From a1b51e98676932d031f5eec1325b2df4bbdc8f26 Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Tue, 6 Jan 2009 03:04:53 +0000 Subject: dm table: drop reference at unbind Move one dm_table_put() so that the last reference in the thread gets dropped in __unbind(). This is required for a following patch, dm-table-rework-reference-counting.patch, which will change the logic in such a way that table destructor is called only at specific points in the code. Signed-off-by: Mikulas Patocka Signed-off-by: Alasdair G Kergon --- drivers/md/dm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/md/dm.c') diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 421c9f02d8ca..82371412029f 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -1330,8 +1330,8 @@ void dm_put(struct mapped_device *md) dm_table_presuspend_targets(map); dm_table_postsuspend_targets(map); } - __unbind(md); dm_table_put(map); + __unbind(md); free_dev(md); } } -- cgit v1.2.2 From 8fbf26ad5b16ad3a826ca7fe3e86700420abed1f Mon Sep 17 00:00:00 2001 From: Kiyoshi Ueda Date: Tue, 6 Jan 2009 03:05:06 +0000 Subject: dm request: add caches This patch prepares some kmem_caches for request-based dm. Signed-off-by: Kiyoshi Ueda Signed-off-by: Jun'ichi Nomura Signed-off-by: Alasdair G Kergon --- drivers/md/dm.c | 41 ++++++++++++++++++++++++++++++++++++++++- 1 file changed, 40 insertions(+), 1 deletion(-) (limited to 'drivers/md/dm.c') diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 82371412029f..4882ce7e88a3 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -32,6 +32,7 @@ static unsigned int _major = 0; static DEFINE_SPINLOCK(_minor_lock); /* + * For bio-based dm. * One of these is allocated per bio. */ struct dm_io { @@ -43,6 +44,7 @@ struct dm_io { }; /* + * For bio-based dm. * One of these is allocated per target within a bio. Hopefully * this will be simplified out one day. */ @@ -54,6 +56,27 @@ struct dm_target_io { DEFINE_TRACE(block_bio_complete); +/* + * For request-based dm. + * One of these is allocated per request. + */ +struct dm_rq_target_io { + struct mapped_device *md; + struct dm_target *ti; + struct request *orig, clone; + int error; + union map_info info; +}; + +/* + * For request-based dm. + * One of these is allocated per bio. + */ +struct dm_rq_clone_bio_info { + struct bio *orig; + struct request *rq; +}; + union map_info *dm_get_mapinfo(struct bio *bio) { if (bio && bio->bi_private) @@ -149,6 +172,8 @@ struct mapped_device { #define MIN_IOS 256 static struct kmem_cache *_io_cache; static struct kmem_cache *_tio_cache; +static struct kmem_cache *_rq_tio_cache; +static struct kmem_cache *_rq_bio_info_cache; static int __init local_init(void) { @@ -164,9 +189,17 @@ static int __init local_init(void) if (!_tio_cache) goto out_free_io_cache; + _rq_tio_cache = KMEM_CACHE(dm_rq_target_io, 0); + if (!_rq_tio_cache) + goto out_free_tio_cache; + + _rq_bio_info_cache = KMEM_CACHE(dm_rq_clone_bio_info, 0); + if (!_rq_bio_info_cache) + goto out_free_rq_tio_cache; + r = dm_uevent_init(); if (r) - goto out_free_tio_cache; + goto out_free_rq_bio_info_cache; _major = major; r = register_blkdev(_major, _name); @@ -180,6 +213,10 @@ static int __init local_init(void) out_uevent_exit: dm_uevent_exit(); +out_free_rq_bio_info_cache: + kmem_cache_destroy(_rq_bio_info_cache); +out_free_rq_tio_cache: + kmem_cache_destroy(_rq_tio_cache); out_free_tio_cache: kmem_cache_destroy(_tio_cache); out_free_io_cache: @@ -190,6 +227,8 @@ out_free_io_cache: static void local_exit(void) { + kmem_cache_destroy(_rq_bio_info_cache); + kmem_cache_destroy(_rq_tio_cache); kmem_cache_destroy(_tio_cache); kmem_cache_destroy(_io_cache); unregister_blkdev(_major, _name); -- cgit v1.2.2 From ab4c1424882be9cd70b89abf2b484add355712fa Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 6 Jan 2009 03:05:09 +0000 Subject: dm: support barriers on simple devices Implement barrier support for single device DM devices This patch implements barrier support in DM for the common case of dm linear just remapping a single underlying device. In this case we can safely pass the barrier through because there can be no reordering between devices. NB. Any DM device might cease to support barriers if it gets reconfigured so code must continue to allow for a possible -EOPNOTSUPP on every barrier bio submitted. - agk Signed-off-by: Andi Kleen Signed-off-by: Mikulas Patocka Signed-off-by: Alasdair G Kergon --- drivers/md/dm.c | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-) (limited to 'drivers/md/dm.c') diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 4882ce7e88a3..dd953b189f45 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -835,7 +835,11 @@ static int __split_bio(struct mapped_device *md, struct bio *bio) ci.map = dm_get_table(md); if (unlikely(!ci.map)) return -EIO; - + if (unlikely(bio_barrier(bio) && !dm_table_barrier_ok(ci.map))) { + dm_table_put(ci.map); + bio_endio(bio, -EOPNOTSUPP); + return 0; + } ci.md = md; ci.bio = bio; ci.io = alloc_io(md); @@ -919,15 +923,6 @@ static int dm_request(struct request_queue *q, struct bio *bio) struct mapped_device *md = q->queuedata; int cpu; - /* - * There is no use in forwarding any barrier request since we can't - * guarantee it is (or can be) handled by the targets correctly. - */ - if (unlikely(bio_barrier(bio))) { - bio_endio(bio, -EOPNOTSUPP); - return 0; - } - down_read(&md->io_lock); cpu = part_stat_lock(); -- cgit v1.2.2 From d58168763f74d1edbc296d7038c60efe6493fdd4 Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Tue, 6 Jan 2009 03:05:10 +0000 Subject: dm table: rework reference counting Rework table reference counting. The existing code uses a reference counter. When the last reference is dropped and the counter reaches zero, the table destructor is called. Table reference counters are acquired/released from upcalls from other kernel code (dm_any_congested, dm_merge_bvec, dm_unplug_all). If the reference counter reaches zero in one of the upcalls, the table destructor is called from almost random kernel code. This leads to various problems: * dm_any_congested being called under a spinlock, which calls the destructor, which calls some sleeping function. * the destructor attempting to take a lock that is already taken by the same process. * stale reference from some other kernel code keeps the table constructed, which keeps some devices open, even after successful return from "dmsetup remove". This can confuse lvm and prevent closing of underlying devices or reusing device minor numbers. The patch changes reference counting so that the table destructor can be called only at predetermined places. The table has always exactly one reference from either mapped_device->map or hash_cell->new_map. After this patch, this reference is not counted in table->holders. A pair of dm_create_table/dm_destroy_table functions is used for table creation/destruction. Temporary references from the other code increase table->holders. A pair of dm_table_get/dm_table_put functions is used to manipulate it. When the table is about to be destroyed, we wait for table->holders to reach 0. Then, we call the table destructor. We use active waiting with msleep(1), because the situation happens rarely (to one user in 5 years) and removing the device isn't performance-critical task: the user doesn't care if it takes one tick more or not. This way, the destructor is called only at specific points (dm_table_destroy function) and the above problems associated with lazy destruction can't happen. Finally remove the temporary protection added to dm_any_congested(). Signed-off-by: Mikulas Patocka Signed-off-by: Alasdair G Kergon --- drivers/md/dm.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) (limited to 'drivers/md/dm.c') diff --git a/drivers/md/dm.c b/drivers/md/dm.c index dd953b189f45..9f9aa64f7336 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -977,8 +977,6 @@ static int dm_any_congested(void *congested_data, int bdi_bits) struct mapped_device *md = congested_data; struct dm_table *map; - atomic_inc(&md->pending); - if (!test_bit(DMF_BLOCK_IO, &md->flags)) { map = dm_get_table(md); if (map) { @@ -987,10 +985,6 @@ static int dm_any_congested(void *congested_data, int bdi_bits) } } - if (!atomic_dec_return(&md->pending)) - /* nudge anyone waiting on suspend queue */ - wake_up(&md->wait); - return r; } @@ -1250,10 +1244,12 @@ static int __bind(struct mapped_device *md, struct dm_table *t) if (md->suspended_bdev) __set_size(md, size); - if (size == 0) + + if (!size) { + dm_table_destroy(t); return 0; + } - dm_table_get(t); dm_table_event_callback(t, event_callback, md); write_lock(&md->map_lock); @@ -1275,7 +1271,7 @@ static void __unbind(struct mapped_device *md) write_lock(&md->map_lock); md->map = NULL; write_unlock(&md->map_lock); - dm_table_put(map); + dm_table_destroy(map); } /* -- cgit v1.2.2 From 784aae735d9b0bba3f8b9faef4c8b30df3bf0128 Mon Sep 17 00:00:00 2001 From: Milan Broz Date: Tue, 6 Jan 2009 03:05:12 +0000 Subject: dm: add name and uuid to sysfs Implement simple read-only sysfs entry for device-mapper block device. This patch adds a simple sysfs directory named "dm" under block device properties and implements - name attribute (string containing mapped device name) - uuid attribute (string containing UUID, or empty string if not set) The kobject is embedded in mapped_device struct, so no additional memory allocation is needed for initializing sysfs entry. During the processing of sysfs attribute we need to lock mapped device which is done by a new function dm_get_from_kobj, which returns the md associated with kobject and increases the usage count. Each 'show attribute' function is responsible for its own locking. Signed-off-by: Milan Broz Signed-off-by: Alasdair G Kergon --- drivers/md/dm.c | 29 ++++++++++++++++++++++++++++- 1 file changed, 28 insertions(+), 1 deletion(-) (limited to 'drivers/md/dm.c') diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 9f9aa64f7336..51ba1db4b3e7 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -1,6 +1,6 @@ /* * 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. */ @@ -167,6 +167,9 @@ struct mapped_device { /* forced geometry settings */ struct hd_geometry geometry; + + /* sysfs handle */ + struct kobject kobj; }; #define MIN_IOS 256 @@ -1285,6 +1288,8 @@ int dm_create(int minor, struct mapped_device **result) if (!md) return -ENXIO; + dm_sysfs_init(md); + *result = md; return 0; } @@ -1360,6 +1365,7 @@ void dm_put(struct mapped_device *md) dm_table_presuspend_targets(map); dm_table_postsuspend_targets(map); } + dm_sysfs_exit(md); dm_table_put(map); __unbind(md); free_dev(md); @@ -1699,6 +1705,27 @@ struct gendisk *dm_disk(struct mapped_device *md) return md->disk; } +struct kobject *dm_kobject(struct mapped_device *md) +{ + return &md->kobj; +} + +/* + * struct mapped_device should not be exported outside of dm.c + * so use this check to verify that kobj is part of md structure + */ +struct mapped_device *dm_get_from_kobject(struct kobject *kobj) +{ + struct mapped_device *md; + + md = container_of(kobj, struct mapped_device, kobj); + if (&md->kobj != kobj) + return NULL; + + dm_get(md); + return md; +} + int dm_suspended(struct mapped_device *md) { return test_bit(DMF_SUSPENDED, &md->flags); -- cgit v1.2.2