[v9,07/11] btrfs: zoned: allow zoned RAID

Message ID 20230914-raid-stripe-tree-v9-7-15d423829637@wdc.com
State New
Headers
Series btrfs: introduce RAID stripe tree |

Commit Message

Johannes Thumshirn Sept. 14, 2023, 4:07 p.m. UTC
  When we have a raid-stripe-tree, we can do RAID0/1/10 on zoned devices for
data block-groups. For meta-data block-groups, we don't actually need
anything special, as all meta-data I/O is protected by the
btrfs_zoned_meta_io_lock() already.

Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
---
 fs/btrfs/raid-stripe-tree.h |   7 ++-
 fs/btrfs/volumes.c          |   2 +
 fs/btrfs/zoned.c            | 144 ++++++++++++++++++++++++++++++++++++++++++--
 3 files changed, 148 insertions(+), 5 deletions(-)
  

Comments

David Sterba Sept. 14, 2023, 5:59 p.m. UTC | #1
On Thu, Sep 14, 2023 at 09:07:02AM -0700, Johannes Thumshirn wrote:
> --- a/fs/btrfs/zoned.c
> +++ b/fs/btrfs/zoned.c
> @@ -1397,9 +1397,11 @@ static int btrfs_load_block_group_dup(struct btrfs_block_group *bg,
>  				      struct zone_info *zone_info,
>  				      unsigned long *active)
>  {
> -	if (map->type & BTRFS_BLOCK_GROUP_DATA) {
> -		btrfs_err(bg->fs_info,
> -			  "zoned: profile DUP not yet supported on data bg");
> +	struct btrfs_fs_info *fs_info = bg->fs_info;
> +
> +	if (map->type & BTRFS_BLOCK_GROUP_DATA &&
> +	    !fs_info->stripe_root) {
> +		btrfs_err(fs_info, "zoned: data DUP profile needs stripe_root");

Using stripe_root for identifier is ok so we don't have overly long ones
but for user messages please use raid-stripe-tree. Fixed.
  

Patch

diff --git a/fs/btrfs/raid-stripe-tree.h b/fs/btrfs/raid-stripe-tree.h
index 5d9629a815c1..f31292ab9030 100644
--- a/fs/btrfs/raid-stripe-tree.h
+++ b/fs/btrfs/raid-stripe-tree.h
@@ -6,6 +6,11 @@ 
 #ifndef BTRFS_RAID_STRIPE_TREE_H
 #define BTRFS_RAID_STRIPE_TREE_H
 
+#define BTRFS_RST_SUPP_BLOCK_GROUP_MASK		(BTRFS_BLOCK_GROUP_DUP |\
+						 BTRFS_BLOCK_GROUP_RAID1_MASK |\
+						 BTRFS_BLOCK_GROUP_RAID0 |\
+						 BTRFS_BLOCK_GROUP_RAID10)
+
 struct btrfs_io_context;
 struct btrfs_io_stripe;
 struct btrfs_ordered_extent;
@@ -32,7 +37,7 @@  static inline bool btrfs_need_stripe_tree_update(struct btrfs_fs_info *fs_info,
 	if (type != BTRFS_BLOCK_GROUP_DATA)
 		return false;
 
-	if (profile & BTRFS_BLOCK_GROUP_RAID1_MASK)
+	if (profile & BTRFS_RST_SUPP_BLOCK_GROUP_MASK)
 		return true;
 
 	return false;
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 2326dbcf85f6..dc311e38eb11 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -6541,6 +6541,8 @@  int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op,
 	 * I/O context structure.
 	 */
 	if (smap && num_alloc_stripes == 1 &&
+	    !(btrfs_need_stripe_tree_update(fs_info, map->type) &&
+	      op != BTRFS_MAP_READ) &&
 	    !((map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) && mirror_num > 1)) {
 		ret = set_io_stripe(fs_info, op, logical, length, smap, map,
 				    stripe_index, stripe_offset, stripe_nr);
diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c
index d05510cb2cb2..ce2846c944d2 100644
--- a/fs/btrfs/zoned.c
+++ b/fs/btrfs/zoned.c
@@ -1397,9 +1397,11 @@  static int btrfs_load_block_group_dup(struct btrfs_block_group *bg,
 				      struct zone_info *zone_info,
 				      unsigned long *active)
 {
-	if (map->type & BTRFS_BLOCK_GROUP_DATA) {
-		btrfs_err(bg->fs_info,
-			  "zoned: profile DUP not yet supported on data bg");
+	struct btrfs_fs_info *fs_info = bg->fs_info;
+
+	if (map->type & BTRFS_BLOCK_GROUP_DATA &&
+	    !fs_info->stripe_root) {
+		btrfs_err(fs_info, "zoned: data DUP profile needs stripe_root");
 		return -EINVAL;
 	}
 
@@ -1433,6 +1435,133 @@  static int btrfs_load_block_group_dup(struct btrfs_block_group *bg,
 	return 0;
 }
 
+static int btrfs_load_block_group_raid1(struct btrfs_block_group *bg,
+					struct map_lookup *map,
+					struct zone_info *zone_info,
+					unsigned long *active)
+{
+	struct btrfs_fs_info *fs_info = bg->fs_info;
+	int i;
+
+	if (map->type & BTRFS_BLOCK_GROUP_DATA &&
+	    !fs_info->stripe_root) {
+		btrfs_err(fs_info, "zoned: data %s needs stripe_root",
+			  btrfs_bg_type_to_raid_name(map->type));
+		return -EINVAL;
+
+	}
+
+	for (i = 0; i < map->num_stripes; i++) {
+		if (zone_info[i].alloc_offset == WP_MISSING_DEV ||
+		    zone_info[i].alloc_offset == WP_CONVENTIONAL)
+			continue;
+
+		if ((zone_info[0].alloc_offset != zone_info[i].alloc_offset) &&
+		    !btrfs_test_opt(fs_info, DEGRADED)) {
+			btrfs_err(fs_info,
+				  "zoned: write pointer offset mismatch of zones in %s profile",
+				  btrfs_bg_type_to_raid_name(map->type));
+			return -EIO;
+		}
+		if (test_bit(0, active) != test_bit(i, active)) {
+			if (!btrfs_test_opt(fs_info, DEGRADED) &&
+			    !btrfs_zone_activate(bg)) {
+				return -EIO;
+			}
+		} else {
+			if (test_bit(0, active))
+				set_bit(BLOCK_GROUP_FLAG_ZONE_IS_ACTIVE,
+					&bg->runtime_flags);
+		}
+		/*
+		 * In case a device is missing we have a cap of 0, so don't
+		 * use it.
+		 */
+		bg->zone_capacity = min_not_zero(zone_info[0].capacity,
+						 zone_info[1].capacity);
+	}
+
+	if (zone_info[0].alloc_offset != WP_MISSING_DEV)
+		bg->alloc_offset = zone_info[0].alloc_offset;
+	else
+		bg->alloc_offset = zone_info[i - 1].alloc_offset;
+
+	return 0;
+}
+
+static int btrfs_load_block_group_raid0(struct btrfs_block_group *bg,
+					struct map_lookup *map,
+					struct zone_info *zone_info,
+					unsigned long *active)
+{
+	struct btrfs_fs_info *fs_info = bg->fs_info;
+
+	if (map->type & BTRFS_BLOCK_GROUP_DATA &&
+	    !fs_info->stripe_root) {
+		btrfs_err(fs_info, "zoned: data %s needs stripe_root",
+			  btrfs_bg_type_to_raid_name(map->type));
+		return -EINVAL;
+
+	}
+
+	for (int i = 0; i < map->num_stripes; i++) {
+		if (zone_info[i].alloc_offset == WP_MISSING_DEV ||
+		    zone_info[i].alloc_offset == WP_CONVENTIONAL)
+			continue;
+
+		if (test_bit(0, active) != test_bit(i, active)) {
+			if (!btrfs_zone_activate(bg))
+				return -EIO;
+		} else {
+			if (test_bit(0, active))
+				set_bit(BLOCK_GROUP_FLAG_ZONE_IS_ACTIVE,
+					&bg->runtime_flags);
+		}
+		bg->zone_capacity += zone_info[i].capacity;
+		bg->alloc_offset += zone_info[i].alloc_offset;
+	}
+
+	return 0;
+}
+
+static int btrfs_load_block_group_raid10(struct btrfs_block_group *bg,
+					struct map_lookup *map,
+					struct zone_info *zone_info,
+					unsigned long *active)
+{
+	struct btrfs_fs_info *fs_info = bg->fs_info;
+
+	if (map->type & BTRFS_BLOCK_GROUP_DATA &&
+	    !fs_info->stripe_root) {
+		btrfs_err(fs_info, "zoned: data %s needs stripe_root",
+			  btrfs_bg_type_to_raid_name(map->type));
+		return -EINVAL;
+
+	}
+
+	for (int i = 0; i < map->num_stripes; i++) {
+		if (zone_info[i].alloc_offset == WP_MISSING_DEV ||
+		    zone_info[i].alloc_offset == WP_CONVENTIONAL)
+			continue;
+
+		if (test_bit(0, active) != test_bit(i, active)) {
+			if (!btrfs_zone_activate(bg))
+				return -EIO;
+		} else {
+			if (test_bit(0, active))
+				set_bit(BLOCK_GROUP_FLAG_ZONE_IS_ACTIVE,
+					&bg->runtime_flags);
+		}
+
+		if ((i % map->sub_stripes) == 0) {
+			bg->zone_capacity += zone_info[i].capacity;
+			bg->alloc_offset += zone_info[i].alloc_offset;
+		}
+	}
+
+	return 0;
+}
+
 int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)
 {
 	struct btrfs_fs_info *fs_info = cache->fs_info;
@@ -1525,11 +1654,18 @@  int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new)
 		ret = btrfs_load_block_group_dup(cache, map, zone_info, active);
 		break;
 	case BTRFS_BLOCK_GROUP_RAID1:
+	case BTRFS_BLOCK_GROUP_RAID1C3:
+	case BTRFS_BLOCK_GROUP_RAID1C4:
+		ret = btrfs_load_block_group_raid1(cache, map, zone_info, active);
+		break;
 	case BTRFS_BLOCK_GROUP_RAID0:
+		ret = btrfs_load_block_group_raid0(cache, map, zone_info, active);
+		break;
 	case BTRFS_BLOCK_GROUP_RAID10:
+		ret = btrfs_load_block_group_raid10(cache, map, zone_info, active);
+		break;
 	case BTRFS_BLOCK_GROUP_RAID5:
 	case BTRFS_BLOCK_GROUP_RAID6:
-		/* non-single profiles are not supported yet */
 	default:
 		btrfs_err(fs_info, "zoned: profile %s not yet supported",
 			  btrfs_bg_type_to_raid_name(map->type));