md/raid1: prevent merging too large request

For SSD, if request size exceeds specific value (optimal io size), request size
isn't important for bandwidth. In such condition, if making request size bigger
will cause some disks idle, the total throughput will actually drop. A good
example is doing a readahead in a two-disk raid1 setup.

So when should we split big requests? We absolutly don't want to split big
request to very small requests. Even in SSD, big request transfer is more
efficient. This patch only considers request with size above optimal io size.

If all disks are busy, is it worth doing a split? Say optimal io size is 16k,
two requests 32k and two disks. We can let each disk run one 32k request, or
split the requests to 4 16k requests and each disk runs two. It's hard to say
which case is better, depending on hardware.

So only consider case where there are idle disks. For readahead, split is
always better in this case. And in my test, below patch can improve > 30%
thoughput. Hmm, not 100%, because disk isn't 100% busy.

Such case can happen not just in readahead, for example, in directio. But I
suppose directio usually will have bigger IO depth and make all disks busy, so
I ignored it.

Note: if the raid uses any hard disk, we don't prevent merging. That will make
performace worse.

Signed-off-by: Shaohua Li <shli@fusionio.com>
Signed-off-by: NeilBrown <neilb@suse.de>

2 files changed, 50 insertions, 7 deletions

diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index d9869f25aa75..7aa958ed2847 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -504,6 +504,7 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect
         unsigned int min_pending;
         struct md_rdev *rdev;
         int choose_first;
+        int choose_next_idle;
 
         rcu_read_lock();
         /*
@@ -520,6 +521,7 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect
         min_pending = UINT_MAX;
         best_good_sectors = 0;
         has_nonrot_disk = 0;
+        choose_next_idle = 0;
 
         if (conf->mddev->recovery_cp < MaxSector &&
             (this_sector + sectors >= conf->next_resync))
@@ -532,6 +534,7 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect
                 sector_t first_bad;
                 int bad_sectors;
                 unsigned int pending;
+                bool nonrot;
 
                 rdev = rcu_dereference(conf->mirrors[disk].rdev);
                 if (r1_bio->bios[disk] == IO_BLOCKED
@@ -590,18 +593,52 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect
                 } else
                         best_good_sectors = sectors;
 
-                has_nonrot_disk |= blk_queue_nonrot(bdev_get_queue(rdev->bdev));
+                nonrot = blk_queue_nonrot(bdev_get_queue(rdev->bdev));
+                has_nonrot_disk |= nonrot;
                 pending = atomic_read(&rdev->nr_pending);
                 dist = abs(this_sector - conf->mirrors[disk].head_position);
-                if (choose_first
-                    /* Don't change to another disk for sequential reads */
-                    || conf->mirrors[disk].next_seq_sect == this_sector
-                    || dist == 0
-                    /* If device is idle, use it */
-                    || pending == 0) {
+                if (choose_first) {
                         best_disk = disk;
                         break;
                 }
+                /* Don't change to another disk for sequential reads */
+                if (conf->mirrors[disk].next_seq_sect == this_sector
+                    || dist == 0) {
+                        int opt_iosize = bdev_io_opt(rdev->bdev) >> 9;
+                        struct raid1_info *mirror = &conf->mirrors[disk];
+
+                        best_disk = disk;
+                        /*
+                         * If buffered sequential IO size exceeds optimal
+                         * iosize, check if there is idle disk. If yes, choose
+                         * the idle disk. read_balance could already choose an
+                         * idle disk before noticing it's a sequential IO in
+                         * this disk. This doesn't matter because this disk
+                         * will idle, next time it will be utilized after the
+                         * first disk has IO size exceeds optimal iosize. In
+                         * this way, iosize of the first disk will be optimal
+                         * iosize at least. iosize of the second disk might be
+                         * small, but not a big deal since when the second disk
+                         * starts IO, the first disk is likely still busy.
+                         */
+                        if (nonrot && opt_iosize > 0 &&
+                            mirror->seq_start != MaxSector &&
+                            mirror->next_seq_sect > opt_iosize &&
+                            mirror->next_seq_sect - opt_iosize >=
+                            mirror->seq_start) {
+                                choose_next_idle = 1;
+                                continue;
+                        }
+                        break;
+                }
+                /* If device is idle, use it */
+                if (pending == 0) {
+                        best_disk = disk;
+                        break;
+                }
+
+                if (choose_next_idle)
+                        continue;
 
                 if (min_pending > pending) {
                         min_pending = pending;
@@ -640,6 +677,10 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect
                         goto retry;
                 }
                 sectors = best_good_sectors;
+
+                if (conf->mirrors[best_disk].next_seq_sect != this_sector)
+                        conf->mirrors[best_disk].seq_start = this_sector;
+
                 conf->mirrors[best_disk].next_seq_sect = this_sector + sectors;
         }
         rcu_read_unlock();
@@ -2605,6 +2646,7 @@ static struct r1conf *setup_conf(struct mddev *mddev)
                         mddev->merge_check_needed = 1;
 
                 disk->head_position = 0;
+                disk->seq_start = MaxSector;
         }
         conf->raid_disks = mddev->raid_disks;
         conf->mddev = mddev;
diff --git a/drivers/md/raid1.h b/drivers/md/raid1.h
index 3770b4a27662..0ff3715fb7eb 100644
--- a/drivers/md/raid1.h
+++ b/drivers/md/raid1.h
@@ -9,6 +9,7 @@ struct raid1_info {
          * we try to keep sequential reads one the same device
          */
         sector_t        next_seq_sect;
+        sector_t        seq_start;
 };
 
 /*