diff mbox series

[3/3] btrfs: split RAID stripes on deletion

Message ID 20240610-b4-rst-updates-v1-3-179c1eec08f2@kernel.org (mailing list archive)
State New, archived
Headers show
Series btrfs: rst: updates for RAID stripe tree | expand

Commit Message

Johannes Thumshirn June 10, 2024, 8:40 a.m. UTC 
From: JohnnesThumshirn <johannes.thumshirn@wdc.com>

The current RAID stripe code assumes, that we will always remove a
whole stripe entry.

But ff we're only removing a part of a RAID stripe we're hitting the
ASSERT()ion checking for this condition.

Instead of assuming the complete deletion of a RAID stripe, split the
stripe if we need to.

Signed-off-by: Johnnes Thumshirn <johannes.thumshirn@wdc.com>
---
 fs/btrfs/raid-stripe-tree.c | 101 +++++++++++++++++++++++++++++++++-----------
 1 file changed, 77 insertions(+), 24 deletions(-)

Comments

Josef Bacik June 10, 2024, 7:45 p.m. UTC | #1 
On Mon, Jun 10, 2024 at 10:40:27AM +0200, Johannes Thumshirn wrote:
> From: JohnnesThumshirn <johannes.thumshirn@wdc.com>
> 
> The current RAID stripe code assumes, that we will always remove a
> whole stripe entry.
> 
> But ff we're only removing a part of a RAID stripe we're hitting the
> ASSERT()ion checking for this condition.
> 
> Instead of assuming the complete deletion of a RAID stripe, split the
> stripe if we need to.
> 
> Signed-off-by: Johnnes Thumshirn <johannes.thumshirn@wdc.com>

I'd like a selftest for this helper, should be relatively straightforward to do,
just to test edgecases and such.  Thanks,

Josef
Johannes Thumshirn June 11, 2024, 6:39 a.m. UTC | #2 
On 10.06.24 21:45, Josef Bacik wrote:
> On Mon, Jun 10, 2024 at 10:40:27AM +0200, Johannes Thumshirn wrote:
>> From: JohnnesThumshirn <johannes.thumshirn@wdc.com>
>>
>> The current RAID stripe code assumes, that we will always remove a
>> whole stripe entry.
>>
>> But ff we're only removing a part of a RAID stripe we're hitting the
>> ASSERT()ion checking for this condition.
>>
>> Instead of assuming the complete deletion of a RAID stripe, split the
>> stripe if we need to.
>>
>> Signed-off-by: Johnnes Thumshirn <johannes.thumshirn@wdc.com>
> 
> I'd like a selftest for this helper, should be relatively straightforward to do,
> just to test edgecases and such.  Thanks,

Sure. Let me see what I can cook up.
diff mbox series

Patch

diff --git a/fs/btrfs/raid-stripe-tree.c b/fs/btrfs/raid-stripe-tree.c
index 3020820dd6e2..41403217c3e6 100644
--- a/fs/btrfs/raid-stripe-tree.c
+++ b/fs/btrfs/raid-stripe-tree.c
@@ -33,42 +33,95 @@  int btrfs_delete_raid_extent(struct btrfs_trans_handle *trans, u64 start, u64 le
 	if (!path)
 		return -ENOMEM;
 
-	while (1) {
-		key.objectid = start;
-		key.type = BTRFS_RAID_STRIPE_KEY;
-		key.offset = length;
+again:
+	key.objectid = start;
+	key.type = BTRFS_RAID_STRIPE_KEY;
+	key.offset = length;
 
-		ret = btrfs_search_slot(trans, stripe_root, &key, path, -1, 1);
-		if (ret < 0)
-			break;
-		if (ret > 0) {
-			ret = 0;
-			if (path->slots[0] == 0)
-				break;
-			path->slots[0]--;
-		}
+	ret = btrfs_search_slot(trans, stripe_root, &key, path, -1, 1);
+	if (ret < 0)
+		goto out;
+	if (ret > 0) {
+		ret = 0;
+		if (path->slots[0] == 0)
+			goto out;
+		path->slots[0]--;
+	}
+
+	leaf = path->nodes[0];
+	slot = path->slots[0];
+	btrfs_item_key_to_cpu(leaf, &key, slot);
+	found_start = key.objectid;
+	found_end = found_start + key.offset;
+
+	/* That stripe ends before we start, we're done. */
+	if (found_end <= start)
+		goto out;
+
+	trace_btrfs_raid_extent_delete(fs_info, start, end,
+				       found_start, found_end);
+
+	if (found_start < start) {
+		u64 diff = start - found_start;
+		struct btrfs_key new_key;
+		int num_stripes;
+		struct btrfs_stripe_extent *stripe_extent;
+
+		new_key.objectid = start;
+		new_key.type = BTRFS_RAID_STRIPE_KEY;
+		new_key.offset = length - diff;
+
+		ret = btrfs_duplicate_item(trans, stripe_root, path,
+					   &new_key);
+		if (ret)
+			goto out;
 
 		leaf = path->nodes[0];
 		slot = path->slots[0];
-		btrfs_item_key_to_cpu(leaf, &key, slot);
-		found_start = key.objectid;
-		found_end = found_start + key.offset;
 
-		/* That stripe ends before we start, we're done. */
-		if (found_end <= start)
-			break;
+		num_stripes =
+			btrfs_num_raid_stripes(btrfs_item_size(leaf, slot));
+		stripe_extent =
+			btrfs_item_ptr(leaf, slot, struct btrfs_stripe_extent);
 
-		trace_btrfs_raid_extent_delete(fs_info, start, end,
-					       found_start, found_end);
+		for (int i = 0; i < num_stripes; i++) {
+			struct btrfs_raid_stride *raid_stride =
+				&stripe_extent->strides[i];
+			u64 physical =
+				btrfs_raid_stride_physical(leaf, raid_stride);
 
-		ASSERT(found_start >= start && found_end <= end);
-		ret = btrfs_del_item(trans, stripe_root, path);
+			btrfs_set_stack_raid_stride_physical(raid_stride,
+							     physical + diff);
+		}
+
+		btrfs_mark_buffer_dirty(trans, leaf);
+		btrfs_release_path(path);
+		goto again;
+	}
+
+	if (found_end > end) {
+		u64 diff = found_end - end;
+		struct btrfs_key new_key;
+
+		new_key.objectid = found_start;
+		new_key.type = BTRFS_RAID_STRIPE_KEY;
+		new_key.offset = length - diff;
+
+		ret = btrfs_duplicate_item(trans, stripe_root, path,
+					   &new_key);
 		if (ret)
-			break;
+			goto out;
 
+		btrfs_mark_buffer_dirty(trans, leaf);
 		btrfs_release_path(path);
+		goto again;
+
 	}
 
+	if (found_start == start && found_end == end)
+		ret = btrfs_del_item(trans, stripe_root, path);
+
+ out:
 	btrfs_free_path(path);
 	return ret;
 }