raid1, raid10 and multipath don't report their 'congested' status
through bdi_*_congested, but should.
This patch adds the appropriate functions which just check the
'congested' status of all active members (with appropriate locking).
raid1 read_balance should be modified to prefer devices where
bdi_read_congested returns false. Then we could use the '&' branch
rather than the '|' branch. However that should would need
some benchmarking first to make sure it is actually a good idea.
Signed-off-by: Neil Brown <[email protected]>
### Diffstat output
./drivers/md/multipath.c | 24 ++++++++++++++++++++++++
./drivers/md/raid1.c | 28 ++++++++++++++++++++++++++++
./drivers/md/raid10.c | 22 ++++++++++++++++++++++
3 files changed, 74 insertions(+)
diff .prev/drivers/md/multipath.c ./drivers/md/multipath.c
--- .prev/drivers/md/multipath.c 2006-08-29 14:52:50.000000000 +1000
+++ ./drivers/md/multipath.c 2006-08-29 14:33:34.000000000 +1000
@@ -228,6 +228,28 @@ static int multipath_issue_flush(request
rcu_read_unlock();
return ret;
}
+static int multipath_congested(void *data, int bits)
+{
+ mddev_t *mddev = data;
+ multipath_conf_t *conf = mddev_to_conf(mddev);
+ int i, ret = 0;
+
+ rcu_read_lock();
+ for (i = 0; i < mddev->raid_disks ; i++) {
+ mdk_rdev_t *rdev = rcu_dereference(conf->multipaths[i].rdev);
+ if (rdev && !test_bit(Faulty, &rdev->flags)) {
+ request_queue_t *q = bdev_get_queue(rdev->bdev);
+
+ ret |= bdi_congested(&q->backing_dev_info, bits);
+ /* Just like multipath_map, we just check the
+ * first available device
+ */
+ break;
+ }
+ }
+ rcu_read_unlock();
+ return ret;
+}
/*
* Careful, this can execute in IRQ contexts as well!
@@ -509,6 +531,8 @@ static int multipath_run (mddev_t *mddev
mddev->queue->unplug_fn = multipath_unplug;
mddev->queue->issue_flush_fn = multipath_issue_flush;
+ mddev->queue->backing_dev_info.congested_fn = multipath_congested;
+ mddev->queue->backing_dev_info.congested_data = mddev;
return 0;
diff .prev/drivers/md/raid1.c ./drivers/md/raid1.c
--- .prev/drivers/md/raid1.c 2006-08-29 14:52:50.000000000 +1000
+++ ./drivers/md/raid1.c 2006-08-29 14:26:59.000000000 +1000
@@ -601,6 +601,32 @@ static int raid1_issue_flush(request_que
return ret;
}
+static int raid1_congested(void *data, int bits)
+{
+ 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; i++) {
+ mdk_rdev_t *rdev = rcu_dereference(conf->mirrors[i].rdev);
+ if (rdev && !test_bit(Faulty, &rdev->flags)) {
+ request_queue_t *q = bdev_get_queue(rdev->bdev);
+
+ /* Note the '|| 1' - when read_balance prefers
+ * non-congested targets, it can be removed
+ */
+ if ((bits & (1<<BDI_write_congested)) || 1)
+ ret |= bdi_congested(&q->backing_dev_info, bits);
+ else
+ ret &= bdi_congested(&q->backing_dev_info, bits);
+ }
+ }
+ rcu_read_unlock();
+ return ret;
+}
+
+
/* Barriers....
* Sometimes we need to suspend IO while we do something else,
* either some resync/recovery, or reconfigure the array.
@@ -1965,6 +1991,8 @@ static int run(mddev_t *mddev)
mddev->queue->unplug_fn = raid1_unplug;
mddev->queue->issue_flush_fn = raid1_issue_flush;
+ mddev->queue->backing_dev_info.congested_fn = raid1_congested;
+ mddev->queue->backing_dev_info.congested_data = mddev;
return 0;
diff .prev/drivers/md/raid10.c ./drivers/md/raid10.c
--- .prev/drivers/md/raid10.c 2006-08-29 14:50:34.000000000 +1000
+++ ./drivers/md/raid10.c 2006-08-29 14:56:51.000000000 +1000
@@ -648,6 +648,26 @@ static int raid10_issue_flush(request_qu
return ret;
}
+static int raid10_congested(void *data, int bits)
+{
+ 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 = rcu_dereference(conf->mirrors[i].rdev);
+ if (rdev && !test_bit(Faulty, &rdev->flags)) {
+ request_queue_t *q = bdev_get_queue(rdev->bdev);
+
+ ret |= bdi_congested(&q->backing_dev_info, bits);
+ }
+ }
+ rcu_read_unlock();
+ return ret;
+}
+
+
/* Barriers....
* Sometimes we need to suspend IO while we do something else,
* either some resync/recovery, or reconfigure the array.
@@ -2094,6 +2114,8 @@ static int run(mddev_t *mddev)
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....
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
[Index of Archives]
[Kernel Newbies]
[Netfilter]
[Bugtraq]
[Photo]
[Stuff]
[Gimp]
[Yosemite News]
[MIPS Linux]
[ARM Linux]
[Linux Security]
[Linux RAID]
[Video 4 Linux]
[Linux for the blind]
[Linux Resources]
