*/
#define NR_RAID10_BIOS 256
-/* When there are this many requests queue to be written by
+/* when we get a read error on a read-only array, we redirect to another
+ * device without failing the first device, or trying to over-write to
+ * correct the read error. To keep track of bad blocks on a per-bio
+ * level, we store IO_BLOCKED in the appropriate 'bios' pointer
+ */
+#define IO_BLOCKED ((struct bio *)1)
+/* When we successfully write to a known bad-block, we need to remove the
+ * bad-block marking which must be done from process context. So we record
+ * the success by setting devs[n].bio to IO_MADE_GOOD
+ */
+#define IO_MADE_GOOD ((struct bio *)2)
+
+#define BIO_SPECIAL(bio) ((unsigned long)bio <= 2)
+
+/* When there are this many requests queued to be written by
* the raid10 thread, we become 'congested' to provide back-pressure
* for writeback.
*/
int sectors = r10_bio->sectors;
int best_good_sectors;
sector_t new_distance, best_dist;
- struct md_rdev *rdev, *best_rdev;
+ struct md_rdev *best_rdev, *rdev = NULL;
int do_balance;
int best_slot;
struct geom *geo = &conf->geo;
return rdev;
}
-static int raid10_congested(void *data, int bits)
+int md_raid10_congested(struct mddev *mddev, int bits)
{
- struct mddev *mddev = data;
struct r10conf *conf = mddev->private;
int i, ret = 0;
conf->pending_count >= max_queued_requests)
return 1;
- if (mddev_congested(mddev, bits))
- return 1;
rcu_read_lock();
for (i = 0;
(i < conf->geo.raid_disks || i < conf->prev.raid_disks)
rcu_read_unlock();
return ret;
}
+EXPORT_SYMBOL_GPL(md_raid10_congested);
+
+static int raid10_congested(void *data, int bits)
+{
+ struct mddev *mddev = data;
+
+ return mddev_congested(mddev, bits) ||
+ md_raid10_congested(mddev, bits);
+}
static void flush_pending_writes(struct r10conf *conf)
{
const unsigned long do_fua = (bio->bi_rw & REQ_FUA);
unsigned long flags;
struct md_rdev *blocked_rdev;
- int plugged;
int sectors_handled;
int max_sectors;
int sectors;
* of r10_bios is recored in bio->bi_phys_segments just as with
* the read case.
*/
- plugged = mddev_check_plugged(mddev);
r10_bio->read_slot = -1; /* make sure repl_bio gets freed */
raid10_find_phys(conf, r10_bio);
bio_list_add(&conf->pending_bio_list, mbio);
conf->pending_count++;
spin_unlock_irqrestore(&conf->device_lock, flags);
+ if (!mddev_check_plugged(mddev))
+ md_wakeup_thread(mddev->thread);
if (!r10_bio->devs[i].repl_bio)
continue;
bio_list_add(&conf->pending_bio_list, mbio);
conf->pending_count++;
spin_unlock_irqrestore(&conf->device_lock, flags);
+ if (!mddev_check_plugged(mddev))
+ md_wakeup_thread(mddev->thread);
}
/* Don't remove the bias on 'remaining' (one_write_done) until
/* In case raid10d snuck in to freeze_array */
wake_up(&conf->wait_barrier);
-
- if (do_sync || !mddev->bitmap || !plugged)
- md_wakeup_thread(mddev->thread);
}
static void status(struct seq_file *seq, struct mddev *mddev)
static void print_conf(struct r10conf *conf)
{
int i;
- struct mirror_info *tmp;
+ struct raid10_info *tmp;
printk(KERN_DEBUG "RAID10 conf printout:\n");
if (!conf) {
{
int i;
struct r10conf *conf = mddev->private;
- struct mirror_info *tmp;
+ struct raid10_info *tmp;
int count = 0;
unsigned long flags;
else
mirror = first;
for ( ; mirror <= last ; mirror++) {
- struct mirror_info *p = &conf->mirrors[mirror];
+ struct raid10_info *p = &conf->mirrors[mirror];
if (p->recovery_disabled == mddev->recovery_disabled)
continue;
if (p->rdev) {
int err = 0;
int number = rdev->raid_disk;
struct md_rdev **rdevp;
- struct mirror_info *p = conf->mirrors + number;
+ struct raid10_info *p = conf->mirrors + number;
print_conf(conf);
if (rdev == p->rdev)
if (r10_sync_page_io(rdev,
r10_bio->devs[sl].addr +
sect,
- s<<9, conf->tmppage, WRITE)
+ s, conf->tmppage, WRITE)
== 0) {
/* Well, this device is dead */
printk(KERN_NOTICE
switch (r10_sync_page_io(rdev,
r10_bio->devs[sl].addr +
sect,
- s<<9, conf->tmppage,
+ s, conf->tmppage,
READ)) {
case 0:
/* Well, this device is dead */
slot = r10_bio->read_slot;
printk_ratelimited(
KERN_ERR
- "md/raid10:%s: %s: redirecting"
+ "md/raid10:%s: %s: redirecting "
"sector %llu to another mirror\n",
mdname(mddev),
bdevname(rdev->bdev, b),
blk_start_plug(&plug);
for (;;) {
- flush_pending_writes(conf);
+ if (atomic_read(&mddev->plug_cnt) == 0)
+ flush_pending_writes(conf);
spin_lock_irqsave(&conf->device_lock, flags);
if (list_empty(head)) {
sector_t sect;
int must_sync;
int any_working;
- struct mirror_info *mirror = &conf->mirrors[i];
+ struct raid10_info *mirror = &conf->mirrors[i];
if ((mirror->rdev == NULL ||
test_bit(In_sync, &mirror->rdev->flags))
/* want to reconstruct this device */
rb2 = r10_bio;
sect = raid10_find_virt(conf, sector_nr, i);
+ if (sect >= mddev->resync_max_sectors) {
+ /* last stripe is not complete - don't
+ * try to recover this sector.
+ */
+ continue;
+ }
/* Unless we are doing a full sync, or a replacement
* we only need to recover the block if it is set in
* the bitmap
goto out;
/* FIXME calc properly */
- conf->mirrors = kzalloc(sizeof(struct mirror_info)*(mddev->raid_disks +
+ conf->mirrors = kzalloc(sizeof(struct raid10_info)*(mddev->raid_disks +
max(0,mddev->delta_disks)),
GFP_KERNEL);
if (!conf->mirrors)
spin_lock_init(&conf->resync_lock);
init_waitqueue_head(&conf->wait_barrier);
- conf->thread = md_register_thread(raid10d, mddev, NULL);
+ conf->thread = md_register_thread(raid10d, mddev, "raid10");
if (!conf->thread)
goto out;
{
struct r10conf *conf;
int i, disk_idx, chunk_size;
- struct mirror_info *disk;
+ struct raid10_info *disk;
struct md_rdev *rdev;
sector_t size;
sector_t min_offset_diff = 0;
conf->thread = NULL;
chunk_size = mddev->chunk_sectors << 9;
- blk_queue_io_min(mddev->queue, chunk_size);
- if (conf->geo.raid_disks % conf->geo.near_copies)
- blk_queue_io_opt(mddev->queue, chunk_size * conf->geo.raid_disks);
- else
- blk_queue_io_opt(mddev->queue, chunk_size *
- (conf->geo.raid_disks / conf->geo.near_copies));
+ if (mddev->queue) {
+ blk_queue_io_min(mddev->queue, chunk_size);
+ if (conf->geo.raid_disks % conf->geo.near_copies)
+ blk_queue_io_opt(mddev->queue, chunk_size * conf->geo.raid_disks);
+ else
+ blk_queue_io_opt(mddev->queue, chunk_size *
+ (conf->geo.raid_disks / conf->geo.near_copies));
+ }
rdev_for_each(rdev, mddev) {
long long diff;
if (first || diff < min_offset_diff)
min_offset_diff = diff;
- disk_stack_limits(mddev->gendisk, rdev->bdev,
- rdev->data_offset << 9);
+ if (mddev->gendisk)
+ disk_stack_limits(mddev->gendisk, rdev->bdev,
+ rdev->data_offset << 9);
disk->head_position = 0;
}
md_set_array_sectors(mddev, size);
mddev->resync_max_sectors = size;
- 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...
- */
- {
+ if (mddev->queue) {
int stripe = conf->geo.raid_disks *
((mddev->chunk_sectors << 9) / PAGE_SIZE);
+ 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...
+ */
stripe /= conf->geo.near_copies;
if (mddev->queue->backing_dev_info.ra_pages < 2 * stripe)
mddev->queue->backing_dev_info.ra_pages = 2 * stripe;
+ blk_queue_merge_bvec(mddev->queue, raid10_mergeable_bvec);
}
- blk_queue_merge_bvec(mddev->queue, raid10_mergeable_bvec);
if (md_integrity_register(mddev))
goto out_free_conf;
lower_barrier(conf);
md_unregister_thread(&mddev->thread);
- blk_sync_queue(mddev->queue); /* the unplug fn references 'conf'*/
+ if (mddev->queue)
+ /* the unplug fn references 'conf'*/
+ blk_sync_queue(mddev->queue);
+
if (conf->r10bio_pool)
mempool_destroy(conf->r10bio_pool);
kfree(conf->mirrors);
if (mddev->delta_disks > 0) {
/* allocate new 'mirrors' list */
conf->mirrors_new = kzalloc(
- sizeof(struct mirror_info)
+ sizeof(struct raid10_info)
*(mddev->raid_disks +
mddev->delta_disks),
GFP_KERNEL);
spin_lock_irq(&conf->device_lock);
if (conf->mirrors_new) {
memcpy(conf->mirrors_new, conf->mirrors,
- sizeof(struct mirror_info)*conf->prev.raid_disks);
+ sizeof(struct raid10_info)*conf->prev.raid_disks);
smp_mb();
kfree(conf->mirrors_old); /* FIXME and elsewhere */
conf->mirrors_old = conf->mirrors;
projects / deliverable / linux.git / blobdiff