From 6283815d1853b7daf31dc4adb83e5c1dc9568251 Mon Sep 17 00:00:00 2001 From: Andre Noll Date: Mon, 13 Oct 2008 11:55:12 +1100 Subject: md: linear: Represent dev_info->size and dev_info->offset in sectors. Rename them to num_sectors and start_sector which is more descriptive. Signed-off-by: Andre Noll Signed-off-by: NeilBrown --- include/linux/raid/linear.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux/raid') diff --git a/include/linux/raid/linear.h b/include/linux/raid/linear.h index 7e375111d007..87090e98529f 100644 --- a/include/linux/raid/linear.h +++ b/include/linux/raid/linear.h @@ -5,8 +5,8 @@ struct dev_info { mdk_rdev_t *rdev; - sector_t size; - sector_t offset; + sector_t num_sectors; + sector_t start_sector; }; typedef struct dev_info dev_info_t; -- cgit v1.2.3 From ab5bd5cbc8d4b868378d062eed3d4240930fbb86 Mon Sep 17 00:00:00 2001 From: Andre Noll Date: Mon, 13 Oct 2008 11:55:12 +1100 Subject: md: Convert remaining 1k representations in linear.c to sectors. This patch renames hash_spacing and preshift to spacing and sector_shift respectively with the following change of semantics: Case 1: (sizeof(sector_t) <= sizeof(u32)). ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In this case, we have sector_shift = preshift = 0 and spacing = 2 * hash_spacing. Hence, the index for the hash table which is computed by the new code in which_dev() as sector / spacing equals the old value which was (sector/2) / hash_spacing. Note also that the value of nb_zone stays the same because both sz and base double. Case 2: (sizeof(sector_t) > sizeof(u32)). ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ (aka the shifting dance case). Here we have sector_shift = preshift + 1 and spacing = 2 * hash_spacing during the computation of nb_zone and curr_sector, but spacing = hash_spacing in which_dev() because in the last hunk of the patch for linear.c we shift down conf->spacing (= 2 * hash_spacing) by one more bit than in the old code. Hence in the computation of nb_zone, sz and base have the same value as before, so nb_zone is not affected. Also curr_sector in the next hunk stays the same. In which_dev() the hash table index is computed as (sector >> sector_shift) / spacing In view of sector_shift = preshift + 1 and spacing = hash_spacing, this equals ((sector/2) >> preshift) / hash_spacing which is the value computed by the old code. Signed-off-by: Andre Noll Signed-off-by: NeilBrown --- drivers/md/linear.c | 35 +++++++++++++++++------------------ include/linux/raid/linear.h | 6 ++++-- 2 files changed, 21 insertions(+), 20 deletions(-) (limited to 'include/linux/raid') diff --git a/drivers/md/linear.c b/drivers/md/linear.c index 13e928bde7cd..09a64b9cbde8 100644 --- a/drivers/md/linear.c +++ b/drivers/md/linear.c @@ -33,14 +33,13 @@ 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 >= hash->num_sectors + hash->start_sector) hash++; @@ -164,25 +163,25 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks) * 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 tmp = 0; int j; for (j = i; j < cnt - 1 && tmp < min_sectors; j++) tmp += conf->disks[j].num_sectors; - if (tmp >= min_sectors && tmp < conf->hash_spacing * 2) - conf->hash_spacing = tmp / 2; + 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 +193,9 @@ 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); } @@ -221,7 +220,7 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks) i = 0; for (curr_sector = 0; curr_sector < conf->array_sectors; - curr_sector += conf->hash_spacing * 2) { + curr_sector += conf->spacing) { while (i < raid_disks-1 && curr_sector >= conf->disks[i+1].start_sector) @@ -230,12 +229,12 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks) *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); diff --git a/include/linux/raid/linear.h b/include/linux/raid/linear.h index 87090e98529f..f38b9c586afb 100644 --- a/include/linux/raid/linear.h +++ b/include/linux/raid/linear.h @@ -15,9 +15,11 @@ struct linear_private_data { struct linear_private_data *prev; /* earlier version */ dev_info_t **hash_table; - sector_t hash_spacing; + sector_t spacing; sector_t array_sectors; - int preshift; /* shift before dividing by hash_spacing */ + int sector_shift; /* shift before dividing + * by spacing + */ dev_info_t disks[0]; }; -- cgit v1.2.3 From fb4d8c76e56a887b9eee99fbc55fe82b18625d30 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 13 Oct 2008 11:55:12 +1100 Subject: md: Remove unnecessary #includes, #defines, and function declarations. A lot of cruft has gathered over the years. Time to remove it. Signed-off-by: NeilBrown --- drivers/md/linear.c | 8 -------- drivers/md/md.c | 21 ++++----------------- drivers/md/multipath.c | 9 --------- drivers/md/raid0.c | 5 ----- drivers/md/raid5.c | 5 ----- drivers/md/raid6.h | 9 --------- include/linux/raid/md.h | 22 ---------------------- 7 files changed, 4 insertions(+), 75 deletions(-) (limited to 'include/linux/raid') diff --git a/drivers/md/linear.c b/drivers/md/linear.c index 09a64b9cbde8..190147c79e79 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 - -#include -#include #include -#define MAJOR_NR MD_MAJOR -#define MD_DRIVER -#define MD_PERSONALITY - /* * find which device holds a particular offset */ diff --git a/drivers/md/md.c b/drivers/md/md.c index fc33082d5ffd..cd97de5982e8 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -32,31 +32,20 @@ Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ -#include -#include #include -#include #include #include #include #include /* for invalidate_bdev */ #include -#include #include -#include - -#include - +#include +#include +#include +#include #include -#ifdef CONFIG_KMOD -#include -#endif - -#include - #define MAJOR_NR MD_MAJOR -#define MD_DRIVER /* 63 partitions with the alternate major number (mdp) */ #define MdpMinorShift 6 @@ -3559,12 +3548,10 @@ 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 diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c index 8bb8794129b3..8744014b9d80 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 -#include -#include #include -#include -#include - -#define MAJOR_NR MD_MAJOR -#define MD_DRIVER -#define MD_PERSONALITY #define MAX_WORK_PER_DISK 128 diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c index 53508a8a981d..8ac6488ad0dc 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 #include -#define MAJOR_NR MD_MAJOR -#define MD_DRIVER -#define MD_PERSONALITY - static void raid0_unplug(struct request_queue *q) { mddev_t *mddev = q->queuedata; diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index ae16794bef20..7e654543c97d 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -43,12 +43,7 @@ * miss any bits. */ -#include -#include -#include -#include #include -#include #include "raid6.h" #include diff --git a/drivers/md/raid6.h b/drivers/md/raid6.h index 31cbee71365f..98dcde88470e 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 -#include -#include -#include -#include -#include -#include -#include -#include #include #include diff --git a/include/linux/raid/md.h b/include/linux/raid/md.h index dc0e3fcb9f28..bb727fa1ce7f 100644 --- a/include/linux/raid/md.h +++ b/include/linux/raid/md.h @@ -19,27 +19,7 @@ #define _MD_H #include -#include -#include -#include -#include -#include -#include -#include #include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include /* * 'md_p.h' holds the 'physical' layout of RAID devices @@ -83,10 +63,8 @@ extern void md_wakeup_thread(mdk_thread_t *thread); extern void md_check_recovery(mddev_t *mddev); extern void md_write_start(mddev_t *mddev, struct bio *bi); extern void md_write_end(mddev_t *mddev); -extern void md_handle_safemode(mddev_t *mddev); extern void md_done_sync(mddev_t *mddev, int blocks, int ok); extern void md_error (mddev_t *mddev, mdk_rdev_t *rdev); -extern void md_unplug_mddev(mddev_t *mddev); extern void md_super_write(mddev_t *mddev, mdk_rdev_t *rdev, sector_t sector, int size, struct page *page); -- cgit v1.2.3 From d710e13812600037a723a673dc5c96a071de98d3 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 13 Oct 2008 11:55:12 +1100 Subject: md: remove space after function name in declaration and call. Having function (args) instead of function(args) make is harder to search for calls of particular functions. So remove all those spaces. Signed-off-by: NeilBrown --- drivers/md/md.c | 26 +++++++++++++------------- drivers/md/raid5.c | 30 +++++++++++++++--------------- include/linux/raid/md.h | 10 +++++----- 3 files changed, 33 insertions(+), 33 deletions(-) (limited to 'include/linux/raid') diff --git a/drivers/md/md.c b/drivers/md/md.c index cd97de5982e8..a29187d1fcf5 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -55,7 +55,7 @@ #ifndef MODULE -static void autostart_arrays (int part); +static void autostart_arrays(int part); #endif static LIST_HEAD(pers_list); @@ -201,7 +201,7 @@ 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; @@ -3962,10 +3962,10 @@ 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); } } @@ -4324,7 +4324,7 @@ static int add_new_disk(mddev_t * mddev, mdu_disk_info_t *info) if (!(info->state & (1<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 "); @@ -6251,7 +6251,7 @@ 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); } /* @@ -6295,7 +6295,7 @@ 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/raid5.c b/drivers/md/raid5.c index 7e654543c97d..d72be4b89e64 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -270,7 +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) { @@ -1146,7 +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; @@ -1178,7 +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]; @@ -1216,10 +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); } } @@ -1315,8 +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; } @@ -1391,8 +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; } @@ -1400,7 +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; @@ -4284,7 +4284,7 @@ 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; @@ -4300,7 +4300,7 @@ 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; @@ -4318,7 +4318,7 @@ 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/include/linux/raid/md.h b/include/linux/raid/md.h index bb727fa1ce7f..82bea14cae1a 100644 --- a/include/linux/raid/md.h +++ b/include/linux/raid/md.h @@ -54,17 +54,17 @@ extern int mdp_major; -extern int register_md_personality (struct mdk_personality *p); -extern int unregister_md_personality (struct mdk_personality *p); -extern mdk_thread_t * md_register_thread (void (*run) (mddev_t *mddev), +extern int register_md_personality(struct mdk_personality *p); +extern int unregister_md_personality(struct mdk_personality *p); +extern mdk_thread_t * md_register_thread(void (*run) (mddev_t *mddev), mddev_t *mddev, const char *name); -extern void md_unregister_thread (mdk_thread_t *thread); +extern void md_unregister_thread(mdk_thread_t *thread); extern void md_wakeup_thread(mdk_thread_t *thread); extern void md_check_recovery(mddev_t *mddev); extern void md_write_start(mddev_t *mddev, struct bio *bi); extern void md_write_end(mddev_t *mddev); extern void md_done_sync(mddev_t *mddev, int blocks, int ok); -extern void md_error (mddev_t *mddev, mdk_rdev_t *rdev); +extern void md_error(mddev_t *mddev, mdk_rdev_t *rdev); extern void md_super_write(mddev_t *mddev, mdk_rdev_t *rdev, sector_t sector, int size, struct page *page); -- cgit v1.2.3 From b62b75905d571c29262a6c38cf9e5f089c203871 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 21 Oct 2008 13:25:21 +1100 Subject: md: use sysfs_notify_dirent to notify changes to md/array_state Now that we have sysfs_notify_dirent, use it to notify changes to md/array_state. As sysfs_notify_dirent can be called in atomic context, we can remove the delayed notify and the MD_NOTIFY_ARRAY_STATE flag. Signed-off-by: NeilBrown --- drivers/md/md.c | 30 ++++++++++++++++-------------- include/linux/raid/md_k.h | 5 ++++- 2 files changed, 20 insertions(+), 15 deletions(-) (limited to 'include/linux/raid') diff --git a/drivers/md/md.c b/drivers/md/md.c index aaa3d465de4e..feea72dc4b69 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -222,6 +222,9 @@ static void mddev_put(mddev_t *mddev) list_del(&mddev->all_mddevs); spin_unlock(&all_mddevs_lock); blk_cleanup_queue(mddev->queue); + if (mddev->sysfs_state) + sysfs_put(mddev->sysfs_state); + mddev->sysfs_state = NULL; kobject_put(&mddev->kobj); } else spin_unlock(&all_mddevs_lock); @@ -2770,7 +2773,7 @@ array_state_store(mddev_t *mddev, const char *buf, size_t len) if (err) return err; else { - sysfs_notify(&mddev->kobj, NULL, "array_state"); + sysfs_notify_dirent(mddev->sysfs_state); return len; } } @@ -3465,8 +3468,10 @@ static struct kobject *md_probe(dev_t dev, int *part, void *data) if (error) printk(KERN_WARNING "md: cannot register %s/md - name in use\n", disk->disk_name); - else + else { kobject_uevent(&mddev->kobj, KOBJ_ADD); + mddev->sysfs_state = sysfs_get_dirent(mddev->kobj.sd, "array_state"); + } return NULL; } @@ -3477,7 +3482,7 @@ static void md_safemode_timeout(unsigned long data) if (!atomic_read(&mddev->writes_pending)) { mddev->safemode = 1; if (mddev->external) - set_bit(MD_NOTIFY_ARRAY_STATE, &mddev->flags); + sysfs_notify_dirent(mddev->sysfs_state); } md_wakeup_thread(mddev->thread); } @@ -3740,7 +3745,7 @@ static int do_md_run(mddev_t * mddev) mddev->changed = 1; md_new_event(mddev); - sysfs_notify(&mddev->kobj, NULL, "array_state"); + sysfs_notify_dirent(mddev->sysfs_state); sysfs_notify(&mddev->kobj, NULL, "sync_action"); sysfs_notify(&mddev->kobj, NULL, "degraded"); kobject_uevent(&disk_to_dev(mddev->gendisk)->kobj, KOBJ_CHANGE); @@ -3767,7 +3772,7 @@ static int restart_array(mddev_t *mddev) set_bit(MD_RECOVERY_NEEDED, &mddev->recovery); md_wakeup_thread(mddev->thread); md_wakeup_thread(mddev->sync_thread); - sysfs_notify(&mddev->kobj, NULL, "array_state"); + sysfs_notify_dirent(mddev->sysfs_state); return 0; } @@ -3847,7 +3852,7 @@ static int do_md_stop(mddev_t * mddev, int mode, int is_open) module_put(mddev->pers->owner); mddev->pers = NULL; /* tell userspace to handle 'inactive' */ - sysfs_notify(&mddev->kobj, NULL, "array_state"); + sysfs_notify_dirent(mddev->sysfs_state); set_capacity(disk, 0); mddev->changed = 1; @@ -3933,7 +3938,7 @@ static int do_md_stop(mddev_t * mddev, int mode, int is_open) mdname(mddev)); err = 0; md_new_event(mddev); - sysfs_notify(&mddev->kobj, NULL, "array_state"); + sysfs_notify_dirent(mddev->sysfs_state); out: return err; } @@ -4938,7 +4943,7 @@ static int md_ioctl(struct inode *inode, struct file *file, if (_IOC_TYPE(cmd) == MD_MAJOR && mddev->ro && mddev->pers) { if (mddev->ro == 2) { mddev->ro = 0; - sysfs_notify(&mddev->kobj, NULL, "array_state"); + sysfs_notify_dirent(mddev->sysfs_state); set_bit(MD_RECOVERY_NEEDED, &mddev->recovery); md_wakeup_thread(mddev->thread); } else { @@ -5612,7 +5617,7 @@ void md_write_start(mddev_t *mddev, struct bio *bi) spin_unlock_irq(&mddev->write_lock); } if (did_change) - sysfs_notify(&mddev->kobj, NULL, "array_state"); + sysfs_notify_dirent(mddev->sysfs_state); wait_event(mddev->sb_wait, !test_bit(MD_CHANGE_CLEAN, &mddev->flags) && !test_bit(MD_CHANGE_PENDING, &mddev->flags)); @@ -5655,7 +5660,7 @@ int md_allow_write(mddev_t *mddev) mddev->safemode = 1; spin_unlock_irq(&mddev->write_lock); md_update_sb(mddev, 0); - sysfs_notify(&mddev->kobj, NULL, "array_state"); + sysfs_notify_dirent(mddev->sysfs_state); } else spin_unlock_irq(&mddev->write_lock); @@ -6048,9 +6053,6 @@ void md_check_recovery(mddev_t *mddev) if (mddev->bitmap) bitmap_daemon_work(mddev->bitmap); - if (test_and_clear_bit(MD_NOTIFY_ARRAY_STATE, &mddev->flags)) - sysfs_notify(&mddev->kobj, NULL, "array_state"); - if (mddev->ro) return; @@ -6103,7 +6105,7 @@ void md_check_recovery(mddev_t *mddev) mddev->safemode = 0; spin_unlock_irq(&mddev->write_lock); if (did_change) - sysfs_notify(&mddev->kobj, NULL, "array_state"); + sysfs_notify_dirent(mddev->sysfs_state); } if (mddev->flags) diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h index c200b9a34aff..b16ad867e944 100644 --- a/include/linux/raid/md_k.h +++ b/include/linux/raid/md_k.h @@ -128,7 +128,6 @@ struct mddev_s #define MD_CHANGE_DEVS 0 /* Some device status has changed */ #define MD_CHANGE_CLEAN 1 /* transition to or from 'clean' */ #define MD_CHANGE_PENDING 2 /* superblock update in progress */ -#define MD_NOTIFY_ARRAY_STATE 3 /* atomic context wants to notify userspace */ int ro; @@ -239,6 +238,10 @@ struct mddev_s sector_t resync_max; /* resync should pause * when it gets here */ + struct sysfs_dirent *sysfs_state; /* handle for 'array_state' + * file in sysfs. + */ + spinlock_t write_lock; wait_queue_head_t sb_wait; /* for waiting on superblock updates */ atomic_t pending_writes; /* number of active superblock writes */ -- cgit v1.2.3 From 3c0ee63a64a20351ed6c16ec797e1f8c850741ea Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 21 Oct 2008 13:25:28 +1100 Subject: md: use sysfs_notify_dirent to notify changes to md/dev-xxx/state The 'state' file for a device reports, for example, when the device has failed. Changes should be reported to userspace ASAP without the possibility of blocking on low-memory. sysfs_notify does have that possibility (as it takes a mutex which can be held across a kmalloc) so use sysfs_notify_dirent instead. Signed-off-by: NeilBrown --- drivers/md/md.c | 21 ++++++++++++--------- include/linux/raid/md_k.h | 3 +++ 2 files changed, 15 insertions(+), 9 deletions(-) (limited to 'include/linux/raid') diff --git a/drivers/md/md.c b/drivers/md/md.c index feea72dc4b69..8b303477c77b 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -1462,6 +1462,8 @@ static int bind_rdev_to_array(mdk_rdev_t * rdev, mddev_t * mddev) kobject_del(&rdev->kobj); goto fail; } + rdev->sysfs_state = sysfs_get_dirent(rdev->kobj.sd, "state"); + list_add_rcu(&rdev->same_set, &mddev->disks); bd_claim_by_disk(rdev->bdev, rdev->bdev->bd_holder, mddev->gendisk); return 0; @@ -1491,7 +1493,8 @@ static void unbind_rdev_from_array(mdk_rdev_t * rdev) printk(KERN_INFO "md: unbind<%s>\n", bdevname(rdev->bdev,b)); rdev->mddev = NULL; sysfs_remove_link(&rdev->kobj, "block"); - + sysfs_put(rdev->sysfs_state); + rdev->sysfs_state = NULL; /* We need to delay this, otherwise we can deadlock when * writing to 'remove' to "dev/state". We also need * to delay it due to rcu usage. @@ -1926,8 +1929,8 @@ state_store(mdk_rdev_t *rdev, const char *buf, size_t len) err = 0; } - if (!err) - sysfs_notify(&rdev->kobj, NULL, "state"); + if (!err && rdev->sysfs_state) + sysfs_notify_dirent(rdev->sysfs_state); return err ? err : len; } static struct rdev_sysfs_entry rdev_state = @@ -2022,7 +2025,7 @@ slot_store(mdk_rdev_t *rdev, const char *buf, size_t len) rdev->raid_disk = -1; return err; } else - sysfs_notify(&rdev->kobj, NULL, "state"); + sysfs_notify_dirent(rdev->sysfs_state); sprintf(nm, "rd%d", rdev->raid_disk); if (sysfs_create_link(&rdev->mddev->kobj, &rdev->kobj, nm)) printk(KERN_WARNING @@ -2039,7 +2042,7 @@ slot_store(mdk_rdev_t *rdev, const char *buf, size_t len) clear_bit(Faulty, &rdev->flags); clear_bit(WriteMostly, &rdev->flags); set_bit(In_sync, &rdev->flags); - sysfs_notify(&rdev->kobj, NULL, "state"); + sysfs_notify_dirent(rdev->sysfs_state); } return len; } @@ -3583,7 +3586,7 @@ static int do_md_run(mddev_t * mddev) return -EINVAL; } } - sysfs_notify(&rdev->kobj, NULL, "state"); + sysfs_notify_dirent(rdev->sysfs_state); } md_probe(mddev->unit, NULL, NULL); @@ -4302,7 +4305,7 @@ static int add_new_disk(mddev_t * mddev, mdu_disk_info_t *info) if (err) export_rdev(rdev); else - sysfs_notify(&rdev->kobj, NULL, "state"); + sysfs_notify_dirent(rdev->sysfs_state); md_update_sb(mddev, 1); if (mddev->degraded) @@ -6113,7 +6116,7 @@ void md_check_recovery(mddev_t *mddev) rdev_for_each(rdev, rtmp, mddev) if (test_and_clear_bit(StateChanged, &rdev->flags)) - sysfs_notify(&rdev->kobj, NULL, "state"); + sysfs_notify_dirent(rdev->sysfs_state); if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) && @@ -6223,7 +6226,7 @@ void md_check_recovery(mddev_t *mddev) void md_wait_for_blocked_rdev(mdk_rdev_t *rdev, mddev_t *mddev) { - sysfs_notify(&rdev->kobj, NULL, "state"); + sysfs_notify_dirent(rdev->sysfs_state); wait_event_timeout(rdev->blocked_wait, !test_bit(Blocked, &rdev->flags), msecs_to_jiffies(5000)); diff --git a/include/linux/raid/md_k.h b/include/linux/raid/md_k.h index b16ad867e944..8fc909ef6787 100644 --- a/include/linux/raid/md_k.h +++ b/include/linux/raid/md_k.h @@ -115,6 +115,9 @@ struct mdk_rdev_s * in superblock. */ struct work_struct del_work; /* used for delayed sysfs removal */ + + struct sysfs_dirent *sysfs_state; /* handle for 'state' + * sysfs entry */ }; struct mddev_s -- cgit v1.2.3