diff options
author | David Vrabel <david.vrabel@csr.com> | 2008-10-20 16:07:19 +0100 |
---|---|---|
committer | David Vrabel <david.vrabel@csr.com> | 2008-10-20 16:07:19 +0100 |
commit | 61e0e79ee3c609eb34edf2fe023708cba6a79b1f (patch) | |
tree | 663deacffd4071120dc9badb70428fe5f124c7b9 /drivers/md | |
parent | c15895ef30c2c03e99802951787183039a349d32 (diff) | |
parent | 0cfd81031a26717fe14380d18275f8e217571615 (diff) |
Merge branch 'master' into for-upstream
Conflicts:
Documentation/ABI/testing/sysfs-bus-usb
drivers/Makefile
Diffstat (limited to 'drivers/md')
-rw-r--r-- | drivers/md/Kconfig | 14 | ||||
-rw-r--r-- | drivers/md/dm-crypt.c | 109 | ||||
-rw-r--r-- | drivers/md/dm-exception-store.c | 29 | ||||
-rw-r--r-- | drivers/md/dm-ioctl.c | 10 | ||||
-rw-r--r-- | drivers/md/dm-mpath.c | 52 | ||||
-rw-r--r-- | drivers/md/dm-mpath.h | 2 | ||||
-rw-r--r-- | drivers/md/dm-raid1.c | 4 | ||||
-rw-r--r-- | drivers/md/dm-stripe.c | 4 | ||||
-rw-r--r-- | drivers/md/dm-table.c | 97 | ||||
-rw-r--r-- | drivers/md/dm.c | 40 | ||||
-rw-r--r-- | drivers/md/dm.h | 10 | ||||
-rw-r--r-- | drivers/md/faulty.c | 2 | ||||
-rw-r--r-- | drivers/md/linear.c | 143 | ||||
-rw-r--r-- | drivers/md/md.c | 96 | ||||
-rw-r--r-- | drivers/md/multipath.c | 21 | ||||
-rw-r--r-- | drivers/md/raid0.c | 15 | ||||
-rw-r--r-- | drivers/md/raid1.c | 14 | ||||
-rw-r--r-- | drivers/md/raid10.c | 20 | ||||
-rw-r--r-- | drivers/md/raid5.c | 117 | ||||
-rw-r--r-- | drivers/md/raid6.h | 9 |
20 files changed, 442 insertions, 366 deletions
diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig index 07d92c11b5d..2281b5098e9 100644 --- a/drivers/md/Kconfig +++ b/drivers/md/Kconfig @@ -30,6 +30,20 @@ config BLK_DEV_MD If unsure, say N. +config MD_AUTODETECT + bool "Autodetect RAID arrays during kernel boot" + depends on BLK_DEV_MD=y + default y + ---help--- + If you say Y here, then the kernel will try to autodetect raid + arrays as part of its boot process. + + If you don't use raid and say Y, this autodetection can cause + a several-second delay in the boot time due to various + synchronisation steps that are part of this step. + + If unsure, say Y. + config MD_LINEAR tristate "Linear (append) mode" depends on BLK_DEV_MD diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index 13956437bc8..682ef9e6acd 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -333,7 +333,6 @@ static void crypt_convert_init(struct crypt_config *cc, ctx->idx_out = bio_out ? bio_out->bi_idx : 0; ctx->sector = sector + cc->iv_offset; init_completion(&ctx->restart); - atomic_set(&ctx->pending, 1); } static int crypt_convert_block(struct crypt_config *cc, @@ -408,6 +407,8 @@ static int crypt_convert(struct crypt_config *cc, { int r; + atomic_set(&ctx->pending, 1); + while(ctx->idx_in < ctx->bio_in->bi_vcnt && ctx->idx_out < ctx->bio_out->bi_vcnt) { @@ -456,9 +457,11 @@ static void dm_crypt_bio_destructor(struct bio *bio) /* * Generate a new unfragmented bio with the given size * This should never violate the device limitations - * May return a smaller bio when running out of pages + * May return a smaller bio when running out of pages, indicated by + * *out_of_pages set to 1. */ -static struct bio *crypt_alloc_buffer(struct dm_crypt_io *io, unsigned size) +static struct bio *crypt_alloc_buffer(struct dm_crypt_io *io, unsigned size, + unsigned *out_of_pages) { struct crypt_config *cc = io->target->private; struct bio *clone; @@ -472,11 +475,14 @@ static struct bio *crypt_alloc_buffer(struct dm_crypt_io *io, unsigned size) return NULL; clone_init(io, clone); + *out_of_pages = 0; for (i = 0; i < nr_iovecs; i++) { page = mempool_alloc(cc->page_pool, gfp_mask); - if (!page) + if (!page) { + *out_of_pages = 1; break; + } /* * if additional pages cannot be allocated without waiting, @@ -517,6 +523,27 @@ static void crypt_free_buffer_pages(struct crypt_config *cc, struct bio *clone) } } +static struct dm_crypt_io *crypt_io_alloc(struct dm_target *ti, + struct bio *bio, sector_t sector) +{ + struct crypt_config *cc = ti->private; + struct dm_crypt_io *io; + + io = mempool_alloc(cc->io_pool, GFP_NOIO); + io->target = ti; + io->base_bio = bio; + io->sector = sector; + io->error = 0; + atomic_set(&io->pending, 0); + + return io; +} + +static void crypt_inc_pending(struct dm_crypt_io *io) +{ + atomic_inc(&io->pending); +} + /* * One of the bios was finished. Check for completion of * the whole request and correctly clean up the buffer. @@ -591,7 +618,7 @@ static void kcryptd_io_read(struct dm_crypt_io *io) struct bio *base_bio = io->base_bio; struct bio *clone; - atomic_inc(&io->pending); + crypt_inc_pending(io); /* * The block layer might modify the bvec array, so always @@ -653,6 +680,7 @@ static void kcryptd_crypt_write_io_submit(struct dm_crypt_io *io, crypt_free_buffer_pages(cc, clone); bio_put(clone); io->error = -EIO; + crypt_dec_pending(io); return; } @@ -664,28 +692,34 @@ static void kcryptd_crypt_write_io_submit(struct dm_crypt_io *io, if (async) kcryptd_queue_io(io); - else { - atomic_inc(&io->pending); + else generic_make_request(clone); - } } -static void kcryptd_crypt_write_convert_loop(struct dm_crypt_io *io) +static void kcryptd_crypt_write_convert(struct dm_crypt_io *io) { struct crypt_config *cc = io->target->private; struct bio *clone; + int crypt_finished; + unsigned out_of_pages = 0; unsigned remaining = io->base_bio->bi_size; int r; /* + * Prevent io from disappearing until this function completes. + */ + crypt_inc_pending(io); + crypt_convert_init(cc, &io->ctx, NULL, io->base_bio, io->sector); + + /* * The allocated buffers can be smaller than the whole bio, * so repeat the whole process until all the data can be handled. */ while (remaining) { - clone = crypt_alloc_buffer(io, remaining); + clone = crypt_alloc_buffer(io, remaining, &out_of_pages); if (unlikely(!clone)) { io->error = -ENOMEM; - return; + break; } io->ctx.bio_out = clone; @@ -693,37 +727,32 @@ static void kcryptd_crypt_write_convert_loop(struct dm_crypt_io *io) remaining -= clone->bi_size; + crypt_inc_pending(io); r = crypt_convert(cc, &io->ctx); + crypt_finished = atomic_dec_and_test(&io->ctx.pending); - if (atomic_dec_and_test(&io->ctx.pending)) { - /* processed, no running async crypto */ + /* Encryption was already finished, submit io now */ + if (crypt_finished) { kcryptd_crypt_write_io_submit(io, r, 0); - if (unlikely(r < 0)) - return; - } else - atomic_inc(&io->pending); - /* out of memory -> run queues */ - if (unlikely(remaining)) { - /* wait for async crypto then reinitialize pending */ - wait_event(cc->writeq, !atomic_read(&io->ctx.pending)); - atomic_set(&io->ctx.pending, 1); - congestion_wait(WRITE, HZ/100); + /* + * If there was an error, do not try next fragments. + * For async, error is processed in async handler. + */ + if (unlikely(r < 0)) + break; } - } -} -static void kcryptd_crypt_write_convert(struct dm_crypt_io *io) -{ - struct crypt_config *cc = io->target->private; - - /* - * Prevent io from disappearing until this function completes. - */ - atomic_inc(&io->pending); + /* + * Out of memory -> run queues + * But don't wait if split was due to the io size restriction + */ + if (unlikely(out_of_pages)) + congestion_wait(WRITE, HZ/100); - crypt_convert_init(cc, &io->ctx, NULL, io->base_bio, io->sector); - kcryptd_crypt_write_convert_loop(io); + if (unlikely(remaining)) + wait_event(cc->writeq, !atomic_read(&io->ctx.pending)); + } crypt_dec_pending(io); } @@ -741,7 +770,7 @@ static void kcryptd_crypt_read_convert(struct dm_crypt_io *io) struct crypt_config *cc = io->target->private; int r = 0; - atomic_inc(&io->pending); + crypt_inc_pending(io); crypt_convert_init(cc, &io->ctx, io->base_bio, io->base_bio, io->sector); @@ -1108,15 +1137,9 @@ static void crypt_dtr(struct dm_target *ti) static int crypt_map(struct dm_target *ti, struct bio *bio, union map_info *map_context) { - struct crypt_config *cc = ti->private; struct dm_crypt_io *io; - io = mempool_alloc(cc->io_pool, GFP_NOIO); - io->target = ti; - io->base_bio = bio; - io->sector = bio->bi_sector - ti->begin; - io->error = 0; - atomic_set(&io->pending, 0); + io = crypt_io_alloc(ti, bio, bio->bi_sector - ti->begin); if (bio_data_dir(io->base_bio) == READ) kcryptd_queue_io(io); diff --git a/drivers/md/dm-exception-store.c b/drivers/md/dm-exception-store.c index 41f408068a7..769ab677f8e 100644 --- a/drivers/md/dm-exception-store.c +++ b/drivers/md/dm-exception-store.c @@ -108,12 +108,12 @@ struct pstore { * Used to keep track of which metadata area the data in * 'chunk' refers to. */ - uint32_t current_area; + chunk_t current_area; /* * The next free chunk for an exception. */ - uint32_t next_free; + chunk_t next_free; /* * The index of next free exception in the current @@ -175,7 +175,7 @@ static void do_metadata(struct work_struct *work) /* * Read or write a chunk aligned and sized block of data from a device. */ -static int chunk_io(struct pstore *ps, uint32_t chunk, int rw, int metadata) +static int chunk_io(struct pstore *ps, chunk_t chunk, int rw, int metadata) { struct dm_io_region where = { .bdev = ps->snap->cow->bdev, @@ -209,16 +209,23 @@ static int chunk_io(struct pstore *ps, uint32_t chunk, int rw, int metadata) } /* + * Convert a metadata area index to a chunk index. + */ +static chunk_t area_location(struct pstore *ps, chunk_t area) +{ + return 1 + ((ps->exceptions_per_area + 1) * area); +} + +/* * Read or write a metadata area. Remembering to skip the first * chunk which holds the header. */ -static int area_io(struct pstore *ps, uint32_t area, int rw) +static int area_io(struct pstore *ps, chunk_t area, int rw) { int r; - uint32_t chunk; + chunk_t chunk; - /* convert a metadata area index to a chunk index */ - chunk = 1 + ((ps->exceptions_per_area + 1) * area); + chunk = area_location(ps, area); r = chunk_io(ps, chunk, rw, 0); if (r) @@ -228,7 +235,7 @@ static int area_io(struct pstore *ps, uint32_t area, int rw) return 0; } -static int zero_area(struct pstore *ps, uint32_t area) +static int zero_area(struct pstore *ps, chunk_t area) { memset(ps->area, 0, ps->snap->chunk_size << SECTOR_SHIFT); return area_io(ps, area, WRITE); @@ -404,7 +411,7 @@ static int insert_exceptions(struct pstore *ps, int *full) static int read_exceptions(struct pstore *ps) { - uint32_t area; + chunk_t area; int r, full = 1; /* @@ -517,6 +524,7 @@ static int persistent_prepare(struct exception_store *store, { struct pstore *ps = get_info(store); uint32_t stride; + chunk_t next_free; sector_t size = get_dev_size(store->snap->cow->bdev); /* Is there enough room ? */ @@ -530,7 +538,8 @@ static int persistent_prepare(struct exception_store *store, * into account the location of the metadata chunks. */ stride = (ps->exceptions_per_area + 1); - if ((++ps->next_free % stride) == 1) + next_free = ++ps->next_free; + if (sector_div(next_free, stride) == 1) ps->next_free++; atomic_inc(&ps->pending_count); diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index b262c0042de..dca401dc70a 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -426,7 +426,7 @@ static int list_devices(struct dm_ioctl *param, size_t param_size) old_nl->next = (uint32_t) ((void *) nl - (void *) old_nl); disk = dm_disk(hc->md); - nl->dev = huge_encode_dev(MKDEV(disk->major, disk->first_minor)); + nl->dev = huge_encode_dev(disk_devt(disk)); nl->next = 0; strcpy(nl->name, hc->name); @@ -539,7 +539,7 @@ static int __dev_status(struct mapped_device *md, struct dm_ioctl *param) if (dm_suspended(md)) param->flags |= DM_SUSPEND_FLAG; - param->dev = huge_encode_dev(MKDEV(disk->major, disk->first_minor)); + param->dev = huge_encode_dev(disk_devt(disk)); /* * Yes, this will be out of date by the time it gets back @@ -548,7 +548,7 @@ static int __dev_status(struct mapped_device *md, struct dm_ioctl *param) */ param->open_count = dm_open_count(md); - if (disk->policy) + if (get_disk_ro(disk)) param->flags |= DM_READONLY_FLAG; param->event_nr = dm_get_event_nr(md); @@ -1131,7 +1131,7 @@ static void retrieve_deps(struct dm_table *table, unsigned int count = 0; struct list_head *tmp; size_t len, needed; - struct dm_dev *dd; + struct dm_dev_internal *dd; struct dm_target_deps *deps; deps = get_result_buffer(param, param_size, &len); @@ -1157,7 +1157,7 @@ static void retrieve_deps(struct dm_table *table, deps->count = count; count = 0; list_for_each_entry (dd, dm_table_get_devices(table), list) - deps->dev[count++] = huge_encode_dev(dd->bdev->bd_dev); + deps->dev[count++] = huge_encode_dev(dd->dm_dev.bdev->bd_dev); param->data_size = param->data_start + needed; } diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index c2fcf28b4c7..9bf3460c554 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -30,9 +30,11 @@ struct pgpath { struct list_head list; struct priority_group *pg; /* Owning PG */ + unsigned is_active; /* Path status */ unsigned fail_count; /* Cumulative failure count */ struct dm_path path; + struct work_struct deactivate_path; }; #define path_to_pgpath(__pgp) container_of((__pgp), struct pgpath, path) @@ -112,6 +114,7 @@ static struct workqueue_struct *kmultipathd, *kmpath_handlerd; static void process_queued_ios(struct work_struct *work); static void trigger_event(struct work_struct *work); static void activate_path(struct work_struct *work); +static void deactivate_path(struct work_struct *work); /*----------------------------------------------- @@ -122,8 +125,10 @@ static struct pgpath *alloc_pgpath(void) { struct pgpath *pgpath = kzalloc(sizeof(*pgpath), GFP_KERNEL); - if (pgpath) - pgpath->path.is_active = 1; + if (pgpath) { + pgpath->is_active = 1; + INIT_WORK(&pgpath->deactivate_path, deactivate_path); + } return pgpath; } @@ -133,6 +138,14 @@ static void free_pgpath(struct pgpath *pgpath) kfree(pgpath); } +static void deactivate_path(struct work_struct *work) +{ + struct pgpath *pgpath = + container_of(work, struct pgpath, deactivate_path); + + blk_abort_queue(pgpath->path.dev->bdev->bd_disk->queue); +} + static struct priority_group *alloc_priority_group(void) { struct priority_group *pg; @@ -563,12 +576,12 @@ static struct pgpath *parse_path(struct arg_set *as, struct path_selector *ps, /* we need at least a path arg */ if (as->argc < 1) { ti->error = "no device given"; - return NULL; + return ERR_PTR(-EINVAL); } p = alloc_pgpath(); if (!p) - return NULL; + return ERR_PTR(-ENOMEM); r = dm_get_device(ti, shift(as), ti->begin, ti->len, dm_table_get_mode(ti->table), &p->path.dev); @@ -596,7 +609,7 @@ static struct pgpath *parse_path(struct arg_set *as, struct path_selector *ps, bad: free_pgpath(p); - return NULL; + return ERR_PTR(r); } static struct priority_group *parse_priority_group(struct arg_set *as, @@ -614,14 +627,14 @@ static struct priority_group *parse_priority_group(struct arg_set *as, if (as->argc < 2) { as->argc = 0; - ti->error = "not enough priority group aruments"; - return NULL; + ti->error = "not enough priority group arguments"; + return ERR_PTR(-EINVAL); } pg = alloc_priority_group(); if (!pg) { ti->error = "couldn't allocate priority group"; - return NULL; + return ERR_PTR(-ENOMEM); } pg->m = m; @@ -654,8 +667,10 @@ static struct priority_group *parse_priority_group(struct arg_set *as, path_args.argv = as->argv; pgpath = parse_path(&path_args, &pg->ps, ti); - if (!pgpath) + if (IS_ERR(pgpath)) { + r = PTR_ERR(pgpath); goto bad; + } pgpath->pg = pg; list_add_tail(&pgpath->list, &pg->pgpaths); @@ -666,7 +681,7 @@ static struct priority_group *parse_priority_group(struct arg_set *as, bad: free_priority_group(pg, ti); - return NULL; + return ERR_PTR(r); } static int parse_hw_handler(struct arg_set *as, struct multipath *m) @@ -785,8 +800,8 @@ static int multipath_ctr(struct dm_target *ti, unsigned int argc, struct priority_group *pg; pg = parse_priority_group(&as, m); - if (!pg) { - r = -EINVAL; + if (IS_ERR(pg)) { + r = PTR_ERR(pg); goto bad; } @@ -834,7 +849,7 @@ static int multipath_map(struct dm_target *ti, struct bio *bio, dm_bio_record(&mpio->details, bio); map_context->ptr = mpio; - bio->bi_rw |= (1 << BIO_RW_FAILFAST); + bio->bi_rw |= (1 << BIO_RW_FAILFAST_TRANSPORT); r = map_io(m, bio, mpio, 0); if (r < 0 || r == DM_MAPIO_REQUEUE) mempool_free(mpio, m->mpio_pool); @@ -852,13 +867,13 @@ static int fail_path(struct pgpath *pgpath) spin_lock_irqsave(&m->lock, flags); - if (!pgpath->path.is_active) + if (!pgpath->is_active) goto out; DMWARN("Failing path %s.", pgpath->path.dev->name); pgpath->pg->ps.type->fail_path(&pgpath->pg->ps, &pgpath->path); - pgpath->path.is_active = 0; + pgpath->is_active = 0; pgpath->fail_count++; m->nr_valid_paths--; @@ -870,6 +885,7 @@ static int fail_path(struct pgpath *pgpath) pgpath->path.dev->name, m->nr_valid_paths); queue_work(kmultipathd, &m->trigger_event); + queue_work(kmultipathd, &pgpath->deactivate_path); out: spin_unlock_irqrestore(&m->lock, flags); @@ -888,7 +904,7 @@ static int reinstate_path(struct pgpath *pgpath) spin_lock_irqsave(&m->lock, flags); - if (pgpath->path.is_active) + if (pgpath->is_active) goto out; if (!pgpath->pg->ps.type->reinstate_path) { @@ -902,7 +918,7 @@ static int reinstate_path(struct pgpath *pgpath) if (r) goto out; - pgpath->path.is_active = 1; + pgpath->is_active = 1; m->current_pgpath = NULL; if (!m->nr_valid_paths++ && m->queue_size) @@ -1290,7 +1306,7 @@ static int multipath_status(struct dm_target *ti, status_type_t type, list_for_each_entry(p, &pg->pgpaths, list) { DMEMIT("%s %s %u ", p->path.dev->name, - p->path.is_active ? "A" : "F", + p->is_active ? "A" : "F", p->fail_count); if (pg->ps.type->status) sz += pg->ps.type->status(&pg->ps, diff --git a/drivers/md/dm-mpath.h b/drivers/md/dm-mpath.h index c198b856a45..e230f719625 100644 --- a/drivers/md/dm-mpath.h +++ b/drivers/md/dm-mpath.h @@ -13,8 +13,6 @@ struct dm_dev; struct dm_path { struct dm_dev *dev; /* Read-only */ - unsigned is_active; /* Read-only */ - void *pscontext; /* For path-selector use */ }; diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c index ff05fe89308..29913e42c4a 100644 --- a/drivers/md/dm-raid1.c +++ b/drivers/md/dm-raid1.c @@ -842,7 +842,9 @@ static int recover(struct mirror_set *ms, struct region *reg) } /* hand to kcopyd */ - set_bit(DM_KCOPYD_IGNORE_ERROR, &flags); + if (!errors_handled(ms)) + set_bit(DM_KCOPYD_IGNORE_ERROR, &flags); + r = dm_kcopyd_copy(ms->kcopyd_client, &from, ms->nr_mirrors - 1, to, flags, recovery_complete, reg); diff --git a/drivers/md/dm-stripe.c b/drivers/md/dm-stripe.c index 4de90ab3968..b745d8ac625 100644 --- a/drivers/md/dm-stripe.c +++ b/drivers/md/dm-stripe.c @@ -284,8 +284,8 @@ static int stripe_end_io(struct dm_target *ti, struct bio *bio, memset(major_minor, 0, sizeof(major_minor)); sprintf(major_minor, "%d:%d", - bio->bi_bdev->bd_disk->major, - bio->bi_bdev->bd_disk->first_minor); + MAJOR(disk_devt(bio->bi_bdev->bd_disk)), + MINOR(disk_devt(bio->bi_bdev->bd_disk))); /* * Test to see which stripe drive triggered the event diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index 61f44140923..a740a6950f5 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -250,7 +250,8 @@ static void free_devices(struct list_head *devices) struct list_head *tmp, *next; list_for_each_safe(tmp, next, devices) { - struct dm_dev *dd = list_entry(tmp, struct dm_dev, list); + struct dm_dev_internal *dd = + list_entry(tmp, struct dm_dev_internal, list); kfree(dd); } } @@ -327,12 +328,12 @@ static int lookup_device(const char *path, dev_t *dev) /* * See if we've already got a device in the list. */ -static struct dm_dev *find_device(struct list_head *l, dev_t dev) +static struct dm_dev_internal *find_device(struct list_head *l, dev_t dev) { - struct dm_dev *dd; + struct dm_dev_internal *dd; list_for_each_entry (dd, l, list) - if (dd->bdev->bd_dev == dev) + if (dd->dm_dev.bdev->bd_dev == dev) return dd; return NULL; @@ -341,45 +342,47 @@ static struct dm_dev *find_device(struct list_head *l, dev_t dev) /* * Open a device so we can use it as a map destination. */ -static int open_dev(struct dm_dev *d, dev_t dev, struct mapped_device *md) +static int open_dev(struct dm_dev_internal *d, dev_t dev, + struct mapped_device *md) { static char *_claim_ptr = "I belong to device-mapper"; struct block_device *bdev; int r; - BUG_ON(d->bdev); + BUG_ON(d->dm_dev.bdev); - bdev = open_by_devnum(dev, d->mode); + bdev = open_by_devnum(dev, d->dm_dev.mode); if (IS_ERR(bdev)) return PTR_ERR(bdev); r = bd_claim_by_disk(bdev, _claim_ptr, dm_disk(md)); if (r) blkdev_put(bdev); else - d->bdev = bdev; + d->dm_dev.bdev = bdev; return r; } /* * Close a device that we've been using. */ -static void close_dev(struct dm_dev *d, struct mapped_device *md) +static void close_dev(struct dm_dev_internal *d, struct mapped_device *md) { - if (!d->bdev) + if (!d->dm_dev.bdev) return; - bd_release_from_disk(d->bdev, dm_disk(md)); - blkdev_put(d->bdev); - d->bdev = NULL; + bd_release_from_disk(d->dm_dev.bdev, dm_disk(md)); + blkdev_put(d->dm_dev.bdev); + d->dm_dev.bdev = NULL; } /* * If possible, this checks an area of a destination device is valid. */ -static int check_device_area(struct dm_dev *dd, sector_t start, sector_t len) +static int check_device_area(struct dm_dev_internal *dd, sector_t start, + sector_t len) { - sector_t dev_size = dd->bdev->bd_inode->i_size >> SECTOR_SHIFT; + sector_t dev_size = dd->dm_dev.bdev->bd_inode->i_size >> SECTOR_SHIFT; if (!dev_size) return 1; @@ -392,16 +395,17 @@ static int check_device_area(struct dm_dev *dd, sector_t start, sector_t len) * careful to leave things as they were if we fail to reopen the * device. */ -static int upgrade_mode(struct dm_dev *dd, int new_mode, struct mapped_device *md) +static int upgrade_mode(struct dm_dev_internal *dd, int new_mode, + struct mapped_device *md) { int r; - struct dm_dev dd_copy; - dev_t dev = dd->bdev->bd_dev; + struct dm_dev_internal dd_copy; + dev_t dev = dd->dm_dev.bdev->bd_dev; dd_copy = *dd; - dd->mode |= new_mode; - dd->bdev = NULL; + dd->dm_dev.mode |= new_mode; + dd->dm_dev.bdev = NULL; r = open_dev(dd, dev, md); if (!r) close_dev(&dd_copy, md); @@ -421,7 +425,7 @@ static int __table_get_device(struct dm_table *t, struct dm_target *ti, { int r; dev_t uninitialized_var(dev); - struct dm_dev *dd; + struct dm_dev_internal *dd; unsigned int major, minor; BUG_ON(!t); @@ -443,20 +447,20 @@ static int __table_get_device(struct dm_table *t, struct dm_target *ti, if (!dd) return -ENOMEM; - dd->mode = mode; - dd->bdev = NULL; + dd->dm_dev.mode = mode; + dd->dm_dev.bdev = NULL; if ((r = open_dev(dd, dev, t->md))) { kfree(dd); return r; } - format_dev_t(dd->name, dev); + format_dev_t(dd->dm_dev.name, dev); atomic_set(&dd->count, 0); list_add(&dd->list, &t->devices); - } else if (dd->mode != (mode | dd->mode)) { + } else if (dd->dm_dev.mode != (mode | dd->dm_dev.mode)) { r = upgrade_mode(dd, mode, t->md); if (r) return r; @@ -465,11 +469,11 @@ static int __table_get_device(struct dm_table *t, struct dm_target *ti, if (!check_device_area(dd, start, len)) { DMWARN("device %s too small for target", path); - dm_put_device(ti, dd); + dm_put_device(ti, &dd->dm_dev); return -EINVAL; } - *result = dd; + *result = &dd->dm_dev; return 0; } @@ -478,6 +482,13 @@ void dm_set_device_limits(struct dm_target *ti, struct block_device *bdev) { struct request_queue *q = bdev_get_queue(bdev); struct io_restrictions *rs = &ti->limits; + char b[BDEVNAME_SIZE]; + + if (unlikely(!q)) { + DMWARN("%s: Cannot set limits for nonexistent device %s", + dm_device_name(ti->table->md), bdevname(bdev, b)); + return; + } /* * Combine the device limits low. @@ -540,8 +551,11 @@ int dm_get_device(struct dm_target *ti, const char *path, sector_t start, /* * Decrement a devices use count and remove it if necessary. */ -void dm_put_device(struct dm_target *ti, struct dm_dev *dd) +void dm_put_device(struct dm_target *ti, struct dm_dev *d) { + struct dm_dev_internal *dd = container_of(d, struct dm_dev_internal, + dm_dev); + if (atomic_dec_and_test(&dd->count)) { close_dev(dd, ti->table->md); list_del(&dd->list); @@ -937,13 +951,20 @@ int dm_table_resume_targets(struct dm_table *t) int dm_table_any_congested(struct dm_table *t, int bdi_bits) { - struct dm_dev *dd; + struct dm_dev_internal *dd; struct list_head *devices = dm_table_get_devices(t); int r = 0; list_for_each_entry(dd, devices, list) { - struct request_queue *q = bdev_get_queue(dd->bdev); - r |= bdi_congested(&q->backing_dev_info, bdi_bits); + struct request_queue *q = bdev_get_queue(dd->dm_dev.bdev); + char b[BDEVNAME_SIZE]; + + if (likely(q)) + r |= bdi_congested(&q->backing_dev_info, bdi_bits); + else + DMWARN_LIMIT("%s: any_congested: nonexistent device %s", + dm_device_name(t->md), + bdevname(dd->dm_dev.bdev, b)); } return r; @@ -951,13 +972,19 @@ int dm_table_any_congested(struct dm_table *t, int bdi_bits) void dm_table_unplug_all(struct dm_table *t) { - struct dm_dev *dd; + struct dm_dev_internal *dd; struct list_head *devices = dm_table_get_devices(t); list_for_each_entry(dd, devices, list) { - struct request_queue *q = bdev_get_queue(dd->bdev); - - blk_unplug(q); + struct request_queue *q = bdev_get_queue(dd->dm_dev.bdev); + char b[BDEVNAME_SIZE]; + + if (likely(q)) + blk_unplug(q); + else + DMWARN_LIMIT("%s: Cannot unplug nonexistent device %s", + dm_device_name(t->md), + bdevname(dd->dm_dev.bdev, b)); } } diff --git a/drivers/md/dm.c b/drivers/md/dm.c index ace998ce59f..327de03a5bd 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -377,13 +377,14 @@ static void free_tio(struct mapped_device *md, struct dm_target_io *tio) static void start_io_acct(struct dm_io *io) { struct mapped_device *md = io->md; + int cpu; 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(); + dm_disk(md)->part0.in_flight = atomic_inc_return(&md->pending); } static int end_io_acct(struct dm_io *io) @@ -391,15 +392,16 @@ static int 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); + dm_disk(md)->part0.in_flight = pending = + atomic_dec_return(&md->pending); return !pending; } @@ -885,6 +887,7 @@ static int 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; /* * There is no use in forwarding any barrier request since we can't @@ -897,8 +900,10 @@ static int dm_request(struct request_queue *q, struct bio *bio) down_read(&md->io_lock); - disk_stat_inc(dm_disk(md), ios[rw]); - disk_stat_add(dm_disk(md), sectors[rw], bio_sectors(bio)); + 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 @@ -1146,7 +1151,7 @@ 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); @@ -1182,7 +1187,7 @@ 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); @@ -1267,7 +1272,7 @@ 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) || test_bit(DMF_FREEING, &md->flags))) { md = NULL; goto out; @@ -1318,7 +1323,8 @@ void dm_put(struct mapped_device *md) 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); + 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)) { @@ -1638,7 +1644,7 @@ out: *---------------------------------------------------------------*/ void dm_kobject_uevent(struct mapped_device *md) { - kobject_uevent(&md->disk->dev.kobj, KOBJ_CHANGE); + kobject_uevent(&disk_to_dev(md->disk)->kobj, KOBJ_CHANGE); } uint32_t dm_next_uevent_seq(struct mapped_device *md) diff --git a/drivers/md/dm.h b/drivers/md/dm.h index 1e59a0b0a78..cd189da2b2f 100644 --- a/drivers/md/dm.h +++ b/drivers/md/dm.h @@ -25,13 +25,10 @@ /* * List of devices that a metadevice uses and should open/close. */ -struct dm_dev { +struct dm_dev_internal { struct list_head list; - atomic_t count; - int mode; - struct block_device *bdev; - char name[16]; + struct dm_dev dm_dev; }; struct dm_table; @@ -49,7 +46,6 @@ void dm_table_presuspend_targets(struct dm_table *t); void dm_table_postsuspend_targets(struct dm_table *t); int dm_table_resume_targets(struct dm_table *t); int dm_table_any_congested(struct dm_table *t, int bdi_bits); -void dm_table_unplug_all(struct dm_table *t); /* * To check the return value from dm_table_find_target(). @@ -93,8 +89,6 @@ void dm_linear_exit(void); int dm_stripe_init(void); void dm_stripe_exit(void); -void *dm_vcalloc(unsigned long nmemb, unsigned long elem_size); -union map_info *dm_get_mapinfo(struct bio *bio); int dm_open_count(struct mapped_device *md); int dm_lock_for_deletion(struct mapped_device *md); diff --git a/drivers/md/faulty.c b/drivers/md/faulty.c index 268547dbfbd..f26c1f9a475 100644 --- a/drivers/md/faulty.c +++ b/drivers/md/faulty.c @@ -287,6 +287,8 @@ static int run(mddev_t *mddev) int i; conf_t *conf = kmalloc(sizeof(*conf), GFP_KERNEL); + if (!conf) + return -ENOMEM; for (i=0; i<Modes; i++) { atomic_set(&conf->counters[i], 0); diff --git a/drivers/md/linear.c b/drivers/md/linear.c index b1eebf88c20..190147c79e7 100644 --- a/drivers/md/linear.c +++ b/drivers/md/linear.c @@ -16,16 +16,8 @@ Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ -#include <linux/module.h> - -#include <linux/raid/md.h> -#include <linux/slab.h> #include <linux/raid/linear.h> -#define MAJOR_NR MD_MAJOR -#define MD_DRIVER -#define MD_PERSONALITY - /* * find which device holds a particular offset */ @@ -33,16 +25,15 @@ static inline dev_info_t *which_dev(mddev_t *mddev, sector_t sector) { dev_info_t *hash; linear_conf_t *conf = mddev_to_conf(mddev); - sector_t block = sector >> 1; /* * sector_div(a,b) returns the remainer and sets a to a/b */ - block >>= conf->preshift; - (void)sector_div(block, conf->hash_spacing); - hash = conf->hash_table[block]; + sector >>= conf->sector_shift; + (void)sector_div(sector, conf->spacing); + hash = conf->hash_table[sector]; - while ((sector>>1) >= (hash->size + hash->offset)) + while (sector >= hash->num_sectors + hash->start_sector) hash++; return hash; } @@ -65,7 +56,7 @@ static int linear_mergeable_bvec(struct request_queue *q, sector_t sector = bvm->bi_sector + get_start_sect(bvm->bi_bdev); dev0 = which_dev(mddev, sector); - maxsectors = (dev0->size << 1) - (sector - (dev0->offset<<1)); + maxsectors = dev0->num_sectors - (sector - dev0->start_sector); if (maxsectors < bio_sectors) maxsectors = 0; @@ -112,8 +103,8 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks) dev_info_t **table; mdk_rdev_t *rdev; int i, nb_zone, cnt; - sector_t min_spacing; - sector_t curr_offset; + sector_t min_sectors; + sector_t curr_sector; struct list_head *tmp; conf = kzalloc (sizeof (*conf) + raid_disks*sizeof(dev_info_t), @@ -145,7 +136,7 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks) mddev->queue->max_sectors > (PAGE_SIZE>>9)) blk_queue_max_sectors(mddev->queue, PAGE_SIZE>>9); - disk->size = rdev->size; + disk->num_sectors = rdev->size * 2; conf->array_sectors += rdev->size * 2; cnt++; @@ -155,34 +146,34 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks) goto out; } - min_spacing = conf->array_sectors / 2; - sector_div(min_spacing, PAGE_SIZE/sizeof(struct dev_info *)); + min_sectors = conf->array_sectors; + sector_div(min_sectors, PAGE_SIZE/sizeof(struct dev_info *)); - /* min_spacing is the minimum spacing that will fit the hash + /* min_sectors is the minimum spacing that will fit the hash * table in one PAGE. This may be much smaller than needed. * We find the smallest non-terminal set of consecutive devices - * that is larger than min_spacing as use the size of that as + * that is larger than min_sectors and use the size of that as * the actual spacing */ - conf->hash_spacing = conf->array_sectors / 2; + conf->spacing = conf->array_sectors; for (i=0; i < cnt-1 ; i++) { - sector_t sz = 0; + sector_t tmp = 0; int j; - for (j = i; j < cnt - 1 && sz < min_spacing; j++) - sz += conf->disks[j].size; - if (sz >= min_spacing && sz < conf->hash_spacing) - conf->hash_spacing = sz; + for (j = i; j < cnt - 1 && tmp < min_sectors; j++) + tmp += conf->disks[j].num_sectors; + if (tmp >= min_sectors && tmp < conf->spacing) + conf->spacing = tmp; } - /* hash_spacing may be too large for sector_div to work with, + /* spacing may be too large for sector_div to work with, * so we might need to pre-shift */ - conf->preshift = 0; + conf->sector_shift = 0; if (sizeof(sector_t) > sizeof(u32)) { - sector_t space = conf->hash_spacing; + sector_t space = conf->spacing; while (space > (sector_t)(~(u32)0)) { space >>= 1; - conf->preshift++; + conf->sector_shift++; } } /* @@ -194,9 +185,9 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks) unsigned round; unsigned long base; - sz = conf->array_sectors >> (conf->preshift + 1); + sz = conf->array_sectors >> conf->sector_shift; sz += 1; /* force round-up */ - base = conf->hash_spacing >> conf->preshift; + base = conf->spacing >> conf->sector_shift; round = sector_div(sz, base); nb_zone = sz + (round ? 1 : 0); } @@ -211,32 +202,31 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks) * Here we generate the linear hash table * First calculate the device offsets. */ - conf->disks[0].offset = 0; + conf->disks[0].start_sector = 0; for (i = 1; i < raid_disks; i++) - conf->disks[i].offset = - conf->disks[i-1].offset + - conf->disks[i-1].size; + conf->disks[i].start_sector = + conf->disks[i-1].start_sector + + conf->disks[i-1].num_sectors; table = conf->hash_table; - curr_offset = 0; i = 0; - for (curr_offset = 0; - curr_offset < conf->array_sectors / 2; - curr_offset += conf->hash_spacing) { + for (curr_sector = 0; + curr_sector < conf->array_sectors; + curr_sector += conf->spacing) { while (i < raid_disks-1 && - curr_offset >= conf->disks[i+1].offset) + curr_sector >= conf->disks[i+1].start_sector) i++; *table ++ = conf->disks + i; } - if (conf->preshift) { - conf->hash_spacing >>= conf->preshift; - /* round hash_spacing up so that when we divide by it, + if (conf->sector_shift) { + conf->spacing >>= conf->sector_shift; + /* round spacing up so that when we divide by it, * we err on the side of "too-low", which is safest. */ - conf->hash_spacing++; + conf->spacing++; } BUG_ON(table - conf->hash_table > nb_zone); @@ -317,40 +307,47 @@ static int linear_make_request (struct request_queue *q, struct bio *bio) const int rw = bio_data_dir(bio); mddev_t *mddev = q->queuedata; dev_info_t *tmp_dev; - sector_t block; + int cpu; if (unlikely(bio_barrier(bio))) { bio_endio(bio, -EOPNOTSUPP); return 0; } - disk_stat_inc(mddev->gendisk, ios[rw]); - disk_stat_add(mddev->gendisk, sectors[rw], bio_sectors(bio)); + cpu = part_stat_lock(); + part_stat_inc(cpu, &mddev->gendisk->part0, ios[rw]); + part_stat_add(cpu, &mddev->gendisk->part0, sectors[rw], + bio_sectors(bio)); + part_stat_unlock(); tmp_dev = which_dev(mddev, bio->bi_sector); - block = bio->bi_sector >> 1; - if (unlikely(block >= (tmp_dev->size + tmp_dev->offset) - || block < tmp_dev->offset)) { + if (unlikely(bio->bi_sector >= (tmp_dev->num_sectors + + tmp_dev->start_sector) + || (bio->bi_sector < + tmp_dev->start_sector))) { char b[BDEVNAME_SIZE]; - printk("linear_make_request: Block %llu out of bounds on " - "dev %s size %llu offset %llu\n", - (unsigned long long)block, + printk("linear_make_request: Sector %llu out of bounds on " + "dev %s: %llu sectors, offset %llu\n", + (unsigned long long)bio->bi_sector, bdevname(tmp_dev->rdev->bdev, b), - (unsigned long long)tmp_dev->size, - (unsigned long long)tmp_dev->offset); + (unsigned long long)tmp_dev->num_sectors, + (unsigned long long)tmp_dev->start_sector); bio_io_error(bio); return 0; } if (unlikely(bio->bi_sector + (bio->bi_size >> 9) > - (tmp_dev->offset + tmp_dev->size)<<1)) { + tmp_dev->start_sector + tmp_dev->num_sectors)) { /* This bio crosses a device boundary, so we have to * split it. */ struct bio_pair *bp; - bp = bio_split(bio, bio_split_pool, - ((tmp_dev->offset + tmp_dev->size)<<1) - bio->bi_sector); + + bp = bio_split(bio, + tmp_dev->start_sector + tmp_dev->num_sectors + - bio->bi_sector); + if (linear_make_request(q, &bp->bio1)) generic_make_request(&bp->bio1); if (linear_make_request(q, &bp->bio2)) @@ -360,7 +357,8 @@ static int linear_make_request (struct request_queue *q, struct bio *bio) } bio->bi_bdev = tmp_dev->rdev->bdev; - bio->bi_sector = bio->bi_sector - (tmp_dev->offset << 1) + tmp_dev->rdev->data_offset; + bio->bi_sector = bio->bi_sector - tmp_dev->start_sector + + tmp_dev->rdev->data_offset; return 1; } @@ -368,29 +366,6 @@ static int linear_make_request (struct request_queue *q, struct bio *bio) static void linear_status (struct seq_file *seq, mddev_t *mddev) { -#undef MD_DEBUG -#ifdef MD_DEBUG - int j; - linear_conf_t *conf = mddev_to_conf(mddev); - sector_t s = 0; - - seq_printf(seq, " "); - for (j = 0; j < mddev->raid_disks; j++) - { - char b[BDEVNAME_SIZE]; - s += conf->smallest_size; - seq_printf(seq, "[%s", - bdevname(conf->hash_table[j][0].rdev->bdev,b)); - - while (s > conf->hash_table[j][0].offset + - conf->hash_table[j][0].size) - seq_printf(seq, "/%s] ", - bdevname(conf->hash_table[j][1].rdev->bdev,b)); - else - seq_printf(seq, "] "); - } - seq_printf(seq, "\n"); -#endif seq_printf(seq, " %dk rounding", mddev->chunk_size/1024); } diff --git a/drivers/md/md.c b/drivers/md/md.c index deeac4b4417..aaa3d465de4 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -32,31 +32,21 @@ Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ -#include <linux/module.h> -#include <linux/kernel.h> #include <linux/kthread.h> -#include <linux/linkage.h> #include <linux/raid/md.h> #include <linux/raid/bitmap.h> #include <linux/sysctl.h> #include <linux/buffer_head.h> /* for invalidate_bdev */ #include <linux/poll.h> -#include <linux/mutex.h> #include <linux/ctype.h> -#include <linux/freezer.h> - -#include <linux/init.h> - +#include <linux/hdreg.h> +#include <linux/proc_fs.h> +#include <linux/random.h> +#include <linux/reboot.h> #include <linux/file.h> - -#ifdef CONFIG_KMOD -#include <linux/kmod.h> -#endif - -#include <asm/unaligned.h> +#include <linux/delay.h> #define MAJOR_NR MD_MAJOR -#define MD_DRIVER /* 63 partitions with the alternate major number (mdp) */ #define MdpMinorShift 6 @@ -66,7 +56,7 @@ #ifndef MODULE -static void autostart_arrays (int part); +static void autostart_arrays(int part); #endif static LIST_HEAD(pers_list); @@ -212,7 +202,7 @@ static DEFINE_SPINLOCK(all_mddevs_lock); ) -static int md_fail_request (struct request_queue *q, struct bio *bio) +static int md_fail_request(struct request_queue *q, struct bio *bio) { bio_io_error(bio); return 0; @@ -1464,10 +1454,7 @@ static int bind_rdev_to_array(mdk_rdev_t * rdev, mddev_t * mddev) if ((err = kobject_add(&rdev->kobj, &mddev->kobj, "dev-%s", b))) goto fail; - if (rdev->bdev->bd_part) - ko = &rdev->bdev->bd_part->dev.kobj; - else - ko = &rdev->bdev->bd_disk->dev.kobj; + ko = &part_to_dev(rdev->bdev->bd_part)->kobj; if ((err = sysfs_create_link(&rdev->kobj, ko, "block"))) { kobject_del(&rdev->kobj); goto fail; @@ -2109,8 +2096,6 @@ rdev_size_store(mdk_rdev_t *rdev, const char *buf, size_t len) if (strict_strtoull(buf, 10, &size) < 0) return -EINVAL; - if (size < my_mddev->size) - return -EINVAL; if (my_mddev->pers && rdev->raid_disk >= 0) { if (my_mddev->persistent) { size = super_types[my_mddev->major_version]. @@ -2121,9 +2106,9 @@ rdev_size_store(mdk_rdev_t *rdev, const char *buf, size_t len) size = (rdev->bdev->bd_inode->i_size >> 10); size -= rdev->data_offset/2; } - if (size < my_mddev->size) - return -EINVAL; /* component must fit device */ } + if (size < my_mddev->size) + return -EINVAL; /* component must fit device */ rdev->size = size; if (size > oldsize && my_mddev->external) { @@ -2409,12 +2394,11 @@ safe_delay_store(mddev_t *mddev, const char *cbuf, size_t len) int i; unsigned long msec; char buf[30]; - char *e; + /* remove a period, and count digits after it */ if (len >= sizeof(buf)) return -EINVAL; - strlcpy(buf, cbuf, len); - buf[len] = 0; + strlcpy(buf, cbuf, sizeof(buf)); for (i=0; i<len; i++) { if (dot) { if (isdigit(buf[i])) { @@ -2427,8 +2411,7 @@ safe_delay_store(mddev_t *mddev, const char *cbuf, size_t len) buf[i] = 0; } } - msec = simple_strtoul(buf, &e, 10); - if (e == buf || (*e && *e != '\n')) + if (strict_strtoul(buf, 10, &msec) < 0) return -EINVAL; msec = (msec * 1000) / scale; if (msec == 0) @@ -2730,9 +2713,9 @@ array_state_store(mddev_t *mddev, const char *buf, size_t len) break; case read_auto: if (mddev->pers) { - if (mddev->ro != 1) + if (mddev->ro == 0) err = do_md_stop(mddev, 1, 0); - else + else if (mddev->ro == 1) err = restart_array(mddev); if (err == 0) { mddev->ro = 2; @@ -2948,7 +2931,13 @@ metadata_store(mddev_t *mddev, const char *buf, size_t len) { int major, minor; char *e; - if (!list_empty(&mddev->disks)) + /* Changing the details of 'external' metadata is + * always permitted. Otherwise there must be + * no devices attached to the array. + */ + if (mddev->external && strncmp(buf, "external:", 9) == 0) + ; + else if (!list_empty(&mddev->disks)) return -EBUSY; if (cmd_match(buf, "none")) { @@ -3470,8 +3459,8 @@ static struct kobject *md_probe(dev_t dev, int *part, void *data) disk->queue = mddev->queue; add_disk(disk); mddev->gendisk = disk; - error = kobject_init_and_add(&mddev->kobj, &md_ktype, &disk->dev.kobj, - "%s", "md"); + error = kobject_init_and_add(&mddev->kobj, &md_ktype, + &disk_to_dev(disk)->kobj, "%s", "md"); mutex_unlock(&disks_mutex); if (error) printk(KERN_WARNING "md: cannot register %s/md - name in use\n", @@ -3530,17 +3519,12 @@ static int do_md_run(mddev_t * mddev) return -EINVAL; } /* - * chunk-size has to be a power of 2 and multiples of PAGE_SIZE + * chunk-size has to be a power of 2 */ if ( (1 << ffz(~chunk_size)) != chunk_size) { printk(KERN_ERR "chunk_size of %d not valid\n", chunk_size); return -EINVAL; } - if (chunk_size < PAGE_SIZE) { - printk(KERN_ERR "too small chunk_size: %d < %ld\n", - chunk_size, PAGE_SIZE); - return -EINVAL; - } /* devices must have minimum size of one chunk */ rdev_for_each(rdev, tmp, mddev) { @@ -3558,12 +3542,10 @@ static int do_md_run(mddev_t * mddev) } } -#ifdef CONFIG_KMOD if (mddev->level != LEVEL_NONE) request_module("md-level-%d", mddev->level); else if (mddev->clevel[0]) request_module("md-%s", mddev->clevel); -#endif /* * Drop all container device buffers, from now on @@ -3761,7 +3743,7 @@ static int do_md_run(mddev_t * mddev) sysfs_notify(&mddev->kobj, NULL, "array_state"); sysfs_notify(&mddev->kobj, NULL, "sync_action"); sysfs_notify(&mddev->kobj, NULL, "degraded"); - kobject_uevent(&mddev->gendisk->dev.kobj, KOBJ_CHANGE); + kobject_uevent(&disk_to_dev(mddev->gendisk)->kobj, KOBJ_CHANGE); return 0; } @@ -3974,10 +3956,10 @@ static void autorun_array(mddev_t *mddev) } printk("\n"); - err = do_md_run (mddev); + err = do_md_run(mddev); if (err) { printk(KERN_WARNING "md: do_md_run() returned %d\n", err); - do_md_stop (mddev, 0, 0); + do_md_stop(mddev, 0, 0); } } @@ -4336,7 +4318,7 @@ static int add_new_disk(mddev_t * mddev, mdu_disk_info_t *info) if (!(info->state & (1<<MD_DISK_FAULTY))) { int err; - rdev = md_import_device (dev, -1, 0); + rdev = md_import_device(dev, -1, 0); if (IS_ERR(rdev)) { printk(KERN_WARNING "md: error, md_import_device() returned %ld\n", @@ -4418,7 +4400,7 @@ static int hot_add_disk(mddev_t * mddev, dev_t dev) return -EINVAL; } - rdev = md_import_device (dev, -1, 0); + rdev = md_import_device(dev, -1, 0); if (IS_ERR(rdev)) { printk(KERN_WARNING "md: error, md_import_device() returned %ld\n", @@ -4937,11 +4919,11 @@ static int md_ioctl(struct inode *inode, struct file *file, goto done_unlock; case STOP_ARRAY: - err = do_md_stop (mddev, 0, 1); + err = do_md_stop(mddev, 0, 1); goto done_unlock; case STOP_ARRAY_RO: - err = do_md_stop (mddev, 1, 1); + err = do_md_stop(mddev, 1, 1); goto done_unlock; } @@ -4990,7 +4972,7 @@ static int md_ioctl(struct inode *inode, struct file *file, goto done_unlock; case RUN_ARRAY: - err = do_md_run (mddev); + err = do_md_run(mddev); goto done_unlock; case SET_BITMAP_FILE: @@ -5428,11 +5410,11 @@ static int md_seq_show(struct seq_file *seq, void *v) seq_printf(seq, " super non-persistent"); if (mddev->pers) { - mddev->pers->status (seq, mddev); + mddev->pers->status(seq, mddev); seq_printf(seq, "\n "); if (mddev->pers->sync_request) { if (mddev->curr_resync > 2) { - status_resync (seq, mddev); + status_resync(seq, mddev); seq_printf(seq, "\n "); } else if (mddev->curr_resync == 1 || mddev->curr_resync == 2) seq_printf(seq, "\tresync=DELAYED\n "); @@ -5549,8 +5531,8 @@ static int is_mddev_idle(mddev_t *mddev) rcu_read_lock(); rdev_for_each_rcu(rdev, mddev) { struct gendisk *disk = rdev->bdev->bd_contains->bd_disk; - curr_events = disk_stat_read(disk, sectors[0]) + - disk_stat_read(disk, sectors[1]) - + curr_events = part_stat_read(&disk->part0, sectors[0]) + + part_stat_read(&disk->part0, sectors[1]) - atomic_read(&disk->sync_io); /* sync IO will cause sync_io to increase before the disk_stats * as sync_io is counted when a request starts, and @@ -6263,7 +6245,7 @@ static int md_notify_reboot(struct notifier_block *this, * appears to still be in use. Hence * the '100'. */ - do_md_stop (mddev, 1, 100); + do_md_stop(mddev, 1, 100); mddev_unlock(mddev); } /* @@ -6307,7 +6289,7 @@ static int __init md_init(void) raid_table_header = register_sysctl_table(raid_root_table); md_geninit(); - return (0); + return 0; } diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c index c4779ccba1c..d4ac47d1127 100644 --- a/drivers/md/multipath.c +++ b/drivers/md/multipath.c @@ -19,16 +19,7 @@ * Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ -#include <linux/module.h> -#include <linux/slab.h> -#include <linux/spinlock.h> #include <linux/raid/multipath.h> -#include <linux/buffer_head.h> -#include <asm/atomic.h> - -#define MAJOR_NR MD_MAJOR -#define MD_DRIVER -#define MD_PERSONALITY #define MAX_WORK_PER_DISK 128 @@ -147,6 +138,7 @@ static int multipath_make_request (struct request_queue *q, struct bio * bio) struct multipath_bh * mp_bh; struct multipath_info *multipath; const int rw = bio_data_dir(bio); + int cpu; if (unlikely(bio_barrier(bio))) { bio_endio(bio, -EOPNOTSUPP); @@ -158,8 +150,11 @@ static int multipath_make_request (struct request_queue *q, struct bio * bio) mp_bh->master_bio = bio; mp_bh->mddev = mddev; - disk_stat_inc(mddev->gendisk, ios[rw]); - disk_stat_add(mddev->gendisk, sectors[rw], bio_sectors(bio)); + cpu = part_stat_lock(); + part_stat_inc(cpu, &mddev->gendisk->part0, ios[rw]); + part_stat_add(cpu, &mddev->gendisk->part0, sectors[rw], + bio_sectors(bio)); + part_stat_unlock(); mp_bh->path = multipath_map(conf); if (mp_bh->path < 0) { @@ -172,7 +167,7 @@ static int multipath_make_request (struct request_queue *q, struct bio * bio) mp_bh->bio = *bio; mp_bh->bio.bi_sector += multipath->rdev->data_offset; mp_bh->bio.bi_bdev = multipath->rdev->bdev; - mp_bh->bio.bi_rw |= (1 << BIO_RW_FAILFAST); + mp_bh->bio.bi_rw |= (1 << BIO_RW_FAILFAST_TRANSPORT); mp_bh->bio.bi_end_io = multipath_end_request; mp_bh->bio.bi_private = mp_bh; generic_make_request(&mp_bh->bio); @@ -398,7 +393,7 @@ static void multipathd (mddev_t *mddev) *bio = *(mp_bh->master_bio); bio->bi_sector += conf->multipaths[mp_bh->path].rdev->data_offset; bio->bi_bdev = conf->multipaths[mp_bh->path].rdev->bdev; - bio->bi_rw |= (1 << BIO_RW_FAILFAST); + bio->bi_rw |= (1 << BIO_RW_FAILFAST_TRANSPORT); bio->bi_end_io = multipath_end_request; bio->bi_private = mp_bh; generic_make_request(bio); diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c index 18361063566..8ac6488ad0d 100644 --- a/drivers/md/raid0.c +++ b/drivers/md/raid0.c @@ -18,13 +18,8 @@ Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ -#include <linux/module.h> #include <linux/raid/raid0.h> -#define MAJOR_NR MD_MAJOR -#define MD_DRIVER -#define MD_PERSONALITY - static void raid0_unplug(struct request_queue *q) { mddev_t *mddev = q->queuedata; @@ -399,14 +394,18 @@ static int raid0_make_request (struct request_queue *q, struct bio *bio) sector_t chunk; sector_t block, rsect; const int rw = bio_data_dir(bio); + int cpu; if (unlikely(bio_barrier(bio))) { bio_endio(bio, -EOPNOTSUPP); return 0; } - disk_stat_inc(mddev->gendisk, ios[rw]); - disk_stat_add(mddev->gendisk, sectors[rw], bio_sectors(bio)); + cpu = part_stat_lock(); + part_stat_inc(cpu, &mddev->gendisk->part0, ios[rw]); + part_stat_add(cpu, &mddev->gendisk->part0, sectors[rw], + bio_sectors(bio)); + part_stat_unlock(); chunk_size = mddev->chunk_size >> 10; chunk_sects = mddev->chunk_size >> 9; @@ -423,7 +422,7 @@ static int raid0_make_request (struct request_queue *q, struct bio *bio) /* This is a one page bio that upper layers * refuse to split for us, so we need to split it. */ - bp = bio_split(bio, bio_split_pool, chunk_sects - (bio->bi_sector & (chunk_sects - 1)) ); + bp = bio_split(bio, chunk_sects - (bio->bi_sector & (chunk_sects - 1))); if (raid0_make_request(q, &bp->bio1)) generic_make_request(&bp->bio1); if (raid0_make_request(q, &bp->bio2)) diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index 03a5ab705c2..9c788e2489b 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -32,6 +32,7 @@ */ #include "dm-bio-list.h" +#include <linux/delay.h> #include <linux/raid/raid1.h> #include <linux/raid/bitmap.h> @@ -779,7 +780,7 @@ static int make_request(struct request_queue *q, struct bio * bio) struct page **behind_pages = NULL; const int rw = bio_data_dir(bio); const int do_sync = bio_sync(bio); - int do_barriers; + int cpu, do_barriers; mdk_rdev_t *blocked_rdev; /* @@ -804,8 +805,11 @@ static int make_request(struct request_queue *q, struct bio * bio) bitmap = mddev->bitmap; - disk_stat_inc(mddev->gendisk, ios[rw]); - disk_stat_add(mddev->gendisk, sectors[rw], bio_sectors(bio)); + cpu = part_stat_lock(); + part_stat_inc(cpu, &mddev->gendisk->part0, ios[rw]); + part_stat_add(cpu, &mddev->gendisk->part0, sectors[rw], + bio_sectors(bio)); + part_stat_unlock(); /* * make_request() can abort the operation when READA is being @@ -1302,9 +1306,6 @@ static void sync_request_write(mddev_t *mddev, r1bio_t *r1_bio) sbio->bi_size = r1_bio->sectors << 9; sbio->bi_idx = 0; sbio->bi_phys_segments = 0; - sbio->bi_hw_segments = 0; - sbio->bi_hw_front_size = 0; - sbio->bi_hw_back_size = 0; sbio->bi_flags &= ~(BIO_POOL_MASK - 1); sbio->bi_flags |= 1 << BIO_UPTODATE; sbio->bi_next = NULL; @@ -1790,7 +1791,6 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i bio->bi_vcnt = 0; bio->bi_idx = 0; bio->bi_phys_segments = 0; - bio->bi_hw_segments = 0; bio->bi_size = 0; bio->bi_end_io = NULL; bio->bi_private = NULL; diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index e34cd0e6247..da5129a24b1 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -19,6 +19,7 @@ */ #include "dm-bio-list.h" +#include <linux/delay.h> #include <linux/raid/raid10.h> #include <linux/raid/bitmap.h> @@ -789,6 +790,7 @@ static int make_request(struct request_queue *q, struct bio * bio) mirror_info_t *mirror; r10bio_t *r10_bio; struct bio *read_bio; + int cpu; int i; int chunk_sects = conf->chunk_mask + 1; const int rw = bio_data_dir(bio); @@ -816,7 +818,7 @@ static int make_request(struct request_queue *q, struct bio * bio) /* This is a one page bio that upper layers * refuse to split for us, so we need to split it. */ - bp = bio_split(bio, bio_split_pool, + bp = bio_split(bio, chunk_sects - (bio->bi_sector & (chunk_sects - 1)) ); if (make_request(q, &bp->bio1)) generic_make_request(&bp->bio1); @@ -843,8 +845,11 @@ static int make_request(struct request_queue *q, struct bio * bio) */ wait_barrier(conf); - disk_stat_inc(mddev->gendisk, ios[rw]); - disk_stat_add(mddev->gendisk, sectors[rw], bio_sectors(bio)); + cpu = part_stat_lock(); + part_stat_inc(cpu, &mddev->gendisk->part0, ios[rw]); + part_stat_add(cpu, &mddev->gendisk->part0, sectors[rw], + bio_sectors(bio)); + part_stat_unlock(); r10_bio = mempool_alloc(conf->r10bio_pool, GFP_NOIO); @@ -1345,9 +1350,6 @@ static void sync_request_write(mddev_t *mddev, r10bio_t *r10_bio) tbio->bi_size = r10_bio->sectors << 9; tbio->bi_idx = 0; tbio->bi_phys_segments = 0; - tbio->bi_hw_segments = 0; - tbio->bi_hw_front_size = 0; - tbio->bi_hw_back_size = 0; tbio->bi_flags &= ~(BIO_POOL_MASK - 1); tbio->bi_flags |= 1 << BIO_UPTODATE; tbio->bi_next = NULL; @@ -1947,7 +1949,6 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i bio->bi_vcnt = 0; bio->bi_idx = 0; bio->bi_phys_segments = 0; - bio->bi_hw_segments = 0; bio->bi_size = 0; } @@ -2028,8 +2029,9 @@ static int run(mddev_t *mddev) int nc, fc, fo; sector_t stride, size; - if (mddev->chunk_size == 0) { - printk(KERN_ERR "md/raid10: non-zero chunk size required.\n"); + if (mddev->chunk_size < PAGE_SIZE) { + printk(KERN_ERR "md/raid10: chunk size must be " + "at least PAGE_SIZE(%ld).\n", PAGE_SIZE); return -EINVAL; } diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 224de022e7c..a36a7435edf 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -43,12 +43,7 @@ * miss any bits. */ -#include <linux/module.h> -#include <linux/slab.h> -#include <linux/highmem.h> -#include <linux/bitops.h> #include <linux/kthread.h> -#include <asm/atomic.h> #include "raid6.h" #include <linux/raid/bitmap.h> @@ -101,6 +96,40 @@ const char raid6_empty_zero_page[PAGE_SIZE] __attribute__((aligned(256))); #endif +/* + * We maintain a biased count of active stripes in the bottom 16 bits of + * bi_phys_segments, and a count of processed stripes in the upper 16 bits + */ +static inline int raid5_bi_phys_segments(struct bio *bio) +{ + return bio->bi_phys_segments & 0xffff; +} + +static inline int raid5_bi_hw_segments(struct bio *bio) +{ + return (bio->bi_phys_segments >> 16) & 0xffff; +} + +static inline int raid5_dec_bi_phys_segments(struct bio *bio) +{ + --bio->bi_phys_segments; + return raid5_bi_phys_segments(bio); +} + +static inline int raid5_dec_bi_hw_segments(struct bio *bio) +{ + unsigned short val = raid5_bi_hw_segments(bio); + + --val; + bio->bi_phys_segments = (val << 16) | raid5_bi_phys_segments(bio); + return val; +} + +static inline void raid5_set_bi_hw_segments(struct bio *bio, unsigned int cnt) +{ + bio->bi_phys_segments = raid5_bi_phys_segments(bio) || (cnt << 16); +} + static inline int raid6_next_disk(int disk, int raid_disks) { disk++; @@ -241,7 +270,7 @@ static int grow_buffers(struct stripe_head *sh, int num) return 0; } -static void raid5_build_block (struct stripe_head *sh, int i); +static void raid5_build_block(struct stripe_head *sh, int i); static void init_stripe(struct stripe_head *sh, sector_t sector, int pd_idx, int disks) { @@ -507,7 +536,7 @@ static void ops_complete_biofill(void *stripe_head_ref) while (rbi && rbi->bi_sector < dev->sector + STRIPE_SECTORS) { rbi2 = r5_next_bio(rbi, dev->sector); - if (--rbi->bi_phys_segments == 0) { + if (!raid5_dec_bi_phys_segments(rbi)) { rbi->bi_next = return_bi; return_bi = rbi; } @@ -1117,7 +1146,7 @@ static void raid5_end_read_request(struct bio * bi, int error) release_stripe(sh); } -static void raid5_end_write_request (struct bio *bi, int error) +static void raid5_end_write_request(struct bio *bi, int error) { struct stripe_head *sh = bi->bi_private; raid5_conf_t *conf = sh->raid_conf; @@ -1149,7 +1178,7 @@ static void raid5_end_write_request (struct bio *bi, int error) static sector_t compute_blocknr(struct stripe_head *sh, int i); -static void raid5_build_block (struct stripe_head *sh, int i) +static void raid5_build_block(struct stripe_head *sh, int i) { struct r5dev *dev = &sh->dev[i]; @@ -1187,10 +1216,10 @@ static void error(mddev_t *mddev, mdk_rdev_t *rdev) set_bit(MD_RECOVERY_INTR, &mddev->recovery); } set_bit(Faulty, &rdev->flags); - printk (KERN_ALERT - "raid5: Disk failure on %s, disabling device.\n" - "raid5: Operation continuing on %d devices.\n", - bdevname(rdev->bdev,b), conf->raid_disks - mddev->degraded); + printk(KERN_ALERT + "raid5: Disk failure on %s, disabling device.\n" + "raid5: Operation continuing on %d devices.\n", + bdevname(rdev->bdev,b), conf->raid_disks - mddev->degraded); } } @@ -1286,8 +1315,8 @@ static sector_t raid5_compute_sector(sector_t r_sector, unsigned int raid_disks, *dd_idx = (*pd_idx + 2 + *dd_idx) % raid_disks; break; default: - printk (KERN_CRIT "raid6: unsupported algorithm %d\n", - conf->algorithm); + printk(KERN_CRIT "raid6: unsupported algorithm %d\n", + conf->algorithm); } break; } @@ -1362,8 +1391,8 @@ static sector_t compute_blocknr(struct stripe_head *sh, int i) } break; default: - printk (KERN_CRIT "raid6: unsupported algorithm %d\n", - conf->algorithm); + printk(KERN_CRIT "raid6: unsupported algorithm %d\n", + conf->algorithm); } break; } @@ -1371,7 +1400,7 @@ static sector_t compute_blocknr(struct stripe_head *sh, int i) chunk_number = stripe * data_disks + i; r_sector = (sector_t)chunk_number * sectors_per_chunk + chunk_offset; - check = raid5_compute_sector (r_sector, raid_disks, data_disks, &dummy1, &dummy2, conf); + check = raid5_compute_sector(r_sector, raid_disks, data_disks, &dummy1, &dummy2, conf); if (check != sh->sector || dummy1 != dd_idx || dummy2 != sh->pd_idx) { printk(KERN_ERR "compute_blocknr: map not correct\n"); return 0; @@ -1725,7 +1754,7 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in if (*bip) bi->bi_next = *bip; *bip = bi; - bi->bi_phys_segments ++; + bi->bi_phys_segments++; spin_unlock_irq(&conf->device_lock); spin_unlock(&sh->lock); @@ -1819,7 +1848,7 @@ handle_failed_stripe(raid5_conf_t *conf, struct stripe_head *sh, sh->dev[i].sector + STRIPE_SECTORS) { struct bio *nextbi = r5_next_bio(bi, sh->dev[i].sector); clear_bit(BIO_UPTODATE, &bi->bi_flags); - if (--bi->bi_phys_segments == 0) { + if (!raid5_dec_bi_phys_segments(bi)) { md_write_end(conf->mddev); bi->bi_next = *return_bi; *return_bi = bi; @@ -1834,7 +1863,7 @@ handle_failed_stripe(raid5_conf_t *conf, struct stripe_head *sh, sh->dev[i].sector + STRIPE_SECTORS) { struct bio *bi2 = r5_next_bio(bi, sh->dev[i].sector); clear_bit(BIO_UPTODATE, &bi->bi_flags); - if (--bi->bi_phys_segments == 0) { + if (!raid5_dec_bi_phys_segments(bi)) { md_write_end(conf->mddev); bi->bi_next = *return_bi; *return_bi = bi; @@ -1858,7 +1887,7 @@ handle_failed_stripe(raid5_conf_t *conf, struct stripe_head *sh, struct bio *nextbi = r5_next_bio(bi, sh->dev[i].sector); clear_bit(BIO_UPTODATE, &bi->bi_flags); - if (--bi->bi_phys_segments == 0) { + if (!raid5_dec_bi_phys_segments(bi)) { bi->bi_next = *return_bi; *return_bi = bi; } @@ -2033,7 +2062,7 @@ static void handle_stripe_clean_event(raid5_conf_t *conf, while (wbi && wbi->bi_sector < dev->sector + STRIPE_SECTORS) { wbi2 = r5_next_bio(wbi, dev->sector); - if (--wbi->bi_phys_segments == 0) { + if (!raid5_dec_bi_phys_segments(wbi)) { md_write_end(conf->mddev); wbi->bi_next = *return_bi; *return_bi = wbi; @@ -2814,7 +2843,7 @@ static bool handle_stripe6(struct stripe_head *sh, struct page *tmp_page) copy_data(0, rbi, dev->page, dev->sector); rbi2 = r5_next_bio(rbi, dev->sector); spin_lock_irq(&conf->device_lock); - if (--rbi->bi_phys_segments == 0) { + if (!raid5_dec_bi_phys_segments(rbi)) { rbi->bi_next = return_bi; return_bi = rbi; } @@ -3155,8 +3184,11 @@ static struct bio *remove_bio_from_retry(raid5_conf_t *conf) if(bi) { conf->retry_read_aligned_list = bi->bi_next; bi->bi_next = NULL; + /* + * this sets the active strip count to 1 and the processed + * strip count to zero (upper 8 bits) + */ bi->bi_phys_segments = 1; /* biased count of active stripes */ - bi->bi_hw_segments = 0; /* count of processed stripes */ } return bi; @@ -3206,8 +3238,7 @@ static int bio_fits_rdev(struct bio *bi) if ((bi->bi_size>>9) > q->max_sectors) return 0; blk_recount_segments(q, bi); - if (bi->bi_phys_segments > q->max_phys_segments || - bi->bi_hw_segments > q->max_hw_segments) + if (bi->bi_phys_segments > q->max_phys_segments) return 0; if (q->merge_bvec_fn) @@ -3351,7 +3382,7 @@ static int make_request(struct request_queue *q, struct bio * bi) sector_t logical_sector, last_sector; struct stripe_head *sh; const int rw = bio_data_dir(bi); - int remaining; + int cpu, remaining; if (unlikely(bio_barrier(bi))) { bio_endio(bi, -EOPNOTSUPP); @@ -3360,8 +3391,11 @@ static int make_request(struct request_queue *q, struct bio * bi) md_write_start(mddev, bi); - disk_stat_inc(mddev->gendisk, ios[rw]); - disk_stat_add(mddev->gendisk, sectors[rw], bio_sectors(bi)); + cpu = part_stat_lock(); + part_stat_inc(cpu, &mddev->gendisk->part0, ios[rw]); + part_stat_add(cpu, &mddev->gendisk->part0, sectors[rw], + bio_sectors(bi)); + part_stat_unlock(); if (rw == READ && mddev->reshape_position == MaxSector && @@ -3468,7 +3502,7 @@ static int make_request(struct request_queue *q, struct bio * bi) } spin_lock_irq(&conf->device_lock); - remaining = --bi->bi_phys_segments; + remaining = raid5_dec_bi_phys_segments(bi); spin_unlock_irq(&conf->device_lock); if (remaining == 0) { @@ -3752,7 +3786,7 @@ static int retry_aligned_read(raid5_conf_t *conf, struct bio *raid_bio) sector += STRIPE_SECTORS, scnt++) { - if (scnt < raid_bio->bi_hw_segments) + if (scnt < raid5_bi_hw_segments(raid_bio)) /* already done this stripe */ continue; @@ -3760,7 +3794,7 @@ static int retry_aligned_read(raid5_conf_t *conf, struct bio *raid_bio) if (!sh) { /* failed to get a stripe - must wait */ - raid_bio->bi_hw_segments = scnt; + raid5_set_bi_hw_segments(raid_bio, scnt); conf->retry_read_aligned = raid_bio; return handled; } @@ -3768,7 +3802,7 @@ static int retry_aligned_read(raid5_conf_t *conf, struct bio *raid_bio) set_bit(R5_ReadError, &sh->dev[dd_idx].flags); if (!add_stripe_bio(sh, raid_bio, dd_idx, 0)) { release_stripe(sh); - raid_bio->bi_hw_segments = scnt; + raid5_set_bi_hw_segments(raid_bio, scnt); conf->retry_read_aligned = raid_bio; return handled; } @@ -3778,7 +3812,7 @@ static int retry_aligned_read(raid5_conf_t *conf, struct bio *raid_bio) handled++; } spin_lock_irq(&conf->device_lock); - remaining = --raid_bio->bi_phys_segments; + remaining = raid5_dec_bi_phys_segments(raid_bio); spin_unlock_irq(&conf->device_lock); if (remaining == 0) bio_endio(raid_bio, 0); @@ -3973,6 +4007,13 @@ static int run(mddev_t *mddev) return -EIO; } + if (mddev->chunk_size < PAGE_SIZE) { + printk(KERN_ERR "md/raid5: chunk_size must be at least " + "PAGE_SIZE but %d < %ld\n", + mddev->chunk_size, PAGE_SIZE); + return -EINVAL; + } + if (mddev->reshape_position != MaxSector) { /* Check that we can continue the reshape. * Currently only disks can change, it must @@ -4250,7 +4291,7 @@ static int stop(mddev_t *mddev) } #ifdef DEBUG -static void print_sh (struct seq_file *seq, struct stripe_head *sh) +static void print_sh(struct seq_file *seq, struct stripe_head *sh) { int i; @@ -4266,7 +4307,7 @@ static void print_sh (struct seq_file *seq, struct stripe_head *sh) seq_printf(seq, "\n"); } -static void printall (struct seq_file *seq, raid5_conf_t *conf) +static void printall(struct seq_file *seq, raid5_conf_t *conf) { struct stripe_head *sh; struct hlist_node *hn; @@ -4284,7 +4325,7 @@ static void printall (struct seq_file *seq, raid5_conf_t *conf) } #endif -static void status (struct seq_file *seq, mddev_t *mddev) +static void status(struct seq_file *seq, mddev_t *mddev) { raid5_conf_t *conf = (raid5_conf_t *) mddev->private; int i; diff --git a/drivers/md/raid6.h b/drivers/md/raid6.h index 31cbee71365..98dcde88470 100644 --- a/drivers/md/raid6.h +++ b/drivers/md/raid6.h @@ -18,15 +18,6 @@ /* Set to 1 to use kernel-wide empty_zero_page */ #define RAID6_USE_EMPTY_ZERO_PAGE 0 -#include <linux/module.h> -#include <linux/stddef.h> -#include <linux/compiler.h> -#include <linux/types.h> -#include <linux/kernel.h> -#include <linux/errno.h> -#include <linux/mempool.h> -#include <linux/list.h> -#include <linux/vmalloc.h> #include <linux/raid/md.h> #include <linux/raid/raid5.h> |