diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2014-10-18 11:39:52 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2014-10-18 11:39:52 -0700 |
commit | 88ed806abb981cc8ec61ee7fab93ecfe63521ebf (patch) | |
tree | f3639f4a6862598fcdbb444ef4b5bbceb63d5c46 | |
parent | e56d9fccb1c78283d7c163bb354be7cbbc93d9fb (diff) | |
parent | b395f75eabb3844c99244928293796ff42feaa3d (diff) | |
download | lwn-88ed806abb981cc8ec61ee7fab93ecfe63521ebf.tar.gz lwn-88ed806abb981cc8ec61ee7fab93ecfe63521ebf.zip |
Merge tag 'md/3.18' of git://neil.brown.name/md
Pull md updates from Neil Brown:
- a few minor bug fixes
- quite a lot of code tidy-up and simplification
- remove PRINT_RAID_DEBUG ioctl. I'm fairly sure it is unused, and it
isn't particularly useful.
* tag 'md/3.18' of git://neil.brown.name/md: (21 commits)
lib/raid6: Add log level to printks
md: move EXPORT_SYMBOL to after function in md.c
md: discard PRINT_RAID_DEBUG ioctl
md: remove MD_BUG()
md: clean up 'exit' labels in md_ioctl().
md: remove unnecessary test for MD_MAJOR in md_ioctl()
md: don't allow "-sync" to be set for device in an active array.
md: remove unwanted white space from md.c
md: don't start resync thread directly from md thread.
md: Just use RCU when checking for overlap between arrays.
md: avoid potential long delay under pers_lock
md: simplify export_array()
md: discard find_rdev_nr in favour of find_rdev_nr_rcu
md: use wait_event() to simplify md_super_wait()
md: be more relaxed about stopping an array which isn't started.
md/raid1: process_checks doesn't use its return value.
md/raid5: fix init_stripe() inconsistencies
md/raid10: another memory leak due to reshape.
md: use set_bit/clear_bit instead of shift/mask for bi_flags changes.
md/raid1: minor typos and reformatting.
...
-rw-r--r-- | drivers/md/bitmap.c | 16 | ||||
-rw-r--r-- | drivers/md/linear.c | 8 | ||||
-rw-r--r-- | drivers/md/md.c | 601 | ||||
-rw-r--r-- | drivers/md/md.h | 34 | ||||
-rw-r--r-- | drivers/md/multipath.c | 28 | ||||
-rw-r--r-- | drivers/md/raid0.c | 9 | ||||
-rw-r--r-- | drivers/md/raid1.c | 37 | ||||
-rw-r--r-- | drivers/md/raid1.h | 2 | ||||
-rw-r--r-- | drivers/md/raid10.c | 18 | ||||
-rw-r--r-- | drivers/md/raid5.c | 29 | ||||
-rw-r--r-- | drivers/md/raid5.h | 4 | ||||
-rw-r--r-- | include/uapi/linux/raid/md_u.h | 1 | ||||
-rw-r--r-- | lib/raid6/algos.c | 12 |
13 files changed, 266 insertions, 533 deletions
diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index 67f8b31e2054..da3604e73e8a 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -879,7 +879,6 @@ void bitmap_unplug(struct bitmap *bitmap) { unsigned long i; int dirty, need_write; - int wait = 0; if (!bitmap || !bitmap->storage.filemap || test_bit(BITMAP_STALE, &bitmap->flags)) @@ -897,16 +896,13 @@ void bitmap_unplug(struct bitmap *bitmap) clear_page_attr(bitmap, i, BITMAP_PAGE_PENDING); write_page(bitmap, bitmap->storage.filemap[i], 0); } - if (dirty) - wait = 1; - } - if (wait) { /* if any writes were performed, we need to wait on them */ - if (bitmap->storage.file) - wait_event(bitmap->write_wait, - atomic_read(&bitmap->pending_writes)==0); - else - md_super_wait(bitmap->mddev); } + if (bitmap->storage.file) + wait_event(bitmap->write_wait, + atomic_read(&bitmap->pending_writes)==0); + else + md_super_wait(bitmap->mddev); + if (test_bit(BITMAP_WRITE_ERROR, &bitmap->flags)) bitmap_file_kick(bitmap); } diff --git a/drivers/md/linear.c b/drivers/md/linear.c index 56f534b4a2d2..64713b77df1c 100644 --- a/drivers/md/linear.c +++ b/drivers/md/linear.c @@ -10,10 +10,10 @@ it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2, or (at your option) any later version. - + You should have received a copy of the GNU General Public License (for example /usr/src/linux/COPYING); if not, write to the Free - Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ #include <linux/blkdev.h> @@ -25,7 +25,7 @@ #include "linear.h" /* - * find which device holds a particular offset + * find which device holds a particular offset */ static inline struct dev_info *which_dev(struct mddev *mddev, sector_t sector) { @@ -355,7 +355,6 @@ static void linear_status (struct seq_file *seq, struct mddev *mddev) seq_printf(seq, " %dk rounding", mddev->chunk_sectors / 2); } - static struct md_personality linear_personality = { .name = "linear", @@ -379,7 +378,6 @@ static void linear_exit (void) unregister_md_personality (&linear_personality); } - module_init(linear_init); module_exit(linear_exit); MODULE_LICENSE("GPL"); diff --git a/drivers/md/md.c b/drivers/md/md.c index 1294238610df..4dfa15da9cb8 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -1,6 +1,6 @@ /* md.c : Multiple Devices driver for Linux - Copyright (C) 1998, 1999, 2000 Ingo Molnar + Copyright (C) 1998, 1999, 2000 Ingo Molnar completely rewritten, based on the MD driver code from Marc Zyngier @@ -66,8 +66,6 @@ static void autostart_arrays(int part); static LIST_HEAD(pers_list); static DEFINE_SPINLOCK(pers_lock); -static void md_print_devices(void); - static DECLARE_WAIT_QUEUE_HEAD(resync_wait); static struct workqueue_struct *md_wq; static struct workqueue_struct *md_misc_wq; @@ -75,8 +73,6 @@ static struct workqueue_struct *md_misc_wq; static int remove_and_add_spares(struct mddev *mddev, struct md_rdev *this); -#define MD_BUG(x...) { printk("md: bug in file %s, line %d\n", __FILE__, __LINE__); md_print_devices(); } - /* * Default number of read corrections we'll attempt on an rdev * before ejecting it from the array. We divide the read error @@ -218,7 +214,6 @@ static void md_new_event_inintr(struct mddev *mddev) static LIST_HEAD(all_mddevs); static DEFINE_SPINLOCK(all_mddevs_lock); - /* * iterates through all used mddevs in the system. * We take care to grab the all_mddevs_lock whenever navigating @@ -228,7 +223,7 @@ static DEFINE_SPINLOCK(all_mddevs_lock); */ #define for_each_mddev(_mddev,_tmp) \ \ - for (({ spin_lock(&all_mddevs_lock); \ + for (({ spin_lock(&all_mddevs_lock); \ _tmp = all_mddevs.next; \ _mddev = NULL;}); \ ({ if (_tmp != &all_mddevs) \ @@ -241,7 +236,6 @@ static DEFINE_SPINLOCK(all_mddevs_lock); _tmp = _tmp->next;}) \ ) - /* Rather than calling directly into the personality make_request function, * IO requests come here first so that we can check if the device is * being suspended pending a reconfiguration. @@ -488,7 +482,7 @@ void mddev_init(struct mddev *mddev) } EXPORT_SYMBOL_GPL(mddev_init); -static struct mddev * mddev_find(dev_t unit) +static struct mddev *mddev_find(dev_t unit) { struct mddev *mddev, *new = NULL; @@ -530,7 +524,7 @@ static struct mddev * mddev_find(dev_t unit) kfree(new); return NULL; } - + is_free = 1; list_for_each_entry(mddev, &all_mddevs, all_mddevs) if (mddev->unit == dev) { @@ -562,7 +556,7 @@ static struct mddev * mddev_find(dev_t unit) goto retry; } -static inline int __must_check mddev_lock(struct mddev * mddev) +static inline int __must_check mddev_lock(struct mddev *mddev) { return mutex_lock_interruptible(&mddev->reconfig_mutex); } @@ -570,7 +564,7 @@ static inline int __must_check mddev_lock(struct mddev * mddev) /* Sometimes we need to take the lock in a situation where * failure due to interrupts is not acceptable. */ -static inline void mddev_lock_nointr(struct mddev * mddev) +static inline void mddev_lock_nointr(struct mddev *mddev) { mutex_lock(&mddev->reconfig_mutex); } @@ -580,14 +574,14 @@ static inline int mddev_is_locked(struct mddev *mddev) return mutex_is_locked(&mddev->reconfig_mutex); } -static inline int mddev_trylock(struct mddev * mddev) +static inline int mddev_trylock(struct mddev *mddev) { return mutex_trylock(&mddev->reconfig_mutex); } static struct attribute_group md_redundancy_group; -static void mddev_unlock(struct mddev * mddev) +static void mddev_unlock(struct mddev *mddev) { if (mddev->to_remove) { /* These cannot be removed under reconfig_mutex as @@ -630,17 +624,6 @@ static void mddev_unlock(struct mddev * mddev) spin_unlock(&pers_lock); } -static struct md_rdev * find_rdev_nr(struct mddev *mddev, int nr) -{ - struct md_rdev *rdev; - - rdev_for_each(rdev, mddev) - if (rdev->desc_nr == nr) - return rdev; - - return NULL; -} - static struct md_rdev *find_rdev_nr_rcu(struct mddev *mddev, int nr) { struct md_rdev *rdev; @@ -693,11 +676,8 @@ static inline sector_t calc_dev_sboffset(struct md_rdev *rdev) return MD_NEW_SIZE_SECTORS(num_sectors); } -static int alloc_disk_sb(struct md_rdev * rdev) +static int alloc_disk_sb(struct md_rdev *rdev) { - if (rdev->sb_page) - MD_BUG(); - rdev->sb_page = alloc_page(GFP_KERNEL); if (!rdev->sb_page) { printk(KERN_ALERT "md: out of memory.\n"); @@ -766,14 +746,7 @@ void md_super_write(struct mddev *mddev, struct md_rdev *rdev, void md_super_wait(struct mddev *mddev) { /* wait for all superblock writes that were scheduled to complete */ - DEFINE_WAIT(wq); - for(;;) { - prepare_to_wait(&mddev->sb_wait, &wq, TASK_UNINTERRUPTIBLE); - if (atomic_read(&mddev->pending_writes)==0) - break; - schedule(); - } - finish_wait(&mddev->sb_wait, &wq); + wait_event(mddev->sb_wait, atomic_read(&mddev->pending_writes)==0); } int sync_page_io(struct md_rdev *rdev, sector_t sector, int size, @@ -801,17 +774,13 @@ int sync_page_io(struct md_rdev *rdev, sector_t sector, int size, } EXPORT_SYMBOL_GPL(sync_page_io); -static int read_disk_sb(struct md_rdev * rdev, int size) +static int read_disk_sb(struct md_rdev *rdev, int size) { char b[BDEVNAME_SIZE]; - if (!rdev->sb_page) { - MD_BUG(); - return -EINVAL; - } + if (rdev->sb_loaded) return 0; - if (!sync_page_io(rdev, 0, size, rdev->sb_page, READ, true)) goto fail; rdev->sb_loaded = 1; @@ -825,7 +794,7 @@ fail: static int uuid_equal(mdp_super_t *sb1, mdp_super_t *sb2) { - return sb1->set_uuid0 == sb2->set_uuid0 && + return sb1->set_uuid0 == sb2->set_uuid0 && sb1->set_uuid1 == sb2->set_uuid1 && sb1->set_uuid2 == sb2->set_uuid2 && sb1->set_uuid3 == sb2->set_uuid3; @@ -861,14 +830,13 @@ abort: return ret; } - static u32 md_csum_fold(u32 csum) { csum = (csum & 0xffff) + (csum >> 16); return (csum & 0xffff) + (csum >> 16); } -static unsigned int calc_sb_csum(mdp_super_t * sb) +static unsigned int calc_sb_csum(mdp_super_t *sb) { u64 newcsum = 0; u32 *sb32 = (u32*)sb; @@ -882,7 +850,6 @@ static unsigned int calc_sb_csum(mdp_super_t * sb) newcsum += sb32[i]; csum = (newcsum & 0xffffffff) + (newcsum>>32); - #ifdef CONFIG_ALPHA /* This used to use csum_partial, which was wrong for several * reasons including that different results are returned on @@ -899,7 +866,6 @@ static unsigned int calc_sb_csum(mdp_super_t * sb) return csum; } - /* * Handle superblock details. * We want to be able to handle multiple superblock formats @@ -965,7 +931,7 @@ int md_check_no_bitmap(struct mddev *mddev) EXPORT_SYMBOL(md_check_no_bitmap); /* - * load_super for 0.90.0 + * load_super for 0.90.0 */ static int super_90_load(struct md_rdev *rdev, struct md_rdev *refdev, int minor_version) { @@ -1044,7 +1010,7 @@ static int super_90_load(struct md_rdev *rdev, struct md_rdev *refdev, int minor ev2 = md_event(refsb); if (ev1 > ev2) ret = 1; - else + else ret = 0; } rdev->sectors = rdev->sb_start; @@ -1118,7 +1084,7 @@ static int super_90_validate(struct mddev *mddev, struct md_rdev *rdev) if (sb->state & (1<<MD_SB_CLEAN)) mddev->recovery_cp = MaxSector; else { - if (sb->events_hi == sb->cp_events_hi && + if (sb->events_hi == sb->cp_events_hi && sb->events_lo == sb->cp_events_lo) { mddev->recovery_cp = sb->recovery_cp; } else @@ -1146,7 +1112,7 @@ static int super_90_validate(struct mddev *mddev, struct md_rdev *rdev) ++ev1; if (sb->disks[rdev->desc_nr].state & ( (1<<MD_DISK_SYNC) | (1 << MD_DISK_ACTIVE))) - if (ev1 < mddev->events) + if (ev1 < mddev->events) return -EINVAL; } else if (mddev->bitmap) { /* if adding to array with a bitmap, then we can accept an @@ -1197,7 +1163,6 @@ static void super_90_sync(struct mddev *mddev, struct md_rdev *rdev) struct md_rdev *rdev2; int next_spare = mddev->raid_disks; - /* make rdev->sb match mddev data.. * * 1/ zero out disks @@ -1366,7 +1331,7 @@ super_90_allow_new_offset(struct md_rdev *rdev, unsigned long long new_offset) * version 1 superblock */ -static __le32 calc_sb_1_csum(struct mdp_superblock_1 * sb) +static __le32 calc_sb_1_csum(struct mdp_superblock_1 *sb) { __le32 disk_csum; u32 csum; @@ -1430,7 +1395,6 @@ static int super_1_load(struct md_rdev *rdev, struct md_rdev *refdev, int minor_ ret = read_disk_sb(rdev, 4096); if (ret) return ret; - sb = page_address(rdev->sb_page); if (sb->magic != cpu_to_le32(MD_SB_MAGIC) || @@ -1817,7 +1781,7 @@ retry: for (i=0; i<max_dev;i++) sb->dev_roles[i] = cpu_to_le16(0xfffe); - + rdev_for_each(rdev2, mddev) { i = rdev2->desc_nr; if (test_bit(Faulty, &rdev2->flags)) @@ -2033,18 +1997,13 @@ void md_integrity_add_rdev(struct md_rdev *rdev, struct mddev *mddev) } EXPORT_SYMBOL(md_integrity_add_rdev); -static int bind_rdev_to_array(struct md_rdev * rdev, struct mddev * mddev) +static int bind_rdev_to_array(struct md_rdev *rdev, struct mddev *mddev) { char b[BDEVNAME_SIZE]; struct kobject *ko; char *s; int err; - if (rdev->mddev) { - MD_BUG(); - return -EINVAL; - } - /* prevent duplicates */ if (find_rdev(mddev, rdev->bdev->bd_dev)) return -EEXIST; @@ -2067,16 +2026,21 @@ static int bind_rdev_to_array(struct md_rdev * rdev, struct mddev * mddev) * If it is -1, assign a free number, else * check number is not in use */ + rcu_read_lock(); if (rdev->desc_nr < 0) { int choice = 0; - if (mddev->pers) choice = mddev->raid_disks; - while (find_rdev_nr(mddev, choice)) + if (mddev->pers) + choice = mddev->raid_disks; + while (find_rdev_nr_rcu(mddev, choice)) choice++; rdev->desc_nr = choice; } else { - if (find_rdev_nr(mddev, rdev->desc_nr)) + if (find_rdev_nr_rcu(mddev, rdev->desc_nr)) { + rcu_read_unlock(); return -EBUSY; + } } + rcu_read_unlock(); if (mddev->max_disks && rdev->desc_nr >= mddev->max_disks) { printk(KERN_WARNING "md: %s: array is limited to %d devices\n", mdname(mddev), mddev->max_disks); @@ -2118,13 +2082,10 @@ static void md_delayed_delete(struct work_struct *ws) kobject_put(&rdev->kobj); } -static void unbind_rdev_from_array(struct md_rdev * rdev) +static void unbind_rdev_from_array(struct md_rdev *rdev) { char b[BDEVNAME_SIZE]; - if (!rdev->mddev) { - MD_BUG(); - return; - } + bd_unlink_disk_holder(rdev->bdev, rdev->mddev->gendisk); list_del_rcu(&rdev->same_set); printk(KERN_INFO "md: unbind<%s>\n", bdevname(rdev->bdev,b)); @@ -2169,20 +2130,17 @@ static void unlock_rdev(struct md_rdev *rdev) { struct block_device *bdev = rdev->bdev; rdev->bdev = NULL; - if (!bdev) - MD_BUG(); blkdev_put(bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL); } void md_autodetect_dev(dev_t dev); -static void export_rdev(struct md_rdev * rdev) +static void export_rdev(struct md_rdev *rdev) { char b[BDEVNAME_SIZE]; + printk(KERN_INFO "md: export_rdev(%s)\n", bdevname(rdev->bdev,b)); - if (rdev->mddev) - MD_BUG(); md_rdev_clear(rdev); #ifndef MODULE if (test_bit(AutoDetected, &rdev->flags)) @@ -2192,7 +2150,7 @@ static void export_rdev(struct md_rdev * rdev) kobject_put(&rdev->kobj); } -static void kick_rdev_from_array(struct md_rdev * rdev) +static void kick_rdev_from_array(struct md_rdev *rdev) { unbind_rdev_from_array(rdev); export_rdev(rdev); @@ -2200,153 +2158,18 @@ static void kick_rdev_from_array(struct md_rdev * rdev) static void export_array(struct mddev *mddev) { - struct md_rdev *rdev, *tmp; + struct md_rdev *rdev; - rdev_for_each_safe(rdev, tmp, mddev) { - if (!rdev->mddev) { - MD_BUG(); - continue; - } + while (!list_empty(&mddev->disks)) { + rdev = list_first_entry(&mddev->disks, struct md_rdev, + same_set); kick_rdev_from_array(rdev); } - if (!list_empty(&mddev->disks)) - MD_BUG(); mddev->raid_disks = 0; mddev->major_version = 0; } -static void print_desc(mdp_disk_t *desc) -{ - printk(" DISK<N:%d,(%d,%d),R:%d,S:%d>\n", desc->number, - desc->major,desc->minor,desc->raid_disk,desc->state); -} - -static void print_sb_90(mdp_super_t *sb) -{ - int i; - - printk(KERN_INFO - "md: SB: (V:%d.%d.%d) ID:<%08x.%08x.%08x.%08x> CT:%08x\n", - sb->major_version, sb->minor_version, sb->patch_version, - sb->set_uuid0, sb->set_uuid1, sb->set_uuid2, sb->set_uuid3, - sb->ctime); - printk(KERN_INFO "md: L%d S%08d ND:%d RD:%d md%d LO:%d CS:%d\n", - sb->level, sb->size, sb->nr_disks, sb->raid_disks, - sb->md_minor, sb->layout, sb->chunk_size); - printk(KERN_INFO "md: UT:%08x ST:%d AD:%d WD:%d" - " FD:%d SD:%d CSUM:%08x E:%08lx\n", - sb->utime, sb->state, sb->active_disks, sb->working_disks, - sb->failed_disks, sb->spare_disks, - sb->sb_csum, (unsigned long)sb->events_lo); - - printk(KERN_INFO); - for (i = 0; i < MD_SB_DISKS; i++) { - mdp_disk_t *desc; - - desc = sb->disks + i; - if (desc->number || desc->major || desc->minor || - desc->raid_disk || (desc->state && (desc->state != 4))) { - printk(" D %2d: ", i); - print_desc(desc); - } - } - printk(KERN_INFO "md: THIS: "); - print_desc(&sb->this_disk); -} - -static void print_sb_1(struct mdp_superblock_1 *sb) -{ - __u8 *uuid; - - uuid = sb->set_uuid; - printk(KERN_INFO - "md: SB: (V:%u) (F:0x%08x) Array-ID:<%pU>\n" - "md: Name: \"%s\" CT:%llu\n", - le32_to_cpu(sb->major_version), - le32_to_cpu(sb->feature_map), - uuid, - sb->set_name, - (unsigned long long)le64_to_cpu(sb->ctime) - & MD_SUPERBLOCK_1_TIME_SEC_MASK); - - uuid = sb->device_uuid; - printk(KERN_INFO - "md: L%u SZ%llu RD:%u LO:%u CS:%u DO:%llu DS:%llu SO:%llu" - " RO:%llu\n" - "md: Dev:%08x UUID: %pU\n" - "md: (F:0x%08x) UT:%llu Events:%llu ResyncOffset:%llu CSUM:0x%08x\n" - "md: (MaxDev:%u) \n", - le32_to_cpu(sb->level), - (unsigned long long)le64_to_cpu(sb->size), - le32_to_cpu(sb->raid_disks), - le32_to_cpu(sb->layout), - le32_to_cpu(sb->chunksize), - (unsigned long long)le64_to_cpu(sb->data_offset), - (unsigned long long)le64_to_cpu(sb->data_size), - (unsigned long long)le64_to_cpu(sb->super_offset), - (unsigned long long)le64_to_cpu(sb->recovery_offset), - le32_to_cpu(sb->dev_number), - uuid, - sb->devflags, - (unsigned long long)le64_to_cpu(sb->utime) & MD_SUPERBLOCK_1_TIME_SEC_MASK, - (unsigned long long)le64_to_cpu(sb->events), - (unsigned long long)le64_to_cpu(sb->resync_offset), - le32_to_cpu(sb->sb_csum), - le32_to_cpu(sb->max_dev) - ); -} - -static void print_rdev(struct md_rdev *rdev, int major_version) -{ - char b[BDEVNAME_SIZE]; - printk(KERN_INFO "md: rdev %s, Sect:%08llu F:%d S:%d DN:%u\n", - bdevname(rdev->bdev, b), (unsigned long long)rdev->sectors, - test_bit(Faulty, &rdev->flags), test_bit(In_sync, &rdev->flags), - rdev->desc_nr); - if (rdev->sb_loaded) { - printk(KERN_INFO "md: rdev superblock (MJ:%d):\n", major_version); - switch (major_version) { - case 0: - print_sb_90(page_address(rdev->sb_page)); - break; - case 1: - print_sb_1(page_address(rdev->sb_page)); - break; - } - } else - printk(KERN_INFO "md: no rdev superblock!\n"); -} - -static void md_print_devices(void) -{ - struct list_head *tmp; - struct md_rdev *rdev; - struct mddev *mddev; - char b[BDEVNAME_SIZE]; - - printk("\n"); - printk("md: **********************************\n"); - printk("md: * <COMPLETE RAID STATE PRINTOUT> *\n"); - printk("md: **********************************\n"); - for_each_mddev(mddev, tmp) { - - if (mddev->bitmap) - bitmap_print_sb(mddev->bitmap); - else - printk("%s: ", mdname(mddev)); - rdev_for_each(rdev, mddev) - printk("<%s>", bdevname(rdev->bdev,b)); - printk("\n"); - - rdev_for_each(rdev, mddev) - print_rdev(rdev, mddev->major_version); - } - printk("md: **********************************\n"); - printk("\n"); -} - - -static void sync_sbs(struct mddev * mddev, int nospares) +static void sync_sbs(struct mddev *mddev, int nospares) { /* Update each superblock (in-memory image), but * if we are allowed to, skip spares which already @@ -2369,7 +2192,7 @@ static void sync_sbs(struct mddev * mddev, int nospares) } } -static void md_update_sb(struct mddev * mddev, int force_change) +static void md_update_sb(struct mddev *mddev, int force_change) { struct md_rdev *rdev; int sync_req; @@ -2390,7 +2213,7 @@ repeat: mddev->curr_resync_completed > rdev->recovery_offset) rdev->recovery_offset = mddev->curr_resync_completed; - } + } if (!mddev->persistent) { clear_bit(MD_CHANGE_CLEAN, &mddev->flags); clear_bit(MD_CHANGE_DEVS, &mddev->flags); @@ -2453,15 +2276,12 @@ repeat: mddev->can_decrease_events = nospares; } - if (!mddev->events) { - /* - * oops, this 64-bit counter should never wrap. - * Either we are in around ~1 trillion A.C., assuming - * 1 reboot per second, or we have a bug: - */ - MD_BUG(); - mddev->events --; - } + /* + * This 64-bit counter should never wrap. + * Either we are in around ~1 trillion A.C., assuming + * 1 reboot per second, or we have a bug... + */ + WARN_ON(mddev->events == 0); rdev_for_each(rdev, mddev) { if (rdev->badblocks.changed) @@ -2668,10 +2488,12 @@ state_store(struct md_rdev *rdev, const char *buf, size_t len) set_bit(In_sync, &rdev->flags); err = 0; } else if (cmd_match(buf, "-insync") && rdev->raid_disk >= 0) { - clear_bit(In_sync, &rdev->flags); - rdev->saved_raid_disk = rdev->raid_disk; - rdev->raid_disk = -1; - err = 0; + if (rdev->mddev->pers == NULL) { + clear_bit(In_sync, &rdev->flags); + rdev->saved_raid_disk = rdev->raid_disk; + rdev->raid_disk = -1; + err = 0; + } } else if (cmd_match(buf, "write_error")) { set_bit(WriteErrorSeen, &rdev->flags); err = 0; @@ -2829,7 +2651,6 @@ slot_store(struct md_rdev *rdev, const char *buf, size_t len) return len; } - static struct rdev_sysfs_entry rdev_slot = __ATTR(slot, S_IRUGO|S_IWUSR, slot_show, slot_store); @@ -2980,20 +2801,20 @@ rdev_size_store(struct md_rdev *rdev, const char *buf, size_t len) rdev->sectors = sectors; if (sectors > oldsectors && my_mddev->external) { - /* need to check that all other rdevs with the same ->bdev - * do not overlap. We need to unlock the mddev to avoid - * a deadlock. We have already changed rdev->sectors, and if - * we have to change it back, we will have the lock again. + /* Need to check that all other rdevs with the same + * ->bdev do not overlap. 'rcu' is sufficient to walk + * the rdev lists safely. + * This check does not provide a hard guarantee, it + * just helps avoid dangerous mistakes. */ struct mddev *mddev; int overlap = 0; struct list_head *tmp; - mddev_unlock(my_mddev); + rcu_read_lock(); for_each_mddev(mddev, tmp) { struct md_rdev *rdev2; - mddev_lock_nointr(mddev); rdev_for_each(rdev2, mddev) if (rdev->bdev == rdev2->bdev && rdev != rdev2 && @@ -3003,13 +2824,12 @@ rdev_size_store(struct md_rdev *rdev, const char *buf, size_t len) overlap = 1; break; } - mddev_unlock(mddev); if (overlap) { mddev_put(mddev); break; } } - mddev_lock_nointr(my_mddev); + rcu_read_unlock(); if (overlap) { /* Someone else could have slipped in a size * change here, but doing so is just silly. @@ -3027,7 +2847,6 @@ rdev_size_store(struct md_rdev *rdev, const char *buf, size_t len) static struct rdev_sysfs_entry rdev_size = __ATTR(size, S_IRUGO|S_IWUSR, rdev_size_show, rdev_size_store); - static ssize_t recovery_start_show(struct md_rdev *rdev, char *page) { unsigned long long recovery_start = rdev->recovery_offset; @@ -3063,7 +2882,6 @@ static ssize_t recovery_start_store(struct md_rdev *rdev, const char *buf, size_ static struct rdev_sysfs_entry rdev_recovery_start = __ATTR(recovery_start, S_IRUGO|S_IWUSR, recovery_start_show, recovery_start_store); - static ssize_t badblocks_show(struct badblocks *bb, char *page, int unack); static ssize_t @@ -3084,7 +2902,6 @@ static ssize_t bb_store(struct md_rdev *rdev, const char *page, size_t len) static struct rdev_sysfs_entry rdev_bad_blocks = __ATTR(bad_blocks, S_IRUGO|S_IWUSR, bb_show, bb_store); - static ssize_t ubb_show(struct md_rdev *rdev, char *page) { return badblocks_show(&rdev->badblocks, page, 1); @@ -3241,7 +3058,7 @@ static struct md_rdev *md_import_device(dev_t newdev, int super_format, int supe size = i_size_read(rdev->bdev->bd_inode) >> BLOCK_SIZE_BITS; if (!size) { - printk(KERN_WARNING + printk(KERN_WARNING "md: %s has zero or unknown size, marking faulty!\n", bdevname(rdev->bdev,b)); err = -EINVAL; @@ -3260,7 +3077,7 @@ static struct md_rdev *md_import_device(dev_t newdev, int super_format, int supe goto abort_free; } if (err < 0) { - printk(KERN_WARNING + printk(KERN_WARNING "md: could not read %s's sb, not importing!\n", bdevname(rdev->bdev,b)); goto abort_free; @@ -3281,8 +3098,7 @@ abort_free: * Check a full RAID array for plausibility */ - -static void analyze_sbs(struct mddev * mddev) +static void analyze_sbs(struct mddev *mddev) { int i; struct md_rdev *rdev, *freshest, *tmp; @@ -3300,12 +3116,11 @@ static void analyze_sbs(struct mddev * mddev) default: printk( KERN_ERR \ "md: fatal superblock inconsistency in %s" - " -- removing from array\n", + " -- removing from array\n", bdevname(rdev->bdev,b)); kick_rdev_from_array(rdev); } - super_types[mddev->major_version]. validate_super(mddev, freshest); @@ -3344,7 +3159,7 @@ static void analyze_sbs(struct mddev * mddev) /* Read a fixed-point number. * Numbers in sysfs attributes should be in "standard" units where * possible, so time should be in seconds. - * However we internally use a a much smaller unit such as + * However we internally use a a much smaller unit such as * milliseconds or jiffies. * This function takes a decimal number with a possible fractional * component, and produces an integer which is the result of @@ -3381,7 +3196,6 @@ int strict_strtoul_scaled(const char *cp, unsigned long *res, int scale) return 0; } - static void md_safemode_timeout(unsigned long data); static ssize_t @@ -3524,7 +3338,7 @@ level_store(struct mddev *mddev, const char *buf, size_t len) /* Looks like we have a winner */ mddev_suspend(mddev); mddev->pers->stop(mddev); - + if (mddev->pers->sync_request == NULL && pers->sync_request != NULL) { /* need to add the md_redundancy_group */ @@ -3533,7 +3347,7 @@ level_store(struct mddev *mddev, const char *buf, size_t len) "md: cannot register extra attributes for %s\n", mdname(mddev)); mddev->sysfs_action = sysfs_get_dirent(mddev->kobj.sd, "sync_action"); - } + } if (mddev->pers->sync_request != NULL && pers->sync_request == NULL) { /* need to remove the md_redundancy_group */ @@ -3611,7 +3425,6 @@ level_store(struct mddev *mddev, const char *buf, size_t len) static struct md_sysfs_entry md_level = __ATTR(level, S_IRUGO|S_IWUSR, level_show, level_store); - static ssize_t layout_show(struct mddev *mddev, char *page) { @@ -3654,7 +3467,6 @@ layout_store(struct mddev *mddev, const char *buf, size_t len) static struct md_sysfs_entry md_layout = __ATTR(layout, S_IRUGO|S_IWUSR, layout_show, layout_store); - static ssize_t raid_disks_show(struct mddev *mddev, char *page) { @@ -3859,9 +3671,9 @@ array_state_show(struct mddev *mddev, char *page) return sprintf(page, "%s\n", array_states[st]); } -static int do_md_stop(struct mddev * mddev, int ro, struct block_device *bdev); -static int md_set_readonly(struct mddev * mddev, struct block_device *bdev); -static int do_md_run(struct mddev * mddev); +static int do_md_stop(struct mddev *mddev, int ro, struct block_device *bdev); +static int md_set_readonly(struct mddev *mddev, struct block_device *bdev); +static int do_md_run(struct mddev *mddev); static int restart_array(struct mddev *mddev); static ssize_t @@ -4012,7 +3824,6 @@ new_dev_store(struct mddev *mddev, const char *buf, size_t len) minor != MINOR(dev)) return -EOVERFLOW; - if (mddev->persistent) { rdev = md_import_device(dev, mddev->major_version, mddev->minor_version); @@ -4108,7 +3919,6 @@ size_store(struct mddev *mddev, const char *buf, size_t len) static struct md_sysfs_entry md_size = __ATTR(component_size, S_IRUGO|S_IWUSR, size_show, size_store); - /* Metadata version. * This is one of * 'none' for arrays with no metadata (good luck...) @@ -4490,7 +4300,7 @@ suspend_lo_store(struct mddev *mddev, const char *buf, size_t len) unsigned long long new = simple_strtoull(buf, &e, 10); unsigned long long old = mddev->suspend_lo; - if (mddev->pers == NULL || + if (mddev->pers == NULL || mddev->pers->quiesce == NULL) return -EINVAL; if (buf == e || (*e && *e != '\n')) @@ -4510,7 +4320,6 @@ suspend_lo_store(struct mddev *mddev, const char *buf, size_t len) static struct md_sysfs_entry md_suspend_lo = __ATTR(suspend_lo, S_IRUGO|S_IWUSR, suspend_lo_show, suspend_lo_store); - static ssize_t suspend_hi_show(struct mddev *mddev, char *page) { @@ -4698,7 +4507,6 @@ static struct attribute_group md_redundancy_group = { .attrs = md_redundancy_attrs, }; - static ssize_t md_attr_show(struct kobject *kobj, struct attribute *attr, char *page) { @@ -5111,7 +4919,7 @@ int md_run(struct mddev *mddev) } else if (mddev->ro == 2) /* auto-readonly not meaningful */ mddev->ro = 0; - atomic_set(&mddev->writes_pending,0); + atomic_set(&mddev->writes_pending,0); atomic_set(&mddev->max_corr_read_errors, MD_DEFAULT_MAX_CORRECTED_READ_ERRORS); mddev->safemode = 0; @@ -5125,9 +4933,9 @@ int md_run(struct mddev *mddev) if (rdev->raid_disk >= 0) if (sysfs_link_rdev(mddev, rdev)) /* failure here is OK */; - + set_bit(MD_RECOVERY_NEEDED, &mddev->recovery); - + if (mddev->flags & MD_UPDATE_SB_FLAGS) md_update_sb(mddev, 0); @@ -5307,7 +5115,7 @@ static int md_set_readonly(struct mddev *mddev, struct block_device *bdev) mddev_lock_nointr(mddev); mutex_lock(&mddev->open_mutex); - if (atomic_read(&mddev->openers) > !!bdev || + if ((mddev->pers && atomic_read(&mddev->openers) > !!bdev) || mddev->sync_thread || (bdev && !test_bit(MD_STILL_CLOSED, &mddev->flags))) { printk("md: %s still in use.\n",mdname(mddev)); @@ -5339,7 +5147,7 @@ out: * 0 - completely stop and dis-assemble array * 2 - stop but do not disassemble array */ -static int do_md_stop(struct mddev * mddev, int mode, +static int do_md_stop(struct mddev *mddev, int mode, struct block_device *bdev) { struct gendisk *disk = mddev->gendisk; @@ -5362,7 +5170,7 @@ static int do_md_stop(struct mddev * mddev, int mode, mddev_lock_nointr(mddev); mutex_lock(&mddev->open_mutex); - if (atomic_read(&mddev->openers) > !!bdev || + if ((mddev->pers && atomic_read(&mddev->openers) > !!bdev) || mddev->sysfs_active || mddev->sync_thread || (bdev && !test_bit(MD_STILL_CLOSED, &mddev->flags))) { @@ -5512,12 +5320,12 @@ static void autorun_devices(int part) "md: cannot allocate memory for md drive.\n"); break; } - if (mddev_lock(mddev)) + if (mddev_lock(mddev)) printk(KERN_WARNING "md: %s locked, cannot run\n", mdname(mddev)); else if (mddev->raid_disks || mddev->major_version || !list_empty(&mddev->disks)) { - printk(KERN_WARNING + printk(KERN_WARNING "md: %s already running, cannot run %s\n", mdname(mddev), bdevname(rdev0->bdev,b)); mddev_unlock(mddev); @@ -5545,7 +5353,7 @@ static void autorun_devices(int part) } #endif /* !MODULE */ -static int get_version(void __user * arg) +static int get_version(void __user *arg) { mdu_version_t ver; @@ -5559,7 +5367,7 @@ static int get_version(void __user * arg) return 0; } -static int get_array_info(struct mddev * mddev, void __user * arg) +static int get_array_info(struct mddev *mddev, void __user *arg) { mdu_array_info_t info; int nr,working,insync,failed,spare; @@ -5574,7 +5382,7 @@ static int get_array_info(struct mddev * mddev, void __user * arg) else { working++; if (test_bit(In_sync, &rdev->flags)) - insync++; + insync++; else spare++; } @@ -5614,7 +5422,7 @@ static int get_array_info(struct mddev * mddev, void __user * arg) return 0; } -static int get_bitmap_file(struct mddev * mddev, void __user * arg) +static int get_bitmap_file(struct mddev *mddev, void __user * arg) { mdu_bitmap_file_t *file = NULL; /* too big for stack allocation */ char *ptr, *buf = NULL; @@ -5652,7 +5460,7 @@ out: return err; } -static int get_disk_info(struct mddev * mddev, void __user * arg) +static int get_disk_info(struct mddev *mddev, void __user * arg) { mdu_disk_info_t info; struct md_rdev *rdev; @@ -5688,7 +5496,7 @@ static int get_disk_info(struct mddev * mddev, void __user * arg) return 0; } -static int add_new_disk(struct mddev * mddev, mdu_disk_info_t *info) +static int add_new_disk(struct mddev *mddev, mdu_disk_info_t *info) { char b[BDEVNAME_SIZE], b2[BDEVNAME_SIZE]; struct md_rdev *rdev; @@ -5702,7 +5510,7 @@ static int add_new_disk(struct mddev * mddev, mdu_disk_info_t *info) /* expecting a device which has a superblock */ rdev = md_import_device(dev, mddev->major_version, mddev->minor_version); if (IS_ERR(rdev)) { - printk(KERN_WARNING + printk(KERN_WARNING "md: md_import_device returned %ld\n", PTR_ERR(rdev)); return PTR_ERR(rdev); @@ -5714,9 +5522,9 @@ static int add_new_disk(struct mddev * mddev, mdu_disk_info_t *info) err = super_types[mddev->major_version] .load_super(rdev, rdev0, mddev->minor_version); if (err < 0) { - printk(KERN_WARNING + printk(KERN_WARNING "md: %s has different UUID to %s\n", - bdevname(rdev->bdev,b), + bdevname(rdev->bdev,b), bdevname(rdev0->bdev,b2)); export_rdev(rdev); return -EINVAL; @@ -5736,7 +5544,7 @@ static int add_new_disk(struct mddev * mddev, mdu_disk_info_t *info) if (mddev->pers) { int err; if (!mddev->pers->hot_add_disk) { - printk(KERN_WARNING + printk(KERN_WARNING "%s: personality does not support diskops!\n", mdname(mddev)); return -EINVAL; @@ -5747,7 +5555,7 @@ static int add_new_disk(struct mddev * mddev, mdu_disk_info_t *info) else rdev = md_import_device(dev, -1, -1); if (IS_ERR(rdev)) { - printk(KERN_WARNING + printk(KERN_WARNING "md: md_import_device returned %ld\n", PTR_ERR(rdev)); return PTR_ERR(rdev); @@ -5821,7 +5629,7 @@ static int add_new_disk(struct mddev * mddev, mdu_disk_info_t *info) int err; rdev = md_import_device(dev, -1, 0); if (IS_ERR(rdev)) { - printk(KERN_WARNING + printk(KERN_WARNING "md: error, md_import_device() returned %ld\n", PTR_ERR(rdev)); return PTR_ERR(rdev); @@ -5856,7 +5664,7 @@ static int add_new_disk(struct mddev * mddev, mdu_disk_info_t *info) return 0; } -static int hot_remove_disk(struct mddev * mddev, dev_t dev) +static int hot_remove_disk(struct mddev *mddev, dev_t dev) { char b[BDEVNAME_SIZE]; struct md_rdev *rdev; @@ -5882,7 +5690,7 @@ busy: return -EBUSY; } -static int hot_add_disk(struct mddev * mddev, dev_t dev) +static int hot_add_disk(struct mddev *mddev, dev_t dev) { char b[BDEVNAME_SIZE]; int err; @@ -5898,7 +5706,7 @@ static int hot_add_disk(struct mddev * mddev, dev_t dev) return -EINVAL; } if (!mddev->pers->hot_add_disk) { - printk(KERN_WARNING + printk(KERN_WARNING "%s: personality does not support diskops!\n", mdname(mddev)); return -EINVAL; @@ -5906,7 +5714,7 @@ static int hot_add_disk(struct mddev * mddev, dev_t dev) rdev = md_import_device(dev, -1, 0); if (IS_ERR(rdev)) { - printk(KERN_WARNING + printk(KERN_WARNING "md: error, md_import_device() returned %ld\n", PTR_ERR(rdev)); return -EINVAL; @@ -5920,7 +5728,7 @@ static int hot_add_disk(struct mddev * mddev, dev_t dev) rdev->sectors = rdev->sb_start; if (test_bit(Faulty, &rdev->flags)) { - printk(KERN_WARNING + printk(KERN_WARNING "md: can not hot-add faulty %s disk to %s!\n", bdevname(rdev->bdev,b), mdname(mddev)); err = -EINVAL; @@ -5968,7 +5776,6 @@ static int set_bitmap_file(struct mddev *mddev, int fd) /* we should be able to change the bitmap.. */ } - if (fd >= 0) { struct inode *inode; if (mddev->bitmap) @@ -6039,7 +5846,7 @@ static int set_bitmap_file(struct mddev *mddev, int fd) * The minor and patch _version numbers are also kept incase the * super_block handler wishes to interpret them. */ -static int set_array_info(struct mddev * mddev, mdu_array_info_t *info) +static int set_array_info(struct mddev *mddev, mdu_array_info_t *info) { if (info->raid_disks == 0) { @@ -6048,7 +5855,7 @@ static int set_array_info(struct mddev * mddev, mdu_array_info_t *info) info->major_version >= ARRAY_SIZE(super_types) || super_types[info->major_version].name == NULL) { /* maybe try to auto-load a module? */ - printk(KERN_INFO + printk(KERN_INFO "md: superblock version %d not known\n", info->major_version); return -EINVAL; @@ -6196,7 +6003,6 @@ static int update_raid_disks(struct mddev *mddev, int raid_disks) return rv; } - /* * update_array_info is used to change the configuration of an * on-line array. @@ -6347,7 +6153,6 @@ static inline bool md_ioctl_valid(unsigned int cmd) case GET_DISK_INFO: case HOT_ADD_DISK: case HOT_REMOVE_DISK: - case PRINT_RAID_DEBUG: case RAID_AUTORUN: case RAID_VERSION: case RESTART_ARRAY_RW: @@ -6391,18 +6196,13 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, switch (cmd) { case RAID_VERSION: err = get_version(argp); - goto done; - - case PRINT_RAID_DEBUG: - err = 0; - md_print_devices(); - goto done; + goto out; #ifndef MODULE case RAID_AUTORUN: err = 0; autostart_arrays(arg); - goto done; + goto out; #endif default:; } @@ -6415,7 +6215,7 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, if (!mddev) { BUG(); - goto abort; + goto out; } /* Some actions do not requires the mutex */ @@ -6425,18 +6225,18 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, err = -ENODEV; else err = get_array_info(mddev, argp); - goto abort; + goto out; case GET_DISK_INFO: if (!mddev->raid_disks && !mddev->external) err = -ENODEV; else err = get_disk_info(mddev, argp); - goto abort; + goto out; case SET_DISK_FAULTY: err = set_disk_faulty(mddev, new_decode_dev(arg)); - goto abort; + goto out; } if (cmd == ADD_NEW_DISK) @@ -6454,10 +6254,10 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, * and writes */ mutex_lock(&mddev->open_mutex); - if (atomic_read(&mddev->openers) > 1) { + if (mddev->pers && atomic_read(&mddev->openers) > 1) { mutex_unlock(&mddev->open_mutex); err = -EBUSY; - goto abort; + goto out; } set_bit(MD_STILL_CLOSED, &mddev->flags); mutex_unlock(&mddev->open_mutex); @@ -6465,10 +6265,10 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, } err = mddev_lock(mddev); if (err) { - printk(KERN_INFO + printk(KERN_INFO "md: ioctl lock interrupted, reason %d, cmd %d\n", err, cmd); - goto abort; + goto out; } if (cmd == SET_ARRAY_INFO) { @@ -6477,38 +6277,38 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, memset(&info, 0, sizeof(info)); else if (copy_from_user(&info, argp, sizeof(info))) { err = -EFAULT; - goto abort_unlock; + goto unlock; } if (mddev->pers) { err = update_array_info(mddev, &info); if (err) { printk(KERN_WARNING "md: couldn't update" " array info. %d\n", err); - goto abort_unlock; + goto unlock; } - goto done_unlock; + goto unlock; } if (!list_empty(&mddev->disks)) { printk(KERN_WARNING "md: array %s already has disks!\n", mdname(mddev)); err = -EBUSY; - goto abort_unlock; + goto unlock; } if (mddev->raid_disks) { printk(KERN_WARNING "md: array %s already initialised!\n", mdname(mddev)); err = -EBUSY; - goto abort_unlock; + goto unlock; } err = set_array_info(mddev, &info); if (err) { printk(KERN_WARNING "md: couldn't set" " array info. %d\n", err); - goto abort_unlock; + goto unlock; } - goto done_unlock; + goto unlock; } /* @@ -6521,7 +6321,7 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, && cmd != RUN_ARRAY && cmd != SET_BITMAP_FILE && cmd != GET_BITMAP_FILE) { err = -ENODEV; - goto abort_unlock; + goto unlock; } /* @@ -6530,23 +6330,23 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, switch (cmd) { case GET_BITMAP_FILE: err = get_bitmap_file(mddev, argp); - goto done_unlock; + goto unlock; case RESTART_ARRAY_RW: err = restart_array(mddev); - goto done_unlock; + goto unlock; case STOP_ARRAY: err = do_md_stop(mddev, 0, bdev); - goto done_unlock; + goto unlock; case STOP_ARRAY_RO: err = md_set_readonly(mddev, bdev); - goto done_unlock; + goto unlock; case HOT_REMOVE_DISK: err = hot_remove_disk(mddev, new_decode_dev(arg)); - goto done_unlock; + goto unlock; case ADD_NEW_DISK: /* We can support ADD_NEW_DISK on read-only arrays @@ -6562,14 +6362,14 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, break; else err = add_new_disk(mddev, &info); - goto done_unlock; + goto unlock; } break; case BLKROSET: if (get_user(ro, (int __user *)(arg))) { err = -EFAULT; - goto done_unlock; + goto unlock; } err = -EINVAL; @@ -6577,11 +6377,11 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, * does not matter, no writes are coming */ if (ro) - goto done_unlock; + goto unlock; /* are we are already prepared for writes? */ if (mddev->ro != 1) - goto done_unlock; + goto unlock; /* transitioning to readauto need only happen for * arrays that call md_write_start @@ -6593,17 +6393,14 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, set_disk_ro(mddev->gendisk, 0); } } - goto done_unlock; + goto unlock; } /* * The remaining ioctls are changing the state of the * superblock, so we do not allow them on read-only arrays. - * However non-MD ioctls (e.g. get-size) will still come through - * here and hit the 'default' below, so only disallow - * 'md' ioctls, and switch to rw mode if started auto-readonly. */ - if (_IOC_TYPE(cmd) == MD_MAJOR && mddev->ro && mddev->pers) { + if (mddev->ro && mddev->pers) { if (mddev->ro == 2) { mddev->ro = 0; sysfs_notify_dirent_safe(mddev->sysfs_state); @@ -6621,7 +6418,7 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, } } else { err = -EROFS; - goto abort_unlock; + goto unlock; } } @@ -6633,38 +6430,32 @@ static int md_ioctl(struct block_device *bdev, fmode_t mode, err = -EFAULT; else err = add_new_disk(mddev, &info); - goto done_unlock; + goto unlock; } case HOT_ADD_DISK: err = hot_add_disk(mddev, new_decode_dev(arg)); - goto done_unlock; + goto unlock; case RUN_ARRAY: err = do_md_run(mddev); - goto done_unlock; + goto unlock; case SET_BITMAP_FILE: err = set_bitmap_file(mddev, (int)arg); - goto done_unlock; + goto unlock; default: err = -EINVAL; - goto abort_unlock; + goto unlock; } -done_unlock: -abort_unlock: +unlock: if (mddev->hold_active == UNTIL_IOCTL && err != -EINVAL) mddev->hold_active = 0; mddev_unlock(mddev); - - return err; -done: - if (err) - MD_BUG(); -abort: +out: return err; } #ifdef CONFIG_COMPAT @@ -6726,7 +6517,7 @@ static int md_open(struct block_device *bdev, fmode_t mode) static void md_release(struct gendisk *disk, fmode_t mode) { - struct mddev *mddev = disk->private_data; + struct mddev *mddev = disk->private_data; BUG_ON(!mddev); atomic_dec(&mddev->openers); @@ -6761,7 +6552,7 @@ static const struct block_device_operations md_fops = .revalidate_disk= md_revalidate, }; -static int md_thread(void * arg) +static int md_thread(void *arg) { struct md_thread *thread = arg; @@ -6810,6 +6601,7 @@ void md_wakeup_thread(struct md_thread *thread) wake_up(&thread->wqueue); } } +EXPORT_SYMBOL(md_wakeup_thread); struct md_thread *md_register_thread(void (*run) (struct md_thread *), struct mddev *mddev, const char *name) @@ -6835,6 +6627,7 @@ struct md_thread *md_register_thread(void (*run) (struct md_thread *), } return thread; } +EXPORT_SYMBOL(md_register_thread); void md_unregister_thread(struct md_thread **threadp) { @@ -6852,14 +6645,10 @@ void md_unregister_thread(struct md_thread **threadp) kthread_stop(thread->tsk); kfree(thread); } +EXPORT_SYMBOL(md_unregister_thread); void md_error(struct mddev *mddev, struct md_rdev *rdev) { - if (!mddev) { - MD_BUG(); - return; - } - if (!rdev || test_bit(Faulty, &rdev->flags)) return; @@ -6876,6 +6665,7 @@ void md_error(struct mddev *mddev, struct md_rdev *rdev) queue_work(md_misc_wq, &mddev->event_work); md_new_event_inintr(mddev); } +EXPORT_SYMBOL(md_error); /* seq_file implementation /proc/mdstat */ @@ -6898,8 +6688,7 @@ static void status_unused(struct seq_file *seq) seq_printf(seq, "\n"); } - -static void status_resync(struct seq_file *seq, struct mddev * mddev) +static void status_resync(struct seq_file *seq, struct mddev *mddev) { sector_t max_sectors, resync, res; unsigned long dt, db; @@ -6919,13 +6708,7 @@ static void status_resync(struct seq_file *seq, struct mddev * mddev) else max_sectors = mddev->dev_sectors; - /* - * Should not happen. - */ - if (!max_sectors) { - MD_BUG(); - return; - } + WARN_ON(max_sectors == 0); /* Pick 'scale' such that (resync>>scale)*1000 will fit * in a sector_t, and (max_sectors>>scale) will fit in a * u32, as those are the requirements for sector_div. @@ -7021,7 +6804,7 @@ static void *md_seq_next(struct seq_file *seq, void *v, loff_t *pos) { struct list_head *tmp; struct mddev *next_mddev, *mddev = v; - + ++*pos; if (v == (void*)2) return NULL; @@ -7036,7 +6819,7 @@ static void *md_seq_next(struct seq_file *seq, void *v, loff_t *pos) else { next_mddev = (void*)2; *pos = 0x10000; - } + } spin_unlock(&all_mddevs_lock); if (v != (void*)1) @@ -7132,7 +6915,7 @@ static int md_seq_show(struct seq_file *seq, void *v) if (mddev->pers) { mddev->pers->status(seq, mddev); - seq_printf(seq, "\n "); + seq_printf(seq, "\n "); if (mddev->pers->sync_request) { if (mddev->curr_resync > 2) { status_resync(seq, mddev); @@ -7150,7 +6933,7 @@ static int md_seq_show(struct seq_file *seq, void *v) seq_printf(seq, "\n"); } mddev_unlock(mddev); - + return 0; } @@ -7204,12 +6987,14 @@ static const struct file_operations md_seq_fops = { int register_md_personality(struct md_personality *p) { + printk(KERN_INFO "md: %s personality registered for level %d\n", + p->name, p->level); spin_lock(&pers_lock); list_add_tail(&p->list, &pers_list); - printk(KERN_INFO "md: %s personality registered for level %d\n", p->name, p->level); spin_unlock(&pers_lock); return 0; } +EXPORT_SYMBOL(register_md_personality); int unregister_md_personality(struct md_personality *p) { @@ -7219,10 +7004,11 @@ int unregister_md_personality(struct md_personality *p) spin_unlock(&pers_lock); return 0; } +EXPORT_SYMBOL(unregister_md_personality); static int is_mddev_idle(struct mddev *mddev, int init) { - struct md_rdev * rdev; + struct md_rdev *rdev; int idle; int curr_events; @@ -7276,7 +7062,7 @@ void md_done_sync(struct mddev *mddev, int blocks, int ok) // stop recovery, signal do_sync .... } } - +EXPORT_SYMBOL(md_done_sync); /* md_write_start(mddev, bi) * If we need to update some array metadata (e.g. 'active' flag @@ -7317,6 +7103,7 @@ void md_write_start(struct mddev *mddev, struct bio *bi) wait_event(mddev->sb_wait, !test_bit(MD_CHANGE_PENDING, &mddev->flags)); } +EXPORT_SYMBOL(md_write_start); void md_write_end(struct mddev *mddev) { @@ -7327,6 +7114,7 @@ void md_write_end(struct mddev *mddev) mod_timer(&mddev->safemode_timer, jiffies + mddev->safemode_delay); } } +EXPORT_SYMBOL(md_write_end); /* md_allow_write(mddev) * Calling this ensures that the array is marked 'active' so that writes @@ -7784,6 +7572,33 @@ no_add: return spares; } +static void md_start_sync(struct work_struct *ws) +{ + struct mddev *mddev = container_of(ws, struct mddev, del_work); + + mddev->sync_thread = md_register_thread(md_do_sync, + mddev, + "resync"); + if (!mddev->sync_thread) { + printk(KERN_ERR "%s: could not start resync" + " thread...\n", + mdname(mddev)); + /* leave the spares where they are, it shouldn't hurt */ + clear_bit(MD_RECOVERY_SYNC, &mddev->recovery); + clear_bit(MD_RECOVERY_RESHAPE, &mddev->recovery); + clear_bit(MD_RECOVERY_REQUESTED, &mddev->recovery); + clear_bit(MD_RECOVERY_CHECK, &mddev->recovery); + clear_bit(MD_RECOVERY_RUNNING, &mddev->recovery); + if (test_and_clear_bit(MD_RECOVERY_RECOVER, + &mddev->recovery)) + if (mddev->sysfs_action) + sysfs_notify_dirent_safe(mddev->sysfs_action); + } else + md_wakeup_thread(mddev->sync_thread); + sysfs_notify_dirent_safe(mddev->sysfs_action); + md_new_event(mddev); +} + /* * This routine is regularly called by all per-raid-array threads to * deal with generic issues like resync and super-block update. @@ -7900,7 +7715,7 @@ void md_check_recovery(struct mddev *mddev) if (!test_and_clear_bit(MD_RECOVERY_NEEDED, &mddev->recovery) || test_bit(MD_RECOVERY_FROZEN, &mddev->recovery)) - goto unlock; + goto not_running; /* no recovery is running. * remove any failed drives, then * add spares if possible. @@ -7912,7 +7727,7 @@ void md_check_recovery(struct mddev *mddev) if (mddev->pers->check_reshape == NULL || mddev->pers->check_reshape(mddev) != 0) /* Cannot proceed */ - goto unlock; + goto not_running; set_bit(MD_RECOVERY_RESHAPE, &mddev->recovery); clear_bit(MD_RECOVERY_RECOVER, &mddev->recovery); } else if ((spares = remove_and_add_spares(mddev, NULL))) { @@ -7925,7 +7740,7 @@ void md_check_recovery(struct mddev *mddev) clear_bit(MD_RECOVERY_RECOVER, &mddev->recovery); } else if (!test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) /* nothing to be done ... */ - goto unlock; + goto not_running; if (mddev->pers->sync_request) { if (spares) { @@ -7935,27 +7750,11 @@ void md_check_recovery(struct mddev *mddev) */ bitmap_write_all(mddev->bitmap); } - mddev->sync_thread = md_register_thread(md_do_sync, - mddev, - "resync"); - if (!mddev->sync_thread) { - printk(KERN_ERR "%s: could not start resync" - " thread...\n", - mdname(mddev)); - /* leave the spares where they are, it shouldn't hurt */ - clear_bit(MD_RECOVERY_RUNNING, &mddev->recovery); - clear_bit(MD_RECOVERY_SYNC, &mddev->recovery); - clear_bit(MD_RECOVERY_RESHAPE, &mddev->recovery); - clear_bit(MD_RECOVERY_REQUESTED, &mddev->recovery); - clear_bit(MD_RECOVERY_CHECK, &mddev->recovery); - } else - md_wakeup_thread(mddev->sync_thread); - sysfs_notify_dirent_safe(mddev->sysfs_action); - md_new_event(mddev); + INIT_WORK(&mddev->del_work, md_start_sync); + queue_work(md_misc_wq, &mddev->del_work); + goto unlock; } - unlock: - wake_up(&mddev->sb_wait); - + not_running: if (!mddev->sync_thread) { clear_bit(MD_RECOVERY_RUNNING, &mddev->recovery); if (test_and_clear_bit(MD_RECOVERY_RECOVER, @@ -7963,9 +7762,12 @@ void md_check_recovery(struct mddev *mddev) if (mddev->sysfs_action) sysfs_notify_dirent_safe(mddev->sysfs_action); } + unlock: + wake_up(&mddev->sb_wait); mddev_unlock(mddev); } } +EXPORT_SYMBOL(md_check_recovery); void md_reap_sync_thread(struct mddev *mddev) { @@ -8008,6 +7810,7 @@ void md_reap_sync_thread(struct mddev *mddev) if (mddev->event_work.func) queue_work(md_misc_wq, &mddev->event_work); } +EXPORT_SYMBOL(md_reap_sync_thread); void md_wait_for_blocked_rdev(struct md_rdev *rdev, struct mddev *mddev) { @@ -8641,7 +8444,6 @@ void md_autodetect_dev(dev_t dev) } } - static void autostart_arrays(int part) { struct md_rdev *rdev; @@ -8665,10 +8467,9 @@ static void autostart_arrays(int part) if (IS_ERR(rdev)) continue; - if (test_bit(Faulty, &rdev->flags)) { - MD_BUG(); + if (test_bit(Faulty, &rdev->flags)) continue; - } + set_bit(AutoDetected, &rdev->flags); list_add(&rdev->same_set, &pending_raid_disks); i_passed++; @@ -8736,20 +8537,8 @@ static int set_ro(const char *val, struct kernel_param *kp) module_param_call(start_ro, set_ro, get_ro, NULL, S_IRUSR|S_IWUSR); module_param(start_dirty_degraded, int, S_IRUGO|S_IWUSR); - module_param_call(new_array, add_named_array, NULL, NULL, S_IWUSR); -EXPORT_SYMBOL(register_md_personality); -EXPORT_SYMBOL(unregister_md_personality); -EXPORT_SYMBOL(md_error); -EXPORT_SYMBOL(md_done_sync); -EXPORT_SYMBOL(md_write_start); -EXPORT_SYMBOL(md_write_end); -EXPORT_SYMBOL(md_register_thread); -EXPORT_SYMBOL(md_unregister_thread); -EXPORT_SYMBOL(md_wakeup_thread); -EXPORT_SYMBOL(md_check_recovery); -EXPORT_SYMBOL(md_reap_sync_thread); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("MD RAID framework"); MODULE_ALIAS("md"); diff --git a/drivers/md/md.h b/drivers/md/md.h index a49d991f3fe1..03cec5bdcaae 100644 --- a/drivers/md/md.h +++ b/drivers/md/md.h @@ -1,15 +1,15 @@ /* md.h : kernel internal structure of the Linux MD driver Copyright (C) 1996-98 Ingo Molnar, Gadi Oxman - + This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2, or (at your option) any later version. - + You should have received a copy of the GNU General Public License (for example /usr/src/linux/COPYING); if not, write to the Free - Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ #ifndef _MD_MD_H @@ -56,7 +56,7 @@ struct md_rdev { __u64 sb_events; sector_t data_offset; /* start of data in array */ sector_t new_data_offset;/* only relevant while reshaping */ - sector_t sb_start; /* offset of the super block (in 512byte sectors) */ + sector_t sb_start; /* offset of the super block (in 512byte sectors) */ int sb_size; /* bytes in the superblock */ int preferred_minor; /* autorun support */ @@ -239,7 +239,7 @@ struct mddev { minor_version, patch_version; int persistent; - int external; /* metadata is + int external; /* metadata is * managed externally */ char metadata_type[17]; /* externally set*/ int chunk_sectors; @@ -248,7 +248,7 @@ struct mddev { char clevel[16]; int raid_disks; int max_disks; - sector_t dev_sectors; /* used size of + sector_t dev_sectors; /* used size of * component devices */ sector_t array_sectors; /* exported array size */ int external_size; /* size managed @@ -312,7 +312,7 @@ struct mddev { int parallel_resync; int ok_start_degraded; - /* recovery/resync flags + /* recovery/resync flags * NEEDED: we might need to start a resync/recover * RUNNING: a thread is running, or about to be started * SYNC: actually doing a resync, not a recovery @@ -392,20 +392,20 @@ struct mddev { unsigned int safemode; /* if set, update "clean" superblock * when no writes pending. - */ + */ unsigned int safemode_delay; struct timer_list safemode_timer; - atomic_t writes_pending; + atomic_t writes_pending; struct request_queue *queue; /* for plugging ... */ - struct bitmap *bitmap; /* the bitmap for the device */ + struct bitmap *bitmap; /* the bitmap for the device */ struct { struct file *file; /* the bitmap file */ loff_t offset; /* offset from superblock of * start of bitmap. May be * negative, but not '0' * For external metadata, offset - * from start of device. + * from start of device. */ unsigned long space; /* space available at this offset */ loff_t default_offset; /* this is the offset to use when @@ -421,7 +421,7 @@ struct mddev { int external; } bitmap_info; - atomic_t max_corr_read_errors; /* max read retries */ + atomic_t max_corr_read_errors; /* max read retries */ struct list_head all_mddevs; struct attribute_group *to_remove; @@ -439,7 +439,6 @@ struct mddev { void (*sync_super)(struct mddev *mddev, struct md_rdev *rdev); }; - static inline void rdev_dec_pending(struct md_rdev *rdev, struct mddev *mddev) { int faulty = test_bit(Faulty, &rdev->flags); @@ -449,7 +448,7 @@ static inline void rdev_dec_pending(struct md_rdev *rdev, struct mddev *mddev) static inline void md_sync_acct(struct block_device *bdev, unsigned long nr_sectors) { - atomic_add(nr_sectors, &bdev->bd_contains->bd_disk->sync_io); + atomic_add(nr_sectors, &bdev->bd_contains->bd_disk->sync_io); } struct md_personality @@ -463,7 +462,7 @@ struct md_personality int (*stop)(struct mddev *mddev); void (*status)(struct seq_file *seq, struct mddev *mddev); /* error_handler must set ->faulty and clear ->in_sync - * if appropriate, and should abort recovery if needed + * if appropriate, and should abort recovery if needed */ void (*error_handler)(struct mddev *mddev, struct md_rdev *rdev); int (*hot_add_disk) (struct mddev *mddev, struct md_rdev *rdev); @@ -493,7 +492,6 @@ struct md_personality void *(*takeover) (struct mddev *mddev); }; - struct md_sysfs_entry { struct attribute attr; ssize_t (*show)(struct mddev *, char *); @@ -560,7 +558,7 @@ struct md_thread { void (*run) (struct md_thread *thread); struct mddev *mddev; wait_queue_head_t wqueue; - unsigned long flags; + unsigned long flags; struct task_struct *tsk; unsigned long timeout; void *private; @@ -594,7 +592,7 @@ extern void md_flush_request(struct mddev *mddev, struct bio *bio); extern void md_super_write(struct mddev *mddev, struct md_rdev *rdev, sector_t sector, int size, struct page *page); extern void md_super_wait(struct mddev *mddev); -extern int sync_page_io(struct md_rdev *rdev, sector_t sector, int size, +extern int sync_page_io(struct md_rdev *rdev, sector_t sector, int size, struct page *page, int rw, bool metadata_op); extern void md_do_sync(struct md_thread *thread); extern void md_new_event(struct mddev *mddev); diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c index 849ad39f547b..399272f9c042 100644 --- a/drivers/md/multipath.c +++ b/drivers/md/multipath.c @@ -31,13 +31,12 @@ #define NR_RESERVED_BUFS 32 - static int multipath_map (struct mpconf *conf) { int i, disks = conf->raid_disks; /* - * Later we do read balancing on the read side + * Later we do read balancing on the read side * now we use the first available disk. */ @@ -68,7 +67,6 @@ static void multipath_reschedule_retry (struct multipath_bh *mp_bh) md_wakeup_thread(mddev->thread); } - /* * multipath_end_bh_io() is called when we have finished servicing a multipathed * operation and are ready to return a success/failure code to the buffer @@ -98,8 +96,8 @@ static void multipath_end_request(struct bio *bio, int error) */ char b[BDEVNAME_SIZE]; md_error (mp_bh->mddev, rdev); - printk(KERN_ERR "multipath: %s: rescheduling sector %llu\n", - bdevname(rdev->bdev,b), + printk(KERN_ERR "multipath: %s: rescheduling sector %llu\n", + bdevname(rdev->bdev,b), (unsigned long long)bio->bi_iter.bi_sector); multipath_reschedule_retry(mp_bh); } else @@ -145,12 +143,12 @@ static void multipath_status (struct seq_file *seq, struct mddev *mddev) { struct mpconf *conf = mddev->private; int i; - + seq_printf (seq, " [%d/%d] [", conf->raid_disks, conf->raid_disks - mddev->degraded); for (i = 0; i < conf->raid_disks; i++) seq_printf (seq, "%s", - conf->multipaths[i].rdev && + conf->multipaths[i].rdev && test_bit(In_sync, &conf->multipaths[i].rdev->flags) ? "U" : "_"); seq_printf (seq, "]"); } @@ -195,7 +193,7 @@ static void multipath_error (struct mddev *mddev, struct md_rdev *rdev) * first check if this is a queued request for a device * which has just failed. */ - printk(KERN_ALERT + printk(KERN_ALERT "multipath: only one IO path left and IO error.\n"); /* leave it active... it's all we have */ return; @@ -242,7 +240,6 @@ static void print_multipath_conf (struct mpconf *conf) } } - static int multipath_add_disk(struct mddev *mddev, struct md_rdev *rdev) { struct mpconf *conf = mddev->private; @@ -325,8 +322,6 @@ abort: return err; } - - /* * This is a kernel thread which: * @@ -356,7 +351,7 @@ static void multipathd(struct md_thread *thread) bio = &mp_bh->bio; bio->bi_iter.bi_sector = mp_bh->master_bio->bi_iter.bi_sector; - + if ((mp_bh->path = multipath_map (conf))<0) { printk(KERN_ALERT "multipath: %s: unrecoverable IO read" " error for block %llu\n", @@ -414,7 +409,7 @@ static int multipath_run (struct mddev *mddev) conf = kzalloc(sizeof(struct mpconf), GFP_KERNEL); mddev->private = conf; if (!conf) { - printk(KERN_ERR + printk(KERN_ERR "multipath: couldn't allocate memory for %s\n", mdname(mddev)); goto out; @@ -423,7 +418,7 @@ static int multipath_run (struct mddev *mddev) conf->multipaths = kzalloc(sizeof(struct multipath_info)*mddev->raid_disks, GFP_KERNEL); if (!conf->multipaths) { - printk(KERN_ERR + printk(KERN_ERR "multipath: couldn't allocate memory for %s\n", mdname(mddev)); goto out_free_conf; @@ -469,7 +464,7 @@ static int multipath_run (struct mddev *mddev) conf->pool = mempool_create_kmalloc_pool(NR_RESERVED_BUFS, sizeof(struct multipath_bh)); if (conf->pool == NULL) { - printk(KERN_ERR + printk(KERN_ERR "multipath: couldn't allocate memory for %s\n", mdname(mddev)); goto out_free_conf; @@ -485,7 +480,7 @@ static int multipath_run (struct mddev *mddev) } } - printk(KERN_INFO + printk(KERN_INFO "multipath: array %s active with %d out of %d IO paths\n", mdname(mddev), conf->raid_disks - mddev->degraded, mddev->raid_disks); @@ -512,7 +507,6 @@ out: return -EIO; } - static int multipath_stop (struct mddev *mddev) { struct mpconf *conf = mddev->private; diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c index cf91f5910c7c..ba6b85de96d2 100644 --- a/drivers/md/raid0.c +++ b/drivers/md/raid0.c @@ -1,10 +1,9 @@ /* raid0.c : Multiple Devices driver for Linux - Copyright (C) 1994-96 Marc ZYNGIER + Copyright (C) 1994-96 Marc ZYNGIER <zyngier@ufr-info-p7.ibp.fr> or <maz@gloups.fdn.fr> - Copyright (C) 1999, 2000 Ingo Molnar, Red Hat - + Copyright (C) 1999, 2000 Ingo Molnar, Red Hat RAID-0 management functions. @@ -12,10 +11,10 @@ it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2, or (at your option) any later version. - + You should have received a copy of the GNU General Public License (for example /usr/src/linux/COPYING); if not, write to the Free - Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ #include <linux/blkdev.h> diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index 55de4f6f7eaf..40b35be34f8d 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -494,7 +494,6 @@ static void raid1_end_write_request(struct bio *bio, int error) bio_put(to_put); } - /* * This routine returns the disk from which the requested read should * be done. There is a per-array 'next expected sequential IO' sector @@ -901,18 +900,18 @@ static sector_t wait_barrier(struct r1conf *conf, struct bio *bio) * However if there are already pending * requests (preventing the barrier from * rising completely), and the - * pre-process bio queue isn't empty, + * per-process bio queue isn't empty, * then don't wait, as we need to empty - * that queue to get the nr_pending - * count down. + * that queue to allow conf->start_next_window + * to increase. */ wait_event_lock_irq(conf->wait_barrier, !conf->array_frozen && (!conf->barrier || - ((conf->start_next_window < - conf->next_resync + RESYNC_SECTORS) && - current->bio_list && - !bio_list_empty(current->bio_list))), + ((conf->start_next_window < + conf->next_resync + RESYNC_SECTORS) && + current->bio_list && + !bio_list_empty(current->bio_list))), conf->resync_lock); conf->nr_waiting--; } @@ -1001,8 +1000,7 @@ static void unfreeze_array(struct r1conf *conf) spin_unlock_irq(&conf->resync_lock); } - -/* duplicate the data pages for behind I/O +/* duplicate the data pages for behind I/O */ static void alloc_behind_pages(struct bio *bio, struct r1bio *r1_bio) { @@ -1471,7 +1469,6 @@ static void status(struct seq_file *seq, struct mddev *mddev) seq_printf(seq, "]"); } - static void error(struct mddev *mddev, struct md_rdev *rdev) { char b[BDEVNAME_SIZE]; @@ -1565,7 +1562,7 @@ static int raid1_spare_active(struct mddev *mddev) unsigned long flags; /* - * Find all failed disks within the RAID1 configuration + * Find all failed disks within the RAID1 configuration * and mark them readable. * Called under mddev lock, so rcu protection not needed. */ @@ -1606,7 +1603,6 @@ static int raid1_spare_active(struct mddev *mddev) return count; } - static int raid1_add_disk(struct mddev *mddev, struct md_rdev *rdev) { struct r1conf *conf = mddev->private; @@ -1735,7 +1731,6 @@ abort: return err; } - static void end_sync_read(struct bio *bio, int error) { struct r1bio *r1_bio = bio->bi_private; @@ -1947,7 +1942,7 @@ static int fix_sync_read_error(struct r1bio *r1_bio) return 1; } -static int process_checks(struct r1bio *r1_bio) +static void process_checks(struct r1bio *r1_bio) { /* We have read all readable devices. If we haven't * got the block, then there is no hope left. @@ -2039,7 +2034,6 @@ static int process_checks(struct r1bio *r1_bio) bio_copy_data(sbio, pbio); } - return 0; } static void sync_request_write(struct mddev *mddev, struct r1bio *r1_bio) @@ -2057,8 +2051,8 @@ static void sync_request_write(struct mddev *mddev, struct r1bio *r1_bio) return; if (test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery)) - if (process_checks(r1_bio) < 0) - return; + process_checks(r1_bio); + /* * schedule writes */ @@ -2458,7 +2452,6 @@ static void raid1d(struct md_thread *thread) blk_finish_plug(&plug); } - static int init_resync(struct r1conf *conf) { int buffs; @@ -2722,7 +2715,7 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, int *skipp /* remove last page from this bio */ bio->bi_vcnt--; bio->bi_iter.bi_size -= len; - bio->bi_flags &= ~(1<< BIO_SEG_VALID); + __clear_bit(BIO_SEG_VALID, &bio->bi_flags); } goto bio_full; } @@ -2947,9 +2940,9 @@ static int run(struct mddev *mddev) printk(KERN_NOTICE "md/raid1:%s: not clean" " -- starting background reconstruction\n", mdname(mddev)); - printk(KERN_INFO + printk(KERN_INFO "md/raid1:%s: active with %d out of %d mirrors\n", - mdname(mddev), mddev->raid_disks - mddev->degraded, + mdname(mddev), mddev->raid_disks - mddev->degraded, mddev->raid_disks); /* diff --git a/drivers/md/raid1.h b/drivers/md/raid1.h index 9bebca7bff2f..33bda55ef9f7 100644 --- a/drivers/md/raid1.h +++ b/drivers/md/raid1.h @@ -90,7 +90,6 @@ struct r1conf { */ int recovery_disabled; - /* poolinfo contains information about the content of the * mempools - it changes when the array grows or shrinks */ @@ -103,7 +102,6 @@ struct r1conf { */ struct page *tmppage; - /* When taking over an array from a different personality, we store * the new thread here until we fully activate the array. */ diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index 6703751d87d7..32e282f4c83c 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -366,7 +366,6 @@ static void raid10_end_read_request(struct bio *bio, int error) struct md_rdev *rdev; struct r10conf *conf = r10_bio->mddev->private; - slot = r10_bio->read_slot; dev = r10_bio->devs[slot].devnum; rdev = r10_bio->devs[slot].rdev; @@ -1559,7 +1558,6 @@ static void make_request(struct mddev *mddev, struct bio *bio) md_write_start(mddev, bio); - do { /* @@ -1782,7 +1780,6 @@ static int raid10_spare_active(struct mddev *mddev) return count; } - static int raid10_add_disk(struct mddev *mddev, struct md_rdev *rdev) { struct r10conf *conf = mddev->private; @@ -1929,7 +1926,6 @@ abort: return err; } - static void end_sync_read(struct bio *bio, int error) { struct r10bio *r10_bio = bio->bi_private; @@ -2295,7 +2291,6 @@ static void recovery_request_write(struct mddev *mddev, struct r10bio *r10_bio) } } - /* * Used by fix_read_error() to decay the per rdev read_errors. * We halve the read error count for every hour that has elapsed @@ -2852,7 +2847,6 @@ static void raid10d(struct md_thread *thread) blk_finish_plug(&plug); } - static int init_resync(struct r10conf *conf) { int buffs; @@ -3388,7 +3382,7 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, /* remove last page from this bio */ bio2->bi_vcnt--; bio2->bi_iter.bi_size -= len; - bio2->bi_flags &= ~(1<< BIO_SEG_VALID); + __clear_bit(BIO_SEG_VALID, &bio2->bi_flags); } goto bio_full; } @@ -3776,7 +3770,6 @@ static int run(struct mddev *mddev) blk_queue_merge_bvec(mddev->queue, raid10_mergeable_bvec); } - if (md_integrity_register(mddev)) goto out_free_conf; @@ -3834,6 +3827,8 @@ static int stop(struct mddev *mddev) mempool_destroy(conf->r10bio_pool); safe_put_page(conf->tmppage); kfree(conf->mirrors); + kfree(conf->mirrors_old); + kfree(conf->mirrors_new); kfree(conf); mddev->private = NULL; return 0; @@ -4121,7 +4116,7 @@ static int raid10_start_reshape(struct mddev *mddev) memcpy(conf->mirrors_new, conf->mirrors, sizeof(struct raid10_info)*conf->prev.raid_disks); smp_mb(); - kfree(conf->mirrors_old); /* FIXME and elsewhere */ + kfree(conf->mirrors_old); conf->mirrors_old = conf->mirrors; conf->mirrors = conf->mirrors_new; conf->mirrors_new = NULL; @@ -4416,7 +4411,7 @@ read_more: read_bio->bi_end_io = end_sync_read; read_bio->bi_rw = READ; read_bio->bi_flags &= (~0UL << BIO_RESET_BITS); - read_bio->bi_flags |= 1 << BIO_UPTODATE; + __set_bit(BIO_UPTODATE, &read_bio->bi_flags); read_bio->bi_vcnt = 0; read_bio->bi_iter.bi_size = 0; r10_bio->master_bio = read_bio; @@ -4473,7 +4468,7 @@ read_more: /* Remove last page from this bio */ bio2->bi_vcnt--; bio2->bi_iter.bi_size -= len; - bio2->bi_flags &= ~(1<<BIO_SEG_VALID); + __clear_bit(BIO_SEG_VALID, &bio2->bi_flags); } goto bio_full; } @@ -4575,7 +4570,6 @@ static void end_reshape(struct r10conf *conf) conf->fullsync = 0; } - static int handle_reshape_read_error(struct mddev *mddev, struct r10bio *r10_bio) { diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 9f0fbecd1eb5..9c66e5997fc8 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -463,7 +463,6 @@ static inline void insert_hash(struct r5conf *conf, struct stripe_head *sh) hlist_add_head(&sh->hash, hp); } - /* find an idle stripe, make sure it is unhashed, and return it. */ static struct stripe_head *get_free_stripe(struct r5conf *conf, int hash) { @@ -531,9 +530,7 @@ static void init_stripe(struct stripe_head *sh, sector_t sector, int previous) BUG_ON(stripe_operations_active(sh)); pr_debug("init_stripe called, stripe %llu\n", - (unsigned long long)sh->sector); - - remove_hash(sh); + (unsigned long long)sector); retry: seq = read_seqcount_begin(&conf->gen_lock); sh->generation = conf->generation - previous; @@ -542,7 +539,6 @@ retry: stripe_set_idx(sector, conf, previous, sh); sh->state = 0; - for (i = sh->disks; i--; ) { struct r5dev *dev = &sh->dev[i]; @@ -1350,7 +1346,6 @@ ops_run_compute6_2(struct stripe_head *sh, struct raid5_percpu *percpu) } } - static void ops_complete_prexor(void *stripe_head_ref) { struct stripe_head *sh = stripe_head_ref; @@ -2419,7 +2414,6 @@ static sector_t raid5_compute_sector(struct r5conf *conf, sector_t r_sector, return new_sector; } - static sector_t compute_blocknr(struct stripe_head *sh, int i, int previous) { struct r5conf *conf = sh->raid_conf; @@ -2437,7 +2431,6 @@ static sector_t compute_blocknr(struct stripe_head *sh, int i, int previous) sector_t r_sector; struct stripe_head sh2; - chunk_offset = sector_div(new_sector, sectors_per_chunk); stripe = new_sector; @@ -2541,7 +2534,6 @@ static sector_t compute_blocknr(struct stripe_head *sh, int i, int previous) return r_sector; } - static void schedule_reconstruction(struct stripe_head *sh, struct stripe_head_state *s, int rcw, int expand) @@ -3013,7 +3005,6 @@ static void handle_stripe_fill(struct stripe_head *sh, set_bit(STRIPE_HANDLE, &sh->state); } - /* handle_stripe_clean_event * any written block on an uptodate or failed drive can be returned. * Note that if we 'wrote' to a failed drive, it will be UPTODATE, but @@ -3304,7 +3295,6 @@ static void handle_parity_checks5(struct r5conf *conf, struct stripe_head *sh, } } - static void handle_parity_checks6(struct r5conf *conf, struct stripe_head *sh, struct stripe_head_state *s, int disks) @@ -3939,7 +3929,6 @@ static void handle_stripe(struct stripe_head *sh) } } - /* Finish reconstruct operations initiated by the expansion process */ if (sh->reconstruct_state == reconstruct_state_result) { struct stripe_head *sh_src @@ -4137,7 +4126,6 @@ static int raid5_mergeable_bvec(struct request_queue *q, return max; } - static int in_chunk_boundary(struct mddev *mddev, struct bio *bio) { sector_t sector = bio->bi_iter.bi_sector + get_start_sect(bio->bi_bdev); @@ -4167,7 +4155,6 @@ static void add_bio_to_retry(struct bio *bi,struct r5conf *conf) md_wakeup_thread(conf->mddev->thread); } - static struct bio *remove_bio_from_retry(struct r5conf *conf) { struct bio *bi; @@ -4191,7 +4178,6 @@ static struct bio *remove_bio_from_retry(struct r5conf *conf) return bi; } - /* * The "raid5_align_endio" should check if the read succeeded and if it * did, call bio_endio on the original bio (having bio_put the new bio @@ -4224,7 +4210,6 @@ static void raid5_align_endio(struct bio *bi, int error) return; } - pr_debug("raid5_align_endio : io error...handing IO for a retry\n"); add_bio_to_retry(raid_bi, conf); @@ -4249,7 +4234,6 @@ static int bio_fits_rdev(struct bio *bi) return 1; } - static int chunk_aligned_read(struct mddev *mddev, struct bio * raid_bio) { struct r5conf *conf = mddev->private; @@ -4301,7 +4285,7 @@ static int chunk_aligned_read(struct mddev *mddev, struct bio * raid_bio) rcu_read_unlock(); raid_bio->bi_next = (void*)rdev; align_bi->bi_bdev = rdev->bdev; - align_bi->bi_flags &= ~(1 << BIO_SEG_VALID); + __clear_bit(BIO_SEG_VALID, &align_bi->bi_flags); if (!bio_fits_rdev(align_bi) || is_badblock(rdev, align_bi->bi_iter.bi_sector, @@ -5446,7 +5430,6 @@ raid5_skip_copy = __ATTR(skip_copy, S_IRUGO | S_IWUSR, raid5_show_skip_copy, raid5_store_skip_copy); - static ssize_t stripe_cache_active_show(struct mddev *mddev, char *page) { @@ -5898,7 +5881,6 @@ static struct r5conf *setup_conf(struct mddev *mddev) return ERR_PTR(-ENOMEM); } - static int only_parity(int raid_disk, int algo, int raid_disks, int max_degraded) { switch (algo) { @@ -5911,7 +5893,7 @@ static int only_parity(int raid_disk, int algo, int raid_disks, int max_degraded return 1; break; case ALGORITHM_PARITY_0_6: - if (raid_disk == 0 || + if (raid_disk == 0 || raid_disk == raid_disks - 1) return 1; break; @@ -6165,7 +6147,6 @@ static int run(struct mddev *mddev) "reshape"); } - /* Ok, everything is just fine now */ if (mddev->to_remove == &raid5_attrs_group) mddev->to_remove = NULL; @@ -6814,7 +6795,6 @@ static void raid5_quiesce(struct mddev *mddev, int state) } } - static void *raid45_takeover_raid0(struct mddev *mddev, int level) { struct r0conf *raid0_conf = mddev->private; @@ -6841,7 +6821,6 @@ static void *raid45_takeover_raid0(struct mddev *mddev, int level) return setup_conf(mddev); } - static void *raid5_takeover_raid1(struct mddev *mddev) { int chunksect; @@ -6902,7 +6881,6 @@ static void *raid5_takeover_raid6(struct mddev *mddev) return setup_conf(mddev); } - static int raid5_check_reshape(struct mddev *mddev) { /* For a 2-drive array, the layout and chunk size can be changed @@ -7051,7 +7029,6 @@ static void *raid6_takeover(struct mddev *mddev) return setup_conf(mddev); } - static struct md_personality raid6_personality = { .name = "raid6", diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h index bc72cd4be5f8..d59f5ca743cd 100644 --- a/drivers/md/raid5.h +++ b/drivers/md/raid5.h @@ -155,7 +155,7 @@ */ /* - * Operations state - intermediate states that are visible outside of + * Operations state - intermediate states that are visible outside of * STRIPE_ACTIVE. * In general _idle indicates nothing is running, _run indicates a data * processing operation is active, and _result means the data processing result @@ -364,7 +364,6 @@ enum { * HANDLE gets cleared if stripe_handle leaves nothing locked. */ - struct disk_info { struct md_rdev *rdev, *replacement; }; @@ -528,7 +527,6 @@ struct r5conf { #define ALGORITHM_ROTATING_N_RESTART 9 /* DDF PRL=6 RLQ=2 */ #define ALGORITHM_ROTATING_N_CONTINUE 10 /*DDF PRL=6 RLQ=3 */ - /* For every RAID5 algorithm we define a RAID6 algorithm * with exactly the same layout for data and parity, and * with the Q block always on the last device (N-1). diff --git a/include/uapi/linux/raid/md_u.h b/include/uapi/linux/raid/md_u.h index 4133e744e4e6..74e7c60c4716 100644 --- a/include/uapi/linux/raid/md_u.h +++ b/include/uapi/linux/raid/md_u.h @@ -39,7 +39,6 @@ #define RAID_VERSION _IOR (MD_MAJOR, 0x10, mdu_version_t) #define GET_ARRAY_INFO _IOR (MD_MAJOR, 0x11, mdu_array_info_t) #define GET_DISK_INFO _IOR (MD_MAJOR, 0x12, mdu_disk_info_t) -#define PRINT_RAID_DEBUG _IO (MD_MAJOR, 0x13) #define RAID_AUTORUN _IO (MD_MAJOR, 0x14) #define GET_BITMAP_FILE _IOR (MD_MAJOR, 0x15, mdu_bitmap_file_t) diff --git a/lib/raid6/algos.c b/lib/raid6/algos.c index f0b1aa3586d1..7d0e5cd7b570 100644 --- a/lib/raid6/algos.c +++ b/lib/raid6/algos.c @@ -121,9 +121,9 @@ static inline const struct raid6_recov_calls *raid6_choose_recov(void) raid6_2data_recov = best->data2; raid6_datap_recov = best->datap; - printk("raid6: using %s recovery algorithm\n", best->name); + pr_info("raid6: using %s recovery algorithm\n", best->name); } else - printk("raid6: Yikes! No recovery algorithm found!\n"); + pr_err("raid6: Yikes! No recovery algorithm found!\n"); return best; } @@ -157,18 +157,18 @@ static inline const struct raid6_calls *raid6_choose_gen( bestperf = perf; best = *algo; } - printk("raid6: %-8s %5ld MB/s\n", (*algo)->name, + pr_info("raid6: %-8s %5ld MB/s\n", (*algo)->name, (perf*HZ) >> (20-16+RAID6_TIME_JIFFIES_LG2)); } } if (best) { - printk("raid6: using algorithm %s (%ld MB/s)\n", + pr_info("raid6: using algorithm %s (%ld MB/s)\n", best->name, (bestperf*HZ) >> (20-16+RAID6_TIME_JIFFIES_LG2)); raid6_call = *best; } else - printk("raid6: Yikes! No algorithm found!\n"); + pr_err("raid6: Yikes! No algorithm found!\n"); return best; } @@ -194,7 +194,7 @@ int __init raid6_select_algo(void) syndromes = (void *) __get_free_pages(GFP_KERNEL, 1); if (!syndromes) { - printk("raid6: Yikes! No memory available.\n"); + pr_err("raid6: Yikes! No memory available.\n"); return -ENOMEM; } |