btrfs: calculate discard delay based on number of extents

An earlier patch keeps track of discardable_extents. These are
undiscarded extents managed by the free space cache. Here, we will use
this to dynamically calculate the discard delay interval.

There are 3 rate to consider. The first is the target convergence rate,
the rate to discard all discardable_extents over the
BTRFS_DISCARD_TARGET_MSEC time frame. This is clamped by the lower
limit, the iops limit or BTRFS_DISCARD_MIN_DELAY (1ms), and the upper
limit, BTRFS_DISCARD_MAX_DELAY (1s). We reevaluate this delay every
transaction commit.
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Dennis Zhou <dennis@kernel.org>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>

fs/btrfs/ctree.h

@@ -468,6 +468,8 @@ struct btrfs_discard_ctl {
 	struct list_head discard_list[BTRFS_NR_DISCARD_LISTS];
 	atomic_t discardable_extents;
 	atomic64_t discardable_bytes;
+	unsigned long delay;
+	u32 iops_limit;
 };
 
 /* delayed seq elem */

fs/btrfs/discard.c

@@ -15,6 +15,12 @@
 #define BTRFS_DISCARD_DELAY		(120ULL * NSEC_PER_SEC)
 #define BTRFS_DISCARD_UNUSED_DELAY	(10ULL * NSEC_PER_SEC)
 
+/* Target completion latency of discarding all discardable extents */
+#define BTRFS_DISCARD_TARGET_MSEC	(6 * 60 * 60UL * MSEC_PER_SEC)
+#define BTRFS_DISCARD_MIN_DELAY_MSEC	(1UL)
+#define BTRFS_DISCARD_MAX_DELAY_MSEC	(1000UL)
+#define BTRFS_DISCARD_MAX_IOPS		(10U)
+
 static struct list_head *get_discard_list(struct btrfs_discard_ctl *discard_ctl,
 					  struct btrfs_block_group *block_group)
 {
@@ -235,11 +241,17 @@ void btrfs_discard_schedule_work(struct btrfs_discard_ctl *discard_ctl,
 
 	block_group = find_next_block_group(discard_ctl, now);
 	if (block_group) {
-		u64 delay = 0;
+		unsigned long delay = discard_ctl->delay;
 
-		if (now < block_group->discard_eligible_time)
-			delay = nsecs_to_jiffies(
-				block_group->discard_eligible_time - now);
+		/*
+		 * This timeout is to hopefully prevent immediate discarding
+		 * in a recently allocated block group.
+		 */
+		if (now < block_group->discard_eligible_time) {
+			u64 bg_timeout = block_group->discard_eligible_time - now;
+
+			delay = max(delay, nsecs_to_jiffies(bg_timeout));
+		}
 
 		mod_delayed_work(discard_ctl->discard_workers,
 				 &discard_ctl->work, delay);
@@ -342,6 +354,39 @@ bool btrfs_run_discard_work(struct btrfs_discard_ctl *discard_ctl)
 		test_bit(BTRFS_FS_DISCARD_RUNNING, &fs_info->flags));
 }
 
+/**
+ * btrfs_discard_calc_delay - recalculate the base delay
+ * @discard_ctl: discard control
+ *
+ * Recalculate the base delay which is based off the total number of
+ * discardable_extents.  Clamp this between the lower_limit (iops_limit or 1ms)
+ * and the upper_limit (BTRFS_DISCARD_MAX_DELAY_MSEC).
+ */
+void btrfs_discard_calc_delay(struct btrfs_discard_ctl *discard_ctl)
+{
+	s32 discardable_extents;
+	u32 iops_limit;
+	unsigned long delay;
+	unsigned long lower_limit = BTRFS_DISCARD_MIN_DELAY_MSEC;
+
+	discardable_extents = atomic_read(&discard_ctl->discardable_extents);
+	if (!discardable_extents)
+		return;
+
+	spin_lock(&discard_ctl->lock);
+
+	iops_limit = READ_ONCE(discard_ctl->iops_limit);
+	if (iops_limit)
+		lower_limit = max_t(unsigned long, lower_limit,
+				    MSEC_PER_SEC / iops_limit);
+
+	delay = BTRFS_DISCARD_TARGET_MSEC / discardable_extents;
+	delay = clamp(delay, lower_limit, BTRFS_DISCARD_MAX_DELAY_MSEC);
+	discard_ctl->delay = msecs_to_jiffies(delay);
+
+	spin_unlock(&discard_ctl->lock);
+}
+
 /**
  * btrfs_discard_update_discardable - propagate discard counters
  * @block_group: block_group of interest
@@ -464,6 +509,8 @@ void btrfs_discard_init(struct btrfs_fs_info *fs_info)
 
 	atomic_set(&discard_ctl->discardable_extents, 0);
 	atomic64_set(&discard_ctl->discardable_bytes, 0);
+	discard_ctl->delay = BTRFS_DISCARD_MAX_DELAY_MSEC;
+	discard_ctl->iops_limit = BTRFS_DISCARD_MAX_IOPS;
 }
 
 void btrfs_discard_cleanup(struct btrfs_fs_info *fs_info)

fs/btrfs/discard.h

@@ -17,6 +17,7 @@ void btrfs_discard_schedule_work(struct btrfs_discard_ctl *discard_ctl,
 bool btrfs_run_discard_work(struct btrfs_discard_ctl *discard_ctl);
 
 /* Update operations */
+void btrfs_discard_calc_delay(struct btrfs_discard_ctl *discard_ctl);
 void btrfs_discard_update_discardable(struct btrfs_block_group *block_group,
 				      struct btrfs_free_space_ctl *ctl);
 

fs/btrfs/extent-tree.c

@@ -2935,8 +2935,10 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans)
 		cond_resched();
 	}
 
-	if (btrfs_test_opt(fs_info, DISCARD_ASYNC))
+	if (btrfs_test_opt(fs_info, DISCARD_ASYNC)) {
+		btrfs_discard_calc_delay(&fs_info->discard_ctl);
 		btrfs_discard_schedule_work(&fs_info->discard_ctl, true);
+	}
 
 	/*
 	 * Transaction is finished.  We don't need the lock anymore.  We

fs/btrfs/sysfs.c

@@ -366,9 +366,40 @@ static ssize_t btrfs_discardable_extents_show(struct kobject *kobj,
 }
 BTRFS_ATTR(discard, discardable_extents, btrfs_discardable_extents_show);
 
+static ssize_t btrfs_discard_iops_limit_show(struct kobject *kobj,
+					     struct kobj_attribute *a,
+					     char *buf)
+{
+	struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj);
+
+	return snprintf(buf, PAGE_SIZE, "%u\n",
+			READ_ONCE(fs_info->discard_ctl.iops_limit));
+}
+
+static ssize_t btrfs_discard_iops_limit_store(struct kobject *kobj,
+					      struct kobj_attribute *a,
+					      const char *buf, size_t len)
+{
+	struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj);
+	struct btrfs_discard_ctl *discard_ctl = &fs_info->discard_ctl;
+	u32 iops_limit;
+	int ret;
+
+	ret = kstrtou32(buf, 10, &iops_limit);
+	if (ret)
+		return -EINVAL;
+
+	WRITE_ONCE(discard_ctl->iops_limit, iops_limit);
+
+	return len;
+}
+BTRFS_ATTR_RW(discard, iops_limit, btrfs_discard_iops_limit_show,
+	      btrfs_discard_iops_limit_store);
+
 static const struct attribute *discard_debug_attrs[] = {
 	BTRFS_ATTR_PTR(discard, discardable_bytes),
 	BTRFS_ATTR_PTR(discard, discardable_extents),
+	BTRFS_ATTR_PTR(discard, iops_limit),
 	NULL,
 };