|FMODE_WRITE|FMODE_EXCL, shared ? (struct md_rdev *)lock_rdev : rdev); if (IS_ERR(bdev)) { printk(KERN_ERR "md: could not open %s.\n", __bdevname(dev, b)); return PTR_ERR(bdev); } rdev->bdev = bdev; return err; } static void unlock_rdev(struct md_rdev *rdev) { struct block_device *bdev = rdev->bdev; rdev->bdev = NULL; if (!bdev) MD_BUG(); blkdev_put(bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL); } void md_autodetect_dev(dev_t dev); static void export_rdev(struct md_rdev * rdev) { char b[BDEVNAME_SIZE]; printk(KERN_INFO "md: export_rdev(%s)\n", bdevname(rdev->bdev,b)); if (rdev->mddev) MD_BUG(); md_rdev_clear(rdev); #ifndef MODULE if (test_bit(AutoDetected, &rdev->flags)) md_autodetect_dev(rdev->bdev->bd_dev); #endif unlock_rdev(rdev); kobject_put(&rdev->kobj); } static void kick_rdev_from_array(struct md_rdev * rdev) { unbind_rdev_from_array(rdev); export_rdev(rdev); } static void export_array(struct mddev *mddev) { struct md_rdev *rdev, *tmp; rdev_for_each_safe(rdev, tmp, mddev) { if (!rdev->mddev) { MD_BUG(); continue; } kick_rdev_from_array(rdev); } if (!list_empty(&mddev->disks)) MD_BUG(); mddev->raid_disks = 0; mddev->major_version = 0; } static void print_desc(mdp_disk_t *desc) { printk(" DISK\n", desc->number, desc->major,desc->minor,desc->raid_disk,desc->state); } static void print_sb_90(mdp_super_t *sb) { int i; printk(KERN_INFO "md: SB: (V:%d.%d.%d) ID:<%08x.%08x.%08x.%08x> CT:%08x\n", sb->major_version, sb->minor_version, sb->patch_version, sb->set_uuid0, sb->set_uuid1, sb->set_uuid2, sb->set_uuid3, sb->ctime); printk(KERN_INFO "md: L%d S%08d ND:%d RD:%d md%d LO:%d CS:%d\n", sb->level, sb->size, sb->nr_disks, sb->raid_disks, sb->md_minor, sb->layout, sb->chunk_size); printk(KERN_INFO "md: UT:%08x ST:%d AD:%d WD:%d" " FD:%d SD:%d CSUM:%08x E:%08lx\n", sb->utime, sb->state, sb->active_disks, sb->working_disks, sb->failed_disks, sb->spare_disks, sb->sb_csum, (unsigned long)sb->events_lo); printk(KERN_INFO); for (i = 0; i < MD_SB_DISKS; i++) { mdp_disk_t *desc; desc = sb->disks + i; if (desc->number || desc->major || desc->minor || desc->raid_disk || (desc->state && (desc->state != 4))) { printk(" D %2d: ", i); print_desc(desc); } } printk(KERN_INFO "md: THIS: "); print_desc(&sb->this_disk); } static void print_sb_1(struct mdp_superblock_1 *sb) { __u8 *uuid; uuid = sb->set_uuid; printk(KERN_INFO "md: SB: (V:%u) (F:0x%08x) Array-ID:<%pU>\n" "md: Name: \"%s\" CT:%llu\n", le32_to_cpu(sb->major_version), le32_to_cpu(sb->feature_map), uuid, sb->set_name, (unsigned long long)le64_to_cpu(sb->ctime) & MD_SUPERBLOCK_1_TIME_SEC_MASK); uuid = sb->device_uuid; printk(KERN_INFO "md: L%u SZ%llu RD:%u LO:%u CS:%u DO:%llu DS:%llu SO:%llu" " RO:%llu\n" "md: Dev:%08x UUID: %pU\n" "md: (F:0x%08x) UT:%llu Events:%llu ResyncOffset:%llu CSUM:0x%08x\n" "md: (MaxDev:%u) \n", le32_to_cpu(sb->level), (unsigned long long)le64_to_cpu(sb->size), le32_to_cpu(sb->raid_disks), le32_to_cpu(sb->layout), le32_to_cpu(sb->chunksize), (unsigned long long)le64_to_cpu(sb->data_offset), (unsigned long long)le64_to_cpu(sb->data_size), (unsigned long long)le64_to_cpu(sb->super_offset), (unsigned long long)le64_to_cpu(sb->recovery_offset), le32_to_cpu(sb->dev_number), uuid, sb->devflags, (unsigned long long)le64_to_cpu(sb->utime) & MD_SUPERBLOCK_1_TIME_SEC_MASK, (unsigned long long)le64_to_cpu(sb->events), (unsigned long long)le64_to_cpu(sb->re