summaryrefslogtreecommitdiff
path: root/drivers/md
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/md')
-rw-r--r--drivers/md/bitmap.c114
-rw-r--r--drivers/md/bitmap.h10
-rw-r--r--drivers/md/dm-io.c27
-rw-r--r--drivers/md/dm-kcopyd.c168
-rw-r--r--drivers/md/dm-log.c3
-rw-r--r--drivers/md/dm-mpath.c2
-rw-r--r--drivers/md/dm-raid1.c10
-rw-r--r--drivers/md/dm-snap-persistent.c13
-rw-r--r--drivers/md/dm-snap.c10
-rw-r--r--drivers/md/dm-table.c23
-rw-r--r--drivers/md/md.c65
-rw-r--r--drivers/md/md.h2
-rw-r--r--drivers/md/multipath.c60
-rw-r--r--drivers/md/multipath.h1
-rw-r--r--drivers/md/raid1.c530
-rw-r--r--drivers/md/raid1.h6
-rw-r--r--drivers/md/raid10.c424
-rw-r--r--drivers/md/raid5.c64
18 files changed, 828 insertions, 704 deletions
diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index 5c9362792f1d..574b09afedd3 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -493,11 +493,11 @@ void bitmap_update_sb(struct bitmap *bitmap)
spin_unlock_irqrestore(&bitmap->lock, flags);
sb = kmap_atomic(bitmap->sb_page, KM_USER0);
sb->events = cpu_to_le64(bitmap->mddev->events);
- if (bitmap->mddev->events < bitmap->events_cleared) {
+ if (bitmap->mddev->events < bitmap->events_cleared)
/* rocking back to read-only */
bitmap->events_cleared = bitmap->mddev->events;
- sb->events_cleared = cpu_to_le64(bitmap->events_cleared);
- }
+ sb->events_cleared = cpu_to_le64(bitmap->events_cleared);
+ sb->state = cpu_to_le32(bitmap->flags);
/* Just in case these have been changed via sysfs: */
sb->daemon_sleep = cpu_to_le32(bitmap->mddev->bitmap_info.daemon_sleep/HZ);
sb->write_behind = cpu_to_le32(bitmap->mddev->bitmap_info.max_write_behind);
@@ -534,6 +534,82 @@ void bitmap_print_sb(struct bitmap *bitmap)
kunmap_atomic(sb, KM_USER0);
}
+/*
+ * bitmap_new_disk_sb
+ * @bitmap
+ *
+ * This function is somewhat the reverse of bitmap_read_sb. bitmap_read_sb
+ * reads and verifies the on-disk bitmap superblock and populates bitmap_info.
+ * This function verifies 'bitmap_info' and populates the on-disk bitmap
+ * structure, which is to be written to disk.
+ *
+ * Returns: 0 on success, -Exxx on error
+ */
+static int bitmap_new_disk_sb(struct bitmap *bitmap)
+{
+ bitmap_super_t *sb;
+ unsigned long chunksize, daemon_sleep, write_behind;
+ int err = -EINVAL;
+
+ bitmap->sb_page = alloc_page(GFP_KERNEL);
+ if (IS_ERR(bitmap->sb_page)) {
+ err = PTR_ERR(bitmap->sb_page);
+ bitmap->sb_page = NULL;
+ return err;
+ }
+ bitmap->sb_page->index = 0;
+
+ sb = kmap_atomic(bitmap->sb_page, KM_USER0);
+
+ sb->magic = cpu_to_le32(BITMAP_MAGIC);
+ sb->version = cpu_to_le32(BITMAP_MAJOR_HI);
+
+ chunksize = bitmap->mddev->bitmap_info.chunksize;
+ BUG_ON(!chunksize);
+ if (!is_power_of_2(chunksize)) {
+ kunmap_atomic(sb, KM_USER0);
+ printk(KERN_ERR "bitmap chunksize not a power of 2\n");
+ return -EINVAL;
+ }
+ sb->chunksize = cpu_to_le32(chunksize);
+
+ daemon_sleep = bitmap->mddev->bitmap_info.daemon_sleep;
+ if (!daemon_sleep ||
+ (daemon_sleep < 1) || (daemon_sleep > MAX_SCHEDULE_TIMEOUT)) {
+ printk(KERN_INFO "Choosing daemon_sleep default (5 sec)\n");
+ daemon_sleep = 5 * HZ;
+ }
+ sb->daemon_sleep = cpu_to_le32(daemon_sleep);
+ bitmap->mddev->bitmap_info.daemon_sleep = daemon_sleep;
+
+ /*
+ * FIXME: write_behind for RAID1. If not specified, what
+ * is a good choice? We choose COUNTER_MAX / 2 arbitrarily.
+ */
+ write_behind = bitmap->mddev->bitmap_info.max_write_behind;
+ if (write_behind > COUNTER_MAX)
+ write_behind = COUNTER_MAX / 2;
+ sb->write_behind = cpu_to_le32(write_behind);
+ bitmap->mddev->bitmap_info.max_write_behind = write_behind;
+
+ /* keep the array size field of the bitmap superblock up to date */
+ sb->sync_size = cpu_to_le64(bitmap->mddev->resync_max_sectors);
+
+ memcpy(sb->uuid, bitmap->mddev->uuid, 16);
+
+ bitmap->flags |= BITMAP_STALE;
+ sb->state |= cpu_to_le32(BITMAP_STALE);
+ bitmap->events_cleared = bitmap->mddev->events;
+ sb->events_cleared = cpu_to_le64(bitmap->mddev->events);
+
+ bitmap->flags |= BITMAP_HOSTENDIAN;
+ sb->version = cpu_to_le32(BITMAP_MAJOR_HOSTENDIAN);
+
+ kunmap_atomic(sb, KM_USER0);
+
+ return 0;
+}
+
/* read the superblock from the bitmap file and initialize some bitmap fields */
static int bitmap_read_sb(struct bitmap *bitmap)
{
@@ -575,7 +651,7 @@ static int bitmap_read_sb(struct bitmap *bitmap)
reason = "unrecognized superblock version";
else if (chunksize < 512)
reason = "bitmap chunksize too small";
- else if ((1 << ffz(~chunksize)) != chunksize)
+ else if (!is_power_of_2(chunksize))
reason = "bitmap chunksize not a power of 2";
else if (daemon_sleep < 1 || daemon_sleep > MAX_SCHEDULE_TIMEOUT)
reason = "daemon sleep period out of range";
@@ -618,7 +694,7 @@ success:
if (le32_to_cpu(sb->version) == BITMAP_MAJOR_HOSTENDIAN)
bitmap->flags |= BITMAP_HOSTENDIAN;
bitmap->events_cleared = le64_to_cpu(sb->events_cleared);
- if (sb->state & cpu_to_le32(BITMAP_STALE))
+ if (bitmap->flags & BITMAP_STALE)
bitmap->events_cleared = bitmap->mddev->events;
err = 0;
out:
@@ -652,9 +728,11 @@ static int bitmap_mask_state(struct bitmap *bitmap, enum bitmap_state bits,
switch (op) {
case MASK_SET:
sb->state |= cpu_to_le32(bits);
+ bitmap->flags |= bits;
break;
case MASK_UNSET:
sb->state &= cpu_to_le32(~bits);
+ bitmap->flags &= ~bits;
break;
default:
BUG();
@@ -1074,8 +1152,8 @@ static int bitmap_init_from_disk(struct bitmap *bitmap, sector_t start)
}
printk(KERN_INFO "%s: bitmap initialized from disk: "
- "read %lu/%lu pages, set %lu bits\n",
- bmname(bitmap), bitmap->file_pages, num_pages, bit_cnt);
+ "read %lu/%lu pages, set %lu of %lu bits\n",
+ bmname(bitmap), bitmap->file_pages, num_pages, bit_cnt, chunks);
return 0;
@@ -1330,7 +1408,7 @@ int bitmap_startwrite(struct bitmap *bitmap, sector_t offset, unsigned long sect
return 0;
}
- if (unlikely((*bmc & COUNTER_MAX) == COUNTER_MAX)) {
+ if (unlikely(COUNTER(*bmc) == COUNTER_MAX)) {
DEFINE_WAIT(__wait);
/* note that it is safe to do the prepare_to_wait
* after the test as long as we do it before dropping
@@ -1402,10 +1480,10 @@ void bitmap_endwrite(struct bitmap *bitmap, sector_t offset, unsigned long secto
sysfs_notify_dirent_safe(bitmap->sysfs_can_clear);
}
- if (!success && ! (*bmc & NEEDED_MASK))
+ if (!success && !NEEDED(*bmc))
*bmc |= NEEDED_MASK;
- if ((*bmc & COUNTER_MAX) == COUNTER_MAX)
+ if (COUNTER(*bmc) == COUNTER_MAX)
wake_up(&bitmap->overflow_wait);
(*bmc)--;
@@ -1726,9 +1804,16 @@ int bitmap_create(mddev_t *mddev)
vfs_fsync(file, 1);
}
/* read superblock from bitmap file (this sets mddev->bitmap_info.chunksize) */
- if (!mddev->bitmap_info.external)
- err = bitmap_read_sb(bitmap);
- else {
+ if (!mddev->bitmap_info.external) {
+ /*
+ * If 'MD_ARRAY_FIRST_USE' is set, then device-mapper is
+ * instructing us to create a new on-disk bitmap instance.
+ */
+ if (test_and_clear_bit(MD_ARRAY_FIRST_USE, &mddev->flags))
+ err = bitmap_new_disk_sb(bitmap);
+ else
+ err = bitmap_read_sb(bitmap);
+ } else {
err = 0;
if (mddev->bitmap_info.chunksize == 0 ||
mddev->bitmap_info.daemon_sleep == 0)
@@ -1752,9 +1837,6 @@ int bitmap_create(mddev_t *mddev)
bitmap->chunks = chunks;
bitmap->pages = pages;
bitmap->missing_pages = pages;
- bitmap->counter_bits = COUNTER_BITS;
-
- bitmap->syncchunk = ~0UL;
#ifdef INJECT_FATAL_FAULT_1
bitmap->bp = NULL;
diff --git a/drivers/md/bitmap.h b/drivers/md/bitmap.h
index d0aeaf46d932..b2a127e891ac 100644
--- a/drivers/md/bitmap.h
+++ b/drivers/md/bitmap.h
@@ -85,7 +85,6 @@
typedef __u16 bitmap_counter_t;
#define COUNTER_BITS 16
#define COUNTER_BIT_SHIFT 4
-#define COUNTER_BYTE_RATIO (COUNTER_BITS / 8)
#define COUNTER_BYTE_SHIFT (COUNTER_BIT_SHIFT - 3)
#define NEEDED_MASK ((bitmap_counter_t) (1 << (COUNTER_BITS - 1)))
@@ -196,19 +195,10 @@ struct bitmap {
mddev_t *mddev; /* the md device that the bitmap is for */
- int counter_bits; /* how many bits per block counter */
-
/* bitmap chunksize -- how much data does each bit represent? */
unsigned long chunkshift; /* chunksize = 2^chunkshift (for bitops) */
unsigned long chunks; /* total number of data chunks for the array */
- /* We hold a count on the chunk currently being synced, and drop
- * it when the last block is started. If the resync is aborted
- * midway, we need to be able to drop that count, so we remember
- * the counted chunk..
- */
- unsigned long syncchunk;
-
__u64 events_cleared;
int need_sync;
diff --git a/drivers/md/dm-io.c b/drivers/md/dm-io.c
index 76a5af00a26b..2067288f61f9 100644
--- a/drivers/md/dm-io.c
+++ b/drivers/md/dm-io.c
@@ -19,6 +19,8 @@
#define DM_MSG_PREFIX "io"
#define DM_IO_MAX_REGIONS BITS_PER_LONG
+#define MIN_IOS 16
+#define MIN_BIOS 16
struct dm_io_client {
mempool_t *pool;
@@ -41,33 +43,21 @@ struct io {
static struct kmem_cache *_dm_io_cache;
/*
- * io contexts are only dynamically allocated for asynchronous
- * io. Since async io is likely to be the majority of io we'll
- * have the same number of io contexts as bios! (FIXME: must reduce this).
- */
-
-static unsigned int pages_to_ios(unsigned int pages)
-{
- return 4 * pages; /* too many ? */
-}
-
-/*
* Create a client with mempool and bioset.
*/
-struct dm_io_client *dm_io_client_create(unsigned num_pages)
+struct dm_io_client *dm_io_client_create(void)
{
- unsigned ios = pages_to_ios(num_pages);
struct dm_io_client *client;
client = kmalloc(sizeof(*client), GFP_KERNEL);
if (!client)
return ERR_PTR(-ENOMEM);
- client->pool = mempool_create_slab_pool(ios, _dm_io_cache);
+ client->pool = mempool_create_slab_pool(MIN_IOS, _dm_io_cache);
if (!client->pool)
goto bad;
- client->bios = bioset_create(16, 0);
+ client->bios = bioset_create(MIN_BIOS, 0);
if (!client->bios)
goto bad;
@@ -81,13 +71,6 @@ struct dm_io_client *dm_io_client_create(unsigned num_pages)
}
EXPORT_SYMBOL(dm_io_client_create);
-int dm_io_client_resize(unsigned num_pages, struct dm_io_client *client)
-{
- return mempool_resize(client->pool, pages_to_ios(num_pages),
- GFP_KERNEL);
-}
-EXPORT_SYMBOL(dm_io_client_resize);
-
void dm_io_client_destroy(struct dm_io_client *client)
{
mempool_destroy(client->pool);
diff --git a/drivers/md/dm-kcopyd.c b/drivers/md/dm-kcopyd.c
index 1bb73a13ca40..819e37eaaeba 100644
--- a/drivers/md/dm-kcopyd.c
+++ b/drivers/md/dm-kcopyd.c
@@ -27,15 +27,19 @@
#include "dm.h"
+#define SUB_JOB_SIZE 128
+#define SPLIT_COUNT 8
+#define MIN_JOBS 8
+#define RESERVE_PAGES (DIV_ROUND_UP(SUB_JOB_SIZE << SECTOR_SHIFT, PAGE_SIZE))
+
/*-----------------------------------------------------------------
* Each kcopyd client has its own little pool of preallocated
* pages for kcopyd io.
*---------------------------------------------------------------*/
struct dm_kcopyd_client {
- spinlock_t lock;
struct page_list *pages;
- unsigned int nr_pages;
- unsigned int nr_free_pages;
+ unsigned nr_reserved_pages;
+ unsigned nr_free_pages;
struct dm_io_client *io_client;
@@ -67,15 +71,18 @@ static void wake(struct dm_kcopyd_client *kc)
queue_work(kc->kcopyd_wq, &kc->kcopyd_work);
}
-static struct page_list *alloc_pl(void)
+/*
+ * Obtain one page for the use of kcopyd.
+ */
+static struct page_list *alloc_pl(gfp_t gfp)
{
struct page_list *pl;
- pl = kmalloc(sizeof(*pl), GFP_KERNEL);
+ pl = kmalloc(sizeof(*pl), gfp);
if (!pl)
return NULL;
- pl->page = alloc_page(GFP_KERNEL);
+ pl->page = alloc_page(gfp);
if (!pl->page) {
kfree(pl);
return NULL;
@@ -90,41 +97,56 @@ static void free_pl(struct page_list *pl)
kfree(pl);
}
-static int kcopyd_get_pages(struct dm_kcopyd_client *kc,
- unsigned int nr, struct page_list **pages)
+/*
+ * Add the provided pages to a client's free page list, releasing
+ * back to the system any beyond the reserved_pages limit.
+ */
+static void kcopyd_put_pages(struct dm_kcopyd_client *kc, struct page_list *pl)
{
- struct page_list *pl;
-
- spin_lock(&kc->lock);
- if (kc->nr_free_pages < nr) {
- spin_unlock(&kc->lock);
- return -ENOMEM;
- }
-
- kc->nr_free_pages -= nr;
- for (*pages = pl = kc->pages; --nr; pl = pl->next)
- ;
+ struct page_list *next;
- kc->pages = pl->next;
- pl->next = NULL;
+ do {
+ next = pl->next;
- spin_unlock(&kc->lock);
+ if (kc->nr_free_pages >= kc->nr_reserved_pages)
+ free_pl(pl);
+ else {
+ pl->next = kc->pages;
+ kc->pages = pl;
+ kc->nr_free_pages++;
+ }
- return 0;
+ pl = next;
+ } while (pl);
}
-static void kcopyd_put_pages(struct dm_kcopyd_client *kc, struct page_list *pl)
+static int kcopyd_get_pages(struct dm_kcopyd_client *kc,
+ unsigned int nr, struct page_list **pages)
{
- struct page_list *cursor;
+ struct page_list *pl;
+
+ *pages = NULL;
+
+ do {
+ pl = alloc_pl(__GFP_NOWARN | __GFP_NORETRY);
+ if (unlikely(!pl)) {
+ /* Use reserved pages */
+ pl = kc->pages;
+ if (unlikely(!pl))
+ goto out_of_memory;
+ kc->pages = pl->next;
+ kc->nr_free_pages--;
+ }
+ pl->next = *pages;
+ *pages = pl;
+ } while (--nr);
- spin_lock(&kc->lock);
- for (cursor = pl; cursor->next; cursor = cursor->next)
- kc->nr_free_pages++;
+ return 0;
- kc->nr_free_pages++;
- cursor->next = kc->pages;
- kc->pages = pl;
- spin_unlock(&kc->lock);
+out_of_memory:
+ if (*pages)
+ kcopyd_put_pages(kc, *pages);
+ return -ENOMEM;
}
/*
@@ -141,13 +163,16 @@ static void drop_pages(struct page_list *pl)
}
}
-static int client_alloc_pages(struct dm_kcopyd_client *kc, unsigned int nr)
+/*
+ * Allocate and reserve nr_pages for the use of a specific client.
+ */
+static int client_reserve_pages(struct dm_kcopyd_client *kc, unsigned nr_pages)
{
- unsigned int i;
+ unsigned i;
struct page_list *pl = NULL, *next;
- for (i = 0; i < nr; i++) {
- next = alloc_pl();
+ for (i = 0; i < nr_pages; i++) {
+ next = alloc_pl(GFP_KERNEL);
if (!next) {
if (pl)
drop_pages(pl);
@@ -157,17 +182,18 @@ static int client_alloc_pages(struct dm_kcopyd_client *kc, unsigned int nr)
pl = next;
}
+ kc->nr_reserved_pages += nr_pages;
kcopyd_put_pages(kc, pl);
- kc->nr_pages += nr;
+
return 0;
}
static void client_free_pages(struct dm_kcopyd_client *kc)
{
- BUG_ON(kc->nr_free_pages != kc->nr_pages);
+ BUG_ON(kc->nr_free_pages != kc->nr_reserved_pages);
drop_pages(kc->pages);
kc->pages = NULL;
- kc->nr_free_pages = kc->nr_pages = 0;
+ kc->nr_free_pages = kc->nr_reserved_pages = 0;
}
/*-----------------------------------------------------------------
@@ -216,16 +242,17 @@ struct kcopyd_job {
struct mutex lock;
atomic_t sub_jobs;
sector_t progress;
-};
-/* FIXME: this should scale with the number of pages */
-#define MIN_JOBS 512
+ struct kcopyd_job *master_job;
+};
static struct kmem_cache *_job_cache;
int __init dm_kcopyd_init(void)
{
- _job_cache = KMEM_CACHE(kcopyd_job, 0);
+ _job_cache = kmem_cache_create("kcopyd_job",
+ sizeof(struct kcopyd_job) * (SPLIT_COUNT + 1),
+ __alignof__(struct kcopyd_job), 0, NULL);
if (!_job_cache)
return -ENOMEM;
@@ -299,7 +326,12 @@ static int run_complete_job(struct kcopyd_job *job)
if (job->pages)
kcopyd_put_pages(kc, job->pages);
- mempool_free(job, kc->job_pool);
+ /*
+ * If this is the master job, the sub jobs have already
+ * completed so we can free everything.
+ */
+ if (job->master_job == job)
+ mempool_free(job, kc->job_pool);
fn(read_err, write_err, context);
if (atomic_dec_and_test(&kc->nr_jobs))
@@ -460,14 +492,14 @@ static void dispatch_job(struct kcopyd_job *job)
wake(kc);
}
-#define SUB_JOB_SIZE 128
static void segment_complete(int read_err, unsigned long write_err,
void *context)
{
/* FIXME: tidy this function */
sector_t progress = 0;
sector_t count = 0;
- struct kcopyd_job *job = (struct kcopyd_job *) context;
+ struct kcopyd_job *sub_job = (struct kcopyd_job *) context;
+ struct kcopyd_job *job = sub_job->master_job;
struct dm_kcopyd_client *kc = job->kc;
mutex_lock(&job->lock);
@@ -498,8 +530,6 @@ static void segment_complete(int read_err, unsigned long write_err,
if (count) {
int i;
- struct kcopyd_job *sub_job = mempool_alloc(kc->job_pool,
- GFP_NOIO);
*sub_job = *job;
sub_job->source.sector += progress;
@@ -511,7 +541,7 @@ static void segment_complete(int read_err, unsigned long write_err,
}
sub_job->fn = segment_complete;
- sub_job->context = job;
+ sub_job->context = sub_job;
dispatch_job(sub_job);
} else if (atomic_dec_and_test(&job->sub_jobs)) {
@@ -531,19 +561,19 @@ static void segment_complete(int read_err, unsigned long write_err,
}
/*
- * Create some little jobs that will do the move between
- * them.
+ * Create some sub jobs to share the work between them.
*/
-#define SPLIT_COUNT 8
-static void split_job(struct kcopyd_job *job)
+static void split_job(struct kcopyd_job *master_job)
{
int i;
- atomic_inc(&job->kc->nr_jobs);
+ atomic_inc(&master_job->kc->nr_jobs);
- atomic_set(&job->sub_jobs, SPLIT_COUNT);
- for (i = 0; i < SPLIT_COUNT; i++)
- segment_complete(0, 0u, job);
+ atomic_set(&master_job->sub_jobs, SPLIT_COUNT);
+ for (i = 0; i < SPLIT_COUNT; i++) {
+ master_job[i + 1].master_job = master_job;
+ segment_complete(0, 0u, &master_job[i + 1]);
+ }
}
int dm_kcopyd_copy(struct dm_kcopyd_client *kc, struct dm_io_region *from,
@@ -553,7 +583,8 @@ int dm_kcopyd_copy(struct dm_kcopyd_client *kc, struct dm_io_region *from,
struct kcopyd_job *job;
/*
- * Allocate a new job.
+ * Allocate an array of jobs consisting of one master job
+ * followed by SPLIT_COUNT sub jobs.
*/
job = mempool_alloc(kc->job_pool, GFP_NOIO);
@@ -577,10 +608,10 @@ int dm_kcopyd_copy(struct dm_kcopyd_client *kc, struct dm_io_region *from,
job->fn = fn;
job->context = context;
+ job->master_job = job;
- if (job->source.count < SUB_JOB_SIZE)
+ if (job->source.count <= SUB_JOB_SIZE)
dispatch_job(job);
-
else {
mutex_init(&job->lock);
job->progress = 0;
@@ -606,17 +637,15 @@ int kcopyd_cancel(struct kcopyd_job *job, int block)
/*-----------------------------------------------------------------
* Client setup
*---------------------------------------------------------------*/
-int dm_kcopyd_client_create(unsigned int nr_pages,
- struct dm_kcopyd_client **result)
+struct dm_kcopyd_client *dm_kcopyd_client_create(void)
{
int r = -ENOMEM;
struct dm_kcopyd_client *kc;
kc = kmalloc(sizeof(*kc), GFP_KERNEL);
if (!kc)
- return -ENOMEM;
+ return ERR_PTR(-ENOMEM);
- spin_lock_init(&kc->lock);
spin_lock_init(&kc->job_lock);
INIT_LIST_HEAD(&kc->complete_jobs);
INIT_LIST_HEAD(&kc->io_jobs);
@@ -633,12 +662,12 @@ int dm_kcopyd_client_create(unsigned int nr_pages,
goto bad_workqueue;
kc->pages = NULL;
- kc->nr_pages = kc->nr_free_pages = 0;
- r = client_alloc_pages(kc, nr_pages);
+ kc->nr_reserved_pages = kc->nr_free_pages = 0;
+ r = client_reserve_pages(kc, RESERVE_PAGES);
if (r)
goto bad_client_pages;
- kc->io_client = dm_io_client_create(nr_pages);
+ kc->io_client = dm_io_client_create();
if (IS_ERR(kc->io_client)) {
r = PTR_ERR(kc->io_client);
goto bad_io_client;
@@ -647,8 +676,7 @@ int dm_kcopyd_client_create(unsigned int nr_pages,
init_waitqueue_head(&kc->destroyq);
atomic_set(&kc->nr_jobs, 0);
- *result = kc;
- return 0;
+ return kc;
bad_io_client:
client_free_pages(kc);
@@ -659,7 +687,7 @@ bad_workqueue:
bad_slab:
kfree(kc);
- return r;
+ return ERR_PTR(r);
}
EXPORT_SYMBOL(dm_kcopyd_client_create);
diff --git a/drivers/md/dm-log.c b/drivers/md/dm-log.c
index a1f321889676..948e3f4925bf 100644
--- a/drivers/md/dm-log.c
+++ b/drivers/md/dm-log.c
@@ -449,8 +449,7 @@ static int create_log_context(struct dm_dirty_log *log, struct dm_target *ti,
lc->io_req.mem.type = DM_IO_VMA;
lc->io_req.notify.fn = NULL;
- lc->io_req.client = dm_io_client_create(dm_div_up(buf_size,
- PAGE_SIZE));
+ lc->io_req.client = dm_io_client_create();
if (IS_ERR(lc->io_req.client)) {
r = PTR_ERR(lc->io_req.client);
DMWARN("couldn't allocate disk io client");
diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index a550a057d991..aa4e570c2cb5 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -1290,7 +1290,7 @@ static int do_end_io(struct multipath *m, struct request *clone,
if (!error && !clone->errors)
return 0; /* I/O complete */
- if (error == -EOPNOTSUPP || error == -EREMOTEIO)
+ if (error == -EOPNOTSUPP || error == -EREMOTEIO || error == -EILSEQ)
return error;
if (mpio->pgpath)
diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c
index 976ad4688afc..9bfd057be686 100644
--- a/drivers/md/dm-raid1.c
+++ b/drivers/md/dm-raid1.c
@@ -22,8 +22,6 @@
#define DM_MSG_PREFIX "raid1"
#define MAX_RECOVERY 1 /* Maximum number of regions recovered in parallel. */
-#define DM_IO_PAGES 64
-#define DM_KCOPYD_PAGES 64
#define DM_RAID1_HANDLE_ERRORS 0x01
#define errors_handled(p) ((p)->features & DM_RAID1_HANDLE_ERRORS)
@@ -887,7 +885,7 @@ static struct mirror_set *alloc_context(unsigned int nr_mirrors,
return NULL;
}
- ms->io_client = dm_io_client_create(DM_IO_PAGES);
+ ms->io_client = dm_io_client_create();
if (IS_ERR(ms->io_client)) {
ti->error = "Error creating dm_io client";
mempool_destroy(ms->read_record_pool);
@@ -1117,9 +1115,11 @@ static int mirror_ctr(struct dm_target *ti, unsigned int argc, char **argv)
goto err_destroy_wq;
}
- r = dm_kcopyd_client_create(DM_KCOPYD_PAGES, &ms->kcopyd_client);
- if (r)
+ ms->kcopyd_client = dm_kcopyd_client_create();
+ if (IS_ERR(ms->kcopyd_client)) {
+ r = PTR_ERR(ms->kcopyd_client);
goto err_destroy_wq;
+ }
wakeup_mirrord(ms);
return 0;
diff --git a/drivers/md/dm-snap-persistent.c b/drivers/md/dm-snap-persistent.c
index 95891dfcbca0..135c2f1fdbfc 100644
--- a/drivers/md/dm-snap-persistent.c
+++ b/drivers/md/dm-snap-persistent.c
@@ -154,11 +154,6 @@ struct pstore {
struct workqueue_struct *metadata_wq;
};
-static unsigned sectors_to_pages(unsigned sectors)
-{
- return DIV_ROUND_UP(sectors, PAGE_SIZE >> 9);
-}
-
static int alloc_area(struct pstore *ps)
{
int r = -ENOMEM;
@@ -318,8 +313,7 @@ static int read_header(struct pstore *ps, int *new_snapshot)
chunk_size_supplied = 0;
}
- ps->io_client = dm_io_client_create(sectors_to_pages(ps->store->
- chunk_size));
+ ps->io_client = dm_io_client_create();
if (IS_ERR(ps->io_client))
return PTR_ERR(ps->io_client);
@@ -368,11 +362,6 @@ static int read_header(struct pstore *ps, int *new_snapshot)
return r;
}
- r = dm_io_client_resize(sectors_to_pages(ps->store->chunk_size),
- ps->io_client);
- if (r)
- return r;
-
r = alloc_area(ps);
return r;
diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c
index a2d330942cb2..9ecff5f3023a 100644
--- a/drivers/md/dm-snap.c
+++ b/drivers/md/dm-snap.c
@@ -40,11 +40,6 @@ static const char dm_snapshot_merge_target_name[] = "snapshot-merge";
#define SNAPSHOT_COPY_PRIORITY 2
/*
- * Reserve 1MB for each snapshot initially (with minimum of 1 page).
- */
-#define SNAPSHOT_PAGES (((1UL << 20) >> PAGE_SHIFT) ? : 1)
-
-/*
* The size of the mempool used to track chunks in use.
*/
#define MIN_IOS 256
@@ -1116,8 +1111,9 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv)
goto bad_hash_tables;
}
- r = dm_kcopyd_client_create(SNAPSHOT_PAGES, &s->kcopyd_client);
- if (r) {
+ s->kcopyd_client = dm_kcopyd_client_create();
+ if (IS_ERR(s->kcopyd_client)) {
+ r = PTR_ERR(s->kcopyd_client);
ti->error = "Could not create kcopyd client";
goto bad_kcopyd;
}
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index cb8380c9767f..451c3bb176d2 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -362,6 +362,7 @@ static void close_dev(struct dm_dev_internal *d, struct mapped_device *md)
static int device_area_is_invalid(struct dm_target *ti, struct dm_dev *dev,
sector_t start, sector_t len, void *data)
{
+ struct request_queue *q;
struct queue_limits *limits = data;
struct block_device *bdev = dev->bdev;
sector_t dev_size =
@@ -370,6 +371,22 @@ static int device_area_is_invalid(struct dm_target *ti, struct dm_dev *dev,
limits->logical_block_size >> SECTOR_SHIFT;
char b[BDEVNAME_SIZE];
+ /*
+ * Some devices exist without request functions,
+ * such as loop devices not yet bound to backing files.
+ * Forbid the use of such devices.
+ */
+ q = bdev_get_queue(bdev);
+ if (!q || !q->make_request_fn) {
+ DMWARN("%s: %s is not yet initialised: "
+ "start=%llu, len=%llu, dev_size=%llu",
+ dm_device_name(ti->table->md), bdevname(bdev, b),
+ (unsigned long long)start,
+ (unsigned long long)len,
+ (unsigned long long)dev_size);
+ return 1;
+ }
+
if (!dev_size)
return 0;
@@ -1346,7 +1363,8 @@ bool dm_table_supports_discards(struct dm_table *t)
return 0;
/*
- * Ensure that at least one underlying device supports discards.
+ * Unless any target used by the table set discards_supported,
+ * require at least one underlying device to support discards.
* t->devices includes internal dm devices such as mirror logs
* so we need to use iterate_devices here, which targets
* supporting discard must provide.
@@ -1354,6 +1372,9 @@ bool dm_table_supports_discards(struct dm_table *t)
while (i < dm_table_get_num_targets(t)) {
ti = dm_table_get_target(t, i++);
+ if (ti->discards_supported)
+ return 1;
+
if (ti->type->iterate_devices &&
ti->type->iterate_devices(ti, device_discard_capable, NULL))
return 1;
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 6e853c61d87e..4332fc2f25d4 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -351,6 +351,9 @@ void mddev_resume(mddev_t *mddev)
mddev->suspended = 0;
wake_up(&mddev->sb_wait);
mddev->pers->quiesce(mddev, 0);
+
+ md_wakeup_thread(mddev->thread);
+ md_wakeup_thread(mddev->sync_thread); /* possibly kick off a reshape */
}
EXPORT_SYMBOL_GPL(mddev_resume);
@@ -1750,6 +1753,18 @@ static struct super_type super_types[] = {
},
};
+static void sync_super(mddev_t *mddev, mdk_rdev_t *rdev)
+{
+ if (mddev->sync_super) {
+ mddev->sync_super(mddev, rdev);
+ return;
+ }
+
+ BUG_ON(mddev->major_version >= ARRAY_SIZE(super_types));
+
+ super_types[mddev->major_version].sync_super(mddev, rdev);
+}
+
static int match_mddev_units(mddev_t *mddev1, mddev_t *mddev2)
{
mdk_rdev_t *rdev, *rdev2;
@@ -1781,8 +1796,8 @@ int md_integrity_register(mddev_t *mddev)
if (list_empty(&mddev->disks))
return 0; /* nothing to do */
- if (blk_get_integrity(mddev->gendisk))
- return 0; /* already registered */
+ if (!mddev->gendisk || blk_get_integrity(mddev->gendisk))
+ return 0; /* shouldn't register, or already is */
list_for_each_entry(rdev, &mddev->disks, same_set) {
/* skip spares and non-functional disks */
if (test_bit(Faulty, &rdev->flags))
@@ -2168,8 +2183,7 @@ static void sync_sbs(mddev_t * mddev, int nospares)
/* Don't update this superblock */
rdev->sb_loaded = 2;
} else {
- super_types[mddev->major_version].
- sync_super(mddev, rdev);
+ sync_super(mddev, rdev);
rdev->sb_loaded = 1;
}
}
@@ -2462,7 +2476,7 @@ slot_store(mdk_rdev_t *rdev, const char *buf, size_t len)
if (rdev->raid_disk == -1)
return -EEXIST;
/* personality does all needed checks */
- if (rdev->mddev->pers->hot_add_disk == NULL)
+ if (rdev->mddev->pers->hot_remove_disk == NULL)
return -EINVAL;
err = rdev->mddev->pers->
hot_remove_disk(rdev->mddev, rdev->raid_disk);
@@ -3170,6 +3184,7 @@ level_store(mddev_t *mddev, const char *buf, size_t len)
mddev->layout = mddev->new_layout;
mddev->chunk_sectors = mddev->new_chunk_sectors;
mddev->delta_disks = 0;
+ mddev->degraded = 0;
if (mddev->pers->sync_request == NULL) {
/* this is now an array without redundancy, so
* it must always be in_sync
@@ -3323,7 +3338,7 @@ resync_start_store(mddev_t *mddev, const char *buf, size_t len)
char *e;
unsigned long long n = simple_strtoull(buf, &e, 10);
- if (mddev->pers)
+ if (mddev->pers && !test_bit(MD_RECOVERY_FROZEN, &mddev->recovery))
return -EBUSY;
if (cmd_match(buf, "none"))
n = MaxSector;
@@ -4346,13 +4361,19 @@ static int md_alloc(dev_t dev, char *name)
disk->fops = &md_fops;
disk->private_data = mddev;
disk->queue = mddev->queue;
+ blk_queue_flush(mddev->queue, REQ_FLUSH | REQ_FUA);
/* Allow extended partitions. This makes the
* 'mdp' device redundant, but we can't really
* remove it now.
*/
disk->flags |= GENHD_FL_EXT_DEVT;
- add_disk(disk);
mddev->gendisk = disk;
+ /* As soon as we call add_disk(), another thread could get
+ * through to md_open, so make sure it doesn't get too far
+ */
+ mutex_lock(&mddev->open_mutex);
+ add_disk(disk);
+
error = kobject_init_and_add(&mddev->kobj, &md_ktype,
&disk_to_dev(disk)->kobj, "%s", "md");
if (error) {
@@ -4366,8 +4387,7 @@ static int md_alloc(dev_t dev, char *name)
if (mddev->kobj.sd &&
sysfs_create_group(&mddev->kobj, &md_bitmap_group))
printk(KERN_DEBUG "pointless warning\n");
-
- blk_queue_flush(mddev->queue, REQ_FLUSH | REQ_FUA);
+ mutex_unlock(&mddev->open_mutex);
abort:
mutex_unlock(&disks_mutex);
if (!error && mddev->kobj.sd) {
@@ -4613,9 +4633,6 @@ int md_run(mddev_t *mddev)
if (mddev->flags)
md_update_sb(mddev, 0);
- md_wakeup_thread(mddev->thread);
- md_wakeup_thread(mddev->sync_thread); /* possibly kick off a reshape */
-
md_new_event(mddev);
sysfs_notify_dirent_safe(mddev->sysfs_state);
sysfs_notify_dirent_safe(mddev->sysfs_action);
@@ -4636,6 +4653,10 @@ static int do_md_run(mddev_t *mddev)
bitmap_destroy(mddev);
goto out;
}
+
+ md_wakeup_thread(mddev->thread);
+ md_wakeup_thread(mddev->sync_thread); /* possibly kick off a reshape */
+
set_capacity(mddev->gendisk, mddev->array_sectors);
revalidate_disk(mddev->gendisk);
mddev->changed = 1;
@@ -5210,6 +5231,16 @@ static int add_new_disk(mddev_t * mddev, mdu_disk_info_t *info)
} else
super_types[mddev->major_version].
validate_super(mddev, rdev);
+ if ((info->state & (1<<MD_DISK_SYNC)) &&
+ (!test_bit(In_sync, &rdev->flags) ||
+ rdev->raid_disk != info->raid_disk)) {
+ /* This was a hot-add request, but events doesn't
+ * match, so reject it.
+ */
+ export_rdev(rdev);
+ return -EINVAL;
+ }
+
if (test_bit(In_sync, &rdev->flags))
rdev->saved_raid_disk = rdev->raid_disk;
else
@@ -5243,6 +5274,8 @@ static int add_new_disk(mddev_t * mddev, mdu_disk_info_t *info)
if (mddev->degraded)
set_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
+ if (!err)
+ md_new_event(mddev);
md_wakeup_thread(mddev->thread);
return err;
}
@@ -6850,8 +6883,8 @@ void md_do_sync(mddev_t *mddev)
* Tune reconstruction:
*/
window = 32*(PAGE_SIZE/512);
- printk(KERN_INFO "md: using %dk window, over a total of %llu blocks.\n",
- window/2,(unsigned long long) max_sectors/2);
+ printk(KERN_INFO "md: using %dk window, over a total of %lluk.\n",
+ window/2, (unsigned long long)max_sectors/2);
atomic_set(&mddev->recovery_active, 0);
last_check = 0;
@@ -7029,7 +7062,6 @@ void md_do_sync(mddev_t *mddev)
}
EXPORT_SYMBOL_GPL(md_do_sync);
-
static int remove_and_add_spares(mddev_t *mddev)
{
mdk_rdev_t *rdev;
@@ -7141,6 +7173,9 @@ static void reap_sync_thread(mddev_t *mddev)
*/
void md_check_recovery(mddev_t *mddev)
{
+ if (mddev->suspended)
+ return;
+
if (mddev->bitmap)
bitmap_daemon_work(mddev);
diff --git a/drivers/md/md.h b/drivers/md/md.h
index 0b1fd3f1d85b..1c26c7a08ae6 100644
--- a/drivers/md/md.h
+++ b/drivers/md/md.h
@@ -124,6 +124,7 @@ 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 /* switch from 'clean' to 'active' in progress */
+#define MD_ARRAY_FIRST_USE 3 /* First use of array, needs initialization */
int suspended;
atomic_t active_io;
@@ -330,6 +331,7 @@ struct mddev_s
atomic_t flush_pending;
struct work_struct flush_work;
struct work_struct event_work; /* used by dm to report failure event */
+ void (*sync_super)(mddev_t *mddev, mdk_rdev_t *rdev);
};
diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c
index c35890990985..3535c23af288 100644
--- a/drivers/md/multipath.c
+++ b/drivers/md/multipath.c
@@ -146,7 +146,7 @@ static void multipath_status (struct seq_file *seq, mddev_t *mddev)
int i;
seq_printf (seq, " [%d/%d] [", conf->raid_disks,
- conf->working_disks);
+ conf->raid_disks - mddev->degraded);
for (i = 0; i < conf->raid_disks; i++)
seq_printf (seq, "%s",
conf->multipaths[i].rdev &&
@@ -186,35 +186,36 @@ static int multipath_congested(void *data, int bits)
static void multipath_error (mddev_t *mddev, mdk_rdev_t *rdev)
{
multipath_conf_t *conf = mddev->private;
+ char b[BDEVNAME_SIZE];
- if (conf->working_disks <= 1) {
+ if (conf->raid_disks - mddev->degraded <= 1) {
/*
* Uh oh, we can do nothing if this is our last path, but
* first check if this is a queued request for a device
* which has just failed.
*/
printk(KERN_ALERT
- "multipath: only one IO path left and IO error.\n");
+ "multipath: only one IO path left and IO error.\n");
/* leave it active... it's all we have */
- } else {
- /*
- * Mark disk as unusable
- */
- if (!test_bit(Faulty, &rdev->flags)) {
- char b[BDEVNAME_SIZE];
- clear_bit(In_sync, &rdev->flags);
- set_bit(Faulty, &rdev->flags);
- set_bit(MD_CHANGE_DEVS, &mddev->flags);
- conf->working_disks--;
- mddev->degraded++;
- printk(KERN_ALERT "multipath: IO failure on %s,"
- " disabling IO path.\n"
- "multipath: Operation continuing"
- " on %d IO paths.\n",
- bdevname (rdev->bdev,b),
- conf->working_disks);
- }
+ return;
+ }
+ /*
+ * Mark disk as unusable
+ */
+ if (test_and_clear_bit(In_sync, &rdev->flags)) {
+ unsigned long flags;
+ spin_lock_irqsave(&conf->device_lock, flags);
+ mddev->degraded++;
+ spin_unlock_irqrestore(&conf->device_lock, flags);
}
+ set_bit(Faulty, &rdev->flags);
+ set_bit(MD_CHANGE_DEVS, &mddev->flags);
+ printk(KERN_ALERT "multipath: IO failure on %s,"
+ " disabling IO path.\n"
+ "multipath: Operation continuing"
+ " on %d IO paths.\n",
+ bdevname(rdev->bdev, b),
+ conf->raid_disks - mddev->degraded);
}
static void print_multipath_conf (multipath_conf_t *conf)
@@ -227,7 +228,7 @@ static void print_multipath_conf (multipath_conf_t *conf)
printk("(conf==NULL)\n");
return;
}
- printk(" --- wd:%d rd:%d\n", conf->working_disks,
+ printk(" --- wd:%d rd:%d\n", conf->raid_disks - conf->mddev->degraded,
conf->raid_disks);
for (i = 0; i < conf->raid_disks; i++) {
@@ -274,10 +275,11 @@ static int multipath_add_disk(mddev_t *mddev, mdk_rdev_t *rdev)
PAGE_CACHE_SIZE - 1);
}
- conf->working_disks++;
+ spin_lock_irq(&conf->device_lock);
mddev->degraded--;
rdev->raid_disk = path;
set_bit(In_sync, &rdev->flags);
+ spin_unlock_irq(&conf->device_lock);
rcu_assign_pointer(p->rdev, rdev);
err = 0;
md_integrity_add_rdev(rdev, mddev);
@@ -391,6 +393,7 @@ static int multipath_run (mddev_t *mddev)
int disk_idx;
struct multipath_info *disk;
mdk_rdev_t *rdev;
+ int working_disks;
if (md_check_no_bitmap(mddev))
return -EINVAL;
@@ -424,7 +427,7 @@ static int multipath_run (mddev_t *mddev)
goto out_free_conf;
}
- conf->working_disks = 0;
+ working_disks = 0;
list_for_each_entry(rdev, &mddev->disks, same_set) {
disk_idx = rdev->raid_disk;
if (disk_idx < 0 ||
@@ -446,7 +449,7 @@ static int multipath_run (mddev_t *mddev)
}
if (!test_bit(Faulty, &rdev->flags))
- conf->working_disks++;
+ working_disks++;
}
conf->raid_disks = mddev->raid_disks;
@@ -454,12 +457,12 @@ static int multipath_run (mddev_t *mddev)
spin_lock_init(&conf->device_lock);
INIT_LIST_HEAD(&conf->retry_list);
- if (!conf->working_disks) {
+ if (!working_disks) {
printk(KERN_ERR "multipath: no operational IO paths for %s\n",
mdname(mddev));
goto out_free_conf;
}
- mddev->degraded = conf->raid_disks - conf->working_disks;
+ mddev->degraded = conf->raid_disks - working_disks;
conf->pool = mempool_create_kmalloc_pool(NR_RESERVED_BUFS,
sizeof(struct multipath_bh));
@@ -481,7 +484,8 @@ static int multipath_run (mddev_t *mddev)
printk(KERN_INFO
"multipath: array %s active with %d out of %d IO paths\n",
- mdname(mddev), conf->working_disks, mddev->raid_disks);
+ mdname(mddev), conf->raid_disks - mddev->degraded,
+ mddev->raid_disks);
/*
* Ok, everything is just fine now
*/
diff --git a/drivers/md/multipath.h b/drivers/md/multipath.h
index d1c2a8d78395..3c5a45eb5f8a 100644
--- a/drivers/md/multipath.h
+++ b/drivers/md/multipath.h
@@ -9,7 +9,6 @@ struct multipath_private_data {
mddev_t *mddev;
struct multipath_info *multipaths;
int raid_disks;
- int working_disks;
spinlock_t device_lock;
struct list_head retry_list;
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 2b7a7ff401dc..f7431b6d8447 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -297,23 +297,24 @@ static void raid1_end_read_request(struct bio *bio, int error)
rdev_dec_pending(conf->mirrors[mirror].rdev, conf->mddev);
}
-static void r1_bio_write_done(r1bio_t *r1_bio, int vcnt, struct bio_vec *bv,
- int behind)
+static void r1_bio_write_done(r1bio_t *r1_bio)
{
if (atomic_dec_and_test(&r1_bio->remaining))
{
/* it really is the end of this request */
if (test_bit(R1BIO_BehindIO, &r1_bio->state)) {
/* free extra copy of the data pages */
- int i = vcnt;
+ int i = r1_bio->behind_page_count;
while (i--)
- safe_put_page(bv[i].bv_page);
+ safe_put_page(r1_bio->behind_pages[i]);
+ kfree(r1_bio->behind_pages);
+ r1_bio->behind_pages = NULL;
}
/* clear the bitmap if all writes complete successfully */
bitmap_endwrite(r1_bio->mddev->bitmap, r1_bio->sector,
r1_bio->sectors,
!test_bit(R1BIO_Degraded, &r1_bio->state),
- behind);
+ test_bit(R1BIO_BehindIO, &r1_bio->state));
md_write_end(r1_bio->mddev);
raid_end_bio_io(r1_bio);
}
@@ -386,7 +387,7 @@ static void raid1_end_write_request(struct bio *bio, int error)
* Let's see if all mirrored write operations have finished
* already.
*/
- r1_bio_write_done(r1_bio, bio->bi_vcnt, bio->bi_io_vec, behind);
+ r1_bio_write_done(r1_bio);
if (to_put)
bio_put(to_put);
@@ -411,10 +412,10 @@ static int read_balance(conf_t *conf, r1bio_t *r1_bio)
{
const sector_t this_sector = r1_bio->sector;
const int sectors = r1_bio->sectors;
- int new_disk = -1;
int start_disk;
+ int best_disk;
int i;
- sector_t new_distance, current_distance;
+ sector_t best_dist;
mdk_rdev_t *rdev;
int choose_first;
@@ -425,6 +426,8 @@ static int read_balance(conf_t *conf, r1bio_t *r1_bio)
* We take the first readable disk when above the resync window.
*/
retry:
+ best_disk = -1;
+ best_dist = MaxSector;
if (conf->mddev->recovery_cp < MaxSector &&
(this_sector + sectors >= conf->next_resync)) {
choose_first = 1;
@@ -434,8 +437,8 @@ static int read_balance(conf_t *conf, r1bio_t *r1_bio)
start_disk = conf->last_used;
}
- /* make sure the disk is operational */
for (i = 0 ; i < conf->raid_disks ; i++) {
+ sector_t dist;
int disk = start_disk + i;
if (disk >= conf->raid_disks)
disk -= conf->raid_disks;
@@ -443,60 +446,43 @@ static int read_balance(conf_t *conf, r1bio_t *r1_bio)
rdev = rcu_dereference(conf->mirrors[disk].rdev);
if (r1_bio->bios[disk] == IO_BLOCKED
|| rdev == NULL
- || !test_bit(In_sync, &rdev->flags))
+ || test_bit(Faulty, &rdev->flags))
continue;
-
- new_disk = disk;
- if (!test_bit(WriteMostly, &rdev->flags))
- break;
- }
-
- if (new_disk < 0 || choose_first)
- goto rb_out;
-
- /*
- * Don't change to another disk for sequential reads:
- */
- if (conf->next_seq_sect == this_sector)
- goto rb_out;
- if (this_sector == conf->mirrors[new_disk].head_position)
- goto rb_out;
-
- current_distance = abs(this_sector
- - conf->mirrors[new_disk].head_position);
-
- /* look for a better disk - i.e. head is closer */
- start_disk = new_disk;
- for (i = 1; i < conf->raid_disks; i++) {
- int disk = start_disk + 1;
- if (disk >= conf->raid_disks)
- disk -= conf->raid_disks;
-
- rdev = rcu_dereference(conf->mirrors[disk].rdev);
- if (r1_bio->bios[disk] == IO_BLOCKED
- || rdev == NULL
- || !test_bit(In_sync, &rdev->flags)
- || test_bit(WriteMostly, &rdev->flags))
+ if (!test_bit(In_sync, &rdev->flags) &&
+ rdev->recovery_offset < this_sector + sectors)
continue;
-
- if (!atomic_read(&rdev->nr_pending)) {
- new_disk = disk;
+ if (test_bit(WriteMostly, &rdev->flags)) {
+ /* Don't balance among write-mostly, just
+ * use the first as a last resort */
+ if (best_disk < 0)
+ best_disk = disk;
+ continue;
+ }
+ /* This is a reasonable device to use. It might
+ * even be best.
+ */
+ dist = abs(this_sector - conf->mirrors[disk].head_position);
+ if (choose_first
+ /* Don't change to another disk for sequential reads */
+ || conf->next_seq_sect == this_sector
+ || dist == 0
+ /* If device is idle, use it */
+ || atomic_read(&rdev->nr_pending) == 0) {
+ best_disk = disk;
break;
}
- new_distance = abs(this_sector - conf->mirrors[disk].head_position);
- if (new_distance < current_distance) {
- current_distance = new_distance;
- new_disk = disk;
+ if (dist < best_dist) {
+ best_dist = dist;
+ best_disk = disk;
}
}
- rb_out:
- if (new_disk >= 0) {
- rdev = rcu_dereference(conf->mirrors[new_disk].rdev);
+ if (best_disk >= 0) {
+ rdev = rcu_dereference(conf->mirrors[best_disk].rdev);
if (!rdev)
goto retry;
atomic_inc(&rdev->nr_pending);
- if (!test_bit(In_sync, &rdev->flags)) {
+ if (test_bit(Faulty, &rdev->flags)) {
/* cannot risk returning a device that failed
* before we inc'ed nr_pending
*/
@@ -504,28 +490,26 @@ static int read_balance(conf_t *conf, r1bio_t *r1_bio)
goto retry;
}
conf->next_seq_sect = this_sector + sectors;
- conf->last_used = new_disk;
+ conf->last_used = best_disk;
}
rcu_read_unlock();
- return new_disk;
+ return best_disk;
}
-static int raid1_congested(void *data, int bits)
+int md_raid1_congested(mddev_t *mddev, int bits)
{
- mddev_t *mddev = data;
conf_t *conf = mddev->private;
int i, ret = 0;
- if (mddev_congested(mddev, bits))
- return 1;
-
rcu_read_lock();
for (i = 0; i < mddev->raid_disks; i++) {
mdk_rdev_t *rdev = rcu_dereference(conf->mirrors[i].rdev);
if (rdev && !test_bit(Faulty, &rdev->flags)) {
struct request_queue *q = bdev_get_queue(rdev->bdev);
+ BUG_ON(!q);
+
/* Note the '|| 1' - when read_balance prefers
* non-congested targets, it can be removed
*/
@@ -538,7 +522,15 @@ static int raid1_congested(void *data, int bits)
rcu_read_unlock();
return ret;
}
+EXPORT_SYMBOL_GPL(md_raid1_congested);
+static int raid1_congested(void *data, int bits)
+{
+ mddev_t *mddev = data;
+
+ return mddev_congested(mddev, bits) ||
+ md_raid1_congested(mddev, bits);
+}
static void flush_pending_writes(conf_t *conf)
{
@@ -675,37 +667,36 @@ static void unfreeze_array(conf_t *conf)
/* duplicate the data pages for behind I/O
- * We return a list of bio_vec rather than just page pointers
- * as it makes freeing easier
*/
-static struct bio_vec *alloc_behind_pages(struct bio *bio)
+static void alloc_behind_pages(struct bio *bio, r1bio_t *r1_bio)
{
int i;
struct bio_vec *bvec;
- struct bio_vec *pages = kzalloc(bio->bi_vcnt * sizeof(struct bio_vec),
+ struct page **pages = kzalloc(bio->bi_vcnt * sizeof(struct page*),
GFP_NOIO);
if (unlikely(!pages))
- goto do_sync_io;
+ return;
bio_for_each_segment(bvec, bio, i) {
- pages[i].bv_page = alloc_page(GFP_NOIO);
- if (unlikely(!pages[i].bv_page))
+ pages[i] = alloc_page(GFP_NOIO);
+ if (unlikely(!pages[i]))
goto do_sync_io;
- memcpy(kmap(pages[i].bv_page) + bvec->bv_offset,
+ memcpy(kmap(pages[i]) + bvec->bv_offset,
kmap(bvec->bv_page) + bvec->bv_offset, bvec->bv_len);
- kunmap(pages[i].bv_page);
+ kunmap(pages[i]);
kunmap(bvec->bv_page);
}
-
- return pages;
+ r1_bio->behind_pages = pages;
+ r1_bio->behind_page_count = bio->bi_vcnt;
+ set_bit(R1BIO_BehindIO, &r1_bio->state);
+ return;
do_sync_io:
- if (pages)
- for (i = 0; i < bio->bi_vcnt && pages[i].bv_page; i++)
- put_page(pages[i].bv_page);
+ for (i = 0; i < bio->bi_vcnt; i++)
+ if (pages[i])
+ put_page(pages[i]);
kfree(pages);
PRINTK("%dB behind alloc failed, doing sync I/O\n", bio->bi_size);
- return NULL;
}
static int make_request(mddev_t *mddev, struct bio * bio)
@@ -717,7 +708,6 @@ static int make_request(mddev_t *mddev, struct bio * bio)
int i, targets = 0, disks;
struct bitmap *bitmap;
unsigned long flags;
- struct bio_vec *behind_pages = NULL;
const int rw = bio_data_dir(bio);
const unsigned long do_sync = (bio->bi_rw & REQ_SYNC);
const unsigned long do_flush_fua = (bio->bi_rw & (REQ_FLUSH | REQ_FUA));
@@ -870,9 +860,8 @@ static int make_request(mddev_t *mddev, struct bio * bio)
if (bitmap &&
(atomic_read(&bitmap->behind_writes)
< mddev->bitmap_info.max_write_behind) &&
- !waitqueue_active(&bitmap->behind_wait) &&
- (behind_pages = alloc_behind_pages(bio)) != NULL)
- set_bit(R1BIO_BehindIO, &r1_bio->state);
+ !waitqueue_active(&bitmap->behind_wait))
+ alloc_behind_pages(bio, r1_bio);
atomic_set(&r1_bio->remaining, 1);
atomic_set(&r1_bio->behind_remaining, 0);
@@ -893,7 +882,7 @@ static int make_request(mddev_t *mddev, struct bio * bio)
mbio->bi_rw = WRITE | do_flush_fua | do_sync;
mbio->bi_private = r1_bio;
- if (behind_pages) {
+ if (r1_bio->behind_pages) {
struct bio_vec *bvec;
int j;
@@ -905,7 +894,7 @@ static int make_request(mddev_t *mddev, struct bio * bio)
* them all
*/
__bio_for_each_segment(bvec, mbio, j, 0)
- bvec->bv_page = behind_pages[j].bv_page;
+ bvec->bv_page = r1_bio->behind_pages[j];
if (test_bit(WriteMostly, &conf->mirrors[i].rdev->flags))
atomic_inc(&r1_bio->behind_remaining);
}
@@ -915,8 +904,7 @@ static int make_request(mddev_t *mddev, struct bio * bio)
bio_list_add(&conf->pending_bio_list, mbio);
spin_unlock_irqrestore(&conf->device_lock, flags);
}
- r1_bio_write_done(r1_bio, bio->bi_vcnt, behind_pages, behind_pages != NULL);
- kfree(behind_pages); /* the behind pages are attached to the bios now */
+ r1_bio_write_done(r1_bio);
/* In case raid1d snuck in to freeze_array */
wake_up(&conf->wait_barrier);
@@ -1196,194 +1184,210 @@ static void end_sync_write(struct bio *bio, int error)
}
}
-static void sync_request_write(mddev_t *mddev, r1bio_t *r1_bio)
+static int fix_sync_read_error(r1bio_t *r1_bio)
{
+ /* Try some synchronous reads of other devices to get
+ * good data, much like with normal read errors. Only
+ * read into the pages we already have so we don't
+ * need to re-issue the read request.
+ * We don't need to freeze the array, because being in an
+ * active sync request, there is no normal IO, and
+ * no overlapping syncs.
+ */
+ mddev_t *mddev = r1_bio->mddev;
conf_t *conf = mddev->private;
- int i;
- int disks = conf->raid_disks;
- struct bio *bio, *wbio;
-
- bio = r1_bio->bios[r1_bio->read_disk];
+ struct bio *bio = r1_bio->bios[r1_bio->read_disk];
+ sector_t sect = r1_bio->sector;
+ int sectors = r1_bio->sectors;
+ int idx = 0;
+ while(sectors) {
+ int s = sectors;
+ int d = r1_bio->read_disk;
+ int success = 0;
+ mdk_rdev_t *rdev;
+ int start;
- if (test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery)) {
- /* We have read all readable devices. If we haven't
- * got the block, then there is no hope left.
- * If we have, then we want to do a comparison
- * and skip the write if everything is the same.
- * If any blocks failed to read, then we need to
- * attempt an over-write
- */
- int primary;
- if (!test_bit(R1BIO_Uptodate, &r1_bio->state)) {
- for (i=0; i<mddev->raid_disks; i++)
- if (r1_bio->bios[i]->bi_end_io == end_sync_read)
- md_error(mddev, conf->mirrors[i].rdev);
+ if (s > (PAGE_SIZE>>9))
+ s = PAGE_SIZE >> 9;
+ do {
+ if (r1_bio->bios[d]->bi_end_io == end_sync_read) {
+ /* No rcu protection needed here devices
+ * can only be removed when no resync is
+ * active, and resync is currently active
+ */
+ rdev = conf->mirrors[d].rdev;
+ if (sync_page_io(rdev,
+ sect,
+ s<<9,
+ bio->bi_io_vec[idx].bv_page,
+ READ, false)) {
+ success = 1;
+ break;
+ }
+ }
+ d++;
+ if (d == conf->raid_disks)
+ d = 0;
+ } while (!success && d != r1_bio->read_disk);
- md_done_sync(mddev, r1_bio->sectors, 1);
+ if (!success) {
+ char b[BDEVNAME_SIZE];
+ /* Cannot read from anywhere, array is toast */
+ md_error(mddev, conf->mirrors[r1_bio->read_disk].rdev);
+ printk(KERN_ALERT "md/raid1:%s: %s: unrecoverable I/O read error"
+ " for block %llu\n",
+ mdname(mddev),
+ bdevname(bio->bi_bdev, b),
+ (unsigned long long)r1_bio->sector);
+ md_done_sync(mddev, r1_bio->sectors, 0);
put_buf(r1_bio);
- return;
+ return 0;
}
- for (primary=0; primary<mddev->raid_disks; primary++)
- if (r1_bio->bios[primary]->bi_end_io == end_sync_read &&
- test_bit(BIO_UPTODATE, &r1_bio->bios[primary]->bi_flags)) {
- r1_bio->bios[primary]->bi_end_io = NULL;
- rdev_dec_pending(conf->mirrors[primary].rdev, mddev);
- break;
- }
- r1_bio->read_disk = primary;
- for (i=0; i<mddev->raid_disks; i++)
- if (r1_bio->bios[i]->bi_end_io == end_sync_read) {
- int j;
- int vcnt = r1_bio->sectors >> (PAGE_SHIFT- 9);
- struct bio *pbio = r1_bio->bios[primary];
- struct bio *sbio = r1_bio->bios[i];
-
- if (test_bit(BIO_UPTODATE, &sbio->bi_flags)) {
- for (j = vcnt; j-- ; ) {
- struct page *p, *s;
- p = pbio->bi_io_vec[j].bv_page;
- s = sbio->bi_io_vec[j].bv_page;
- if (memcmp(page_address(p),
- page_address(s),
- PAGE_SIZE))
- break;
- }
- } else
- j = 0;
- if (j >= 0)
- mddev->resync_mismatches += r1_bio->sectors;
- if (j < 0 || (test_bit(MD_RECOVERY_CHECK, &mddev->recovery)
- && test_bit(BIO_UPTODATE, &sbio->bi_flags))) {
- sbio->bi_end_io = NULL;
- rdev_dec_pending(conf->mirrors[i].rdev, mddev);
- } else {
- /* fixup the bio for reuse */
- int size;
- sbio->bi_vcnt = vcnt;
- sbio->bi_size = r1_bio->sectors << 9;
- sbio->bi_idx = 0;
- sbio->bi_phys_segments = 0;
- sbio->bi_flags &= ~(BIO_POOL_MASK - 1);
- sbio->bi_flags |= 1 << BIO_UPTODATE;
- sbio->bi_next = NULL;
- sbio->bi_sector = r1_bio->sector +
- conf->mirrors[i].rdev->data_offset;
- sbio->bi_bdev = conf->mirrors[i].rdev->bdev;
- size = sbio->bi_size;
- for (j = 0; j < vcnt ; j++) {
- struct bio_vec *bi;
- bi = &sbio->bi_io_vec[j];
- bi->bv_offset = 0;
- if (size > PAGE_SIZE)
- bi->bv_len = PAGE_SIZE;
- else
- bi->bv_len = size;
- size -= PAGE_SIZE;
- memcpy(page_address(bi->bv_page),
- page_address(pbio->bi_io_vec[j].bv_page),
- PAGE_SIZE);
- }
- }
- }
+ start = d;
+ /* write it back and re-read */
+ while (d != r1_bio->read_disk) {
+ if (d == 0)
+ d = conf->raid_disks;
+ d--;
+ if (r1_bio->bios[d]->bi_end_io != end_sync_read)
+ continue;
+ rdev = conf->mirrors[d].rdev;
+ if (sync_page_io(rdev,
+ sect,
+ s<<9,
+ bio->bi_io_vec[idx].bv_page,
+ WRITE, false) == 0) {
+ r1_bio->bios[d]->bi_end_io = NULL;
+ rdev_dec_pending(rdev, mddev);
+ md_error(mddev, rdev);
+ } else
+ atomic_add(s, &rdev->corrected_errors);
+ }
+ d = start;
+ while (d != r1_bio->read_disk) {
+ if (d == 0)
+ d = conf->raid_disks;
+ d--;
+ if (r1_bio->bios[d]->bi_end_io != end_sync_read)
+ continue;
+ rdev = conf->mirrors[d].rdev;
+ if (sync_page_io(rdev,
+ sect,
+ s<<9,
+ bio->bi_io_vec[idx].bv_page,
+ READ, false) == 0)
+ md_error(mddev, rdev);
+ }
+ sectors -= s;
+ sect += s;
+ idx ++;
}
- if (!test_bit(R1BIO_Uptodate, &r1_bio->state)) {
- /* ouch - failed to read all of that.
- * Try some synchronous reads of other devices to get
- * good data, much like with normal read errors. Only
- * read into the pages we already have so we don't
- * need to re-issue the read request.
- * We don't need to freeze the array, because being in an
- * active sync request, there is no normal IO, and
- * no overlapping syncs.
- */
- sector_t sect = r1_bio->sector;
- int sectors = r1_bio->sectors;
- int idx = 0;
-
- while(sectors) {
- int s = sectors;
- int d = r1_bio->read_disk;
- int success = 0;
- mdk_rdev_t *rdev;
-
- if (s > (PAGE_SIZE>>9))
- s = PAGE_SIZE >> 9;
- do {
- if (r1_bio->bios[d]->bi_end_io == end_sync_read) {
- /* No rcu protection needed here devices
- * can only be removed when no resync is
- * active, and resync is currently active
- */
- rdev = conf->mirrors[d].rdev;
- if (sync_page_io(rdev,
- sect,
- s<<9,
- bio->bi_io_vec[idx].bv_page,
- READ, false)) {
- success = 1;
- break;
- }
- }
- d++;
- if (d == conf->raid_disks)
- d = 0;
- } while (!success && d != r1_bio->read_disk);
-
- if (success) {
- int start = d;
- /* write it back and re-read */
- set_bit(R1BIO_Uptodate, &r1_bio->state);
- while (d != r1_bio->read_disk) {
- if (d == 0)
- d = conf->raid_disks;
- d--;
- if (r1_bio->bios[d]->bi_end_io != end_sync_read)
- continue;
- rdev = conf->mirrors[d].rdev;
- atomic_add(s, &rdev->corrected_errors);
- if (sync_page_io(rdev,
- sect,
- s<<9,
- bio->bi_io_vec[idx].bv_page,
- WRITE, false) == 0)
- md_error(mddev, rdev);
- }
- d = start;
- while (d != r1_bio->read_disk) {
- if (d == 0)
- d = conf->raid_disks;
- d--;
- if (r1_bio->bios[d]->bi_end_io != end_sync_read)
- continue;
- rdev = conf->mirrors[d].rdev;
- if (sync_page_io(rdev,
- sect,
- s<<9,
- bio->bi_io_vec[idx].bv_page,
- READ, false) == 0)
- md_error(mddev, rdev);
- }
- } else {
- char b[BDEVNAME_SIZE];
- /* Cannot read from anywhere, array is toast */
- md_error(mddev, conf->mirrors[r1_bio->read_disk].rdev);
- printk(KERN_ALERT "md/raid1:%s: %s: unrecoverable I/O read error"
- " for block %llu\n",
- mdname(mddev),
- bdevname(bio->bi_bdev, b),
- (unsigned long long)r1_bio->sector);
- md_done_sync(mddev, r1_bio->sectors, 0);
- put_buf(r1_bio);
- return;
+ set_bit(R1BIO_Uptodate, &r1_bio->state);
+ set_bit(BIO_UPTODATE, &bio->bi_flags);
+ return 1;
+}
+
+static int process_checks(r1bio_t *r1_bio)
+{
+ /* We have read all readable devices. If we haven't
+ * got the block, then there is no hope left.
+ * If we have, then we want to do a comparison
+ * and skip the write if everything is the same.
+ * If any blocks failed to read, then we need to
+ * attempt an over-write
+ */
+ mddev_t *mddev = r1_bio->mddev;
+ conf_t *conf = mddev->private;
+ int primary;
+ int i;
+
+ for (primary = 0; primary < conf->raid_disks; primary++)
+ if (r1_bio->bios[primary]->bi_end_io == end_sync_read &&
+ test_bit(BIO_UPTODATE, &r1_bio->bios[primary]->bi_flags)) {
+ r1_bio->bios[primary]->bi_end_io = NULL;
+ rdev_dec_pending(conf->mirrors[primary].rdev, mddev);
+ break;
+ }
+ r1_bio->read_disk = primary;
+ for (i = 0; i < conf->raid_disks; i++) {
+ int j;
+ int vcnt = r1_bio->sectors >> (PAGE_SHIFT- 9);
+ struct bio *pbio = r1_bio->bios[primary];
+ struct bio *sbio = r1_bio->bios[i];
+ int size;
+
+ if (r1_bio->bios[i]->bi_end_io != end_sync_read)
+ continue;
+
+ if (test_bit(BIO_UPTODATE, &sbio->bi_flags)) {
+ for (j = vcnt; j-- ; ) {
+ struct page *p, *s;
+ p = pbio->bi_io_vec[j].bv_page;
+ s = sbio->bi_io_vec[j].bv_page;
+ if (memcmp(page_address(p),
+ page_address(s),
+ PAGE_SIZE))
+ break;
}
- sectors -= s;
- sect += s;
- idx ++;
+ } else
+ j = 0;
+ if (j >= 0)
+ mddev->resync_mismatches += r1_bio->sectors;
+ if (j < 0 || (test_bit(MD_RECOVERY_CHECK, &mddev->recovery)
+ && test_bit(BIO_UPTODATE, &sbio->bi_flags))) {
+ /* No need to write to this device. */
+ sbio->bi_end_io = NULL;
+ rdev_dec_pending(conf->mirrors[i].rdev, mddev);
+ continue;
+ }
+ /* fixup the bio for reuse */
+ sbio->bi_vcnt = vcnt;
+ sbio->bi_size = r1_bio->sectors << 9;
+ sbio->bi_idx = 0;
+ sbio->bi_phys_segments = 0;
+ sbio->bi_flags &= ~(BIO_POOL_MASK - 1);
+ sbio->bi_flags |= 1 << BIO_UPTODATE;
+ sbio->bi_next = NULL;
+ sbio->bi_sector = r1_bio->sector +
+ conf->mirrors[i].rdev->data_offset;
+ sbio->bi_bdev = conf->mirrors[i].rdev->bdev;
+ size = sbio->bi_size;
+ for (j = 0; j < vcnt ; j++) {
+ struct bio_vec *bi;
+ bi = &sbio->bi_io_vec[j];
+ bi->bv_offset = 0;
+ if (size > PAGE_SIZE)
+ bi->bv_len = PAGE_SIZE;
+ else
+ bi->bv_len = size;
+ size -= PAGE_SIZE;
+ memcpy(page_address(bi->bv_page),
+ page_address(pbio->bi_io_vec[j].bv_page),
+ PAGE_SIZE);
}
}
+ return 0;
+}
+
+static void sync_request_write(mddev_t *mddev, r1bio_t *r1_bio)
+{
+ conf_t *conf = mddev->private;
+ int i;
+ int disks = conf->raid_disks;
+ struct bio *bio, *wbio;
+ bio = r1_bio->bios[r1_bio->read_disk];
+
+ if (!test_bit(R1BIO_Uptodate, &r1_bio->state))
+ /* ouch - failed to read all of that. */
+ if (!fix_sync_read_error(r1_bio))
+ return;
+
+ if (test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery))
+ if (process_checks(r1_bio) < 0)
+ return;
/*
* schedule writes
*/
@@ -1974,6 +1978,8 @@ static int run(mddev_t *mddev)
return PTR_ERR(conf);
list_for_each_entry(rdev, &mddev->disks, same_set) {
+ if (!mddev->gendisk)
+ continue;
disk_stack_limits(mddev->gendisk, rdev->bdev,
rdev->data_offset << 9);
/* as we don't honour merge_bvec_fn, we must never risk
@@ -2015,8 +2021,10 @@ static int run(mddev_t *mddev)
md_set_array_sectors(mddev, raid1_size(mddev, 0, 0));
- mddev->queue->backing_dev_info.congested_fn = raid1_congested;
- mddev->queue->backing_dev_info.congested_data = mddev;
+ if (mddev->queue) {
+ mddev->queue->backing_dev_info.congested_fn = raid1_congested;
+ mddev->queue->backing_dev_info.congested_data = mddev;
+ }
return md_integrity_register(mddev);
}
@@ -2063,7 +2071,7 @@ static int raid1_resize(mddev_t *mddev, sector_t sectors)
set_capacity(mddev->gendisk, mddev->array_sectors);
revalidate_disk(mddev->gendisk);
if (sectors > mddev->dev_sectors &&
- mddev->recovery_cp == MaxSector) {
+ mddev->recovery_cp > mddev->dev_sectors) {
mddev->recovery_cp = mddev->dev_sectors;
set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
}
diff --git a/drivers/md/raid1.h b/drivers/md/raid1.h
index cbfdf1a6acd9..e743a64fac4f 100644
--- a/drivers/md/raid1.h
+++ b/drivers/md/raid1.h
@@ -94,7 +94,9 @@ struct r1bio_s {
int read_disk;
struct list_head retry_list;
- struct bitmap_update *bitmap_update;
+ /* Next two are only valid when R1BIO_BehindIO is set */
+ struct page **behind_pages;
+ int behind_page_count;
/*
* if the IO is in WRITE direction, then multiple bios are used.
* We choose the number when they are allocated.
@@ -124,4 +126,6 @@ struct r1bio_s {
*/
#define R1BIO_Returned 6
+extern int md_raid1_congested(mddev_t *mddev, int bits);
+
#endif
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 8e9462626ec5..6e846688962f 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -271,9 +271,10 @@ static void raid10_end_read_request(struct bio *bio, int error)
*/
set_bit(R10BIO_Uptodate, &r10_bio->state);
raid_end_bio_io(r10_bio);
+ rdev_dec_pending(conf->mirrors[dev].rdev, conf->mddev);
} else {
/*
- * oops, read error:
+ * oops, read error - keep the refcount on the rdev
*/
char b[BDEVNAME_SIZE];
if (printk_ratelimit())
@@ -282,8 +283,6 @@ static void raid10_end_read_request(struct bio *bio, int error)
bdevname(conf->mirrors[dev].rdev->bdev,b), (unsigned long long)r10_bio->sector);
reschedule_retry(r10_bio);
}
-
- rdev_dec_pending(conf->mirrors[dev].rdev, conf->mddev);
}
static void raid10_end_write_request(struct bio *bio, int error)
@@ -488,13 +487,19 @@ static int raid10_mergeable_bvec(struct request_queue *q,
static int read_balance(conf_t *conf, r10bio_t *r10_bio)
{
const sector_t this_sector = r10_bio->sector;
- int disk, slot, nslot;
+ int disk, slot;
const int sectors = r10_bio->sectors;
- sector_t new_distance, current_distance;
+ sector_t new_distance, best_dist;
mdk_rdev_t *rdev;
+ int do_balance;
+ int best_slot;
raid10_find_phys(conf, r10_bio);
rcu_read_lock();
+retry:
+ best_slot = -1;
+ best_dist = MaxSector;
+ do_balance = 1;
/*
* Check if we can balance. We can balance on the whole
* device if no resync is going on (recovery is ok), or below
@@ -502,86 +507,58 @@ static int read_balance(conf_t *conf, r10bio_t *r10_bio)
* above the resync window.
*/
if (conf->mddev->recovery_cp < MaxSector
- && (this_sector + sectors >= conf->next_resync)) {
- /* make sure that disk is operational */
- slot = 0;
- disk = r10_bio->devs[slot].devnum;
-
- while ((rdev = rcu_dereference(conf->mirrors[disk].rdev)) == NULL ||
- r10_bio->devs[slot].bio == IO_BLOCKED ||
- !test_bit(In_sync, &rdev->flags)) {
- slot++;
- if (slot == conf->copies) {
- slot = 0;
- disk = -1;
- break;
- }
- disk = r10_bio->devs[slot].devnum;
- }
- goto rb_out;
- }
-
+ && (this_sector + sectors >= conf->next_resync))
+ do_balance = 0;
- /* make sure the disk is operational */
- slot = 0;
- disk = r10_bio->devs[slot].devnum;
- while ((rdev=rcu_dereference(conf->mirrors[disk].rdev)) == NULL ||
- r10_bio->devs[slot].bio == IO_BLOCKED ||
- !test_bit(In_sync, &rdev->flags)) {
- slot ++;
- if (slot == conf->copies) {
- disk = -1;
- goto rb_out;
- }
+ for (slot = 0; slot < conf->copies ; slot++) {
+ if (r10_bio->devs[slot].bio == IO_BLOCKED)
+ continue;
disk = r10_bio->devs[slot].devnum;
- }
-
-
- current_distance = abs(r10_bio->devs[slot].addr -
- conf->mirrors[disk].head_position);
-
- /* Find the disk whose head is closest,
- * or - for far > 1 - find the closest to partition beginning */
-
- for (nslot = slot; nslot < conf->copies; nslot++) {
- int ndisk = r10_bio->devs[nslot].devnum;
-
-
- if ((rdev=rcu_dereference(conf->mirrors[ndisk].rdev)) == NULL ||
- r10_bio->devs[nslot].bio == IO_BLOCKED ||
- !test_bit(In_sync, &rdev->flags))
+ rdev = rcu_dereference(conf->mirrors[disk].rdev);
+ if (rdev == NULL)
continue;
+ if (!test_bit(In_sync, &rdev->flags))
+ continue;
+
+ if (!do_balance)
+ break;
/* This optimisation is debatable, and completely destroys
* sequential read speed for 'far copies' arrays. So only
* keep it for 'near' arrays, and review those later.
*/
- if (conf->near_copies > 1 && !atomic_read(&rdev->nr_pending)) {
- disk = ndisk;
- slot = nslot;
+ if (conf->near_copies > 1 && !atomic_read(&rdev->nr_pending))
break;
- }
/* for far > 1 always use the lowest address */
if (conf->far_copies > 1)
- new_distance = r10_bio->devs[nslot].addr;
+ new_distance = r10_bio->devs[slot].addr;
else
- new_distance = abs(r10_bio->devs[nslot].addr -
- conf->mirrors[ndisk].head_position);
- if (new_distance < current_distance) {
- current_distance = new_distance;
- disk = ndisk;
- slot = nslot;
+ new_distance = abs(r10_bio->devs[slot].addr -
+ conf->mirrors[disk].head_position);
+ if (new_distance < best_dist) {
+ best_dist = new_distance;
+ best_slot = slot;
}
}
+ if (slot == conf->copies)
+ slot = best_slot;
-rb_out:
- r10_bio->read_slot = slot;
-/* conf->next_seq_sect = this_sector + sectors;*/
-
- if (disk >= 0 && (rdev=rcu_dereference(conf->mirrors[disk].rdev))!= NULL)
- atomic_inc(&conf->mirrors[disk].rdev->nr_pending);
- else
+ if (slot >= 0) {
+ disk = r10_bio->devs[slot].devnum;
+ rdev = rcu_dereference(conf->mirrors[disk].rdev);
+ if (!rdev)
+ goto retry;
+ atomic_inc(&rdev->nr_pending);
+ if (test_bit(Faulty, &rdev->flags)) {
+ /* Cannot risk returning a device that failed
+ * before we inc'ed nr_pending
+ */
+ rdev_dec_pending(rdev, conf->mddev);
+ goto retry;
+ }
+ r10_bio->read_slot = slot;
+ } else
disk = -1;
rcu_read_unlock();
@@ -1460,40 +1437,33 @@ static void fix_read_error(conf_t *conf, mddev_t *mddev, r10bio_t *r10_bio)
int max_read_errors = atomic_read(&mddev->max_corr_read_errors);
int d = r10_bio->devs[r10_bio->read_slot].devnum;
- rcu_read_lock();
- rdev = rcu_dereference(conf->mirrors[d].rdev);
- if (rdev) { /* If rdev is not NULL */
- char b[BDEVNAME_SIZE];
- int cur_read_error_count = 0;
+ /* still own a reference to this rdev, so it cannot
+ * have been cleared recently.
+ */
+ rdev = conf->mirrors[d].rdev;
- bdevname(rdev->bdev, b);
+ if (test_bit(Faulty, &rdev->flags))
+ /* drive has already been failed, just ignore any
+ more fix_read_error() attempts */
+ return;
- if (test_bit(Faulty, &rdev->flags)) {
- rcu_read_unlock();
- /* drive has already been failed, just ignore any
- more fix_read_error() attempts */
- return;
- }
+ check_decay_read_errors(mddev, rdev);
+ atomic_inc(&rdev->read_errors);
+ if (atomic_read(&rdev->read_errors) > max_read_errors) {
+ char b[BDEVNAME_SIZE];
+ bdevname(rdev->bdev, b);
- check_decay_read_errors(mddev, rdev);
- atomic_inc(&rdev->read_errors);
- cur_read_error_count = atomic_read(&rdev->read_errors);
- if (cur_read_error_count > max_read_errors) {
- rcu_read_unlock();
- printk(KERN_NOTICE
- "md/raid10:%s: %s: Raid device exceeded "
- "read_error threshold "
- "[cur %d:max %d]\n",
- mdname(mddev),
- b, cur_read_error_count, max_read_errors);
- printk(KERN_NOTICE
- "md/raid10:%s: %s: Failing raid "
- "device\n", mdname(mddev), b);
- md_error(mddev, conf->mirrors[d].rdev);
- return;
- }
+ printk(KERN_NOTICE
+ "md/raid10:%s: %s: Raid device exceeded "
+ "read_error threshold [cur %d:max %d]\n",
+ mdname(mddev), b,
+ atomic_read(&rdev->read_errors), max_read_errors);
+ printk(KERN_NOTICE
+ "md/raid10:%s: %s: Failing raid device\n",
+ mdname(mddev), b);
+ md_error(mddev, conf->mirrors[d].rdev);
+ return;
}
- rcu_read_unlock();
while(sectors) {
int s = sectors;
@@ -1562,8 +1532,8 @@ static void fix_read_error(conf_t *conf, mddev_t *mddev, r10bio_t *r10_bio)
"write failed"
" (%d sectors at %llu on %s)\n",
mdname(mddev), s,
- (unsigned long long)(sect+
- rdev->data_offset),
+ (unsigned long long)(
+ sect + rdev->data_offset),
bdevname(rdev->bdev, b));
printk(KERN_NOTICE "md/raid10:%s: %s: failing "
"drive\n",
@@ -1599,8 +1569,8 @@ static void fix_read_error(conf_t *conf, mddev_t *mddev, r10bio_t *r10_bio)
"corrected sectors"
" (%d sectors at %llu on %s)\n",
mdname(mddev), s,
- (unsigned long long)(sect+
- rdev->data_offset),
+ (unsigned long long)(
+ sect + rdev->data_offset),
bdevname(rdev->bdev, b));
printk(KERN_NOTICE "md/raid10:%s: %s: failing drive\n",
mdname(mddev),
@@ -1612,8 +1582,8 @@ static void fix_read_error(conf_t *conf, mddev_t *mddev, r10bio_t *r10_bio)
"md/raid10:%s: read error corrected"
" (%d sectors at %llu on %s)\n",
mdname(mddev), s,
- (unsigned long long)(sect+
- rdev->data_offset),
+ (unsigned long long)(
+ sect + rdev->data_offset),
bdevname(rdev->bdev, b));
}
@@ -1663,7 +1633,8 @@ static void raid10d(mddev_t *mddev)
else if (test_bit(R10BIO_IsRecover, &r10_bio->state))
recovery_request_write(mddev, r10_bio);
else {
- int mirror;
+ int slot = r10_bio->read_slot;
+ int mirror = r10_bio->devs[slot].devnum;
/* we got a read error. Maybe the drive is bad. Maybe just
* the block and we can fix it.
* We freeze all other IO, and try reading the block from
@@ -1677,9 +1648,10 @@ static void raid10d(mddev_t *mddev)
fix_read_error(conf, mddev, r10_bio);
unfreeze_array(conf);
}
+ rdev_dec_pending(conf->mirrors[mirror].rdev, mddev);
- bio = r10_bio->devs[r10_bio->read_slot].bio;
- r10_bio->devs[r10_bio->read_slot].bio =
+ bio = r10_bio->devs[slot].bio;
+ r10_bio->devs[slot].bio =
mddev->ro ? IO_BLOCKED : NULL;
mirror = read_balance(conf, r10_bio);
if (mirror == -1) {
@@ -1693,6 +1665,7 @@ static void raid10d(mddev_t *mddev)
} else {
const unsigned long do_sync = (r10_bio->master_bio->bi_rw & REQ_SYNC);
bio_put(bio);
+ slot = r10_bio->read_slot;
rdev = conf->mirrors[mirror].rdev;
if (printk_ratelimit())
printk(KERN_ERR "md/raid10:%s: %s: redirecting sector %llu to"
@@ -1702,8 +1675,8 @@ static void raid10d(mddev_t *mddev)
(unsigned long long)r10_bio->sector);
bio = bio_clone_mddev(r10_bio->master_bio,
GFP_NOIO, mddev);
- r10_bio->devs[r10_bio->read_slot].bio = bio;
- bio->bi_sector = r10_bio->devs[r10_bio->read_slot].addr
+ r10_bio->devs[slot].bio = bio;
+ bio->bi_sector = r10_bio->devs[slot].addr
+ rdev->data_offset;
bio->bi_bdev = rdev->bdev;
bio->bi_rw = READ | do_sync;
@@ -1763,13 +1736,13 @@ static int init_resync(conf_t *conf)
*
*/
-static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, int go_faster)
+static sector_t sync_request(mddev_t *mddev, sector_t sector_nr,
+ int *skipped, int go_faster)
{
conf_t *conf = mddev->private;
r10bio_t *r10_bio;
struct bio *biolist = NULL, *bio;
sector_t max_sector, nr_sectors;
- int disk;
int i;
int max_sync;
sector_t sync_blocks;
@@ -1858,108 +1831,114 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i
int j, k;
r10_bio = NULL;
- for (i=0 ; i<conf->raid_disks; i++)
- if (conf->mirrors[i].rdev &&
- !test_bit(In_sync, &conf->mirrors[i].rdev->flags)) {
- int still_degraded = 0;
- /* want to reconstruct this device */
- r10bio_t *rb2 = r10_bio;
- sector_t sect = raid10_find_virt(conf, sector_nr, i);
- int must_sync;
- /* Unless we are doing a full sync, we only need
- * to recover the block if it is set in the bitmap
- */
- must_sync = bitmap_start_sync(mddev->bitmap, sect,
- &sync_blocks, 1);
- if (sync_blocks < max_sync)
- max_sync = sync_blocks;
- if (!must_sync &&
- !conf->fullsync) {
- /* yep, skip the sync_blocks here, but don't assume
- * that there will never be anything to do here
- */
- chunks_skipped = -1;
- continue;
- }
+ for (i=0 ; i<conf->raid_disks; i++) {
+ int still_degraded;
+ r10bio_t *rb2;
+ sector_t sect;
+ int must_sync;
- r10_bio = mempool_alloc(conf->r10buf_pool, GFP_NOIO);
- raise_barrier(conf, rb2 != NULL);
- atomic_set(&r10_bio->remaining, 0);
+ if (conf->mirrors[i].rdev == NULL ||
+ test_bit(In_sync, &conf->mirrors[i].rdev->flags))
+ continue;
- r10_bio->master_bio = (struct bio*)rb2;
- if (rb2)
- atomic_inc(&rb2->remaining);
- r10_bio->mddev = mddev;
- set_bit(R10BIO_IsRecover, &r10_bio->state);
- r10_bio->sector = sect;
+ still_degraded = 0;
+ /* want to reconstruct this device */
+ rb2 = r10_bio;
+ sect = raid10_find_virt(conf, sector_nr, i);
+ /* Unless we are doing a full sync, we only need
+ * to recover the block if it is set in the bitmap
+ */
+ must_sync = bitmap_start_sync(mddev->bitmap, sect,
+ &sync_blocks, 1);
+ if (sync_blocks < max_sync)
+ max_sync = sync_blocks;
+ if (!must_sync &&
+ !conf->fullsync) {
+ /* yep, skip the sync_blocks here, but don't assume
+ * that there will never be anything to do here
+ */
+ chunks_skipped = -1;
+ continue;
+ }
- raid10_find_phys(conf, r10_bio);
+ r10_bio = mempool_alloc(conf->r10buf_pool, GFP_NOIO);
+ raise_barrier(conf, rb2 != NULL);
+ atomic_set(&r10_bio->remaining, 0);
- /* Need to check if the array will still be
- * degraded
- */
- for (j=0; j<conf->raid_disks; j++)
- if (conf->mirrors[j].rdev == NULL ||
- test_bit(Faulty, &conf->mirrors[j].rdev->flags)) {
- still_degraded = 1;
- break;
- }
-
- must_sync = bitmap_start_sync(mddev->bitmap, sect,
- &sync_blocks, still_degraded);
-
- for (j=0; j<conf->copies;j++) {
- int d = r10_bio->devs[j].devnum;
- if (conf->mirrors[d].rdev &&
- test_bit(In_sync, &conf->mirrors[d].rdev->flags)) {
- /* This is where we read from */
- bio = r10_bio->devs[0].bio;
- bio->bi_next = biolist;
- biolist = bio;
- bio->bi_private = r10_bio;
- bio->bi_end_io = end_sync_read;
- bio->bi_rw = READ;
- bio->bi_sector = r10_bio->devs[j].addr +
- conf->mirrors[d].rdev->data_offset;
- bio->bi_bdev = conf->mirrors[d].rdev->bdev;
- atomic_inc(&conf->mirrors[d].rdev->nr_pending);
- atomic_inc(&r10_bio->remaining);
- /* and we write to 'i' */
-
- for (k=0; k<conf->copies; k++)
- if (r10_bio->devs[k].devnum == i)
- break;
- BUG_ON(k == conf->copies);
- bio = r10_bio->devs[1].bio;
- bio->bi_next = biolist;
- biolist = bio;
- bio->bi_private = r10_bio;
- bio->bi_end_io = end_sync_write;
- bio->bi_rw = WRITE;
- bio->bi_sector = r10_bio->devs[k].addr +
- conf->mirrors[i].rdev->data_offset;
- bio->bi_bdev = conf->mirrors[i].rdev->bdev;
-
- r10_bio->devs[0].devnum = d;
- r10_bio->devs[1].devnum = i;
+ r10_bio->master_bio = (struct bio*)rb2;
+ if (rb2)
+ atomic_inc(&rb2->remaining);
+ r10_bio->mddev = mddev;
+ set_bit(R10BIO_IsRecover, &r10_bio->state);
+ r10_bio->sector = sect;
- break;
- }
- }
- if (j == conf->copies) {
- /* Cannot recover, so abort the recovery */
- put_buf(r10_bio);
- if (rb2)
- atomic_dec(&rb2->remaining);
- r10_bio = rb2;
- if (!test_and_set_bit(MD_RECOVERY_INTR,
- &mddev->recovery))
- printk(KERN_INFO "md/raid10:%s: insufficient "
- "working devices for recovery.\n",
- mdname(mddev));
+ raid10_find_phys(conf, r10_bio);
+
+ /* Need to check if the array will still be
+ * degraded
+ */
+ for (j=0; j<conf->raid_disks; j++)
+ if (conf->mirrors[j].rdev == NULL ||
+ test_bit(Faulty, &conf->mirrors[j].rdev->flags)) {
+ still_degraded = 1;
break;
}
+
+ must_sync = bitmap_start_sync(mddev->bitmap, sect,
+ &sync_blocks, still_degraded);
+
+ for (j=0; j<conf->copies;j++) {
+ int d = r10_bio->devs[j].devnum;
+ if (!conf->mirrors[d].rdev ||
+ !test_bit(In_sync, &conf->mirrors[d].rdev->flags))
+ continue;
+ /* This is where we read from */
+ bio = r10_bio->devs[0].bio;
+ bio->bi_next = biolist;
+ biolist = bio;
+ bio->bi_private = r10_bio;
+ bio->bi_end_io = end_sync_read;
+ bio->bi_rw = READ;
+ bio->bi_sector = r10_bio->devs[j].addr +
+ conf->mirrors[d].rdev->data_offset;
+ bio->bi_bdev = conf->mirrors[d].rdev->bdev;
+ atomic_inc(&conf->mirrors[d].rdev->nr_pending);
+ atomic_inc(&r10_bio->remaining);
+ /* and we write to 'i' */
+
+ for (k=0; k<conf->copies; k++)
+ if (r10_bio->devs[k].devnum == i)
+ break;
+ BUG_ON(k == conf->copies);
+ bio = r10_bio->devs[1].bio;
+ bio->bi_next = biolist;
+ biolist = bio;
+ bio->bi_private = r10_bio;
+ bio->bi_end_io = end_sync_write;
+ bio->bi_rw = WRITE;
+ bio->bi_sector = r10_bio->devs[k].addr +
+ conf->mirrors[i].rdev->data_offset;
+ bio->bi_bdev = conf->mirrors[i].rdev->bdev;
+
+ r10_bio->devs[0].devnum = d;
+ r10_bio->devs[1].devnum = i;
+
+ break;
+ }
+ if (j == conf->copies) {
+ /* Cannot recover, so abort the recovery */
+ put_buf(r10_bio);
+ if (rb2)
+ atomic_dec(&rb2->remaining);
+ r10_bio = rb2;
+ if (!test_and_set_bit(MD_RECOVERY_INTR,
+ &mddev->recovery))
+ printk(KERN_INFO "md/raid10:%s: insufficient "
+ "working devices for recovery.\n",
+ mdname(mddev));
+ break;
}
+ }
if (biolist == NULL) {
while (r10_bio) {
r10bio_t *rb2 = r10_bio;
@@ -1977,7 +1956,8 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i
if (!bitmap_start_sync(mddev->bitmap, sector_nr,
&sync_blocks, mddev->degraded) &&
- !conf->fullsync && !test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery)) {
+ !conf->fullsync && !test_bit(MD_RECOVERY_REQUESTED,
+ &mddev->recovery)) {
/* We can skip this block */
*skipped = 1;
return sync_blocks + sectors_skipped;
@@ -2022,7 +2002,8 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i
for (i=0; i<conf->copies; i++) {
int d = r10_bio->devs[i].devnum;
if (r10_bio->devs[i].bio->bi_end_io)
- rdev_dec_pending(conf->mirrors[d].rdev, mddev);
+ rdev_dec_pending(conf->mirrors[d].rdev,
+ mddev);
}
put_buf(r10_bio);
biolist = NULL;
@@ -2047,26 +2028,27 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i
do {
struct page *page;
int len = PAGE_SIZE;
- disk = 0;
if (sector_nr + (len>>9) > max_sector)
len = (max_sector - sector_nr) << 9;
if (len == 0)
break;
for (bio= biolist ; bio ; bio=bio->bi_next) {
+ struct bio *bio2;
page = bio->bi_io_vec[bio->bi_vcnt].bv_page;
- if (bio_add_page(bio, page, len, 0) == 0) {
- /* stop here */
- struct bio *bio2;
- bio->bi_io_vec[bio->bi_vcnt].bv_page = page;
- for (bio2 = biolist; bio2 && bio2 != bio; bio2 = bio2->bi_next) {
- /* remove last page from this bio */
- bio2->bi_vcnt--;
- bio2->bi_size -= len;
- bio2->bi_flags &= ~(1<< BIO_SEG_VALID);
- }
- goto bio_full;
+ if (bio_add_page(bio, page, len, 0))
+ continue;
+
+ /* stop here */
+ bio->bi_io_vec[bio->bi_vcnt].bv_page = page;
+ for (bio2 = biolist;
+ bio2 && bio2 != bio;
+ bio2 = bio2->bi_next) {
+ /* remove last page from this bio */
+ bio2->bi_vcnt--;
+ bio2->bi_size -= len;
+ bio2->bi_flags &= ~(1<< BIO_SEG_VALID);
}
- disk = i;
+ goto bio_full;
}
nr_sectors += len>>9;
sector_nr += len>>9;
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index f301e6ae220c..b72edf35ec54 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -129,7 +129,7 @@ static inline int raid5_dec_bi_hw_segments(struct bio *bio)
static inline void raid5_set_bi_hw_segments(struct bio *bio, unsigned int cnt)
{
- bio->bi_phys_segments = raid5_bi_phys_segments(bio) || (cnt << 16);
+ bio->bi_phys_segments = raid5_bi_phys_segments(bio) | (cnt << 16);
}
/* Find first data disk in a raid6 stripe */
@@ -514,7 +514,7 @@ static void ops_run_io(struct stripe_head *sh, struct stripe_head_state *s)
bi = &sh->dev[i].req;
bi->bi_rw = rw;
- if (rw == WRITE)
+ if (rw & WRITE)
bi->bi_end_io = raid5_end_write_request;
else
bi->bi_end_io = raid5_end_read_request;
@@ -548,13 +548,13 @@ static void ops_run_io(struct stripe_head *sh, struct stripe_head_state *s)
bi->bi_io_vec[0].bv_offset = 0;
bi->bi_size = STRIPE_SIZE;
bi->bi_next = NULL;
- if (rw == WRITE &&
+ if ((rw & WRITE) &&
test_bit(R5_ReWrite, &sh->dev[i].flags))
atomic_add(STRIPE_SECTORS,
&rdev->corrected_errors);
generic_make_request(bi);
} else {
- if (rw == WRITE)
+ if (rw & WRITE)
set_bit(STRIPE_DEGRADED, &sh->state);
pr_debug("skip op %ld on disc %d for sector %llu\n",
bi->bi_rw, i, (unsigned long long)sh->sector);
@@ -585,7 +585,7 @@ async_copy_data(int frombio, struct bio *bio, struct page *page,
init_async_submit(&submit, flags, tx, NULL, NULL, NULL);
bio_for_each_segment(bvl, bio, i) {
- int len = bio_iovec_idx(bio, i)->bv_len;
+ int len = bvl->bv_len;
int clen;
int b_offset = 0;
@@ -601,8 +601,8 @@ async_copy_data(int frombio, struct bio *bio, struct page *page,
clen = len;
if (clen > 0) {
- b_offset += bio_iovec_idx(bio, i)->bv_offset;
- bio_page = bio_iovec_idx(bio, i)->bv_page;
+ b_offset += bvl->bv_offset;
+ bio_page = bvl->bv_page;
if (frombio)
tx = async_memcpy(page, bio_page, page_offset,
b_offset, clen, &submit);
@@ -1700,27 +1700,25 @@ static void error(mddev_t *mddev, mdk_rdev_t *rdev)
raid5_conf_t *conf = mddev->private;
pr_debug("raid456: error called\n");
- if (!test_bit(Faulty, &rdev->flags)) {
- set_bit(MD_CHANGE_DEVS, &mddev->flags);
- if (test_and_clear_bit(In_sync, &rdev->flags)) {
- unsigned long flags;
- spin_lock_irqsave(&conf->device_lock, flags);
- mddev->degraded++;
- spin_unlock_irqrestore(&conf->device_lock, flags);
- /*
- * if recovery was running, make sure it aborts.
- */
- set_bit(MD_RECOVERY_INTR, &mddev->recovery);
- }
- set_bit(Faulty, &rdev->flags);
- printk(KERN_ALERT
- "md/raid:%s: Disk failure on %s, disabling device.\n"
- "md/raid:%s: Operation continuing on %d devices.\n",
- mdname(mddev),
- bdevname(rdev->bdev, b),
- mdname(mddev),
- conf->raid_disks - mddev->degraded);
+ if (test_and_clear_bit(In_sync, &rdev->flags)) {
+ unsigned long flags;
+ spin_lock_irqsave(&conf->device_lock, flags);
+ mddev->degraded++;
+ spin_unlock_irqrestore(&conf->device_lock, flags);
+ /*
+ * if recovery was running, make sure it aborts.
+ */
+ set_bit(MD_RECOVERY_INTR, &mddev->recovery);
}
+ set_bit(Faulty, &rdev->flags);
+ set_bit(MD_CHANGE_DEVS, &mddev->flags);
+ printk(KERN_ALERT
+ "md/raid:%s: Disk failure on %s, disabling device.\n"
+ "md/raid:%s: Operation continuing on %d devices.\n",
+ mdname(mddev),
+ bdevname(rdev->bdev, b),
+ mdname(mddev),
+ conf->raid_disks - mddev->degraded);
}
/*
@@ -3960,7 +3958,7 @@ static int make_request(mddev_t *mddev, struct bio * bi)
/* spinlock is needed as reshape_progress may be
* 64bit on a 32bit platform, and so it might be
* possible to see a half-updated value
- * Ofcourse reshape_progress could change after
+ * Of course reshape_progress could change after
* the lock is dropped, so once we get a reference
* to the stripe that we think it is, we will have
* to check again.
@@ -4860,7 +4858,7 @@ static raid5_conf_t *setup_conf(mddev_t *mddev)
printk(KERN_INFO "md/raid:%s: device %s operational as raid"
" disk %d\n",
mdname(mddev), bdevname(rdev->bdev, b), raid_disk);
- } else
+ } else if (rdev->saved_raid_disk != raid_disk)
/* Cannot rely on bitmap to complete recovery */
conf->fullsync = 1;
}
@@ -5151,7 +5149,6 @@ static int run(mddev_t *mddev)
mddev->queue->backing_dev_info.congested_data = mddev;
mddev->queue->backing_dev_info.congested_fn = raid5_congested;
- mddev->queue->queue_lock = &conf->device_lock;
chunk_size = mddev->chunk_sectors << 9;
blk_queue_io_min(mddev->queue, chunk_size);
@@ -5392,7 +5389,8 @@ static int raid5_resize(mddev_t *mddev, sector_t sectors)
return -EINVAL;
set_capacity(mddev->gendisk, mddev->array_sectors);
revalidate_disk(mddev->gendisk);
- if (sectors > mddev->dev_sectors && mddev->recovery_cp == MaxSector) {
+ if (sectors > mddev->dev_sectors &&
+ mddev->recovery_cp > mddev->dev_sectors) {
mddev->recovery_cp = mddev->dev_sectors;
set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
}
@@ -5679,6 +5677,7 @@ static void raid5_quiesce(mddev_t *mddev, int state)
static void *raid45_takeover_raid0(mddev_t *mddev, int level)
{
struct raid0_private_data *raid0_priv = mddev->private;
+ sector_t sectors;
/* for raid0 takeover only one zone is supported */
if (raid0_priv->nr_strip_zones > 1) {
@@ -5687,6 +5686,9 @@ static void *raid45_takeover_raid0(mddev_t *mddev, int level)
return ERR_PTR(-EINVAL);
}
+ sectors = raid0_priv->strip_zone[0].zone_end;
+ sector_div(sectors, raid0_priv->strip_zone[0].nb_dev);
+ mddev->dev_sectors = sectors;
mddev->new_level = level;
mddev->new_layout = ALGORITHM_PARITY_N;
mddev->new_chunk_sectors = mddev->chunk_sectors;
generated by cgit v1.2.3 (git 2.39.0) at 2024-11-15 12:46:29 +0300