From de01dfadf25bf83cfe3d85c163005c4320532658 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 6 Feb 2009 18:02:46 +1100 Subject: md: Ensure an md array never has too many devices. Each different metadata format supported by md supports a different maximum number of devices. We really should be enforcing this maximum in the kernel, but we aren't quite doing that properly. We currently only enforce it at the 'hot_add' point, which is an older interface which is not used by current userspace. We need to also enforce it at 'add_new_disk' time for active arrays and at 'do_md_run' time when starting a new array. So move the test from 'hot_add' into 'bind_rdev_to_array' which is called from both 'hot_add' and 'add_new_disk, and add a new test in 'analyse_sbs' which is called from 'do_md_run'. This bug (or missing feature) has been around "forever" and so the patch is suitable for any -stable that is currently maintained. Cc: stable@kernel.org Signed-off-by: NeilBrown --- drivers/md/md.c | 24 ++++++++++++++---------- 1 file changed, 14 insertions(+), 10 deletions(-) (limited to 'drivers/md/md.c') diff --git a/drivers/md/md.c b/drivers/md/md.c index 41e2509bf89..4495104f6c9 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -1481,6 +1481,11 @@ static int bind_rdev_to_array(mdk_rdev_t * rdev, mddev_t * mddev) if (find_rdev_nr(mddev, rdev->desc_nr)) return -EBUSY; } + 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); + return -EBUSY; + } bdevname(rdev->bdev,b); while ( (s=strchr(b, '/')) != NULL) *s = '!'; @@ -2441,6 +2446,15 @@ static void analyze_sbs(mddev_t * mddev) i = 0; rdev_for_each(rdev, tmp, mddev) { + if (rdev->desc_nr >= mddev->max_disks || + i > mddev->max_disks) { + printk(KERN_WARNING + "md: %s: %s: only %d devices permitted\n", + mdname(mddev), bdevname(rdev->bdev, b), + mddev->max_disks); + kick_rdev_from_array(rdev); + continue; + } if (rdev != freshest) if (super_types[mddev->major_version]. validate_super(mddev, rdev)) { @@ -4614,13 +4628,6 @@ static int hot_add_disk(mddev_t * mddev, dev_t dev) * noticed in interrupt contexts ... */ - if (rdev->desc_nr == mddev->max_disks) { - printk(KERN_WARNING "%s: can not hot-add to full array!\n", - mdname(mddev)); - err = -EBUSY; - goto abort_unbind_export; - } - rdev->raid_disk = -1; md_update_sb(mddev, 1); @@ -4634,9 +4641,6 @@ static int hot_add_disk(mddev_t * mddev, dev_t dev) md_new_event(mddev); return 0; -abort_unbind_export: - unbind_rdev_from_array(rdev); - abort_export: export_rdev(rdev); return err; -- cgit v1.2.3-70-g09d2 From 93dbb393503d53cd226e5e1f0088fe8f4dbaa2b8 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 16 Feb 2009 10:25:40 +0100 Subject: block: fix bad definition of BIO_RW_SYNC We can't OR shift values, so get rid of BIO_RW_SYNC and use BIO_RW_SYNCIO and BIO_RW_UNPLUG explicitly. This brings back the behaviour from before 213d9417fec62ef4c3675621b9364a667954d4dd. Signed-off-by: Jens Axboe --- block/blktrace.c | 2 +- drivers/md/dm-io.c | 2 +- drivers/md/dm-kcopyd.c | 2 +- drivers/md/md.c | 4 ++-- include/linux/bio.h | 2 -- include/linux/blktrace_api.h | 1 + include/linux/fs.h | 6 +++--- kernel/power/swap.c | 5 +++-- mm/page_io.c | 2 +- 9 files changed, 13 insertions(+), 13 deletions(-) (limited to 'drivers/md/md.c') diff --git a/block/blktrace.c b/block/blktrace.c index 39cc3bfe56e..7cf9d1ff45a 100644 --- a/block/blktrace.c +++ b/block/blktrace.c @@ -142,7 +142,7 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, what |= ddir_act[rw & WRITE]; what |= MASK_TC_BIT(rw, BARRIER); - what |= MASK_TC_BIT(rw, SYNC); + what |= MASK_TC_BIT(rw, SYNCIO); what |= MASK_TC_BIT(rw, AHEAD); what |= MASK_TC_BIT(rw, META); what |= MASK_TC_BIT(rw, DISCARD); diff --git a/drivers/md/dm-io.c b/drivers/md/dm-io.c index a34338567a2..f14813be4ef 100644 --- a/drivers/md/dm-io.c +++ b/drivers/md/dm-io.c @@ -328,7 +328,7 @@ static void dispatch_io(int rw, unsigned int num_regions, struct dpages old_pages = *dp; if (sync) - rw |= (1 << BIO_RW_SYNC); + rw |= (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG); /* * For multiple regions we need to be careful to rewind diff --git a/drivers/md/dm-kcopyd.c b/drivers/md/dm-kcopyd.c index 3073618269e..0a225da2127 100644 --- a/drivers/md/dm-kcopyd.c +++ b/drivers/md/dm-kcopyd.c @@ -344,7 +344,7 @@ static int run_io_job(struct kcopyd_job *job) { int r; struct dm_io_request io_req = { - .bi_rw = job->rw | (1 << BIO_RW_SYNC), + .bi_rw = job->rw | (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG), .mem.type = DM_IO_PAGE_LIST, .mem.ptr.pl = job->pages, .mem.offset = job->offset, diff --git a/drivers/md/md.c b/drivers/md/md.c index 4495104f6c9..03b4cd0a634 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -474,7 +474,7 @@ void md_super_write(mddev_t *mddev, mdk_rdev_t *rdev, * causes ENOTSUPP, we allocate a spare bio... */ struct bio *bio = bio_alloc(GFP_NOIO, 1); - int rw = (1<bi_bdev = rdev->bdev; bio->bi_sector = sector; @@ -531,7 +531,7 @@ int sync_page_io(struct block_device *bdev, sector_t sector, int size, struct completion event; int ret; - rw |= (1 << BIO_RW_SYNC); + rw |= (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG); bio->bi_bdev = bdev; bio->bi_sector = sector; diff --git a/include/linux/bio.h b/include/linux/bio.h index 2aa283ab062..1b16108a541 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -171,8 +171,6 @@ struct bio { #define BIO_RW_FAILFAST_TRANSPORT 8 #define BIO_RW_FAILFAST_DRIVER 9 -#define BIO_RW_SYNC (BIO_RW_SYNCIO | BIO_RW_UNPLUG) - #define bio_rw_flagged(bio, flag) ((bio)->bi_rw & (1 << (flag))) /* diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h index 25379cba237..6e915878e88 100644 --- a/include/linux/blktrace_api.h +++ b/include/linux/blktrace_api.h @@ -15,6 +15,7 @@ enum blktrace_cat { BLK_TC_WRITE = 1 << 1, /* writes */ BLK_TC_BARRIER = 1 << 2, /* barrier */ BLK_TC_SYNC = 1 << 3, /* sync IO */ + BLK_TC_SYNCIO = BLK_TC_SYNC, BLK_TC_QUEUE = 1 << 4, /* queueing/merging */ BLK_TC_REQUEUE = 1 << 5, /* requeueing */ BLK_TC_ISSUE = 1 << 6, /* issue */ diff --git a/include/linux/fs.h b/include/linux/fs.h index 6022f44043f..67857dc1636 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -87,10 +87,10 @@ struct inodes_stat_t { #define WRITE 1 #define READA 2 /* read-ahead - don't block if no resources */ #define SWRITE 3 /* for ll_rw_block() - wait for buffer lock */ -#define READ_SYNC (READ | (1 << BIO_RW_SYNC)) +#define READ_SYNC (READ | (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG)) #define READ_META (READ | (1 << BIO_RW_META)) -#define WRITE_SYNC (WRITE | (1 << BIO_RW_SYNC)) -#define SWRITE_SYNC (SWRITE | (1 << BIO_RW_SYNC)) +#define WRITE_SYNC (WRITE | (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG)) +#define SWRITE_SYNC (SWRITE | (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG)) #define WRITE_BARRIER (WRITE | (1 << BIO_RW_BARRIER)) #define DISCARD_NOBARRIER (1 << BIO_RW_DISCARD) #define DISCARD_BARRIER ((1 << BIO_RW_DISCARD) | (1 << BIO_RW_BARRIER)) diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 6da14358537..505f319e489 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -60,6 +60,7 @@ static struct block_device *resume_bdev; static int submit(int rw, pgoff_t page_off, struct page *page, struct bio **bio_chain) { + const int bio_rw = rw | (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG); struct bio *bio; bio = bio_alloc(__GFP_WAIT | __GFP_HIGH, 1); @@ -80,7 +81,7 @@ static int submit(int rw, pgoff_t page_off, struct page *page, bio_get(bio); if (bio_chain == NULL) { - submit_bio(rw | (1 << BIO_RW_SYNC), bio); + submit_bio(bio_rw, bio); wait_on_page_locked(page); if (rw == READ) bio_set_pages_dirty(bio); @@ -90,7 +91,7 @@ static int submit(int rw, pgoff_t page_off, struct page *page, get_page(page); /* These pages are freed later */ bio->bi_private = *bio_chain; *bio_chain = bio; - submit_bio(rw | (1 << BIO_RW_SYNC), bio); + submit_bio(bio_rw, bio); } return 0; } diff --git a/mm/page_io.c b/mm/page_io.c index dc6ce0afbde..3023c475e04 100644 --- a/mm/page_io.c +++ b/mm/page_io.c @@ -111,7 +111,7 @@ int swap_writepage(struct page *page, struct writeback_control *wbc) goto out; } if (wbc->sync_mode == WB_SYNC_ALL) - rw |= (1 << BIO_RW_SYNC); + rw |= (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG); count_vm_event(PSWPOUT); set_page_writeback(page); unlock_page(page); -- cgit v1.2.3-70-g09d2 From 5fd3a17ed456637a224cf4ca82b9ad9d005bc8d4 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Wed, 4 Mar 2009 00:57:25 -0700 Subject: md: fix deadlock when stopping arrays Resolve a deadlock when stopping redundant arrays, i.e. ones that require a call to sysfs_remove_group when shutdown. The deadlock is summarized below: Thread1 Thread2 ------- ------- read sysfs attribute stop array take mddev lock sysfs_remove_group sysfs_get_active wait for mddev lock wait for active Sysrq-w: -------- mdmon S 00000017 2212 4163 1 f1982ea8 00000046 2dcf6b85 00000017 c0b23100 f2f83ed0 c0b23100 f2f8413c c0b23100 c0b23100 c0b1fb98 f2f8413c 00000000 f2f8413c c0b23100 f2291ecc 00000002 c0b23100 00000000 00000017 f2f83ed0 f1982eac 00000046 c044d9dd Call Trace: [] ? debug_mutex_add_waiter+0x1d/0x58 [] __mutex_lock_common+0x1d9/0x338 [] ? __mutex_lock_common+0x1d9/0x338 [] mutex_lock_interruptible_nested+0x33/0x3a [] ? mddev_lock+0x14/0x16 [] mddev_lock+0x14/0x16 [] md_attr_show+0x2a/0x49 [] sysfs_read_file+0x93/0xf9 mdadm D 00000017 2812 4177 1 f0401d78 00000046 430456f8 00000017 f0401d58 f0401d20 c0b23100 f2da2c4c c0b23100 c0b23100 c0b1fb98 f2da2c4c 0a10fc36 00000000 c0b23100 f0401d70 00000003 c0b23100 00000000 00000017 f2da29e0 00000001 00000002 00000000 Call Trace: [] schedule_timeout+0x1b/0x95 [] ? schedule_timeout+0x1b/0x95 [] ? wait_for_common+0x34/0xdc [] ? trace_hardirqs_on_caller+0x18/0x145 [] ? trace_hardirqs_on+0xb/0xd [] wait_for_common+0xa0/0xdc [] ? default_wake_function+0x0/0x12 [] wait_for_completion+0x17/0x19 [] sysfs_addrm_finish+0x19f/0x1d1 [] sysfs_hash_and_remove+0x42/0x55 [] sysfs_remove_group+0x57/0x86 [] do_md_stop+0x13a/0x499 This has been there for a while, but is easier to trigger now that mdmon is closely watching sysfs. Cc: Reported-by: Jacek Danecki Signed-off-by: Dan Williams --- drivers/md/md.c | 30 ++++++++++++++++++------------ 1 file changed, 18 insertions(+), 12 deletions(-) (limited to 'drivers/md/md.c') diff --git a/drivers/md/md.c b/drivers/md/md.c index 03b4cd0a634..a307f87eb90 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -214,12 +214,7 @@ static inline mddev_t *mddev_get(mddev_t *mddev) return mddev; } -static void mddev_delayed_delete(struct work_struct *ws) -{ - mddev_t *mddev = container_of(ws, mddev_t, del_work); - kobject_del(&mddev->kobj); - kobject_put(&mddev->kobj); -} +static void mddev_delayed_delete(struct work_struct *ws); static void mddev_put(mddev_t *mddev) { @@ -3542,6 +3537,21 @@ static struct kobj_type md_ktype = { int mdp_major = 0; +static void mddev_delayed_delete(struct work_struct *ws) +{ + mddev_t *mddev = container_of(ws, mddev_t, del_work); + + if (mddev->private == &md_redundancy_group) { + sysfs_remove_group(&mddev->kobj, &md_redundancy_group); + if (mddev->sysfs_action) + sysfs_put(mddev->sysfs_action); + mddev->sysfs_action = NULL; + mddev->private = NULL; + } + kobject_del(&mddev->kobj); + kobject_put(&mddev->kobj); +} + static int md_alloc(dev_t dev, char *name) { static DEFINE_MUTEX(disks_mutex); @@ -4033,13 +4043,9 @@ static int do_md_stop(mddev_t * mddev, int mode, int is_open) mddev->queue->merge_bvec_fn = NULL; mddev->queue->unplug_fn = NULL; mddev->queue->backing_dev_info.congested_fn = NULL; - if (mddev->pers->sync_request) { - sysfs_remove_group(&mddev->kobj, &md_redundancy_group); - if (mddev->sysfs_action) - sysfs_put(mddev->sysfs_action); - mddev->sysfs_action = NULL; - } module_put(mddev->pers->owner); + if (mddev->pers->sync_request) + mddev->private = &md_redundancy_group; mddev->pers = NULL; /* tell userspace to handle 'inactive' */ sysfs_notify_dirent(mddev->sysfs_state); -- cgit v1.2.3-70-g09d2