* Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*/
+#include "dm-bio-list.h"
#include <linux/raid/raid10.h>
+#include <linux/raid/bitmap.h>
/*
* RAID10 provides a combination of RAID0 and RAID1 functionality.
* raid_disks
* near_copies (stored in low byte of layout)
* far_copies (stored in second byte of layout)
+ * far_offset (stored in bit 16 of layout )
*
* The data to be stored is divided into chunks using chunksize.
* Each device is divided into far_copies sections.
* near_copies copies of each chunk is stored (each on a different drive).
* The starting device for each section is offset near_copies from the starting
* device of the previous section.
- * Thus there are (near_copies*far_copies) of each chunk, and each is on a different
+ * Thus they are (near_copies*far_copies) of each chunk, and each is on a different
* drive.
* near_copies and far_copies must be at least one, and their product is at most
* raid_disks.
+ *
+ * If far_offset is true, then the far_copies are handled a bit differently.
+ * The copies are still in different stripes, but instead of be very far apart
+ * on disk, there are adjacent stripes.
*/
/*
static void unplug_slaves(mddev_t *mddev);
-static void * r10bio_pool_alloc(unsigned int __nocast gfp_flags, void *data)
+static void allow_barrier(conf_t *conf);
+static void lower_barrier(conf_t *conf);
+
+static void * r10bio_pool_alloc(gfp_t gfp_flags, void *data)
{
conf_t *conf = data;
r10bio_t *r10_bio;
int size = offsetof(struct r10bio_s, devs[conf->copies]);
/* allocate a r10bio with room for raid_disks entries in the bios array */
- r10_bio = kmalloc(size, gfp_flags);
- if (r10_bio)
- memset(r10_bio, 0, size);
- else
+ r10_bio = kzalloc(size, gfp_flags);
+ if (!r10_bio)
unplug_slaves(conf->mddev);
return r10_bio;
* one for write (we recover only one drive per r10buf)
*
*/
-static void * r10buf_pool_alloc(unsigned int __nocast gfp_flags, void *data)
+static void * r10buf_pool_alloc(gfp_t gfp_flags, void *data)
{
conf_t *conf = data;
struct page *page;
out_free_pages:
for ( ; i > 0 ; i--)
- __free_page(bio->bi_io_vec[i-1].bv_page);
+ safe_put_page(bio->bi_io_vec[i-1].bv_page);
while (j--)
for (i = 0; i < RESYNC_PAGES ; i++)
- __free_page(r10_bio->devs[j].bio->bi_io_vec[i].bv_page);
+ safe_put_page(r10_bio->devs[j].bio->bi_io_vec[i].bv_page);
j = -1;
out_free_bio:
while ( ++j < nalloc )
struct bio *bio = r10bio->devs[j].bio;
if (bio) {
for (i = 0; i < RESYNC_PAGES; i++) {
- __free_page(bio->bi_io_vec[i].bv_page);
+ safe_put_page(bio->bi_io_vec[i].bv_page);
bio->bi_io_vec[i].bv_page = NULL;
}
bio_put(bio);
for (i = 0; i < conf->copies; i++) {
struct bio **bio = & r10_bio->devs[i].bio;
- if (*bio)
+ if (*bio && *bio != IO_BLOCKED)
bio_put(*bio);
*bio = NULL;
}
}
-static inline void free_r10bio(r10bio_t *r10_bio)
+static void free_r10bio(r10bio_t *r10_bio)
{
- unsigned long flags;
-
conf_t *conf = mddev_to_conf(r10_bio->mddev);
/*
* Wake up any possible resync thread that waits for the device
* to go idle.
*/
- spin_lock_irqsave(&conf->resync_lock, flags);
- if (!--conf->nr_pending) {
- wake_up(&conf->wait_idle);
- wake_up(&conf->wait_resume);
- }
- spin_unlock_irqrestore(&conf->resync_lock, flags);
+ allow_barrier(conf);
put_all_bios(conf, r10_bio);
mempool_free(r10_bio, conf->r10bio_pool);
}
-static inline void put_buf(r10bio_t *r10_bio)
+static void put_buf(r10bio_t *r10_bio)
{
conf_t *conf = mddev_to_conf(r10_bio->mddev);
- unsigned long flags;
mempool_free(r10_bio, conf->r10buf_pool);
- spin_lock_irqsave(&conf->resync_lock, flags);
- if (!conf->barrier)
- BUG();
- --conf->barrier;
- wake_up(&conf->wait_resume);
- wake_up(&conf->wait_idle);
-
- if (!--conf->nr_pending) {
- wake_up(&conf->wait_idle);
- wake_up(&conf->wait_resume);
- }
- spin_unlock_irqrestore(&conf->resync_lock, flags);
+ lower_barrier(conf);
}
static void reschedule_retry(r10bio_t *r10_bio)
spin_lock_irqsave(&conf->device_lock, flags);
list_add(&r10_bio->retry_list, &conf->retry_list);
+ conf->nr_queued ++;
spin_unlock_irqrestore(&conf->device_lock, flags);
md_wakeup_thread(mddev->thread);
{
struct bio *bio = r10_bio->master_bio;
- bio_endio(bio, bio->bi_size,
+ bio_endio(bio,
test_bit(R10BIO_Uptodate, &r10_bio->state) ? 0 : -EIO);
free_r10bio(r10_bio);
}
r10_bio->devs[slot].addr + (r10_bio->sectors);
}
-static int raid10_end_read_request(struct bio *bio, unsigned int bytes_done, int error)
+static void raid10_end_read_request(struct bio *bio, int error)
{
int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
r10bio_t * r10_bio = (r10bio_t *)(bio->bi_private);
int slot, dev;
conf_t *conf = mddev_to_conf(r10_bio->mddev);
- if (bio->bi_size)
- return 1;
slot = r10_bio->read_slot;
dev = r10_bio->devs[slot].devnum;
/*
* this branch is our 'one mirror IO has finished' event handler:
*/
- if (!uptodate)
- md_error(r10_bio->mddev, conf->mirrors[dev].rdev);
- else
+ update_head_pos(slot, r10_bio);
+
+ if (uptodate) {
/*
* Set R10BIO_Uptodate in our master bio, so that
* we will return a good error code to the higher
* wait for the 'master' bio.
*/
set_bit(R10BIO_Uptodate, &r10_bio->state);
-
- update_head_pos(slot, r10_bio);
-
- /*
- * we have only one bio on the read side
- */
- if (uptodate)
raid_end_bio_io(r10_bio);
- else {
+ } else {
/*
* oops, read error:
*/
}
rdev_dec_pending(conf->mirrors[dev].rdev, conf->mddev);
- return 0;
}
-static int raid10_end_write_request(struct bio *bio, unsigned int bytes_done, int error)
+static void raid10_end_write_request(struct bio *bio, int error)
{
int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
r10bio_t * r10_bio = (r10bio_t *)(bio->bi_private);
int slot, dev;
conf_t *conf = mddev_to_conf(r10_bio->mddev);
- if (bio->bi_size)
- return 1;
-
for (slot = 0; slot < conf->copies; slot++)
if (r10_bio->devs[slot].bio == bio)
break;
/*
* this branch is our 'one mirror IO has finished' event handler:
*/
- if (!uptodate)
+ if (!uptodate) {
md_error(r10_bio->mddev, conf->mirrors[dev].rdev);
- else
+ /* an I/O failed, we can't clear the bitmap */
+ set_bit(R10BIO_Degraded, &r10_bio->state);
+ } else
/*
* Set R10BIO_Uptodate in our master bio, so that
* we will return a good error code for to the higher
* already.
*/
if (atomic_dec_and_test(&r10_bio->remaining)) {
+ /* clear the bitmap if all writes complete successfully */
+ bitmap_endwrite(r10_bio->mddev->bitmap, r10_bio->sector,
+ r10_bio->sectors,
+ !test_bit(R10BIO_Degraded, &r10_bio->state),
+ 0);
md_write_end(r10_bio->mddev);
raid_end_bio_io(r10_bio);
}
rdev_dec_pending(conf->mirrors[dev].rdev, conf->mddev);
- return 0;
}
* With this layout, and block is never stored twice on the one device.
*
* raid10_find_phys finds the sector offset of a given virtual sector
- * on each device that it is on. If a block isn't on a device,
- * that entry in the array is set to MaxSector.
+ * on each device that it is on.
*
* raid10_find_virt does the reverse mapping, from a device and a
* sector offset to a virtual address
chunk *= conf->near_copies;
stripe = chunk;
dev = sector_div(stripe, conf->raid_disks);
+ if (conf->far_offset)
+ stripe *= conf->far_copies;
sector += stripe << conf->chunk_shift;
{
sector_t offset, chunk, vchunk;
- while (sector > conf->stride) {
- sector -= conf->stride;
- if (dev < conf->near_copies)
- dev += conf->raid_disks - conf->near_copies;
- else
- dev -= conf->near_copies;
- }
-
offset = sector & conf->chunk_mask;
- chunk = sector >> conf->chunk_shift;
+ if (conf->far_offset) {
+ int fc;
+ chunk = sector >> conf->chunk_shift;
+ fc = sector_div(chunk, conf->far_copies);
+ dev -= fc * conf->near_copies;
+ if (dev < 0)
+ dev += conf->raid_disks;
+ } else {
+ while (sector >= conf->stride) {
+ sector -= conf->stride;
+ if (dev < conf->near_copies)
+ dev += conf->raid_disks - conf->near_copies;
+ else
+ dev -= conf->near_copies;
+ }
+ chunk = sector >> conf->chunk_shift;
+ }
vchunk = chunk * conf->raid_disks + dev;
sector_div(vchunk, conf->near_copies);
return (vchunk << conf->chunk_shift) + offset;
* If near_copies == raid_disk, there are no striping issues,
* but in that case, the function isn't called at all.
*/
-static int raid10_mergeable_bvec(request_queue_t *q, struct bio *bio,
+static int raid10_mergeable_bvec(struct request_queue *q, struct bio *bio,
struct bio_vec *bio_vec)
{
mddev_t *mddev = q->queuedata;
int disk, slot, nslot;
const int sectors = r10_bio->sectors;
sector_t new_distance, current_distance;
+ mdk_rdev_t *rdev;
raid10_find_phys(conf, r10_bio);
rcu_read_lock();
/*
* Check if we can balance. We can balance on the whole
- * device if no resync is going on, or below the resync window.
- * We take the first readable disk when above the resync window.
+ * device if no resync is going on (recovery is ok), or below
+ * the resync window. We take the first readable disk when
+ * above the resync window.
*/
if (conf->mddev->recovery_cp < MaxSector
&& (this_sector + sectors >= conf->next_resync)) {
slot = 0;
disk = r10_bio->devs[slot].devnum;
- while (!conf->mirrors[disk].rdev ||
- !conf->mirrors[disk].rdev->in_sync) {
+ 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;
/* make sure the disk is operational */
slot = 0;
disk = r10_bio->devs[slot].devnum;
- while (!conf->mirrors[disk].rdev ||
- !conf->mirrors[disk].rdev->in_sync) {
+ 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;
}
- current_distance = abs(this_sector - conf->mirrors[disk].head_position);
+ current_distance = abs(r10_bio->devs[slot].addr -
+ conf->mirrors[disk].head_position);
- /* Find the disk whose head is closest */
+ /* 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 (!conf->mirrors[ndisk].rdev ||
- !conf->mirrors[ndisk].rdev->in_sync)
+ if ((rdev=rcu_dereference(conf->mirrors[ndisk].rdev)) == NULL ||
+ r10_bio->devs[nslot].bio == IO_BLOCKED ||
+ !test_bit(In_sync, &rdev->flags))
continue;
- if (!atomic_read(&conf->mirrors[ndisk].rdev->nr_pending)) {
+ /* 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;
break;
}
- new_distance = abs(r10_bio->devs[nslot].addr -
- conf->mirrors[ndisk].head_position);
+
+ /* for far > 1 always use the lowest address */
+ if (conf->far_copies > 1)
+ new_distance = r10_bio->devs[nslot].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;
r10_bio->read_slot = slot;
/* conf->next_seq_sect = this_sector + sectors;*/
- if (disk >= 0 && conf->mirrors[disk].rdev)
+ if (disk >= 0 && (rdev=rcu_dereference(conf->mirrors[disk].rdev))!= NULL)
atomic_inc(&conf->mirrors[disk].rdev->nr_pending);
+ else
+ disk = -1;
rcu_read_unlock();
return disk;
rcu_read_lock();
for (i=0; i<mddev->raid_disks; i++) {
- mdk_rdev_t *rdev = conf->mirrors[i].rdev;
- if (rdev && !rdev->faulty && atomic_read(&rdev->nr_pending)) {
- request_queue_t *r_queue = bdev_get_queue(rdev->bdev);
+ mdk_rdev_t *rdev = rcu_dereference(conf->mirrors[i].rdev);
+ if (rdev && !test_bit(Faulty, &rdev->flags) && atomic_read(&rdev->nr_pending)) {
+ struct request_queue *r_queue = bdev_get_queue(rdev->bdev);
atomic_inc(&rdev->nr_pending);
rcu_read_unlock();
- if (r_queue->unplug_fn)
- r_queue->unplug_fn(r_queue);
+ blk_unplug(r_queue);
rdev_dec_pending(rdev, mddev);
rcu_read_lock();
rcu_read_unlock();
}
-static void raid10_unplug(request_queue_t *q)
+static void raid10_unplug(struct request_queue *q)
{
+ mddev_t *mddev = q->queuedata;
+
unplug_slaves(q->queuedata);
+ md_wakeup_thread(mddev->thread);
}
-static int raid10_issue_flush(request_queue_t *q, struct gendisk *disk,
- sector_t *error_sector)
+static int raid10_congested(void *data, int bits)
{
- mddev_t *mddev = q->queuedata;
+ mddev_t *mddev = data;
conf_t *conf = mddev_to_conf(mddev);
int i, ret = 0;
rcu_read_lock();
- for (i=0; i<mddev->raid_disks && ret == 0; i++) {
- mdk_rdev_t *rdev = conf->mirrors[i].rdev;
- if (rdev && !rdev->faulty) {
- struct block_device *bdev = rdev->bdev;
- request_queue_t *r_queue = bdev_get_queue(bdev);
-
- if (!r_queue->issue_flush_fn)
- ret = -EOPNOTSUPP;
- else {
- atomic_inc(&rdev->nr_pending);
- rcu_read_unlock();
- ret = r_queue->issue_flush_fn(r_queue, bdev->bd_disk,
- error_sector);
- rdev_dec_pending(rdev, mddev);
- rcu_read_lock();
- }
+ for (i = 0; i < mddev->raid_disks && ret == 0; 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);
+
+ ret |= bdi_congested(&q->backing_dev_info, bits);
}
}
rcu_read_unlock();
return ret;
}
-/*
- * Throttle resync depth, so that we can both get proper overlapping of
- * requests, but are still able to handle normal requests quickly.
+static int flush_pending_writes(conf_t *conf)
+{
+ /* Any writes that have been queued but are awaiting
+ * bitmap updates get flushed here.
+ * We return 1 if any requests were actually submitted.
+ */
+ int rv = 0;
+
+ spin_lock_irq(&conf->device_lock);
+
+ if (conf->pending_bio_list.head) {
+ struct bio *bio;
+ bio = bio_list_get(&conf->pending_bio_list);
+ blk_remove_plug(conf->mddev->queue);
+ spin_unlock_irq(&conf->device_lock);
+ /* flush any pending bitmap writes to disk
+ * before proceeding w/ I/O */
+ bitmap_unplug(conf->mddev->bitmap);
+
+ while (bio) { /* submit pending writes */
+ struct bio *next = bio->bi_next;
+ bio->bi_next = NULL;
+ generic_make_request(bio);
+ bio = next;
+ }
+ rv = 1;
+ } else
+ spin_unlock_irq(&conf->device_lock);
+ return rv;
+}
+/* Barriers....
+ * Sometimes we need to suspend IO while we do something else,
+ * either some resync/recovery, or reconfigure the array.
+ * To do this we raise a 'barrier'.
+ * The 'barrier' is a counter that can be raised multiple times
+ * to count how many activities are happening which preclude
+ * normal IO.
+ * We can only raise the barrier if there is no pending IO.
+ * i.e. if nr_pending == 0.
+ * We choose only to raise the barrier if no-one is waiting for the
+ * barrier to go down. This means that as soon as an IO request
+ * is ready, no other operations which require a barrier will start
+ * until the IO request has had a chance.
+ *
+ * So: regular IO calls 'wait_barrier'. When that returns there
+ * is no backgroup IO happening, It must arrange to call
+ * allow_barrier when it has finished its IO.
+ * backgroup IO calls must call raise_barrier. Once that returns
+ * there is no normal IO happeing. It must arrange to call
+ * lower_barrier when the particular background IO completes.
*/
#define RESYNC_DEPTH 32
-static void device_barrier(conf_t *conf, sector_t sect)
+static void raise_barrier(conf_t *conf, int force)
{
+ BUG_ON(force && !conf->barrier);
spin_lock_irq(&conf->resync_lock);
- wait_event_lock_irq(conf->wait_idle, !waitqueue_active(&conf->wait_resume),
- conf->resync_lock, unplug_slaves(conf->mddev));
-
- if (!conf->barrier++) {
- wait_event_lock_irq(conf->wait_idle, !conf->nr_pending,
- conf->resync_lock, unplug_slaves(conf->mddev));
- if (conf->nr_pending)
- BUG();
+
+ /* Wait until no block IO is waiting (unless 'force') */
+ wait_event_lock_irq(conf->wait_barrier, force || !conf->nr_waiting,
+ conf->resync_lock,
+ raid10_unplug(conf->mddev->queue));
+
+ /* block any new IO from starting */
+ conf->barrier++;
+
+ /* No wait for all pending IO to complete */
+ wait_event_lock_irq(conf->wait_barrier,
+ !conf->nr_pending && conf->barrier < RESYNC_DEPTH,
+ conf->resync_lock,
+ raid10_unplug(conf->mddev->queue));
+
+ spin_unlock_irq(&conf->resync_lock);
+}
+
+static void lower_barrier(conf_t *conf)
+{
+ unsigned long flags;
+ spin_lock_irqsave(&conf->resync_lock, flags);
+ conf->barrier--;
+ spin_unlock_irqrestore(&conf->resync_lock, flags);
+ wake_up(&conf->wait_barrier);
+}
+
+static void wait_barrier(conf_t *conf)
+{
+ spin_lock_irq(&conf->resync_lock);
+ if (conf->barrier) {
+ conf->nr_waiting++;
+ wait_event_lock_irq(conf->wait_barrier, !conf->barrier,
+ conf->resync_lock,
+ raid10_unplug(conf->mddev->queue));
+ conf->nr_waiting--;
}
- wait_event_lock_irq(conf->wait_resume, conf->barrier < RESYNC_DEPTH,
- conf->resync_lock, unplug_slaves(conf->mddev));
- conf->next_resync = sect;
+ conf->nr_pending++;
+ spin_unlock_irq(&conf->resync_lock);
+}
+
+static void allow_barrier(conf_t *conf)
+{
+ unsigned long flags;
+ spin_lock_irqsave(&conf->resync_lock, flags);
+ conf->nr_pending--;
+ spin_unlock_irqrestore(&conf->resync_lock, flags);
+ wake_up(&conf->wait_barrier);
+}
+
+static void freeze_array(conf_t *conf)
+{
+ /* stop syncio and normal IO and wait for everything to
+ * go quiet.
+ * We increment barrier and nr_waiting, and then
+ * wait until nr_pending match nr_queued+1
+ * This is called in the context of one normal IO request
+ * that has failed. Thus any sync request that might be pending
+ * will be blocked by nr_pending, and we need to wait for
+ * pending IO requests to complete or be queued for re-try.
+ * Thus the number queued (nr_queued) plus this request (1)
+ * must match the number of pending IOs (nr_pending) before
+ * we continue.
+ */
+ spin_lock_irq(&conf->resync_lock);
+ conf->barrier++;
+ conf->nr_waiting++;
+ wait_event_lock_irq(conf->wait_barrier,
+ conf->nr_pending == conf->nr_queued+1,
+ conf->resync_lock,
+ ({ flush_pending_writes(conf);
+ raid10_unplug(conf->mddev->queue); }));
+ spin_unlock_irq(&conf->resync_lock);
+}
+
+static void unfreeze_array(conf_t *conf)
+{
+ /* reverse the effect of the freeze */
+ spin_lock_irq(&conf->resync_lock);
+ conf->barrier--;
+ conf->nr_waiting--;
+ wake_up(&conf->wait_barrier);
spin_unlock_irq(&conf->resync_lock);
}
-static int make_request(request_queue_t *q, struct bio * bio)
+static int make_request(struct request_queue *q, struct bio * bio)
{
mddev_t *mddev = q->queuedata;
conf_t *conf = mddev_to_conf(mddev);
struct bio *read_bio;
int i;
int chunk_sects = conf->chunk_mask + 1;
+ const int rw = bio_data_dir(bio);
+ const int do_sync = bio_sync(bio);
+ struct bio_list bl;
+ unsigned long flags;
+ mdk_rdev_t *blocked_rdev;
+
+ if (unlikely(bio_barrier(bio))) {
+ bio_endio(bio, -EOPNOTSUPP);
+ return 0;
+ }
/* If this request crosses a chunk boundary, we need to
* split it. This will only happen for 1 PAGE (or less) requests.
" or bigger than %dk %llu %d\n", chunk_sects/2,
(unsigned long long)bio->bi_sector, bio->bi_size >> 10);
- bio_io_error(bio, bio->bi_size);
+ bio_io_error(bio);
return 0;
}
+ md_write_start(mddev, bio);
+
/*
* Register the new request and wait if the reconstruction
* thread has put up a bar for new requests.
* Continue immediately if no resync is active currently.
*/
- spin_lock_irq(&conf->resync_lock);
- wait_event_lock_irq(conf->wait_resume, !conf->barrier, conf->resync_lock, );
- conf->nr_pending++;
- spin_unlock_irq(&conf->resync_lock);
+ wait_barrier(conf);
- if (bio_data_dir(bio)==WRITE) {
- disk_stat_inc(mddev->gendisk, writes);
- disk_stat_add(mddev->gendisk, write_sectors, bio_sectors(bio));
- } else {
- disk_stat_inc(mddev->gendisk, reads);
- disk_stat_add(mddev->gendisk, read_sectors, bio_sectors(bio));
- }
+ disk_stat_inc(mddev->gendisk, ios[rw]);
+ disk_stat_add(mddev->gendisk, sectors[rw], bio_sectors(bio));
r10_bio = mempool_alloc(conf->r10bio_pool, GFP_NOIO);
r10_bio->mddev = mddev;
r10_bio->sector = bio->bi_sector;
+ r10_bio->state = 0;
- if (bio_data_dir(bio) == READ) {
+ if (rw == READ) {
/*
* read balancing logic:
*/
mirror->rdev->data_offset;
read_bio->bi_bdev = mirror->rdev->bdev;
read_bio->bi_end_io = raid10_end_read_request;
- read_bio->bi_rw = READ;
+ read_bio->bi_rw = READ | do_sync;
read_bio->bi_private = r10_bio;
generic_make_request(read_bio);
/*
* WRITE:
*/
- /* first select target devices under spinlock and
+ /* first select target devices under rcu_lock and
* inc refcount on their rdev. Record them by setting
* bios[x] to bio
*/
raid10_find_phys(conf, r10_bio);
+ retry_write:
+ blocked_rdev = 0;
rcu_read_lock();
for (i = 0; i < conf->copies; i++) {
int d = r10_bio->devs[i].devnum;
- if (conf->mirrors[d].rdev &&
- !conf->mirrors[d].rdev->faulty) {
- atomic_inc(&conf->mirrors[d].rdev->nr_pending);
+ mdk_rdev_t *rdev = rcu_dereference(conf->mirrors[d].rdev);
+ if (rdev && unlikely(test_bit(Blocked, &rdev->flags))) {
+ atomic_inc(&rdev->nr_pending);
+ blocked_rdev = rdev;
+ break;
+ }
+ if (rdev && !test_bit(Faulty, &rdev->flags)) {
+ atomic_inc(&rdev->nr_pending);
r10_bio->devs[i].bio = bio;
- } else
+ } else {
r10_bio->devs[i].bio = NULL;
+ set_bit(R10BIO_Degraded, &r10_bio->state);
+ }
}
rcu_read_unlock();
- atomic_set(&r10_bio->remaining, 1);
- md_write_start(mddev);
+ if (unlikely(blocked_rdev)) {
+ /* Have to wait for this device to get unblocked, then retry */
+ int j;
+ int d;
+
+ for (j = 0; j < i; j++)
+ if (r10_bio->devs[j].bio) {
+ d = r10_bio->devs[j].devnum;
+ rdev_dec_pending(conf->mirrors[d].rdev, mddev);
+ }
+ allow_barrier(conf);
+ md_wait_for_blocked_rdev(blocked_rdev, mddev);
+ wait_barrier(conf);
+ goto retry_write;
+ }
+
+ atomic_set(&r10_bio->remaining, 0);
+
+ bio_list_init(&bl);
for (i = 0; i < conf->copies; i++) {
struct bio *mbio;
int d = r10_bio->devs[i].devnum;
conf->mirrors[d].rdev->data_offset;
mbio->bi_bdev = conf->mirrors[d].rdev->bdev;
mbio->bi_end_io = raid10_end_write_request;
- mbio->bi_rw = WRITE;
+ mbio->bi_rw = WRITE | do_sync;
mbio->bi_private = r10_bio;
atomic_inc(&r10_bio->remaining);
- generic_make_request(mbio);
+ bio_list_add(&bl, mbio);
}
- if (atomic_dec_and_test(&r10_bio->remaining)) {
+ if (unlikely(!atomic_read(&r10_bio->remaining))) {
+ /* the array is dead */
md_write_end(mddev);
raid_end_bio_io(r10_bio);
+ return 0;
}
+ bitmap_startwrite(mddev->bitmap, bio->bi_sector, r10_bio->sectors, 0);
+ spin_lock_irqsave(&conf->device_lock, flags);
+ bio_list_merge(&conf->pending_bio_list, &bl);
+ blk_plug_device(mddev->queue);
+ spin_unlock_irqrestore(&conf->device_lock, flags);
+
+ /* In case raid10d snuck in to freeze_array */
+ wake_up(&conf->wait_barrier);
+
+ if (do_sync)
+ md_wakeup_thread(mddev->thread);
+
return 0;
}
seq_printf(seq, " %dK chunks", mddev->chunk_size/1024);
if (conf->near_copies > 1)
seq_printf(seq, " %d near-copies", conf->near_copies);
- if (conf->far_copies > 1)
- seq_printf(seq, " %d far-copies", conf->far_copies);
-
+ if (conf->far_copies > 1) {
+ if (conf->far_offset)
+ seq_printf(seq, " %d offset-copies", conf->far_copies);
+ else
+ seq_printf(seq, " %d far-copies", conf->far_copies);
+ }
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->mirrors[i].rdev &&
- conf->mirrors[i].rdev->in_sync ? "U" : "_");
+ test_bit(In_sync, &conf->mirrors[i].rdev->flags) ? "U" : "_");
seq_printf(seq, "]");
}
* next level up know.
* else mark the drive as failed
*/
- if (rdev->in_sync
- && conf->working_disks == 1)
+ if (test_bit(In_sync, &rdev->flags)
+ && conf->raid_disks-mddev->degraded == 1)
/*
* Don't fail the drive, just return an IO error.
* The test should really be more sophisticated than
* really dead" tests...
*/
return;
- if (rdev->in_sync) {
+ if (test_and_clear_bit(In_sync, &rdev->flags)) {
+ unsigned long flags;
+ spin_lock_irqsave(&conf->device_lock, flags);
mddev->degraded++;
- conf->working_disks--;
+ spin_unlock_irqrestore(&conf->device_lock, flags);
/*
* if recovery is running, make sure it aborts.
*/
set_bit(MD_RECOVERY_ERR, &mddev->recovery);
}
- rdev->in_sync = 0;
- rdev->faulty = 1;
- mddev->sb_dirty = 1;
- printk(KERN_ALERT "raid10: Disk failure on %s, disabling device. \n"
- " Operation continuing on %d devices\n",
- bdevname(rdev->bdev,b), conf->working_disks);
+ set_bit(Faulty, &rdev->flags);
+ set_bit(MD_CHANGE_DEVS, &mddev->flags);
+ printk(KERN_ALERT "raid10: Disk failure on %s, disabling device.\n"
+ "raid10: Operation continuing on %d devices.\n",
+ bdevname(rdev->bdev,b), conf->raid_disks - mddev->degraded);
}
static void print_conf(conf_t *conf)
printk("(!conf)\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++) {
tmp = conf->mirrors + i;
if (tmp->rdev)
printk(" disk %d, wo:%d, o:%d, dev:%s\n",
- i, !tmp->rdev->in_sync, !tmp->rdev->faulty,
+ i, !test_bit(In_sync, &tmp->rdev->flags),
+ !test_bit(Faulty, &tmp->rdev->flags),
bdevname(tmp->rdev->bdev,b));
}
}
static void close_sync(conf_t *conf)
{
- spin_lock_irq(&conf->resync_lock);
- wait_event_lock_irq(conf->wait_resume, !conf->barrier,
- conf->resync_lock, unplug_slaves(conf->mddev));
- spin_unlock_irq(&conf->resync_lock);
-
- if (conf->barrier) BUG();
- if (waitqueue_active(&conf->wait_idle)) BUG();
+ wait_barrier(conf);
+ allow_barrier(conf);
mempool_destroy(conf->r10buf_pool);
conf->r10buf_pool = NULL;
}
+/* check if there are enough drives for
+ * every block to appear on atleast one
+ */
+static int enough(conf_t *conf)
+{
+ int first = 0;
+
+ do {
+ int n = conf->copies;
+ int cnt = 0;
+ while (n--) {
+ if (conf->mirrors[first].rdev)
+ cnt++;
+ first = (first+1) % conf->raid_disks;
+ }
+ if (cnt == 0)
+ return 0;
+ } while (first != 0);
+ return 1;
+}
+
static int raid10_spare_active(mddev_t *mddev)
{
int i;
for (i = 0; i < conf->raid_disks; i++) {
tmp = conf->mirrors + i;
if (tmp->rdev
- && !tmp->rdev->faulty
- && !tmp->rdev->in_sync) {
- conf->working_disks++;
+ && !test_bit(Faulty, &tmp->rdev->flags)
+ && !test_and_set_bit(In_sync, &tmp->rdev->flags)) {
+ unsigned long flags;
+ spin_lock_irqsave(&conf->device_lock, flags);
mddev->degraded--;
- tmp->rdev->in_sync = 1;
+ spin_unlock_irqrestore(&conf->device_lock, flags);
}
}
* very different from resync
*/
return 0;
+ if (!enough(conf))
+ return 0;
- for (mirror=0; mirror < mddev->raid_disks; mirror++)
+ if (rdev->saved_raid_disk >= 0 &&
+ conf->mirrors[rdev->saved_raid_disk].rdev == NULL)
+ mirror = rdev->saved_raid_disk;
+ else
+ mirror = 0;
+ for ( ; mirror < mddev->raid_disks; mirror++)
if ( !(p=conf->mirrors+mirror)->rdev) {
blk_queue_stack_limits(mddev->queue,
p->head_position = 0;
rdev->raid_disk = mirror;
found = 1;
- p->rdev = rdev;
+ if (rdev->saved_raid_disk != mirror)
+ conf->fullsync = 1;
+ rcu_assign_pointer(p->rdev, rdev);
break;
}
print_conf(conf);
rdev = p->rdev;
if (rdev) {
- if (rdev->in_sync ||
+ if (test_bit(In_sync, &rdev->flags) ||
atomic_read(&rdev->nr_pending)) {
err = -EBUSY;
goto abort;
}
-static int end_sync_read(struct bio *bio, unsigned int bytes_done, int error)
+static void end_sync_read(struct bio *bio, int error)
{
- int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
r10bio_t * r10_bio = (r10bio_t *)(bio->bi_private);
conf_t *conf = mddev_to_conf(r10_bio->mddev);
int i,d;
- if (bio->bi_size)
- return 1;
-
for (i=0; i<conf->copies; i++)
if (r10_bio->devs[i].bio == bio)
break;
- if (i == conf->copies)
- BUG();
+ BUG_ON(i == conf->copies);
update_head_pos(i, r10_bio);
d = r10_bio->devs[i].devnum;
- if (!uptodate)
- md_error(r10_bio->mddev,
- conf->mirrors[d].rdev);
+
+ if (test_bit(BIO_UPTODATE, &bio->bi_flags))
+ set_bit(R10BIO_Uptodate, &r10_bio->state);
+ else {
+ atomic_add(r10_bio->sectors,
+ &conf->mirrors[d].rdev->corrected_errors);
+ if (!test_bit(MD_RECOVERY_SYNC, &conf->mddev->recovery))
+ md_error(r10_bio->mddev,
+ conf->mirrors[d].rdev);
+ }
/* for reconstruct, we always reschedule after a read.
* for resync, only after all reads
reschedule_retry(r10_bio);
}
rdev_dec_pending(conf->mirrors[d].rdev, conf->mddev);
- return 0;
}
-static int end_sync_write(struct bio *bio, unsigned int bytes_done, int error)
+static void end_sync_write(struct bio *bio, int error)
{
int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
r10bio_t * r10_bio = (r10bio_t *)(bio->bi_private);
conf_t *conf = mddev_to_conf(mddev);
int i,d;
- if (bio->bi_size)
- return 1;
-
for (i = 0; i < conf->copies; i++)
if (r10_bio->devs[i].bio == bio)
break;
}
}
rdev_dec_pending(conf->mirrors[d].rdev, mddev);
- return 0;
}
/*
fbio = r10_bio->devs[i].bio;
/* now find blocks with errors */
- for (i=first+1 ; i < conf->copies ; i++) {
- int vcnt, j, d;
+ for (i=0 ; i < conf->copies ; i++) {
+ int j, d;
+ int vcnt = r10_bio->sectors >> (PAGE_SHIFT-9);
- if (!test_bit(BIO_UPTODATE, &r10_bio->devs[i].bio->bi_flags))
- continue;
- /* We know that the bi_io_vec layout is the same for
- * both 'first' and 'i', so we just compare them.
- * All vec entries are PAGE_SIZE;
- */
tbio = r10_bio->devs[i].bio;
- vcnt = r10_bio->sectors >> (PAGE_SHIFT-9);
- for (j = 0; j < vcnt; j++)
- if (memcmp(page_address(fbio->bi_io_vec[j].bv_page),
- page_address(tbio->bi_io_vec[j].bv_page),
- PAGE_SIZE))
- break;
- if (j == vcnt)
+
+ if (tbio->bi_end_io != end_sync_read)
+ continue;
+ if (i == first)
+ continue;
+ if (test_bit(BIO_UPTODATE, &r10_bio->devs[i].bio->bi_flags)) {
+ /* We know that the bi_io_vec layout is the same for
+ * both 'first' and 'i', so we just compare them.
+ * All vec entries are PAGE_SIZE;
+ */
+ for (j = 0; j < vcnt; j++)
+ if (memcmp(page_address(fbio->bi_io_vec[j].bv_page),
+ page_address(tbio->bi_io_vec[j].bv_page),
+ PAGE_SIZE))
+ break;
+ if (j == vcnt)
+ continue;
+ mddev->resync_mismatches += r10_bio->sectors;
+ }
+ if (test_bit(MD_RECOVERY_CHECK, &mddev->recovery))
+ /* Don't fix anything. */
continue;
/* Ok, we need to write this bio
* First we need to fixup bv_offset, bv_len and
atomic_inc(&conf->mirrors[d].rdev->nr_pending);
md_sync_acct(conf->mirrors[d].rdev->bdev, wbio->bi_size >> 9);
- generic_make_request(wbio);
+ if (test_bit(R10BIO_Uptodate, &r10_bio->state))
+ generic_make_request(wbio);
+ else
+ bio_endio(wbio, -EIO);
}
*
* 1. Retries failed read operations on working mirrors.
* 2. Updates the raid superblock when problems encounter.
- * 3. Performs writes following reads for array syncronising.
+ * 3. Performs writes following reads for array synchronising.
*/
+static void fix_read_error(conf_t *conf, mddev_t *mddev, r10bio_t *r10_bio)
+{
+ int sect = 0; /* Offset from r10_bio->sector */
+ int sectors = r10_bio->sectors;
+ mdk_rdev_t*rdev;
+ while(sectors) {
+ int s = sectors;
+ int sl = r10_bio->read_slot;
+ int success = 0;
+ int start;
+
+ if (s > (PAGE_SIZE>>9))
+ s = PAGE_SIZE >> 9;
+
+ rcu_read_lock();
+ do {
+ int d = r10_bio->devs[sl].devnum;
+ rdev = rcu_dereference(conf->mirrors[d].rdev);
+ if (rdev &&
+ test_bit(In_sync, &rdev->flags)) {
+ atomic_inc(&rdev->nr_pending);
+ rcu_read_unlock();
+ success = sync_page_io(rdev->bdev,
+ r10_bio->devs[sl].addr +
+ sect + rdev->data_offset,
+ s<<9,
+ conf->tmppage, READ);
+ rdev_dec_pending(rdev, mddev);
+ rcu_read_lock();
+ if (success)
+ break;
+ }
+ sl++;
+ if (sl == conf->copies)
+ sl = 0;
+ } while (!success && sl != r10_bio->read_slot);
+ rcu_read_unlock();
+
+ if (!success) {
+ /* Cannot read from anywhere -- bye bye array */
+ int dn = r10_bio->devs[r10_bio->read_slot].devnum;
+ md_error(mddev, conf->mirrors[dn].rdev);
+ break;
+ }
+
+ start = sl;
+ /* write it back and re-read */
+ rcu_read_lock();
+ while (sl != r10_bio->read_slot) {
+ int d;
+ if (sl==0)
+ sl = conf->copies;
+ sl--;
+ d = r10_bio->devs[sl].devnum;
+ rdev = rcu_dereference(conf->mirrors[d].rdev);
+ if (rdev &&
+ test_bit(In_sync, &rdev->flags)) {
+ atomic_inc(&rdev->nr_pending);
+ rcu_read_unlock();
+ atomic_add(s, &rdev->corrected_errors);
+ if (sync_page_io(rdev->bdev,
+ r10_bio->devs[sl].addr +
+ sect + rdev->data_offset,
+ s<<9, conf->tmppage, WRITE)
+ == 0)
+ /* Well, this device is dead */
+ md_error(mddev, rdev);
+ rdev_dec_pending(rdev, mddev);
+ rcu_read_lock();
+ }
+ }
+ sl = start;
+ while (sl != r10_bio->read_slot) {
+ int d;
+ if (sl==0)
+ sl = conf->copies;
+ sl--;
+ d = r10_bio->devs[sl].devnum;
+ rdev = rcu_dereference(conf->mirrors[d].rdev);
+ if (rdev &&
+ test_bit(In_sync, &rdev->flags)) {
+ char b[BDEVNAME_SIZE];
+ atomic_inc(&rdev->nr_pending);
+ rcu_read_unlock();
+ if (sync_page_io(rdev->bdev,
+ r10_bio->devs[sl].addr +
+ sect + rdev->data_offset,
+ s<<9, conf->tmppage, READ) == 0)
+ /* Well, this device is dead */
+ md_error(mddev, rdev);
+ else
+ printk(KERN_INFO
+ "raid10:%s: read error corrected"
+ " (%d sectors at %llu on %s)\n",
+ mdname(mddev), s,
+ (unsigned long long)(sect+
+ rdev->data_offset),
+ bdevname(rdev->bdev, b));
+
+ rdev_dec_pending(rdev, mddev);
+ rcu_read_lock();
+ }
+ }
+ rcu_read_unlock();
+
+ sectors -= s;
+ sect += s;
+ }
+}
+
static void raid10d(mddev_t *mddev)
{
r10bio_t *r10_bio;
mdk_rdev_t *rdev;
md_check_recovery(mddev);
- md_handle_safemode(mddev);
for (;;) {
char b[BDEVNAME_SIZE];
+
+ unplug += flush_pending_writes(conf);
+
spin_lock_irqsave(&conf->device_lock, flags);
- if (list_empty(head))
+ if (list_empty(head)) {
+ spin_unlock_irqrestore(&conf->device_lock, flags);
break;
+ }
r10_bio = list_entry(head->prev, r10bio_t, retry_list);
list_del(head->prev);
+ conf->nr_queued--;
spin_unlock_irqrestore(&conf->device_lock, flags);
mddev = r10_bio->mddev;
unplug = 1;
} else {
int mirror;
+ /* 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
+ * other devices. When we find one, we re-write
+ * and check it that fixes the read error.
+ * This is all done synchronously while the array is
+ * frozen.
+ */
+ if (mddev->ro == 0) {
+ freeze_array(conf);
+ fix_read_error(conf, mddev, r10_bio);
+ unfreeze_array(conf);
+ }
+
bio = r10_bio->devs[r10_bio->read_slot].bio;
- r10_bio->devs[r10_bio->read_slot].bio = NULL;
- bio_put(bio);
+ r10_bio->devs[r10_bio->read_slot].bio =
+ mddev->ro ? IO_BLOCKED : NULL;
mirror = read_balance(conf, r10_bio);
if (mirror == -1) {
printk(KERN_ALERT "raid10: %s: unrecoverable I/O"
bdevname(bio->bi_bdev,b),
(unsigned long long)r10_bio->sector);
raid_end_bio_io(r10_bio);
+ bio_put(bio);
} else {
+ const int do_sync = bio_sync(r10_bio->master_bio);
+ bio_put(bio);
rdev = conf->mirrors[mirror].rdev;
if (printk_ratelimit())
printk(KERN_ERR "raid10: %s: redirecting sector %llu to"
bio->bi_sector = r10_bio->devs[r10_bio->read_slot].addr
+ rdev->data_offset;
bio->bi_bdev = rdev->bdev;
- bio->bi_rw = READ;
+ bio->bi_rw = READ | do_sync;
bio->bi_private = r10_bio;
bio->bi_end_io = raid10_end_read_request;
unplug = 1;
}
}
}
- spin_unlock_irqrestore(&conf->device_lock, flags);
if (unplug)
unplug_slaves(mddev);
}
int buffs;
buffs = RESYNC_WINDOW / RESYNC_BLOCK_SIZE;
- if (conf->r10buf_pool)
- BUG();
+ BUG_ON(conf->r10buf_pool);
conf->r10buf_pool = mempool_create(buffs, r10buf_pool_alloc, r10buf_pool_free, conf);
if (!conf->r10buf_pool)
return -ENOMEM;
*
*/
-static int sync_request(mddev_t *mddev, sector_t sector_nr, int go_faster)
+static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, int go_faster)
{
conf_t *conf = mddev_to_conf(mddev);
r10bio_t *r10_bio;
sector_t max_sector, nr_sectors;
int disk;
int i;
+ int max_sync;
+ int sync_blocks;
sector_t sectors_skipped = 0;
int chunks_skipped = 0;
if (!conf->r10buf_pool)
if (init_resync(conf))
- return -ENOMEM;
+ return 0;
skipped:
max_sector = mddev->size << 1;
if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery))
max_sector = mddev->resync_max_sectors;
if (sector_nr >= max_sector) {
+ /* If we aborted, we need to abort the
+ * sync on the 'current' bitmap chucks (there can
+ * be several when recovering multiple devices).
+ * as we may have started syncing it but not finished.
+ * We can find the current address in
+ * mddev->curr_resync, but for recovery,
+ * we need to convert that to several
+ * virtual addresses.
+ */
+ if (mddev->curr_resync < max_sector) { /* aborted */
+ if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery))
+ bitmap_end_sync(mddev->bitmap, mddev->curr_resync,
+ &sync_blocks, 1);
+ else for (i=0; i<conf->raid_disks; i++) {
+ sector_t sect =
+ raid10_find_virt(conf, mddev->curr_resync, i);
+ bitmap_end_sync(mddev->bitmap, sect,
+ &sync_blocks, 1);
+ }
+ } else /* completed sync */
+ conf->fullsync = 0;
+
+ bitmap_close_sync(mddev->bitmap);
close_sync(conf);
+ *skipped = 1;
return sectors_skipped;
}
if (chunks_skipped >= conf->raid_disks) {
/* if there has been nothing to do on any drive,
* then there is nothing to do at all..
*/
- sector_t sec = max_sector - sector_nr;
- md_done_sync(mddev, sec, 1);
- return sec + sectors_skipped;
+ *skipped = 1;
+ return (max_sector - sector_nr) + sectors_skipped;
}
+ if (max_sector > mddev->resync_max)
+ max_sector = mddev->resync_max; /* Don't do IO beyond here */
+
/* make sure whole request will fit in a chunk - if chunks
* are meaningful
*/
* If there is non-resync activity waiting for us then
* put in a delay to throttle resync.
*/
- if (!go_faster && waitqueue_active(&conf->wait_resume))
+ if (!go_faster && conf->nr_waiting)
msleep_interruptible(1000);
- device_barrier(conf, sector_nr + RESYNC_SECTORS);
+
+ bitmap_cond_end_sync(mddev->bitmap, sector_nr);
/* Again, very different code for resync and recovery.
* Both must result in an r10bio with a list of bios that
* end_sync_write if we will want to write.
*/
+ max_sync = RESYNC_PAGES << (PAGE_SHIFT-9);
if (!test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) {
/* recovery... the complicated one */
int i, j, k;
for (i=0 ; i<conf->raid_disks; i++)
if (conf->mirrors[i].rdev &&
- !conf->mirrors[i].rdev->in_sync) {
+ !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;
+ }
r10_bio = mempool_alloc(conf->r10buf_pool, GFP_NOIO);
- spin_lock_irq(&conf->resync_lock);
- conf->nr_pending++;
- if (rb2) conf->barrier++;
- spin_unlock_irq(&conf->resync_lock);
+ raise_barrier(conf, rb2 != NULL);
atomic_set(&r10_bio->remaining, 0);
r10_bio->master_bio = (struct bio*)rb2;
atomic_inc(&rb2->remaining);
r10_bio->mddev = mddev;
set_bit(R10BIO_IsRecover, &r10_bio->state);
- r10_bio->sector = raid10_find_virt(conf, sector_nr, i);
+ r10_bio->sector = sect;
+
raid10_find_phys(conf, r10_bio);
+ /* Need to check if this section will still be
+ * degraded
+ */
+ for (j=0; j<conf->copies;j++) {
+ int d = r10_bio->devs[j].devnum;
+ if (conf->mirrors[d].rdev == NULL ||
+ test_bit(Faulty, &conf->mirrors[d].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 &&
- conf->mirrors[d].rdev->in_sync) {
+ 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 = 0;
+ 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;
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 = 1;
+ 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;
}
}
if (j == conf->copies) {
- BUG();
+ /* 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_ERR, &mddev->recovery))
+ printk(KERN_INFO "raid10: %s: insufficient working devices for recovery.\n",
+ mdname(mddev));
+ break;
}
}
if (biolist == NULL) {
} else {
/* resync. Schedule a read for every block at this virt offset */
int count = 0;
- r10_bio = mempool_alloc(conf->r10buf_pool, GFP_NOIO);
- spin_lock_irq(&conf->resync_lock);
- conf->nr_pending++;
- spin_unlock_irq(&conf->resync_lock);
+ if (!bitmap_start_sync(mddev->bitmap, sector_nr,
+ &sync_blocks, mddev->degraded) &&
+ !conf->fullsync && !test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery)) {
+ /* We can skip this block */
+ *skipped = 1;
+ return sync_blocks + sectors_skipped;
+ }
+ if (sync_blocks < max_sync)
+ max_sync = sync_blocks;
+ r10_bio = mempool_alloc(conf->r10buf_pool, GFP_NOIO);
r10_bio->mddev = mddev;
atomic_set(&r10_bio->remaining, 0);
+ raise_barrier(conf, 0);
+ conf->next_resync = sector_nr;
r10_bio->master_bio = NULL;
r10_bio->sector = sector_nr;
int d = r10_bio->devs[i].devnum;
bio = r10_bio->devs[i].bio;
bio->bi_end_io = NULL;
+ clear_bit(BIO_UPTODATE, &bio->bi_flags);
if (conf->mirrors[d].rdev == NULL ||
- conf->mirrors[d].rdev->faulty)
+ test_bit(Faulty, &conf->mirrors[d].rdev->flags))
continue;
atomic_inc(&conf->mirrors[d].rdev->nr_pending);
atomic_inc(&r10_bio->remaining);
biolist = bio;
bio->bi_private = r10_bio;
bio->bi_end_io = end_sync_read;
- bio->bi_rw = 0;
+ bio->bi_rw = READ;
bio->bi_sector = r10_bio->devs[i].addr +
conf->mirrors[d].rdev->data_offset;
bio->bi_bdev = conf->mirrors[d].rdev->bdev;
}
nr_sectors = 0;
+ if (sector_nr + max_sync < max_sector)
+ max_sector = sector_nr + max_sync;
do {
struct page *page;
int len = PAGE_SIZE;
}
}
+ if (sectors_skipped)
+ /* pretend they weren't skipped, it makes
+ * no important difference in this case
+ */
+ md_done_sync(mddev, sectors_skipped, 1);
+
return sectors_skipped + nr_sectors;
giveup:
/* There is nowhere to write, so all non-sync
* drives must be failed, so try the next chunk...
*/
{
- int sec = max_sector - sector_nr;
+ sector_t sec = max_sector - sector_nr;
sectors_skipped += sec;
chunks_skipped ++;
sector_nr = max_sector;
- md_done_sync(mddev, sec, 1);
goto skipped;
}
}
mirror_info_t *disk;
mdk_rdev_t *rdev;
struct list_head *tmp;
- int nc, fc;
+ int nc, fc, fo;
sector_t stride, size;
- if (mddev->level != 10) {
- printk(KERN_ERR "raid10: %s: raid level not set correctly... (%d)\n",
- mdname(mddev), mddev->level);
- goto out;
+ if (mddev->chunk_size == 0) {
+ printk(KERN_ERR "md/raid10: non-zero chunk size required.\n");
+ return -EINVAL;
}
+
nc = mddev->layout & 255;
fc = (mddev->layout >> 8) & 255;
+ fo = mddev->layout & (1<<16);
if ((nc*fc) <2 || (nc*fc) > mddev->raid_disks ||
- (mddev->layout >> 16)) {
+ (mddev->layout >> 17)) {
printk(KERN_ERR "raid10: %s: unsupported raid10 layout: 0x%8x\n",
mdname(mddev), mddev->layout);
goto out;
* bookkeeping area. [whatever we allocate in run(),
* should be freed in stop()]
*/
- conf = kmalloc(sizeof(conf_t), GFP_KERNEL);
+ conf = kzalloc(sizeof(conf_t), GFP_KERNEL);
mddev->private = conf;
if (!conf) {
printk(KERN_ERR "raid10: couldn't allocate memory for %s\n",
mdname(mddev));
goto out;
}
- memset(conf, 0, sizeof(*conf));
- conf->mirrors = kmalloc(sizeof(struct mirror_info)*mddev->raid_disks,
+ conf->mirrors = kzalloc(sizeof(struct mirror_info)*mddev->raid_disks,
GFP_KERNEL);
if (!conf->mirrors) {
printk(KERN_ERR "raid10: couldn't allocate memory for %s\n",
mdname(mddev));
goto out_free_conf;
}
- memset(conf->mirrors, 0, sizeof(struct mirror_info)*mddev->raid_disks);
+ conf->tmppage = alloc_page(GFP_KERNEL);
+ if (!conf->tmppage)
+ goto out_free_conf;
+
+ conf->mddev = mddev;
+ conf->raid_disks = mddev->raid_disks;
conf->near_copies = nc;
conf->far_copies = fc;
conf->copies = nc*fc;
+ conf->far_offset = fo;
conf->chunk_mask = (sector_t)(mddev->chunk_size>>9)-1;
conf->chunk_shift = ffz(~mddev->chunk_size) - 9;
- stride = mddev->size >> (conf->chunk_shift-1);
- sector_div(stride, fc);
+ size = mddev->size >> (conf->chunk_shift-1);
+ sector_div(size, fc);
+ size = size * conf->raid_disks;
+ sector_div(size, nc);
+ /* 'size' is now the number of chunks in the array */
+ /* calculate "used chunks per device" in 'stride' */
+ stride = size * conf->copies;
+
+ /* We need to round up when dividing by raid_disks to
+ * get the stride size.
+ */
+ stride += conf->raid_disks - 1;
+ sector_div(stride, conf->raid_disks);
+ mddev->size = stride << (conf->chunk_shift-1);
+
+ if (fo)
+ stride = 1;
+ else
+ sector_div(stride, fc);
conf->stride = stride << conf->chunk_shift;
conf->r10bio_pool = mempool_create(NR_RAID10_BIOS, r10bio_pool_alloc,
goto out_free_conf;
}
- ITERATE_RDEV(mddev, rdev, tmp) {
+ rdev_for_each(rdev, tmp, mddev) {
disk_idx = rdev->raid_disk;
if (disk_idx >= mddev->raid_disks
|| disk_idx < 0)
mddev->queue->max_sectors = (PAGE_SIZE>>9);
disk->head_position = 0;
- if (!rdev->faulty && rdev->in_sync)
- conf->working_disks++;
}
- conf->raid_disks = mddev->raid_disks;
- conf->mddev = mddev;
spin_lock_init(&conf->device_lock);
INIT_LIST_HEAD(&conf->retry_list);
spin_lock_init(&conf->resync_lock);
- init_waitqueue_head(&conf->wait_idle);
- init_waitqueue_head(&conf->wait_resume);
+ init_waitqueue_head(&conf->wait_barrier);
- if (!conf->working_disks) {
- printk(KERN_ERR "raid10: no operational mirrors for %s\n",
- mdname(mddev));
+ /* need to check that every block has at least one working mirror */
+ if (!enough(conf)) {
+ printk(KERN_ERR "raid10: not enough operational mirrors for %s\n",
+ mdname(mddev));
goto out_free_conf;
}
disk = conf->mirrors + i;
- if (!disk->rdev) {
+ if (!disk->rdev ||
+ !test_bit(In_sync, &disk->rdev->flags)) {
disk->head_position = 0;
mddev->degraded++;
}
/*
* Ok, everything is just fine now
*/
- size = conf->stride * conf->raid_disks;
- sector_div(size, conf->near_copies);
- mddev->array_size = size/2;
- mddev->resync_max_sectors = size;
+ mddev->array_size = size << (conf->chunk_shift-1);
+ mddev->resync_max_sectors = size << conf->chunk_shift;
mddev->queue->unplug_fn = raid10_unplug;
- mddev->queue->issue_flush_fn = raid10_issue_flush;
+ mddev->queue->backing_dev_info.congested_fn = raid10_congested;
+ mddev->queue->backing_dev_info.congested_data = mddev;
/* Calculate max read-ahead size.
* We need to readahead at least twice a whole stripe....
* maybe...
*/
{
- int stripe = conf->raid_disks * mddev->chunk_size / PAGE_CACHE_SIZE;
+ int stripe = conf->raid_disks * (mddev->chunk_size / PAGE_SIZE);
stripe /= conf->near_copies;
if (mddev->queue->backing_dev_info.ra_pages < 2* stripe)
mddev->queue->backing_dev_info.ra_pages = 2* stripe;
out_free_conf:
if (conf->r10bio_pool)
mempool_destroy(conf->r10bio_pool);
- if (conf->mirrors)
- kfree(conf->mirrors);
+ safe_put_page(conf->tmppage);
+ kfree(conf->mirrors);
kfree(conf);
mddev->private = NULL;
out:
blk_sync_queue(mddev->queue); /* the unplug fn references 'conf'*/
if (conf->r10bio_pool)
mempool_destroy(conf->r10bio_pool);
- if (conf->mirrors)
- kfree(conf->mirrors);
+ kfree(conf->mirrors);
kfree(conf);
mddev->private = NULL;
return 0;
}
+static void raid10_quiesce(mddev_t *mddev, int state)
+{
+ conf_t *conf = mddev_to_conf(mddev);
+
+ switch(state) {
+ case 1:
+ raise_barrier(conf, 0);
+ break;
+ case 0:
+ lower_barrier(conf);
+ break;
+ }
+ if (mddev->thread) {
+ if (mddev->bitmap)
+ mddev->thread->timeout = mddev->bitmap->daemon_sleep * HZ;
+ else
+ mddev->thread->timeout = MAX_SCHEDULE_TIMEOUT;
+ md_wakeup_thread(mddev->thread);
+ }
+}
-static mdk_personality_t raid10_personality =
+static struct mdk_personality raid10_personality =
{
.name = "raid10",
+ .level = 10,
.owner = THIS_MODULE,
.make_request = make_request,
.run = run,
.hot_remove_disk= raid10_remove_disk,
.spare_active = raid10_spare_active,
.sync_request = sync_request,
+ .quiesce = raid10_quiesce,
};
static int __init raid_init(void)
{
- return register_md_personality(RAID10, &raid10_personality);
+ return register_md_personality(&raid10_personality);
}
static void raid_exit(void)
{
- unregister_md_personality(RAID10);
+ unregister_md_personality(&raid10_personality);
}
module_init(raid_init);
module_exit(raid_exit);
MODULE_LICENSE("GPL");
MODULE_ALIAS("md-personality-9"); /* RAID10 */
+MODULE_ALIAS("md-raid10");
+MODULE_ALIAS("md-level-10");
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff