diff options
Diffstat (limited to 'drivers/md/dm.c')
| -rw-r--r-- | drivers/md/dm.c | 2501 |
1 files changed, 1896 insertions, 605 deletions
diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 6617ce4af09..32b958dbc49 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -1,12 +1,11 @@ /* * Copyright (C) 2001, 2002 Sistina Software (UK) Limited. - * Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved. + * Copyright (C) 2004-2008 Red Hat, Inc. All rights reserved. * * This file is released under the GPL. */ #include "dm.h" -#include "dm-bio-list.h" #include "dm-uevent.h" #include <linux/init.h> @@ -15,86 +14,143 @@ #include <linux/moduleparam.h> #include <linux/blkpg.h> #include <linux/bio.h> -#include <linux/buffer_head.h> #include <linux/mempool.h> #include <linux/slab.h> #include <linux/idr.h> #include <linux/hdreg.h> -#include <linux/blktrace_api.h> -#include <linux/smp_lock.h> +#include <linux/delay.h> + +#include <trace/events/block.h> #define DM_MSG_PREFIX "core" +#ifdef CONFIG_PRINTK +/* + * ratelimit state to be used in DMXXX_LIMIT(). + */ +DEFINE_RATELIMIT_STATE(dm_ratelimit_state, + DEFAULT_RATELIMIT_INTERVAL, + DEFAULT_RATELIMIT_BURST); +EXPORT_SYMBOL(dm_ratelimit_state); +#endif + +/* + * Cookies are numeric values sent with CHANGE and REMOVE + * uevents while resuming, removing or renaming the device. + */ +#define DM_COOKIE_ENV_VAR_NAME "DM_COOKIE" +#define DM_COOKIE_LENGTH 24 + static const char *_name = DM_NAME; static unsigned int major = 0; static unsigned int _major = 0; +static DEFINE_IDR(_minor_idr); + static DEFINE_SPINLOCK(_minor_lock); + +static void do_deferred_remove(struct work_struct *w); + +static DECLARE_WORK(deferred_remove_work, do_deferred_remove); + +static struct workqueue_struct *deferred_remove_workqueue; + /* + * For bio-based dm. * One of these is allocated per bio. */ struct dm_io { struct mapped_device *md; int error; - struct bio *bio; atomic_t io_count; + struct bio *bio; unsigned long start_time; + spinlock_t endio_lock; + struct dm_stats_aux stats_aux; }; /* - * One of these is allocated per target within a bio. Hopefully - * this will be simplified out one day. + * For request-based dm. + * One of these is allocated per request. */ -struct dm_target_io { - struct dm_io *io; +struct dm_rq_target_io { + struct mapped_device *md; struct dm_target *ti; + struct request *orig, clone; + int error; union map_info info; }; -union map_info *dm_get_mapinfo(struct bio *bio) +/* + * For request-based dm - the bio clones we allocate are embedded in these + * structs. + * + * We allocate these with bio_alloc_bioset, using the front_pad parameter when + * the bioset is created - this means the bio has to come at the end of the + * struct. + */ +struct dm_rq_clone_bio_info { + struct bio *orig; + struct dm_rq_target_io *tio; + struct bio clone; +}; + +union map_info *dm_get_rq_mapinfo(struct request *rq) { - if (bio && bio->bi_private) - return &((struct dm_target_io *)bio->bi_private)->info; + if (rq && rq->end_io_data) + return &((struct dm_rq_target_io *)rq->end_io_data)->info; return NULL; } +EXPORT_SYMBOL_GPL(dm_get_rq_mapinfo); #define MINOR_ALLOCED ((void *)-1) /* * Bits for the md->flags field. */ -#define DMF_BLOCK_IO 0 +#define DMF_BLOCK_IO_FOR_SUSPEND 0 #define DMF_SUSPENDED 1 #define DMF_FROZEN 2 #define DMF_FREEING 3 #define DMF_DELETING 4 #define DMF_NOFLUSH_SUSPENDING 5 +#define DMF_MERGE_IS_OPTIONAL 6 +#define DMF_DEFERRED_REMOVE 7 /* - * Work processed by per-device workqueue. + * A dummy definition to make RCU happy. + * struct dm_table should never be dereferenced in this file. */ -struct dm_wq_req { - enum { - DM_WQ_FLUSH_ALL, - DM_WQ_FLUSH_DEFERRED, - } type; - struct work_struct work; - struct mapped_device *md; - void *context; +struct dm_table { + int undefined__; }; +/* + * Work processed by per-device workqueue. + */ struct mapped_device { - struct rw_semaphore io_lock; + struct srcu_struct io_barrier; struct mutex suspend_lock; - spinlock_t pushback_lock; - rwlock_t map_lock; atomic_t holders; atomic_t open_count; + /* + * The current mapping. + * Use dm_get_live_table{_fast} or take suspend_lock for + * dereference. + */ + struct dm_table *map; + unsigned long flags; struct request_queue *queue; + unsigned type; + /* Protect queue and type against concurrent access. */ + struct mutex type_lock; + + struct target_type *immutable_target_type; + struct gendisk *disk; char name[16]; @@ -103,26 +159,21 @@ struct mapped_device { /* * A list of ios that arrived while we were suspended. */ - atomic_t pending; + atomic_t pending[2]; wait_queue_head_t wait; + struct work_struct work; struct bio_list deferred; - struct bio_list pushback; + spinlock_t deferred_lock; /* - * Processing queue (flush/barriers) + * Processing queue (flush) */ struct workqueue_struct *wq; /* - * The current mapping. - */ - struct dm_table *map; - - /* * io objects are allocated from here. */ mempool_t *io_pool; - mempool_t *tio_pool; struct bio_set *bs; @@ -139,57 +190,128 @@ struct mapped_device { * freeze/thaw support require holding onto a super block */ struct super_block *frozen_sb; - struct block_device *suspended_bdev; + struct block_device *bdev; /* forced geometry settings */ struct hd_geometry geometry; + + /* kobject and completion */ + struct dm_kobject_holder kobj_holder; + + /* zero-length flush that will be cloned and submitted to targets */ + struct bio flush_bio; + + struct dm_stats stats; }; -#define MIN_IOS 256 +/* + * For mempools pre-allocation at the table loading time. + */ +struct dm_md_mempools { + mempool_t *io_pool; + struct bio_set *bs; +}; + +#define RESERVED_BIO_BASED_IOS 16 +#define RESERVED_REQUEST_BASED_IOS 256 +#define RESERVED_MAX_IOS 1024 static struct kmem_cache *_io_cache; -static struct kmem_cache *_tio_cache; +static struct kmem_cache *_rq_tio_cache; + +/* + * Bio-based DM's mempools' reserved IOs set by the user. + */ +static unsigned reserved_bio_based_ios = RESERVED_BIO_BASED_IOS; + +/* + * Request-based DM's mempools' reserved IOs set by the user. + */ +static unsigned reserved_rq_based_ios = RESERVED_REQUEST_BASED_IOS; + +static unsigned __dm_get_reserved_ios(unsigned *reserved_ios, + unsigned def, unsigned max) +{ + unsigned ios = ACCESS_ONCE(*reserved_ios); + unsigned modified_ios = 0; + + if (!ios) + modified_ios = def; + else if (ios > max) + modified_ios = max; + + if (modified_ios) { + (void)cmpxchg(reserved_ios, ios, modified_ios); + ios = modified_ios; + } + + return ios; +} + +unsigned dm_get_reserved_bio_based_ios(void) +{ + return __dm_get_reserved_ios(&reserved_bio_based_ios, + RESERVED_BIO_BASED_IOS, RESERVED_MAX_IOS); +} +EXPORT_SYMBOL_GPL(dm_get_reserved_bio_based_ios); + +unsigned dm_get_reserved_rq_based_ios(void) +{ + return __dm_get_reserved_ios(&reserved_rq_based_ios, + RESERVED_REQUEST_BASED_IOS, RESERVED_MAX_IOS); +} +EXPORT_SYMBOL_GPL(dm_get_reserved_rq_based_ios); static int __init local_init(void) { - int r; + int r = -ENOMEM; /* allocate a slab for the dm_ios */ _io_cache = KMEM_CACHE(dm_io, 0); if (!_io_cache) - return -ENOMEM; + return r; - /* allocate a slab for the target ios */ - _tio_cache = KMEM_CACHE(dm_target_io, 0); - if (!_tio_cache) { - kmem_cache_destroy(_io_cache); - return -ENOMEM; - } + _rq_tio_cache = KMEM_CACHE(dm_rq_target_io, 0); + if (!_rq_tio_cache) + goto out_free_io_cache; r = dm_uevent_init(); - if (r) { - kmem_cache_destroy(_tio_cache); - kmem_cache_destroy(_io_cache); - return r; + if (r) + goto out_free_rq_tio_cache; + + deferred_remove_workqueue = alloc_workqueue("kdmremove", WQ_UNBOUND, 1); + if (!deferred_remove_workqueue) { + r = -ENOMEM; + goto out_uevent_exit; } _major = major; r = register_blkdev(_major, _name); - if (r < 0) { - kmem_cache_destroy(_tio_cache); - kmem_cache_destroy(_io_cache); - dm_uevent_exit(); - return r; - } + if (r < 0) + goto out_free_workqueue; if (!_major) _major = r; return 0; + +out_free_workqueue: + destroy_workqueue(deferred_remove_workqueue); +out_uevent_exit: + dm_uevent_exit(); +out_free_rq_tio_cache: + kmem_cache_destroy(_rq_tio_cache); +out_free_io_cache: + kmem_cache_destroy(_io_cache); + + return r; } static void local_exit(void) { - kmem_cache_destroy(_tio_cache); + flush_scheduled_work(); + destroy_workqueue(deferred_remove_workqueue); + + kmem_cache_destroy(_rq_tio_cache); kmem_cache_destroy(_io_cache); unregister_blkdev(_major, _name); dm_uevent_exit(); @@ -204,7 +326,10 @@ static int (*_inits[])(void) __initdata = { dm_target_init, dm_linear_init, dm_stripe_init, + dm_io_init, + dm_kcopyd_init, dm_interface_init, + dm_statistics_init, }; static void (*_exits[])(void) = { @@ -212,7 +337,10 @@ static void (*_exits[])(void) = { dm_target_exit, dm_linear_exit, dm_stripe_exit, + dm_io_exit, + dm_kcopyd_exit, dm_interface_exit, + dm_statistics_exit, }; static int __init dm_init(void) @@ -242,23 +370,33 @@ static void __exit dm_exit(void) while (i--) _exits[i](); + + /* + * Should be empty by this point. + */ + idr_destroy(&_minor_idr); } /* * Block device functions */ -static int dm_blk_open(struct inode *inode, struct file *file) +int dm_deleting_md(struct mapped_device *md) +{ + return test_bit(DMF_DELETING, &md->flags); +} + +static int dm_blk_open(struct block_device *bdev, fmode_t mode) { struct mapped_device *md; spin_lock(&_minor_lock); - md = inode->i_bdev->bd_disk->private_data; + md = bdev->bd_disk->private_data; if (!md) goto out; if (test_bit(DMF_FREEING, &md->flags) || - test_bit(DMF_DELETING, &md->flags)) { + dm_deleting_md(md)) { md = NULL; goto out; } @@ -272,14 +410,19 @@ out: return md ? 0 : -ENXIO; } -static int dm_blk_close(struct inode *inode, struct file *file) +static void dm_blk_close(struct gendisk *disk, fmode_t mode) { - struct mapped_device *md; + struct mapped_device *md = disk->private_data; + + spin_lock(&_minor_lock); + + if (atomic_dec_and_test(&md->open_count) && + (test_bit(DMF_DEFERRED_REMOVE, &md->flags))) + queue_work(deferred_remove_workqueue, &deferred_remove_work); - md = inode->i_bdev->bd_disk->private_data; - atomic_dec(&md->open_count); dm_put(md); - return 0; + + spin_unlock(&_minor_lock); } int dm_open_count(struct mapped_device *md) @@ -290,14 +433,18 @@ int dm_open_count(struct mapped_device *md) /* * Guarantees nothing is using the device before it's deleted. */ -int dm_lock_for_deletion(struct mapped_device *md) +int dm_lock_for_deletion(struct mapped_device *md, bool mark_deferred, bool only_deferred) { int r = 0; spin_lock(&_minor_lock); - if (dm_open_count(md)) + if (dm_open_count(md)) { r = -EBUSY; + if (mark_deferred) + set_bit(DMF_DEFERRED_REMOVE, &md->flags); + } else if (only_deferred && !test_bit(DMF_DEFERRED_REMOVE, &md->flags)) + r = -EEXIST; else set_bit(DMF_DELETING, &md->flags); @@ -306,6 +453,42 @@ int dm_lock_for_deletion(struct mapped_device *md) return r; } +int dm_cancel_deferred_remove(struct mapped_device *md) +{ + int r = 0; + + spin_lock(&_minor_lock); + + if (test_bit(DMF_DELETING, &md->flags)) + r = -EBUSY; + else + clear_bit(DMF_DEFERRED_REMOVE, &md->flags); + + spin_unlock(&_minor_lock); + + return r; +} + +static void do_deferred_remove(struct work_struct *w) +{ + dm_deferred_remove(); +} + +sector_t dm_get_size(struct mapped_device *md) +{ + return get_capacity(md->disk); +} + +struct request_queue *dm_get_md_queue(struct mapped_device *md) +{ + return md->queue; +} + +struct dm_stats *dm_get_stats(struct mapped_device *md) +{ + return &md->stats; +} + static int dm_blk_getgeo(struct block_device *bdev, struct hd_geometry *geo) { struct mapped_device *md = bdev->bd_disk->private_data; @@ -313,20 +496,17 @@ static int dm_blk_getgeo(struct block_device *bdev, struct hd_geometry *geo) return dm_get_geometry(md, geo); } -static int dm_blk_ioctl(struct inode *inode, struct file *file, +static int dm_blk_ioctl(struct block_device *bdev, fmode_t mode, unsigned int cmd, unsigned long arg) { - struct mapped_device *md; + struct mapped_device *md = bdev->bd_disk->private_data; + int srcu_idx; struct dm_table *map; struct dm_target *tgt; int r = -ENOTTY; - /* We don't really need this lock, but we do need 'inode'. */ - unlock_kernel(); - - md = inode->i_bdev->bd_disk->private_data; - - map = dm_get_table(md); +retry: + map = dm_get_live_table(md, &srcu_idx); if (!map || !dm_table_get_size(map)) goto out; @@ -337,18 +517,22 @@ static int dm_blk_ioctl(struct inode *inode, struct file *file, tgt = dm_table_get_target(map, 0); - if (dm_suspended(md)) { + if (dm_suspended_md(md)) { r = -EAGAIN; goto out; } if (tgt->type->ioctl) - r = tgt->type->ioctl(tgt, inode, file, cmd, arg); + r = tgt->type->ioctl(tgt, cmd, arg); out: - dm_table_put(map); + dm_put_live_table(md, srcu_idx); + + if (r == -ENOTCONN) { + msleep(10); + goto retry; + } - lock_kernel(); return r; } @@ -362,80 +546,127 @@ static void free_io(struct mapped_device *md, struct dm_io *io) mempool_free(io, md->io_pool); } -static struct dm_target_io *alloc_tio(struct mapped_device *md) +static void free_tio(struct mapped_device *md, struct dm_target_io *tio) +{ + bio_put(&tio->clone); +} + +static struct dm_rq_target_io *alloc_rq_tio(struct mapped_device *md, + gfp_t gfp_mask) { - return mempool_alloc(md->tio_pool, GFP_NOIO); + return mempool_alloc(md->io_pool, gfp_mask); } -static void free_tio(struct mapped_device *md, struct dm_target_io *tio) +static void free_rq_tio(struct dm_rq_target_io *tio) +{ + mempool_free(tio, tio->md->io_pool); +} + +static int md_in_flight(struct mapped_device *md) { - mempool_free(tio, md->tio_pool); + return atomic_read(&md->pending[READ]) + + atomic_read(&md->pending[WRITE]); } static void start_io_acct(struct dm_io *io) { struct mapped_device *md = io->md; + struct bio *bio = io->bio; + int cpu; + int rw = bio_data_dir(bio); io->start_time = jiffies; - preempt_disable(); - disk_round_stats(dm_disk(md)); - preempt_enable(); - dm_disk(md)->in_flight = atomic_inc_return(&md->pending); + cpu = part_stat_lock(); + part_round_stats(cpu, &dm_disk(md)->part0); + part_stat_unlock(); + atomic_set(&dm_disk(md)->part0.in_flight[rw], + atomic_inc_return(&md->pending[rw])); + + if (unlikely(dm_stats_used(&md->stats))) + dm_stats_account_io(&md->stats, bio->bi_rw, bio->bi_iter.bi_sector, + bio_sectors(bio), false, 0, &io->stats_aux); } -static int end_io_acct(struct dm_io *io) +static void end_io_acct(struct dm_io *io) { struct mapped_device *md = io->md; struct bio *bio = io->bio; unsigned long duration = jiffies - io->start_time; - int pending; + int pending, cpu; int rw = bio_data_dir(bio); - preempt_disable(); - disk_round_stats(dm_disk(md)); - preempt_enable(); - dm_disk(md)->in_flight = pending = atomic_dec_return(&md->pending); + cpu = part_stat_lock(); + part_round_stats(cpu, &dm_disk(md)->part0); + part_stat_add(cpu, &dm_disk(md)->part0, ticks[rw], duration); + part_stat_unlock(); - disk_stat_add(dm_disk(md), ticks[rw], duration); + if (unlikely(dm_stats_used(&md->stats))) + dm_stats_account_io(&md->stats, bio->bi_rw, bio->bi_iter.bi_sector, + bio_sectors(bio), true, duration, &io->stats_aux); - return !pending; + /* + * After this is decremented the bio must not be touched if it is + * a flush. + */ + pending = atomic_dec_return(&md->pending[rw]); + atomic_set(&dm_disk(md)->part0.in_flight[rw], pending); + pending += atomic_read(&md->pending[rw^0x1]); + + /* nudge anyone waiting on suspend queue */ + if (!pending) + wake_up(&md->wait); } /* * Add the bio to the list of deferred io. */ -static int queue_io(struct mapped_device *md, struct bio *bio) +static void queue_io(struct mapped_device *md, struct bio *bio) { - down_write(&md->io_lock); - - if (!test_bit(DMF_BLOCK_IO, &md->flags)) { - up_write(&md->io_lock); - return 1; - } + unsigned long flags; + spin_lock_irqsave(&md->deferred_lock, flags); bio_list_add(&md->deferred, bio); - - up_write(&md->io_lock); - return 0; /* deferred successfully */ + spin_unlock_irqrestore(&md->deferred_lock, flags); + queue_work(md->wq, &md->work); } /* * Everyone (including functions in this file), should use this * function to access the md->map field, and make sure they call - * dm_table_put() when finished. + * dm_put_live_table() when finished. */ -struct dm_table *dm_get_table(struct mapped_device *md) +struct dm_table *dm_get_live_table(struct mapped_device *md, int *srcu_idx) __acquires(md->io_barrier) { - struct dm_table *t; + *srcu_idx = srcu_read_lock(&md->io_barrier); - read_lock(&md->map_lock); - t = md->map; - if (t) - dm_table_get(t); - read_unlock(&md->map_lock); + return srcu_dereference(md->map, &md->io_barrier); +} - return t; +void dm_put_live_table(struct mapped_device *md, int srcu_idx) __releases(md->io_barrier) +{ + srcu_read_unlock(&md->io_barrier, srcu_idx); +} + +void dm_sync_table(struct mapped_device *md) +{ + synchronize_srcu(&md->io_barrier); + synchronize_rcu_expedited(); +} + +/* + * A fast alternative to dm_get_live_table/dm_put_live_table. + * The caller must not block between these two functions. + */ +static struct dm_table *dm_get_live_table_fast(struct mapped_device *md) __acquires(RCU) +{ + rcu_read_lock(); + return rcu_dereference(md->map); +} + +static void dm_put_live_table_fast(struct mapped_device *md) __releases(RCU) +{ + rcu_read_unlock(); } /* @@ -486,46 +717,68 @@ static int __noflush_suspending(struct mapped_device *md) static void dec_pending(struct dm_io *io, int error) { unsigned long flags; + int io_error; + struct bio *bio; + struct mapped_device *md = io->md; /* Push-back supersedes any I/O errors */ - if (error && !(io->error > 0 && __noflush_suspending(io->md))) - io->error = error; + if (unlikely(error)) { + spin_lock_irqsave(&io->endio_lock, flags); + if (!(io->error > 0 && __noflush_suspending(md))) + io->error = error; + spin_unlock_irqrestore(&io->endio_lock, flags); + } if (atomic_dec_and_test(&io->io_count)) { if (io->error == DM_ENDIO_REQUEUE) { /* * Target requested pushing back the I/O. - * This must be handled before the sleeper on - * suspend queue merges the pushback list. */ - spin_lock_irqsave(&io->md->pushback_lock, flags); - if (__noflush_suspending(io->md)) - bio_list_add(&io->md->pushback, io->bio); + spin_lock_irqsave(&md->deferred_lock, flags); + if (__noflush_suspending(md)) + bio_list_add_head(&md->deferred, io->bio); else /* noflush suspend was interrupted. */ io->error = -EIO; - spin_unlock_irqrestore(&io->md->pushback_lock, flags); + spin_unlock_irqrestore(&md->deferred_lock, flags); } - if (end_io_acct(io)) - /* nudge anyone waiting on suspend queue */ - wake_up(&io->md->wait); + io_error = io->error; + bio = io->bio; + end_io_acct(io); + free_io(md, io); - if (io->error != DM_ENDIO_REQUEUE) { - blk_add_trace_bio(io->md->queue, io->bio, - BLK_TA_COMPLETE); + if (io_error == DM_ENDIO_REQUEUE) + return; - bio_endio(io->bio, io->error); + if ((bio->bi_rw & REQ_FLUSH) && bio->bi_iter.bi_size) { + /* + * Preflush done for flush with data, reissue + * without REQ_FLUSH. + */ + bio->bi_rw &= ~REQ_FLUSH; + queue_io(md, bio); + } else { + /* done with normal IO or empty flush */ + trace_block_bio_complete(md->queue, bio, io_error); + bio_endio(bio, io_error); } - - free_io(io->md, io); } } +static void disable_write_same(struct mapped_device *md) +{ + struct queue_limits *limits = dm_get_queue_limits(md); + + /* device doesn't really support WRITE SAME, disable it */ + limits->max_write_same_sectors = 0; +} + static void clone_endio(struct bio *bio, int error) { int r = 0; - struct dm_target_io *tio = bio->bi_private; + struct dm_target_io *tio = container_of(bio, struct dm_target_io, clone); + struct dm_io *io = tio->io; struct mapped_device *md = tio->io->md; dm_endio_fn endio = tio->ti->type->end_io; @@ -533,7 +786,7 @@ static void clone_endio(struct bio *bio, int error) error = -EIO; if (endio) { - r = endio(tio->ti, bio, error, &tio->info); + r = endio(tio->ti, bio, error); if (r < 0 || r == DM_ENDIO_REQUEUE) /* * error and requeue request are handled @@ -549,51 +802,390 @@ static void clone_endio(struct bio *bio, int error) } } - dec_pending(tio->io, error); + if (unlikely(r == -EREMOTEIO && (bio->bi_rw & REQ_WRITE_SAME) && + !bdev_get_queue(bio->bi_bdev)->limits.max_write_same_sectors)) + disable_write_same(md); + + free_tio(md, tio); + dec_pending(io, error); +} + +/* + * Partial completion handling for request-based dm + */ +static void end_clone_bio(struct bio *clone, int error) +{ + struct dm_rq_clone_bio_info *info = + container_of(clone, struct dm_rq_clone_bio_info, clone); + struct dm_rq_target_io *tio = info->tio; + struct bio *bio = info->orig; + unsigned int nr_bytes = info->orig->bi_iter.bi_size; + + bio_put(clone); + + if (tio->error) + /* + * An error has already been detected on the request. + * Once error occurred, just let clone->end_io() handle + * the remainder. + */ + return; + else if (error) { + /* + * Don't notice the error to the upper layer yet. + * The error handling decision is made by the target driver, + * when the request is completed. + */ + tio->error = error; + return; + } + + /* + * I/O for the bio successfully completed. + * Notice the data completion to the upper layer. + */ /* - * Store md for cleanup instead of tio which is about to get freed. + * bios are processed from the head of the list. + * So the completing bio should always be rq->bio. + * If it's not, something wrong is happening. */ - bio->bi_private = md->bs; + if (tio->orig->bio != bio) + DMERR("bio completion is going in the middle of the request"); - bio_put(bio); - free_tio(md, tio); + /* + * Update the original request. + * Do not use blk_end_request() here, because it may complete + * the original request before the clone, and break the ordering. + */ + blk_update_request(tio->orig, 0, nr_bytes); } -static sector_t max_io_len(struct mapped_device *md, - sector_t sector, struct dm_target *ti) +/* + * Don't touch any member of the md after calling this function because + * the md may be freed in dm_put() at the end of this function. + * Or do dm_get() before calling this function and dm_put() later. + */ +static void rq_completed(struct mapped_device *md, int rw, int run_queue) { - sector_t offset = sector - ti->begin; - sector_t len = ti->len - offset; + atomic_dec(&md->pending[rw]); + + /* nudge anyone waiting on suspend queue */ + if (!md_in_flight(md)) + wake_up(&md->wait); + + /* + * Run this off this callpath, as drivers could invoke end_io while + * inside their request_fn (and holding the queue lock). Calling + * back into ->request_fn() could deadlock attempting to grab the + * queue lock again. + */ + if (run_queue) + blk_run_queue_async(md->queue); /* - * Does the target need to split even further ? + * dm_put() must be at the end of this function. See the comment above */ - if (ti->split_io) { - sector_t boundary; - boundary = ((offset + ti->split_io) & ~(ti->split_io - 1)) - - offset; - if (len > boundary) - len = boundary; + dm_put(md); +} + +static void free_rq_clone(struct request *clone) +{ + struct dm_rq_target_io *tio = clone->end_io_data; + + blk_rq_unprep_clone(clone); + free_rq_tio(tio); +} + +/* + * Complete the clone and the original request. + * Must be called without queue lock. + */ +static void dm_end_request(struct request *clone, int error) +{ + int rw = rq_data_dir(clone); + struct dm_rq_target_io *tio = clone->end_io_data; + struct mapped_device *md = tio->md; + struct request *rq = tio->orig; + + if (rq->cmd_type == REQ_TYPE_BLOCK_PC) { + rq->errors = clone->errors; + rq->resid_len = clone->resid_len; + + if (rq->sense) + /* + * We are using the sense buffer of the original + * request. + * So setting the length of the sense data is enough. + */ + rq->sense_len = clone->sense_len; + } + + free_rq_clone(clone); + blk_end_request_all(rq, error); + rq_completed(md, rw, true); +} + +static void dm_unprep_request(struct request *rq) +{ + struct request *clone = rq->special; + + rq->special = NULL; + rq->cmd_flags &= ~REQ_DONTPREP; + + free_rq_clone(clone); +} + +/* + * Requeue the original request of a clone. + */ +void dm_requeue_unmapped_request(struct request *clone) +{ + int rw = rq_data_dir(clone); + struct dm_rq_target_io *tio = clone->end_io_data; + struct mapped_device *md = tio->md; + struct request *rq = tio->orig; + struct request_queue *q = rq->q; + unsigned long flags; + + dm_unprep_request(rq); + + spin_lock_irqsave(q->queue_lock, flags); + blk_requeue_request(q, rq); + spin_unlock_irqrestore(q->queue_lock, flags); + + rq_completed(md, rw, 0); +} +EXPORT_SYMBOL_GPL(dm_requeue_unmapped_request); + +static void __stop_queue(struct request_queue *q) +{ + blk_stop_queue(q); +} + +static void stop_queue(struct request_queue *q) +{ + unsigned long flags; + + spin_lock_irqsave(q->queue_lock, flags); + __stop_queue(q); + spin_unlock_irqrestore(q->queue_lock, flags); +} + +static void __start_queue(struct request_queue *q) +{ + if (blk_queue_stopped(q)) + blk_start_queue(q); +} + +static void start_queue(struct request_queue *q) +{ + unsigned long flags; + + spin_lock_irqsave(q->queue_lock, flags); + __start_queue(q); + spin_unlock_irqrestore(q->queue_lock, flags); +} + +static void dm_done(struct request *clone, int error, bool mapped) +{ + int r = error; + struct dm_rq_target_io *tio = clone->end_io_data; + dm_request_endio_fn rq_end_io = NULL; + + if (tio->ti) { + rq_end_io = tio->ti->type->rq_end_io; + + if (mapped && rq_end_io) + r = rq_end_io(tio->ti, clone, error, &tio->info); + } + + if (unlikely(r == -EREMOTEIO && (clone->cmd_flags & REQ_WRITE_SAME) && + !clone->q->limits.max_write_same_sectors)) + disable_write_same(tio->md); + + if (r <= 0) + /* The target wants to complete the I/O */ + dm_end_request(clone, r); + else if (r == DM_ENDIO_INCOMPLETE) + /* The target will handle the I/O */ + return; + else if (r == DM_ENDIO_REQUEUE) + /* The target wants to requeue the I/O */ + dm_requeue_unmapped_request(clone); + else { + DMWARN("unimplemented target endio return value: %d", r); + BUG(); + } +} + +/* + * Request completion handler for request-based dm + */ +static void dm_softirq_done(struct request *rq) +{ + bool mapped = true; + struct request *clone = rq->completion_data; + struct dm_rq_target_io *tio = clone->end_io_data; + + if (rq->cmd_flags & REQ_FAILED) + mapped = false; + + dm_done(clone, tio->error, mapped); +} + +/* + * Complete the clone and the original request with the error status + * through softirq context. + */ +static void dm_complete_request(struct request *clone, int error) +{ + struct dm_rq_target_io *tio = clone->end_io_data; + struct request *rq = tio->orig; + + tio->error = error; + rq->completion_data = clone; + blk_complete_request(rq); +} + +/* + * Complete the not-mapped clone and the original request with the error status + * through softirq context. + * Target's rq_end_io() function isn't called. + * This may be used when the target's map_rq() function fails. + */ +void dm_kill_unmapped_request(struct request *clone, int error) +{ + struct dm_rq_target_io *tio = clone->end_io_data; + struct request *rq = tio->orig; + + rq->cmd_flags |= REQ_FAILED; + dm_complete_request(clone, error); +} +EXPORT_SYMBOL_GPL(dm_kill_unmapped_request); + +/* + * Called with the queue lock held + */ +static void end_clone_request(struct request *clone, int error) +{ + /* + * For just cleaning up the information of the queue in which + * the clone was dispatched. + * The clone is *NOT* freed actually here because it is alloced from + * dm own mempool and REQ_ALLOCED isn't set in clone->cmd_flags. + */ + __blk_put_request(clone->q, clone); + + /* + * Actual request completion is done in a softirq context which doesn't + * hold the queue lock. Otherwise, deadlock could occur because: + * - another request may be submitted by the upper level driver + * of the stacking during the completion + * - the submission which requires queue lock may be done + * against this queue + */ + dm_complete_request(clone, error); +} + +/* + * Return maximum size of I/O possible at the supplied sector up to the current + * target boundary. + */ +static sector_t max_io_len_target_boundary(sector_t sector, struct dm_target *ti) +{ + sector_t target_offset = dm_target_offset(ti, sector); + + return ti->len - target_offset; +} + +static sector_t max_io_len(sector_t sector, struct dm_target *ti) +{ + sector_t len = max_io_len_target_boundary(sector, ti); + sector_t offset, max_len; + + /* + * Does the target need to split even further? + */ + if (ti->max_io_len) { + offset = dm_target_offset(ti, sector); + if (unlikely(ti->max_io_len & (ti->max_io_len - 1))) + max_len = sector_div(offset, ti->max_io_len); + else + max_len = offset & (ti->max_io_len - 1); + max_len = ti->max_io_len - max_len; + + if (len > max_len) + len = max_len; } return len; } -static void __map_bio(struct dm_target *ti, struct bio *clone, - struct dm_target_io *tio) +int dm_set_target_max_io_len(struct dm_target *ti, sector_t len) +{ + if (len > UINT_MAX) { + DMERR("Specified maximum size of target IO (%llu) exceeds limit (%u)", + (unsigned long long)len, UINT_MAX); + ti->error = "Maximum size of target IO is too large"; + return -EINVAL; + } + + ti->max_io_len = (uint32_t) len; + + return 0; +} +EXPORT_SYMBOL_GPL(dm_set_target_max_io_len); + +/* + * A target may call dm_accept_partial_bio only from the map routine. It is + * allowed for all bio types except REQ_FLUSH. + * + * dm_accept_partial_bio informs the dm that the target only wants to process + * additional n_sectors sectors of the bio and the rest of the data should be + * sent in a next bio. + * + * A diagram that explains the arithmetics: + * +--------------------+---------------+-------+ + * | 1 | 2 | 3 | + * +--------------------+---------------+-------+ + * + * <-------------- *tio->len_ptr ---------------> + * <------- bi_size -------> + * <-- n_sectors --> + * + * Region 1 was already iterated over with bio_advance or similar function. + * (it may be empty if the target doesn't use bio_advance) + * Region 2 is the remaining bio size that the target wants to process. + * (it may be empty if region 1 is non-empty, although there is no reason + * to make it empty) + * The target requires that region 3 is to be sent in the next bio. + * + * If the target wants to receive multiple copies of the bio (via num_*bios, etc), + * the partially processed part (the sum of regions 1+2) must be the same for all + * copies of the bio. + */ +void dm_accept_partial_bio(struct bio *bio, unsigned n_sectors) +{ + struct dm_target_io *tio = container_of(bio, struct dm_target_io, clone); + unsigned bi_size = bio->bi_iter.bi_size >> SECTOR_SHIFT; + BUG_ON(bio->bi_rw & REQ_FLUSH); + BUG_ON(bi_size > *tio->len_ptr); + BUG_ON(n_sectors > bi_size); + *tio->len_ptr -= bi_size - n_sectors; + bio->bi_iter.bi_size = n_sectors << SECTOR_SHIFT; +} +EXPORT_SYMBOL_GPL(dm_accept_partial_bio); + +static void __map_bio(struct dm_target_io *tio) { int r; sector_t sector; struct mapped_device *md; - - /* - * Sanity checks. - */ - BUG_ON(!clone->bi_size); + struct bio *clone = &tio->clone; + struct dm_target *ti = tio->ti; clone->bi_end_io = clone_endio; - clone->bi_private = tio; /* * Map the clone. If r == 0 we don't need to do @@ -601,25 +1193,19 @@ static void __map_bio(struct dm_target *ti, struct bio *clone, * this io. */ atomic_inc(&tio->io->io_count); - sector = clone->bi_sector; - r = ti->type->map(ti, clone, &tio->info); + sector = clone->bi_iter.bi_sector; + r = ti->type->map(ti, clone); if (r == DM_MAPIO_REMAPPED) { /* the bio has been remapped so dispatch it */ - blk_add_trace_remap(bdev_get_queue(clone->bi_bdev), clone, - tio->io->bio->bi_bdev->bd_dev, - clone->bi_sector, sector); + trace_block_bio_remap(bdev_get_queue(clone->bi_bdev), clone, + tio->io->bio->bi_bdev->bd_dev, sector); generic_make_request(clone); } else if (r < 0 || r == DM_MAPIO_REQUEUE) { /* error the io and bail out, or requeue it if needed */ md = tio->io->md; dec_pending(tio->io, r); - /* - * Store bio_set for cleanup. - */ - clone->bi_private = md->bs; - bio_put(clone); free_tio(md, tio); } else if (r) { DMWARN("unimplemented target map return value: %d", r); @@ -633,285 +1219,633 @@ struct clone_info { struct bio *bio; struct dm_io *io; sector_t sector; - sector_t sector_count; - unsigned short idx; + unsigned sector_count; }; -static void dm_bio_destructor(struct bio *bio) +static void bio_setup_sector(struct bio *bio, sector_t sector, unsigned len) { - struct bio_set *bs = bio->bi_private; - - bio_free(bio, bs); + bio->bi_iter.bi_sector = sector; + bio->bi_iter.bi_size = to_bytes(len); } /* - * Creates a little bio that is just does part of a bvec. + * Creates a bio that consists of range of complete bvecs. */ -static struct bio *split_bvec(struct bio *bio, sector_t sector, - unsigned short idx, unsigned int offset, - unsigned int len, struct bio_set *bs) +static void clone_bio(struct dm_target_io *tio, struct bio *bio, + sector_t sector, unsigned len) { - struct bio *clone; - struct bio_vec *bv = bio->bi_io_vec + idx; + struct bio *clone = &tio->clone; - clone = bio_alloc_bioset(GFP_NOIO, 1, bs); - clone->bi_destructor = dm_bio_destructor; - *clone->bi_io_vec = *bv; + __bio_clone_fast(clone, bio); - clone->bi_sector = sector; - clone->bi_bdev = bio->bi_bdev; - clone->bi_rw = bio->bi_rw; - clone->bi_vcnt = 1; - clone->bi_size = to_bytes(len); - clone->bi_io_vec->bv_offset = offset; - clone->bi_io_vec->bv_len = clone->bi_size; + if (bio_integrity(bio)) + bio_integrity_clone(clone, bio, GFP_NOIO); - return clone; + bio_advance(clone, to_bytes(sector - clone->bi_iter.bi_sector)); + clone->bi_iter.bi_size = to_bytes(len); + + if (bio_integrity(bio)) + bio_integrity_trim(clone, 0, len); } -/* - * Creates a bio that consists of range of complete bvecs. - */ -static struct bio *clone_bio(struct bio *bio, sector_t sector, - unsigned short idx, unsigned short bv_count, - unsigned int len, struct bio_set *bs) +static struct dm_target_io *alloc_tio(struct clone_info *ci, + struct dm_target *ti, int nr_iovecs, + unsigned target_bio_nr) { + struct dm_target_io *tio; struct bio *clone; - clone = bio_alloc_bioset(GFP_NOIO, bio->bi_max_vecs, bs); - __bio_clone(clone, bio); - clone->bi_destructor = dm_bio_destructor; - clone->bi_sector = sector; - clone->bi_idx = idx; - clone->bi_vcnt = idx + bv_count; - clone->bi_size = to_bytes(len); - clone->bi_flags &= ~(1 << BIO_SEG_VALID); + clone = bio_alloc_bioset(GFP_NOIO, nr_iovecs, ci->md->bs); + tio = container_of(clone, struct dm_target_io, clone); - return clone; + tio->io = ci->io; + tio->ti = ti; + tio->target_bio_nr = target_bio_nr; + + return tio; +} + +static void __clone_and_map_simple_bio(struct clone_info *ci, + struct dm_target *ti, + unsigned target_bio_nr, unsigned *len) +{ + struct dm_target_io *tio = alloc_tio(ci, ti, ci->bio->bi_max_vecs, target_bio_nr); + struct bio *clone = &tio->clone; + + tio->len_ptr = len; + + /* + * Discard requests require the bio's inline iovecs be initialized. + * ci->bio->bi_max_vecs is BIO_INLINE_VECS anyway, for both flush + * and discard, so no need for concern about wasted bvec allocations. + */ + __bio_clone_fast(clone, ci->bio); + if (len) + bio_setup_sector(clone, ci->sector, *len); + + __map_bio(tio); +} + +static void __send_duplicate_bios(struct clone_info *ci, struct dm_target *ti, + unsigned num_bios, unsigned *len) +{ + unsigned target_bio_nr; + + for (target_bio_nr = 0; target_bio_nr < num_bios; target_bio_nr++) + __clone_and_map_simple_bio(ci, ti, target_bio_nr, len); } -static int __clone_and_map(struct clone_info *ci) +static int __send_empty_flush(struct clone_info *ci) { - struct bio *clone, *bio = ci->bio; + unsigned target_nr = 0; struct dm_target *ti; - sector_t len = 0, max; - struct dm_target_io *tio; - ti = dm_table_find_target(ci->map, ci->sector); - if (!dm_target_is_valid(ti)) - return -EIO; + BUG_ON(bio_has_data(ci->bio)); + while ((ti = dm_table_get_target(ci->map, target_nr++))) + __send_duplicate_bios(ci, ti, ti->num_flush_bios, NULL); - max = max_io_len(ci->md, ci->sector, ti); + return 0; +} + +static void __clone_and_map_data_bio(struct clone_info *ci, struct dm_target *ti, + sector_t sector, unsigned *len) +{ + struct bio *bio = ci->bio; + struct dm_target_io *tio; + unsigned target_bio_nr; + unsigned num_target_bios = 1; /* - * Allocate a target io object. + * Does the target want to receive duplicate copies of the bio? */ - tio = alloc_tio(ci->md); - tio->io = ci->io; - tio->ti = ti; - memset(&tio->info, 0, sizeof(tio->info)); + if (bio_data_dir(bio) == WRITE && ti->num_write_bios) + num_target_bios = ti->num_write_bios(ti, bio); + + for (target_bio_nr = 0; target_bio_nr < num_target_bios; target_bio_nr++) { + tio = alloc_tio(ci, ti, 0, target_bio_nr); + tio->len_ptr = len; + clone_bio(tio, bio, sector, *len); + __map_bio(tio); + } +} - if (ci->sector_count <= max) { - /* - * Optimise for the simple case where we can do all of - * the remaining io with a single clone. - */ - clone = clone_bio(bio, ci->sector, ci->idx, - bio->bi_vcnt - ci->idx, ci->sector_count, - ci->md->bs); - __map_bio(ti, clone, tio); - ci->sector_count = 0; +typedef unsigned (*get_num_bios_fn)(struct dm_target *ti); - } else if (to_sector(bio->bi_io_vec[ci->idx].bv_len) <= max) { - /* - * There are some bvecs that don't span targets. - * Do as many of these as possible. - */ - int i; - sector_t remaining = max; - sector_t bv_len; +static unsigned get_num_discard_bios(struct dm_target *ti) +{ + return ti->num_discard_bios; +} - for (i = ci->idx; remaining && (i < bio->bi_vcnt); i++) { - bv_len = to_sector(bio->bi_io_vec[i].bv_len); +static unsigned get_num_write_same_bios(struct dm_target *ti) +{ + return ti->num_write_same_bios; +} - if (bv_len > remaining) - break; +typedef bool (*is_split_required_fn)(struct dm_target *ti); - remaining -= bv_len; - len += bv_len; - } +static bool is_split_required_for_discard(struct dm_target *ti) +{ + return ti->split_discard_bios; +} - clone = clone_bio(bio, ci->sector, ci->idx, i - ci->idx, len, - ci->md->bs); - __map_bio(ti, clone, tio); +static int __send_changing_extent_only(struct clone_info *ci, + get_num_bios_fn get_num_bios, + is_split_required_fn is_split_required) +{ + struct dm_target *ti; + unsigned len; + unsigned num_bios; - ci->sector += len; - ci->sector_count -= len; - ci->idx = i; + do { + ti = dm_table_find_target(ci->map, ci->sector); + if (!dm_target_is_valid(ti)) + return -EIO; - } else { /* - * Handle a bvec that must be split between two or more targets. + * Even though the device advertised support for this type of + * request, that does not mean every target supports it, and + * reconfiguration might also have changed that since the + * check was performed. */ - struct bio_vec *bv = bio->bi_io_vec + ci->idx; - sector_t remaining = to_sector(bv->bv_len); - unsigned int offset = 0; + num_bios = get_num_bios ? get_num_bios(ti) : 0; + if (!num_bios) + return -EOPNOTSUPP; - do { - if (offset) { - ti = dm_table_find_target(ci->map, ci->sector); - if (!dm_target_is_valid(ti)) - return -EIO; + if (is_split_required && !is_split_required(ti)) + len = min((sector_t)ci->sector_count, max_io_len_target_boundary(ci->sector, ti)); + else + len = min((sector_t)ci->sector_count, max_io_len(ci->sector, ti)); - max = max_io_len(ci->md, ci->sector, ti); + __send_duplicate_bios(ci, ti, num_bios, &len); - tio = alloc_tio(ci->md); - tio->io = ci->io; - tio->ti = ti; - memset(&tio->info, 0, sizeof(tio->info)); - } + ci->sector += len; + } while (ci->sector_count -= len); - len = min(remaining, max); + return 0; +} - clone = split_bvec(bio, ci->sector, ci->idx, - bv->bv_offset + offset, len, - ci->md->bs); +static int __send_discard(struct clone_info *ci) +{ + return __send_changing_extent_only(ci, get_num_discard_bios, + is_split_required_for_discard); +} - __map_bio(ti, clone, tio); +static int __send_write_same(struct clone_info *ci) +{ + return __send_changing_extent_only(ci, get_num_write_same_bios, NULL); +} - ci->sector += len; - ci->sector_count -= len; - offset += to_bytes(len); - } while (remaining -= len); +/* + * Select the correct strategy for processing a non-flush bio. + */ +static int __split_and_process_non_flush(struct clone_info *ci) +{ + struct bio *bio = ci->bio; + struct dm_target *ti; + unsigned len; - ci->idx++; - } + if (unlikely(bio->bi_rw & REQ_DISCARD)) + return __send_discard(ci); + else if (unlikely(bio->bi_rw & REQ_WRITE_SAME)) + return __send_write_same(ci); + + ti = dm_table_find_target(ci->map, ci->sector); + if (!dm_target_is_valid(ti)) + return -EIO; + + len = min_t(sector_t, max_io_len(ci->sector, ti), ci->sector_count); + + __clone_and_map_data_bio(ci, ti, ci->sector, &len); + + ci->sector += len; + ci->sector_count -= len; return 0; } /* - * Split the bio into several clones. + * Entry point to split a bio into clones and submit them to the targets. */ -static int __split_bio(struct mapped_device *md, struct bio *bio) +static void __split_and_process_bio(struct mapped_device *md, + struct dm_table *map, struct bio *bio) { struct clone_info ci; int error = 0; - ci.map = dm_get_table(md); - if (unlikely(!ci.map)) - return -EIO; + if (unlikely(!map)) { + bio_io_error(bio); + return; + } + ci.map = map; ci.md = md; - ci.bio = bio; ci.io = alloc_io(md); ci.io->error = 0; atomic_set(&ci.io->io_count, 1); ci.io->bio = bio; ci.io->md = md; - ci.sector = bio->bi_sector; - ci.sector_count = bio_sectors(bio); - ci.idx = bio->bi_idx; + spin_lock_init(&ci.io->endio_lock); + ci.sector = bio->bi_iter.bi_sector; start_io_acct(ci.io); - while (ci.sector_count && !error) - error = __clone_and_map(&ci); + + if (bio->bi_rw & REQ_FLUSH) { + ci.bio = &ci.md->flush_bio; + ci.sector_count = 0; + error = __send_empty_flush(&ci); + /* dec_pending submits any data associated with flush */ + } else { + ci.bio = bio; + ci.sector_count = bio_sectors(bio); + while (ci.sector_count && !error) + error = __split_and_process_non_flush(&ci); + } /* drop the extra reference count */ dec_pending(ci.io, error); - dm_table_put(ci.map); - - return 0; } /*----------------------------------------------------------------- * CRUD END *---------------------------------------------------------------*/ +static int dm_merge_bvec(struct request_queue *q, + struct bvec_merge_data *bvm, + struct bio_vec *biovec) +{ + struct mapped_device *md = q->queuedata; + struct dm_table *map = dm_get_live_table_fast(md); + struct dm_target *ti; + sector_t max_sectors; + int max_size = 0; + + if (unlikely(!map)) + goto out; + + ti = dm_table_find_target(map, bvm->bi_sector); + if (!dm_target_is_valid(ti)) + goto out; + + /* + * Find maximum amount of I/O that won't need splitting + */ + max_sectors = min(max_io_len(bvm->bi_sector, ti), + (sector_t) BIO_MAX_SECTORS); + max_size = (max_sectors << SECTOR_SHIFT) - bvm->bi_size; + if (max_size < 0) + max_size = 0; + + /* + * merge_bvec_fn() returns number of bytes + * it can accept at this offset + * max is precomputed maximal io size + */ + if (max_size && ti->type->merge) + max_size = ti->type->merge(ti, bvm, biovec, max_size); + /* + * If the target doesn't support merge method and some of the devices + * provided their merge_bvec method (we know this by looking at + * queue_max_hw_sectors), then we can't allow bios with multiple vector + * entries. So always set max_size to 0, and the code below allows + * just one page. + */ + else if (queue_max_hw_sectors(q) <= PAGE_SIZE >> 9) + max_size = 0; + +out: + dm_put_live_table_fast(md); + /* + * Always allow an entire first page + */ + if (max_size <= biovec->bv_len && !(bvm->bi_size >> SECTOR_SHIFT)) + max_size = biovec->bv_len; + + return max_size; +} + /* * The request function that just remaps the bio built up by * dm_merge_bvec. */ -static int dm_request(struct request_queue *q, struct bio *bio) +static void _dm_request(struct request_queue *q, struct bio *bio) { - int r = -EIO; int rw = bio_data_dir(bio); struct mapped_device *md = q->queuedata; + int cpu; + int srcu_idx; + struct dm_table *map; + + map = dm_get_live_table(md, &srcu_idx); + + cpu = part_stat_lock(); + part_stat_inc(cpu, &dm_disk(md)->part0, ios[rw]); + part_stat_add(cpu, &dm_disk(md)->part0, sectors[rw], bio_sectors(bio)); + part_stat_unlock(); + + /* if we're suspended, we have to queue this io for later */ + if (unlikely(test_bit(DMF_BLOCK_IO_FOR_SUSPEND, &md->flags))) { + dm_put_live_table(md, srcu_idx); + + if (bio_rw(bio) != READA) + queue_io(md, bio); + else + bio_io_error(bio); + return; + } + + __split_and_process_bio(md, map, bio); + dm_put_live_table(md, srcu_idx); + return; +} + +int dm_request_based(struct mapped_device *md) +{ + return blk_queue_stackable(md->queue); +} + +static void dm_request(struct request_queue *q, struct bio *bio) +{ + struct mapped_device *md = q->queuedata; + + if (dm_request_based(md)) + blk_queue_bio(q, bio); + else + _dm_request(q, bio); +} + +void dm_dispatch_request(struct request *rq) +{ + int r; + + if (blk_queue_io_stat(rq->q)) + rq->cmd_flags |= REQ_IO_STAT; + + rq->start_time = jiffies; + r = blk_insert_cloned_request(rq->q, rq); + if (r) + dm_complete_request(rq, r); +} +EXPORT_SYMBOL_GPL(dm_dispatch_request); + +static int dm_rq_bio_constructor(struct bio *bio, struct bio *bio_orig, + void *data) +{ + struct dm_rq_target_io *tio = data; + struct dm_rq_clone_bio_info *info = + container_of(bio, struct dm_rq_clone_bio_info, clone); + + info->orig = bio_orig; + info->tio = tio; + bio->bi_end_io = end_clone_bio; + + return 0; +} + +static int setup_clone(struct request *clone, struct request *rq, + struct dm_rq_target_io *tio) +{ + int r; + + r = blk_rq_prep_clone(clone, rq, tio->md->bs, GFP_ATOMIC, + dm_rq_bio_constructor, tio); + if (r) + return r; + + clone->cmd = rq->cmd; + clone->cmd_len = rq->cmd_len; + clone->sense = rq->sense; + clone->end_io = end_clone_request; + clone->end_io_data = tio; + + return 0; +} + +static struct request *clone_rq(struct request *rq, struct mapped_device *md, + gfp_t gfp_mask) +{ + struct request *clone; + struct dm_rq_target_io *tio; + + tio = alloc_rq_tio(md, gfp_mask); + if (!tio) + return NULL; + + tio->md = md; + tio->ti = NULL; + tio->orig = rq; + tio->error = 0; + memset(&tio->info, 0, sizeof(tio->info)); + + clone = &tio->clone; + if (setup_clone(clone, rq, tio)) { + /* -ENOMEM */ + free_rq_tio(tio); + return NULL; + } + + return clone; +} + +/* + * Called with the queue lock held. + */ +static int dm_prep_fn(struct request_queue *q, struct request *rq) +{ + struct mapped_device *md = q->queuedata; + struct request *clone; + + if (unlikely(rq->special)) { + DMWARN("Already has something in rq->special."); + return BLKPREP_KILL; + } + + clone = clone_rq(rq, md, GFP_ATOMIC); + if (!clone) + return BLKPREP_DEFER; + + rq->special = clone; + rq->cmd_flags |= REQ_DONTPREP; + + return BLKPREP_OK; +} + +/* + * Returns: + * 0 : the request has been processed (not requeued) + * !0 : the request has been requeued + */ +static int map_request(struct dm_target *ti, struct request *clone, + struct mapped_device *md) +{ + int r, requeued = 0; + struct dm_rq_target_io *tio = clone->end_io_data; + + tio->ti = ti; + r = ti->type->map_rq(ti, clone, &tio->info); + switch (r) { + case DM_MAPIO_SUBMITTED: + /* The target has taken the I/O to submit by itself later */ + break; + case DM_MAPIO_REMAPPED: + /* The target has remapped the I/O so dispatch it */ + trace_block_rq_remap(clone->q, clone, disk_devt(dm_disk(md)), + blk_rq_pos(tio->orig)); + dm_dispatch_request(clone); + break; + case DM_MAPIO_REQUEUE: + /* The target wants to requeue the I/O */ + dm_requeue_unmapped_request(clone); + requeued = 1; + break; + default: + if (r > 0) { + DMWARN("unimplemented target map return value: %d", r); + BUG(); + } + + /* The target wants to complete the I/O */ + dm_kill_unmapped_request(clone, r); + break; + } + + return requeued; +} + +static struct request *dm_start_request(struct mapped_device *md, struct request *orig) +{ + struct request *clone; + + blk_start_request(orig); + clone = orig->special; + atomic_inc(&md->pending[rq_data_dir(clone)]); /* - * There is no use in forwarding any barrier request since we can't - * guarantee it is (or can be) handled by the targets correctly. + * Hold the md reference here for the in-flight I/O. + * We can't rely on the reference count by device opener, + * because the device may be closed during the request completion + * when all bios are completed. + * See the comment in rq_completed() too. */ - if (unlikely(bio_barrier(bio))) { - bio_endio(bio, -EOPNOTSUPP); - return 0; - } + dm_get(md); - down_read(&md->io_lock); + return clone; +} - disk_stat_inc(dm_disk(md), ios[rw]); - disk_stat_add(dm_disk(md), sectors[rw], bio_sectors(bio)); +/* + * q->request_fn for request-based dm. + * Called with the queue lock held. + */ +static void dm_request_fn(struct request_queue *q) +{ + struct mapped_device *md = q->queuedata; + int srcu_idx; + struct dm_table *map = dm_get_live_table(md, &srcu_idx); + struct dm_target *ti; + struct request *rq, *clone; + sector_t pos; /* - * If we're suspended we have to queue - * this io for later. + * For suspend, check blk_queue_stopped() and increment + * ->pending within a single queue_lock not to increment the + * number of in-flight I/Os after the queue is stopped in + * dm_suspend(). */ - while (test_bit(DMF_BLOCK_IO, &md->flags)) { - up_read(&md->io_lock); + while (!blk_queue_stopped(q)) { + rq = blk_peek_request(q); + if (!rq) + goto delay_and_out; + + /* always use block 0 to find the target for flushes for now */ + pos = 0; + if (!(rq->cmd_flags & REQ_FLUSH)) + pos = blk_rq_pos(rq); + + ti = dm_table_find_target(map, pos); + if (!dm_target_is_valid(ti)) { + /* + * Must perform setup, that dm_done() requires, + * before calling dm_kill_unmapped_request + */ + DMERR_LIMIT("request attempted access beyond the end of device"); + clone = dm_start_request(md, rq); + dm_kill_unmapped_request(clone, -EIO); + continue; + } - if (bio_rw(bio) != READA) - r = queue_io(md, bio); + if (ti->type->busy && ti->type->busy(ti)) + goto delay_and_out; - if (r <= 0) - goto out_req; + clone = dm_start_request(md, rq); - /* - * We're in a while loop, because someone could suspend - * before we get to the following read lock. - */ - down_read(&md->io_lock); + spin_unlock(q->queue_lock); + if (map_request(ti, clone, md)) + goto requeued; + + BUG_ON(!irqs_disabled()); + spin_lock(q->queue_lock); } - r = __split_bio(md, bio); - up_read(&md->io_lock); + goto out; -out_req: - if (r < 0) - bio_io_error(bio); +requeued: + BUG_ON(!irqs_disabled()); + spin_lock(q->queue_lock); - return 0; +delay_and_out: + blk_delay_queue(q, HZ / 10); +out: + dm_put_live_table(md, srcu_idx); +} + +int dm_underlying_device_busy(struct request_queue *q) +{ + return blk_lld_busy(q); } +EXPORT_SYMBOL_GPL(dm_underlying_device_busy); -static void dm_unplug_all(struct request_queue *q) +static int dm_lld_busy(struct request_queue *q) { + int r; struct mapped_device *md = q->queuedata; - struct dm_table *map = dm_get_table(md); + struct dm_table *map = dm_get_live_table_fast(md); - if (map) { - dm_table_unplug_all(map); - dm_table_put(map); - } + if (!map || test_bit(DMF_BLOCK_IO_FOR_SUSPEND, &md->flags)) + r = 1; + else + r = dm_table_any_busy_target(map); + + dm_put_live_table_fast(md); + + return r; } static int dm_any_congested(void *congested_data, int bdi_bits) { - int r; - struct mapped_device *md = (struct mapped_device *) congested_data; - struct dm_table *map = dm_get_table(md); + int r = bdi_bits; + struct mapped_device *md = congested_data; + struct dm_table *map; - if (!map || test_bit(DMF_BLOCK_IO, &md->flags)) - r = bdi_bits; - else - r = dm_table_any_congested(map, bdi_bits); + if (!test_bit(DMF_BLOCK_IO_FOR_SUSPEND, &md->flags)) { + map = dm_get_live_table_fast(md); + if (map) { + /* + * Request-based dm cares about only own queue for + * the query about congestion status of request_queue + */ + if (dm_request_based(md)) + r = md->queue->backing_dev_info.state & + bdi_bits; + else + r = dm_table_any_congested(map, bdi_bits); + } + dm_put_live_table_fast(md); + } - dm_table_put(map); return r; } /*----------------------------------------------------------------- * An IDR is used to keep track of allocated minor numbers. *---------------------------------------------------------------*/ -static DEFINE_IDR(_minor_idr); - static void free_minor(int minor) { spin_lock(&_minor_lock); @@ -922,68 +1856,66 @@ static void free_minor(int minor) /* * See if the device with a specific minor # is free. */ -static int specific_minor(struct mapped_device *md, int minor) +static int specific_minor(int minor) { - int r, m; + int r; if (minor >= (1 << MINORBITS)) return -EINVAL; - r = idr_pre_get(&_minor_idr, GFP_KERNEL); - if (!r) - return -ENOMEM; - + idr_preload(GFP_KERNEL); spin_lock(&_minor_lock); - if (idr_find(&_minor_idr, minor)) { - r = -EBUSY; - goto out; - } - - r = idr_get_new_above(&_minor_idr, MINOR_ALLOCED, minor, &m); - if (r) - goto out; - - if (m != minor) { - idr_remove(&_minor_idr, m); - r = -EBUSY; - goto out; - } + r = idr_alloc(&_minor_idr, MINOR_ALLOCED, minor, minor + 1, GFP_NOWAIT); -out: spin_unlock(&_minor_lock); - return r; + idr_preload_end(); + if (r < 0) + return r == -ENOSPC ? -EBUSY : r; + return 0; } -static int next_free_minor(struct mapped_device *md, int *minor) +static int next_free_minor(int *minor) { - int r, m; - - r = idr_pre_get(&_minor_idr, GFP_KERNEL); - if (!r) - return -ENOMEM; + int r; + idr_preload(GFP_KERNEL); spin_lock(&_minor_lock); - r = idr_get_new(&_minor_idr, MINOR_ALLOCED, &m); - if (r) { - goto out; - } + r = idr_alloc(&_minor_idr, MINOR_ALLOCED, 0, 1 << MINORBITS, GFP_NOWAIT); - if (m >= (1 << MINORBITS)) { - idr_remove(&_minor_idr, m); - r = -ENOSPC; - goto out; - } - - *minor = m; - -out: spin_unlock(&_minor_lock); - return r; + idr_preload_end(); + if (r < 0) + return r; + *minor = r; + return 0; } -static struct block_device_operations dm_blk_dops; +static const struct block_device_operations dm_blk_dops; + +static void dm_wq_work(struct work_struct *work); + +static void dm_init_md_queue(struct mapped_device *md) +{ + /* + * Request-based dm devices cannot be stacked on top of bio-based dm + * devices. The type of this dm device has not been decided yet. + * The type is decided at the first table loading time. + * To prevent problematic device stacking, clear the queue flag + * for request stacking support until then. + * + * This queue is new, so no concurrency on the queue_flags. + */ + queue_flag_clear_unlocked(QUEUE_FLAG_STACKABLE, md->queue); + + md->queue->queuedata = md; + md->queue->backing_dev_info.congested_fn = dm_any_congested; + md->queue->backing_dev_info.congested_data = md; + blk_queue_make_request(md->queue, dm_request); + blk_queue_bounce_limit(md->queue, BLK_BOUNCE_ANY); + blk_queue_merge_bvec(md->queue, dm_merge_bvec); +} /* * Allocate and initialise a blank device with a given minor. @@ -991,7 +1923,7 @@ static struct block_device_operations dm_blk_dops; static struct mapped_device *alloc_dev(int minor) { int r; - struct mapped_device *md = kmalloc(sizeof(*md), GFP_KERNEL); + struct mapped_device *md = kzalloc(sizeof(*md), GFP_KERNEL); void *old_md; if (!md) { @@ -1004,17 +1936,20 @@ static struct mapped_device *alloc_dev(int minor) /* get a minor number for the dev */ if (minor == DM_ANY_MINOR) - r = next_free_minor(md, &minor); + r = next_free_minor(&minor); else - r = specific_minor(md, minor); + r = specific_minor(minor); if (r < 0) goto bad_minor; - memset(md, 0, sizeof(*md)); - init_rwsem(&md->io_lock); + r = init_srcu_struct(&md->io_barrier); + if (r < 0) + goto bad_io_barrier; + + md->type = DM_TYPE_NONE; mutex_init(&md->suspend_lock); - spin_lock_init(&md->pushback_lock); - rwlock_init(&md->map_lock); + mutex_init(&md->type_lock); + spin_lock_init(&md->deferred_lock); atomic_set(&md->holders, 1); atomic_set(&md->open_count, 0); atomic_set(&md->event_nr, 0); @@ -1026,32 +1961,18 @@ static struct mapped_device *alloc_dev(int minor) if (!md->queue) goto bad_queue; - md->queue->queuedata = md; - md->queue->backing_dev_info.congested_fn = dm_any_congested; - md->queue->backing_dev_info.congested_data = md; - blk_queue_make_request(md->queue, dm_request); - blk_queue_bounce_limit(md->queue, BLK_BOUNCE_ANY); - md->queue->unplug_fn = dm_unplug_all; - - md->io_pool = mempool_create_slab_pool(MIN_IOS, _io_cache); - if (!md->io_pool) - goto bad_io_pool; - - md->tio_pool = mempool_create_slab_pool(MIN_IOS, _tio_cache); - if (!md->tio_pool) - goto bad_tio_pool; - - md->bs = bioset_create(16, 16); - if (!md->bs) - goto bad_no_bioset; + dm_init_md_queue(md); md->disk = alloc_disk(1); if (!md->disk) goto bad_disk; - atomic_set(&md->pending, 0); + atomic_set(&md->pending[0], 0); + atomic_set(&md->pending[1], 0); init_waitqueue_head(&md->wait); + INIT_WORK(&md->work, dm_wq_work); init_waitqueue_head(&md->eventq); + init_completion(&md->kobj_holder.completion); md->disk->major = _major; md->disk->first_minor = minor; @@ -1062,10 +1983,20 @@ static struct mapped_device *alloc_dev(int minor) add_disk(md->disk); format_dev_t(md->name, MKDEV(_major, minor)); - md->wq = create_singlethread_workqueue("kdmflush"); + md->wq = alloc_workqueue("kdmflush", WQ_MEM_RECLAIM, 0); if (!md->wq) goto bad_thread; + md->bdev = bdget_disk(md->disk, 0); + if (!md->bdev) + goto bad_bdev; + + bio_init(&md->flush_bio); + md->flush_bio.bi_bdev = md->bdev; + md->flush_bio.bi_rw = WRITE_FLUSH; + + dm_stats_init(&md->stats); + /* Populate the mapping, nobody knows we exist yet */ spin_lock(&_minor_lock); old_md = idr_replace(&_minor_idr, md, minor); @@ -1075,17 +2006,16 @@ static struct mapped_device *alloc_dev(int minor) return md; +bad_bdev: + destroy_workqueue(md->wq); bad_thread: + del_gendisk(md->disk); put_disk(md->disk); bad_disk: - bioset_free(md->bs); -bad_no_bioset: - mempool_destroy(md->tio_pool); -bad_tio_pool: - mempool_destroy(md->io_pool); -bad_io_pool: blk_cleanup_queue(md->queue); bad_queue: + cleanup_srcu_struct(&md->io_barrier); +bad_io_barrier: free_minor(minor); bad_minor: module_put(THIS_MODULE); @@ -1098,17 +2028,18 @@ static void unlock_fs(struct mapped_device *md); static void free_dev(struct mapped_device *md) { - int minor = md->disk->first_minor; + int minor = MINOR(disk_devt(md->disk)); - if (md->suspended_bdev) { - unlock_fs(md); - bdput(md->suspended_bdev); - } + unlock_fs(md); + bdput(md->bdev); destroy_workqueue(md->wq); - mempool_destroy(md->tio_pool); - mempool_destroy(md->io_pool); - bioset_free(md->bs); + if (md->io_pool) + mempool_destroy(md->io_pool); + if (md->bs) + bioset_free(md->bs); + blk_integrity_unregister(md->disk); del_gendisk(md->disk); + cleanup_srcu_struct(&md->io_barrier); free_minor(minor); spin_lock(&_minor_lock); @@ -1117,10 +2048,50 @@ static void free_dev(struct mapped_device *md) put_disk(md->disk); blk_cleanup_queue(md->queue); + dm_stats_cleanup(&md->stats); module_put(THIS_MODULE); kfree(md); } +static void __bind_mempools(struct mapped_device *md, struct dm_table *t) +{ + struct dm_md_mempools *p = dm_table_get_md_mempools(t); + + if (md->io_pool && md->bs) { + /* The md already has necessary mempools. */ + if (dm_table_get_type(t) == DM_TYPE_BIO_BASED) { + /* + * Reload bioset because front_pad may have changed + * because a different table was loaded. + */ + bioset_free(md->bs); + md->bs = p->bs; + p->bs = NULL; + } else if (dm_table_get_type(t) == DM_TYPE_REQUEST_BASED) { + /* + * There's no need to reload with request-based dm + * because the size of front_pad doesn't change. + * Note for future: If you are to reload bioset, + * prep-ed requests in the queue may refer + * to bio from the old bioset, so you must walk + * through the queue to unprep. + */ + } + goto out; + } + + BUG_ON(!p || md->io_pool || md->bs); + + md->io_pool = p->io_pool; + p->io_pool = NULL; + md->bs = p->bs; + p->bs = NULL; + +out: + /* mempool bind completed, now no need any mempools in the table */ + dm_table_free_md_mempools(t); +} + /* * Bind a table to the device. */ @@ -1134,62 +2105,141 @@ static void event_callback(void *context) list_splice_init(&md->uevent_list, &uevents); spin_unlock_irqrestore(&md->uevent_lock, flags); - dm_send_uevents(&uevents, &md->disk->dev.kobj); + dm_send_uevents(&uevents, &disk_to_dev(md->disk)->kobj); atomic_inc(&md->event_nr); wake_up(&md->eventq); } +/* + * Protected by md->suspend_lock obtained by dm_swap_table(). + */ static void __set_size(struct mapped_device *md, sector_t size) { set_capacity(md->disk, size); - mutex_lock(&md->suspended_bdev->bd_inode->i_mutex); - i_size_write(md->suspended_bdev->bd_inode, (loff_t)size << SECTOR_SHIFT); - mutex_unlock(&md->suspended_bdev->bd_inode->i_mutex); + i_size_write(md->bdev->bd_inode, (loff_t)size << SECTOR_SHIFT); } -static int __bind(struct mapped_device *md, struct dm_table *t) +/* + * Return 1 if the queue has a compulsory merge_bvec_fn function. + * + * If this function returns 0, then the device is either a non-dm + * device without a merge_bvec_fn, or it is a dm device that is + * able to split any bios it receives that are too big. + */ +int dm_queue_merge_is_compulsory(struct request_queue *q) { + struct mapped_device *dev_md; + + if (!q->merge_bvec_fn) + return 0; + + if (q->make_request_fn == dm_request) { + dev_md = q->queuedata; + if (test_bit(DMF_MERGE_IS_OPTIONAL, &dev_md->flags)) + return 0; + } + + return 1; +} + +static int dm_device_merge_is_compulsory(struct dm_target *ti, + struct dm_dev *dev, sector_t start, + sector_t len, void *data) +{ + struct block_device *bdev = dev->bdev; + struct request_queue *q = bdev_get_queue(bdev); + + return dm_queue_merge_is_compulsory(q); +} + +/* + * Return 1 if it is acceptable to ignore merge_bvec_fn based + * on the properties of the underlying devices. + */ +static int dm_table_merge_is_optional(struct dm_table *table) +{ + unsigned i = 0; + struct dm_target *ti; + + while (i < dm_table_get_num_targets(table)) { + ti = dm_table_get_target(table, i++); + + if (ti->type->iterate_devices && + ti->type->iterate_devices(ti, dm_device_merge_is_compulsory, NULL)) + return 0; + } + + return 1; +} + +/* + * Returns old map, which caller must destroy. + */ +static struct dm_table *__bind(struct mapped_device *md, struct dm_table *t, + struct queue_limits *limits) +{ + struct dm_table *old_map; struct request_queue *q = md->queue; sector_t size; + int merge_is_optional; size = dm_table_get_size(t); /* * Wipe any geometry if the size of the table changed. */ - if (size != get_capacity(md->disk)) + if (size != dm_get_size(md)) memset(&md->geometry, 0, sizeof(md->geometry)); - if (md->suspended_bdev) - __set_size(md, size); - if (size == 0) - return 0; + __set_size(md, size); - dm_table_get(t); dm_table_event_callback(t, event_callback, md); - write_lock(&md->map_lock); - md->map = t; - dm_table_set_restrictions(t, q); - write_unlock(&md->map_lock); + /* + * The queue hasn't been stopped yet, if the old table type wasn't + * for request-based during suspension. So stop it to prevent + * I/O mapping before resume. + * This must be done before setting the queue restrictions, + * because request-based dm may be run just after the setting. + */ + if (dm_table_request_based(t) && !blk_queue_stopped(q)) + stop_queue(q); - return 0; + __bind_mempools(md, t); + + merge_is_optional = dm_table_merge_is_optional(t); + + old_map = md->map; + rcu_assign_pointer(md->map, t); + md->immutable_target_type = dm_table_get_immutable_target_type(t); + + dm_table_set_restrictions(t, q, limits); + if (merge_is_optional) + set_bit(DMF_MERGE_IS_OPTIONAL, &md->flags); + else + clear_bit(DMF_MERGE_IS_OPTIONAL, &md->flags); + dm_sync_table(md); + + return old_map; } -static void __unbind(struct mapped_device *md) +/* + * Returns unbound table for the caller to free. + */ +static struct dm_table *__unbind(struct mapped_device *md) { struct dm_table *map = md->map; if (!map) - return; + return NULL; dm_table_event_callback(map, NULL, NULL); - write_lock(&md->map_lock); - md->map = NULL; - write_unlock(&md->map_lock); - dm_table_put(map); + RCU_INIT_POINTER(md->map, NULL); + dm_sync_table(md); + + return map; } /* @@ -1203,10 +2253,94 @@ int dm_create(int minor, struct mapped_device **result) if (!md) return -ENXIO; + dm_sysfs_init(md); + *result = md; return 0; } +/* + * Functions to manage md->type. + * All are required to hold md->type_lock. + */ +void dm_lock_md_type(struct mapped_device *md) +{ + mutex_lock(&md->type_lock); +} + +void dm_unlock_md_type(struct mapped_device *md) +{ + mutex_unlock(&md->type_lock); +} + +void dm_set_md_type(struct mapped_device *md, unsigned type) +{ + BUG_ON(!mutex_is_locked(&md->type_lock)); + md->type = type; +} + +unsigned dm_get_md_type(struct mapped_device *md) +{ + BUG_ON(!mutex_is_locked(&md->type_lock)); + return md->type; +} + +struct target_type *dm_get_immutable_target_type(struct mapped_device *md) +{ + return md->immutable_target_type; +} + +/* + * The queue_limits are only valid as long as you have a reference + * count on 'md'. + */ +struct queue_limits *dm_get_queue_limits(struct mapped_device *md) +{ + BUG_ON(!atomic_read(&md->holders)); + return &md->queue->limits; +} +EXPORT_SYMBOL_GPL(dm_get_queue_limits); + +/* + * Fully initialize a request-based queue (->elevator, ->request_fn, etc). + */ +static int dm_init_request_based_queue(struct mapped_device *md) +{ + struct request_queue *q = NULL; + + if (md->queue->elevator) + return 1; + + /* Fully initialize the queue */ + q = blk_init_allocated_queue(md->queue, dm_request_fn, NULL); + if (!q) + return 0; + + md->queue = q; + dm_init_md_queue(md); + blk_queue_softirq_done(md->queue, dm_softirq_done); + blk_queue_prep_rq(md->queue, dm_prep_fn); + blk_queue_lld_busy(md->queue, dm_lld_busy); + + elv_register_queue(md->queue); + + return 1; +} + +/* + * Setup the DM device's queue based on md's type + */ +int dm_setup_md_queue(struct mapped_device *md) +{ + if ((dm_get_md_type(md) == DM_TYPE_REQUEST_BASED) && + !dm_init_request_based_queue(md)) { + DMWARN("Cannot initialize queue for request-based mapped device"); + return -EINVAL; + } + + return 0; +} + static struct mapped_device *dm_find_md(dev_t dev) { struct mapped_device *md; @@ -1219,7 +2353,8 @@ static struct mapped_device *dm_find_md(dev_t dev) md = idr_find(&_minor_idr, minor); if (md && (md == MINOR_ALLOCED || - (dm_disk(md)->first_minor != minor) || + (MINOR(disk_devt(dm_disk(md))) != minor) || + dm_deleting_md(md) || test_bit(DMF_FREEING, &md->flags))) { md = NULL; goto out; @@ -1240,6 +2375,7 @@ struct mapped_device *dm_get_md(dev_t dev) return md; } +EXPORT_SYMBOL_GPL(dm_get_md); void *dm_get_mdptr(struct mapped_device *md) { @@ -1254,6 +2390,7 @@ void dm_set_mdptr(struct mapped_device *md, void *ptr) void dm_get(struct mapped_device *md) { atomic_inc(&md->holders); + BUG_ON(test_bit(DMF_FREEING, &md->flags)); } const char *dm_device_name(struct mapped_device *md) @@ -1262,40 +2399,76 @@ const char *dm_device_name(struct mapped_device *md) } EXPORT_SYMBOL_GPL(dm_device_name); -void dm_put(struct mapped_device *md) +static void __dm_destroy(struct mapped_device *md, bool wait) { struct dm_table *map; + int srcu_idx; - BUG_ON(test_bit(DMF_FREEING, &md->flags)); + might_sleep(); - if (atomic_dec_and_lock(&md->holders, &_minor_lock)) { - map = dm_get_table(md); - idr_replace(&_minor_idr, MINOR_ALLOCED, dm_disk(md)->first_minor); - set_bit(DMF_FREEING, &md->flags); - spin_unlock(&_minor_lock); - if (!dm_suspended(md)) { - dm_table_presuspend_targets(map); - dm_table_postsuspend_targets(map); - } - __unbind(md); - dm_table_put(map); - free_dev(md); + spin_lock(&_minor_lock); + map = dm_get_live_table(md, &srcu_idx); + idr_replace(&_minor_idr, MINOR_ALLOCED, MINOR(disk_devt(dm_disk(md)))); + set_bit(DMF_FREEING, &md->flags); + spin_unlock(&_minor_lock); + + if (!dm_suspended_md(md)) { + dm_table_presuspend_targets(map); + dm_table_postsuspend_targets(map); } + + /* dm_put_live_table must be before msleep, otherwise deadlock is possible */ + dm_put_live_table(md, srcu_idx); + + /* + * Rare, but there may be I/O requests still going to complete, + * for example. Wait for all references to disappear. + * No one should increment the reference count of the mapped_device, + * after the mapped_device state becomes DMF_FREEING. + */ + if (wait) + while (atomic_read(&md->holders)) + msleep(1); + else if (atomic_read(&md->holders)) + DMWARN("%s: Forcibly removing mapped_device still in use! (%d users)", + dm_device_name(md), atomic_read(&md->holders)); + + dm_sysfs_exit(md); + dm_table_destroy(__unbind(md)); + free_dev(md); +} + +void dm_destroy(struct mapped_device *md) +{ + __dm_destroy(md, true); +} + +void dm_destroy_immediate(struct mapped_device *md) +{ + __dm_destroy(md, false); +} + +void dm_put(struct mapped_device *md) +{ + atomic_dec(&md->holders); } EXPORT_SYMBOL_GPL(dm_put); -static int dm_wait_for_completion(struct mapped_device *md) +static int dm_wait_for_completion(struct mapped_device *md, int interruptible) { int r = 0; + DECLARE_WAITQUEUE(wait, current); + + add_wait_queue(&md->wait, &wait); while (1) { - set_current_state(TASK_INTERRUPTIBLE); + set_current_state(interruptible); - smp_mb(); - if (!atomic_read(&md->pending)) + if (!md_in_flight(md)) break; - if (signal_pending(current)) { + if (interruptible == TASK_INTERRUPTIBLE && + signal_pending(current)) { r = -EINTR; break; } @@ -1304,97 +2477,89 @@ static int dm_wait_for_completion(struct mapped_device *md) } set_current_state(TASK_RUNNING); + remove_wait_queue(&md->wait, &wait); + return r; } /* * Process the deferred bios */ -static void __flush_deferred_io(struct mapped_device *md) +static void dm_wq_work(struct work_struct *work) { + struct mapped_device *md = container_of(work, struct mapped_device, + work); struct bio *c; + int srcu_idx; + struct dm_table *map; - while ((c = bio_list_pop(&md->deferred))) { - if (__split_bio(md, c)) - bio_io_error(c); - } - - clear_bit(DMF_BLOCK_IO, &md->flags); -} - -static void __merge_pushback_list(struct mapped_device *md) -{ - unsigned long flags; + map = dm_get_live_table(md, &srcu_idx); - spin_lock_irqsave(&md->pushback_lock, flags); - clear_bit(DMF_NOFLUSH_SUSPENDING, &md->flags); - bio_list_merge_head(&md->deferred, &md->pushback); - bio_list_init(&md->pushback); - spin_unlock_irqrestore(&md->pushback_lock, flags); -} + while (!test_bit(DMF_BLOCK_IO_FOR_SUSPEND, &md->flags)) { + spin_lock_irq(&md->deferred_lock); + c = bio_list_pop(&md->deferred); + spin_unlock_irq(&md->deferred_lock); -static void dm_wq_work(struct work_struct *work) -{ - struct dm_wq_req *req = container_of(work, struct dm_wq_req, work); - struct mapped_device *md = req->md; + if (!c) + break; - down_write(&md->io_lock); - switch (req->type) { - case DM_WQ_FLUSH_ALL: - __merge_pushback_list(md); - /* pass through */ - case DM_WQ_FLUSH_DEFERRED: - __flush_deferred_io(md); - break; - default: - DMERR("dm_wq_work: unrecognised work type %d", req->type); - BUG(); + if (dm_request_based(md)) + generic_make_request(c); + else + __split_and_process_bio(md, map, c); } - up_write(&md->io_lock); -} -static void dm_wq_queue(struct mapped_device *md, int type, void *context, - struct dm_wq_req *req) -{ - req->type = type; - req->md = md; - req->context = context; - INIT_WORK(&req->work, dm_wq_work); - queue_work(md->wq, &req->work); + dm_put_live_table(md, srcu_idx); } -static void dm_queue_flush(struct mapped_device *md, int type, void *context) +static void dm_queue_flush(struct mapped_device *md) { - struct dm_wq_req req; - - dm_wq_queue(md, type, context, &req); - flush_workqueue(md->wq); + clear_bit(DMF_BLOCK_IO_FOR_SUSPEND, &md->flags); + smp_mb__after_atomic(); + queue_work(md->wq, &md->work); } /* - * Swap in a new table (destroying old one). + * Swap in a new table, returning the old one for the caller to destroy. */ -int dm_swap_table(struct mapped_device *md, struct dm_table *table) +struct dm_table *dm_swap_table(struct mapped_device *md, struct dm_table *table) { - int r = -EINVAL; + struct dm_table *live_map = NULL, *map = ERR_PTR(-EINVAL); + struct queue_limits limits; + int r; mutex_lock(&md->suspend_lock); /* device must be suspended */ - if (!dm_suspended(md)) + if (!dm_suspended_md(md)) goto out; - /* without bdev, the device size cannot be changed */ - if (!md->suspended_bdev) - if (get_capacity(md->disk) != dm_table_get_size(table)) + /* + * If the new table has no data devices, retain the existing limits. + * This helps multipath with queue_if_no_path if all paths disappear, + * then new I/O is queued based on these limits, and then some paths + * reappear. + */ + if (dm_table_has_no_data_devices(table)) { + live_map = dm_get_live_table_fast(md); + if (live_map) + limits = md->queue->limits; + dm_put_live_table_fast(md); + } + + if (!live_map) { + r = dm_calculate_queue_limits(table, &limits); + if (r) { + map = ERR_PTR(r); goto out; + } + } - __unbind(md); - r = __bind(md, table); + map = __bind(md, table, &limits); out: mutex_unlock(&md->suspend_lock); - return r; + return map; } /* @@ -1407,7 +2572,7 @@ static int lock_fs(struct mapped_device *md) WARN_ON(md->frozen_sb); - md->frozen_sb = freeze_bdev(md->suspended_bdev); + md->frozen_sb = freeze_bdev(md->bdev); if (IS_ERR(md->frozen_sb)) { r = PTR_ERR(md->frozen_sb); md->frozen_sb = NULL; @@ -1416,9 +2581,6 @@ static int lock_fs(struct mapped_device *md) set_bit(DMF_FROZEN, &md->flags); - /* don't bdput right now, we don't want the bdev - * to go away while it is locked. - */ return 0; } @@ -1427,7 +2589,7 @@ static void unlock_fs(struct mapped_device *md) if (!test_bit(DMF_FROZEN, &md->flags)) return; - thaw_bdev(md->suspended_bdev, md->frozen_sb); + thaw_bdev(md->bdev, md->frozen_sb); md->frozen_sb = NULL; clear_bit(DMF_FROZEN, &md->flags); } @@ -1439,22 +2601,30 @@ static void unlock_fs(struct mapped_device *md) * dm_bind_table, dm_suspend must be called to flush any in * flight bios and ensure that any further io gets deferred. */ +/* + * Suspend mechanism in request-based dm. + * + * 1. Flush all I/Os by lock_fs() if needed. + * 2. Stop dispatching any I/O by stopping the request_queue. + * 3. Wait for all in-flight I/Os to be completed or requeued. + * + * To abort suspend, start the request_queue. + */ int dm_suspend(struct mapped_device *md, unsigned suspend_flags) { struct dm_table *map = NULL; - DECLARE_WAITQUEUE(wait, current); int r = 0; int do_lockfs = suspend_flags & DM_SUSPEND_LOCKFS_FLAG ? 1 : 0; int noflush = suspend_flags & DM_SUSPEND_NOFLUSH_FLAG ? 1 : 0; mutex_lock(&md->suspend_lock); - if (dm_suspended(md)) { + if (dm_suspended_md(md)) { r = -EINVAL; goto out_unlock; } - map = dm_get_table(md); + map = md->map; /* * DMF_NOFLUSH_SUSPENDING must be set before presuspend. @@ -1466,78 +2636,73 @@ int dm_suspend(struct mapped_device *md, unsigned suspend_flags) /* This does not get reverted if there's an error later. */ dm_table_presuspend_targets(map); - /* bdget() can stall if the pending I/Os are not flushed */ - if (!noflush) { - md->suspended_bdev = bdget_disk(md->disk, 0); - if (!md->suspended_bdev) { - DMWARN("bdget failed in dm_suspend"); - r = -ENOMEM; - goto flush_and_out; - } - - /* - * Flush I/O to the device. noflush supersedes do_lockfs, - * because lock_fs() needs to flush I/Os. - */ - if (do_lockfs) { - r = lock_fs(md); - if (r) - goto out; - } + /* + * Flush I/O to the device. + * Any I/O submitted after lock_fs() may not be flushed. + * noflush takes precedence over do_lockfs. + * (lock_fs() flushes I/Os and waits for them to complete.) + */ + if (!noflush && do_lockfs) { + r = lock_fs(md); + if (r) + goto out_unlock; } /* - * First we set the BLOCK_IO flag so no more ios will be mapped. + * Here we must make sure that no processes are submitting requests + * to target drivers i.e. no one may be executing + * __split_and_process_bio. This is called from dm_request and + * dm_wq_work. + * + * To get all processes out of __split_and_process_bio in dm_request, + * we take the write lock. To prevent any process from reentering + * __split_and_process_bio from dm_request and quiesce the thread + * (dm_wq_work), we set BMF_BLOCK_IO_FOR_SUSPEND and call + * flush_workqueue(md->wq). */ - down_write(&md->io_lock); - set_bit(DMF_BLOCK_IO, &md->flags); + set_bit(DMF_BLOCK_IO_FOR_SUSPEND, &md->flags); + synchronize_srcu(&md->io_barrier); - add_wait_queue(&md->wait, &wait); - up_write(&md->io_lock); + /* + * Stop md->queue before flushing md->wq in case request-based + * dm defers requests to md->wq from md->queue. + */ + if (dm_request_based(md)) + stop_queue(md->queue); - /* unplug */ - if (map) - dm_table_unplug_all(map); + flush_workqueue(md->wq); /* - * Wait for the already-mapped ios to complete. + * At this point no more requests are entering target request routines. + * We call dm_wait_for_completion to wait for all existing requests + * to finish. */ - r = dm_wait_for_completion(md); - - down_write(&md->io_lock); - remove_wait_queue(&md->wait, &wait); + r = dm_wait_for_completion(md, TASK_INTERRUPTIBLE); if (noflush) - __merge_pushback_list(md); - up_write(&md->io_lock); + clear_bit(DMF_NOFLUSH_SUSPENDING, &md->flags); + synchronize_srcu(&md->io_barrier); /* were we interrupted ? */ if (r < 0) { - dm_queue_flush(md, DM_WQ_FLUSH_DEFERRED, NULL); + dm_queue_flush(md); + + if (dm_request_based(md)) + start_queue(md->queue); unlock_fs(md); - goto out; /* pushback list is already flushed, so skip flush */ + goto out_unlock; /* pushback list is already flushed, so skip flush */ } - dm_table_postsuspend_targets(map); + /* + * If dm_wait_for_completion returned 0, the device is completely + * quiescent now. There is no request-processing activity. All new + * requests are being added to md->deferred list. + */ set_bit(DMF_SUSPENDED, &md->flags); -flush_and_out: - if (r && noflush) - /* - * Because there may be already I/Os in the pushback list, - * flush them before return. - */ - dm_queue_flush(md, DM_WQ_FLUSH_ALL, NULL); - -out: - if (r && md->suspended_bdev) { - bdput(md->suspended_bdev); - md->suspended_bdev = NULL; - } - - dm_table_put(map); + dm_table_postsuspend_targets(map); out_unlock: mutex_unlock(&md->suspend_lock); @@ -1550,10 +2715,10 @@ int dm_resume(struct mapped_device *md) struct dm_table *map = NULL; mutex_lock(&md->suspend_lock); - if (!dm_suspended(md)) + if (!dm_suspended_md(md)) goto out; - map = dm_get_table(md); + map = md->map; if (!map || !dm_table_get_size(map)) goto out; @@ -1561,36 +2726,76 @@ int dm_resume(struct mapped_device *md) if (r) goto out; - dm_queue_flush(md, DM_WQ_FLUSH_DEFERRED, NULL); + dm_queue_flush(md); - unlock_fs(md); + /* + * Flushing deferred I/Os must be done after targets are resumed + * so that mapping of targets can work correctly. + * Request-based dm is queueing the deferred I/Os in its request_queue. + */ + if (dm_request_based(md)) + start_queue(md->queue); - if (md->suspended_bdev) { - bdput(md->suspended_bdev); - md->suspended_bdev = NULL; - } + unlock_fs(md); clear_bit(DMF_SUSPENDED, &md->flags); - dm_table_unplug_all(map); - - dm_kobject_uevent(md); - r = 0; - out: - dm_table_put(map); mutex_unlock(&md->suspend_lock); return r; } +/* + * Internal suspend/resume works like userspace-driven suspend. It waits + * until all bios finish and prevents issuing new bios to the target drivers. + * It may be used only from the kernel. + * + * Internal suspend holds md->suspend_lock, which prevents interaction with + * userspace-driven suspend. + */ + +void dm_internal_suspend(struct mapped_device *md) +{ + mutex_lock(&md->suspend_lock); + if (dm_suspended_md(md)) + return; + + set_bit(DMF_BLOCK_IO_FOR_SUSPEND, &md->flags); + synchronize_srcu(&md->io_barrier); + flush_workqueue(md->wq); + dm_wait_for_completion(md, TASK_UNINTERRUPTIBLE); +} + +void dm_internal_resume(struct mapped_device *md) +{ + if (dm_suspended_md(md)) + goto done; + + dm_queue_flush(md); + +done: + mutex_unlock(&md->suspend_lock); +} + /*----------------------------------------------------------------- * Event notification. *---------------------------------------------------------------*/ -void dm_kobject_uevent(struct mapped_device *md) -{ - kobject_uevent(&md->disk->dev.kobj, KOBJ_CHANGE); +int dm_kobject_uevent(struct mapped_device *md, enum kobject_action action, + unsigned cookie) +{ + char udev_cookie[DM_COOKIE_LENGTH]; + char *envp[] = { udev_cookie, NULL }; + + if (!cookie) + return kobject_uevent(&disk_to_dev(md->disk)->kobj, action); + else { + snprintf(udev_cookie, DM_COOKIE_LENGTH, "%s=%u", + DM_COOKIE_ENV_VAR_NAME, cookie); + return kobject_uevent_env(&disk_to_dev(md->disk)->kobj, + action, envp); + } } uint32_t dm_next_uevent_seq(struct mapped_device *md) @@ -1627,23 +2832,104 @@ struct gendisk *dm_disk(struct mapped_device *md) return md->disk; } -int dm_suspended(struct mapped_device *md) +struct kobject *dm_kobject(struct mapped_device *md) +{ + return &md->kobj_holder.kobj; +} + +struct mapped_device *dm_get_from_kobject(struct kobject *kobj) +{ + struct mapped_device *md; + + md = container_of(kobj, struct mapped_device, kobj_holder.kobj); + + if (test_bit(DMF_FREEING, &md->flags) || + dm_deleting_md(md)) + return NULL; + + dm_get(md); + return md; +} + +int dm_suspended_md(struct mapped_device *md) { return test_bit(DMF_SUSPENDED, &md->flags); } -int dm_noflush_suspending(struct dm_target *ti) +int dm_test_deferred_remove_flag(struct mapped_device *md) { - struct mapped_device *md = dm_table_get_md(ti->table); - int r = __noflush_suspending(md); + return test_bit(DMF_DEFERRED_REMOVE, &md->flags); +} - dm_put(md); +int dm_suspended(struct dm_target *ti) +{ + return dm_suspended_md(dm_table_get_md(ti->table)); +} +EXPORT_SYMBOL_GPL(dm_suspended); - return r; +int dm_noflush_suspending(struct dm_target *ti) +{ + return __noflush_suspending(dm_table_get_md(ti->table)); } EXPORT_SYMBOL_GPL(dm_noflush_suspending); -static struct block_device_operations dm_blk_dops = { +struct dm_md_mempools *dm_alloc_md_mempools(unsigned type, unsigned integrity, unsigned per_bio_data_size) +{ + struct dm_md_mempools *pools = kzalloc(sizeof(*pools), GFP_KERNEL); + struct kmem_cache *cachep; + unsigned int pool_size; + unsigned int front_pad; + + if (!pools) + return NULL; + + if (type == DM_TYPE_BIO_BASED) { + cachep = _io_cache; + pool_size = dm_get_reserved_bio_based_ios(); + front_pad = roundup(per_bio_data_size, __alignof__(struct dm_target_io)) + offsetof(struct dm_target_io, clone); + } else if (type == DM_TYPE_REQUEST_BASED) { + cachep = _rq_tio_cache; + pool_size = dm_get_reserved_rq_based_ios(); + front_pad = offsetof(struct dm_rq_clone_bio_info, clone); + /* per_bio_data_size is not used. See __bind_mempools(). */ + WARN_ON(per_bio_data_size != 0); + } else + goto out; + + pools->io_pool = mempool_create_slab_pool(pool_size, cachep); + if (!pools->io_pool) + goto out; + + pools->bs = bioset_create(pool_size, front_pad); + if (!pools->bs) + goto out; + + if (integrity && bioset_integrity_create(pools->bs, pool_size)) + goto out; + + return pools; + +out: + dm_free_md_mempools(pools); + + return NULL; +} + +void dm_free_md_mempools(struct dm_md_mempools *pools) +{ + if (!pools) + return; + + if (pools->io_pool) + mempool_destroy(pools->io_pool); + + if (pools->bs) + bioset_free(pools->bs); + + kfree(pools); +} + +static const struct block_device_operations dm_blk_dops = { .open = dm_blk_open, .release = dm_blk_close, .ioctl = dm_blk_ioctl, @@ -1651,8 +2937,6 @@ static struct block_device_operations dm_blk_dops = { .owner = THIS_MODULE }; -EXPORT_SYMBOL(dm_get_mapinfo); - /* * module hooks */ @@ -1661,6 +2945,13 @@ module_exit(dm_exit); module_param(major, uint, 0); MODULE_PARM_DESC(major, "The major number of the device mapper"); + +module_param(reserved_bio_based_ios, uint, S_IRUGO | S_IWUSR); +MODULE_PARM_DESC(reserved_bio_based_ios, "Reserved IOs in bio-based mempools"); + +module_param(reserved_rq_based_ios, uint, S_IRUGO | S_IWUSR); +MODULE_PARM_DESC(reserved_rq_based_ios, "Reserved IOs in request-based mempools"); + MODULE_DESCRIPTION(DM_NAME " driver"); MODULE_AUTHOR("Joe Thornber <dm-devel@redhat.com>"); MODULE_LICENSE("GPL"); |