linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Naohiro Aota <naohiro.aota@wdc.com>
To: linux-btrfs@vger.kernel.org, dsterba@suse.com
Cc: hare@suse.com, linux-fsdevel@vger.kernel.org,
	Naohiro Aota <naohiro.aota@wdc.com>
Subject: [PATCH v8 38/41] btrfs: extend zoned allocator to use dedicated tree-log block group
Date: Fri,  2 Oct 2020 03:36:45 +0900	[thread overview]
Message-ID: <17f8b62a6fc896598378ecf88bdab5f6b3d3b9cc.1601574234.git.naohiro.aota@wdc.com> (raw)
In-Reply-To: <dece91bca322ce44bed19f2b0f460fa5ded2e512.1601574234.git.naohiro.aota@wdc.com>

This is the 1/3 patch to enable tree log on ZONED mode.

The tree-log feature does not work on ZONED mode as is. Blocks for a
tree-log tree are allocated mixed with other metadata blocks, and btrfs
writes and syncs the tree-log blocks to devices at the time of fsync(),
which is different timing from a global transaction commit. As a result,
both writing tree-log blocks and writing other metadata blocks become
non-sequential writes that ZONED mode must avoid.

We can introduce a dedicated block group for tree-log blocks so that
tree-log blocks and other metadata blocks can be separated write streams.
As a result, each write stream can now be written to devices separately.
"fs_info->treelog_bg" tracks the dedicated block group and btrfs assign
"treelog_bg" on-demand on tree-log block allocation time.

This commit extends the zoned block allocator to use the block group.

Signed-off-by: Naohiro Aota <naohiro.aota@wdc.com>
---
 fs/btrfs/block-group.c |  7 +++++
 fs/btrfs/ctree.h       |  2 ++
 fs/btrfs/extent-tree.c | 65 ++++++++++++++++++++++++++++++++++++++----
 3 files changed, 69 insertions(+), 5 deletions(-)

diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c
index d39fa80d3d90..64253d4a7bfc 100644
--- a/fs/btrfs/block-group.c
+++ b/fs/btrfs/block-group.c
@@ -939,6 +939,13 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	btrfs_return_cluster_to_free_space(block_group, cluster);
 	spin_unlock(&cluster->refill_lock);
 
+	if (btrfs_fs_incompat(fs_info, ZONED)) {
+		spin_lock(&fs_info->treelog_bg_lock);
+		if (fs_info->treelog_bg == block_group->start)
+			fs_info->treelog_bg = 0;
+		spin_unlock(&fs_info->treelog_bg_lock);
+	}
+
 	path = btrfs_alloc_path();
 	if (!path) {
 		ret = -ENOMEM;
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index d021bc4a92cd..81e2f5b78917 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -954,6 +954,8 @@ struct btrfs_fs_info {
 	unsigned long exclusive_operation;
 
 	struct mutex zoned_meta_io_lock;
+	spinlock_t treelog_bg_lock;
+	u64 treelog_bg;
 
 #ifdef CONFIG_BTRFS_FS_REF_VERIFY
 	spinlock_t ref_verify_lock;
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index f44faaf7aca2..c4d382e5b45f 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -3656,6 +3656,9 @@ struct find_free_extent_ctl {
 
 	/* Allocation policy */
 	enum btrfs_extent_allocation_policy policy;
+
+	/* Allocation is called for tree-log */
+	bool for_treelog;
 };
 
 
@@ -3856,23 +3859,54 @@ static int do_allocation_zoned(struct btrfs_block_group *block_group,
 			       struct find_free_extent_ctl *ffe_ctl,
 			       struct btrfs_block_group **bg_ret)
 {
+	struct btrfs_fs_info *fs_info = block_group->fs_info;
 	struct btrfs_space_info *space_info = block_group->space_info;
 	struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
 	u64 start = block_group->start;
 	u64 num_bytes = ffe_ctl->num_bytes;
 	u64 avail;
+	u64 bytenr = block_group->start;
+	u64 log_bytenr;
 	int ret = 0;
+	bool skip;
 
 	ASSERT(btrfs_fs_incompat(block_group->fs_info, ZONED));
 
+	/*
+	 * Do not allow non-tree-log blocks in the dedicated tree-log block
+	 * group, and vice versa.
+	 */
+	spin_lock(&fs_info->treelog_bg_lock);
+	log_bytenr = fs_info->treelog_bg;
+	skip = log_bytenr && ((ffe_ctl->for_treelog && bytenr != log_bytenr) ||
+			      (!ffe_ctl->for_treelog && bytenr == log_bytenr));
+	spin_unlock(&fs_info->treelog_bg_lock);
+	if (skip)
+		return 1;
+
 	spin_lock(&space_info->lock);
 	spin_lock(&block_group->lock);
+	spin_lock(&fs_info->treelog_bg_lock);
+
+	ASSERT(!ffe_ctl->for_treelog ||
+	       block_group->start == fs_info->treelog_bg ||
+	       fs_info->treelog_bg == 0);
 
 	if (block_group->ro) {
 		ret = 1;
 		goto out;
 	}
 
+	/*
+	 * Do not allow currently using block group to be tree-log dedicated
+	 * block group.
+	 */
+	if (ffe_ctl->for_treelog && !fs_info->treelog_bg &&
+	    (block_group->used || block_group->reserved)) {
+		ret = 1;
+		goto out;
+	}
+
 	avail = block_group->length - block_group->alloc_offset;
 	if (avail < num_bytes) {
 		ffe_ctl->max_extent_size = avail;
@@ -3880,6 +3914,9 @@ static int do_allocation_zoned(struct btrfs_block_group *block_group,
 		goto out;
 	}
 
+	if (ffe_ctl->for_treelog && !fs_info->treelog_bg)
+		fs_info->treelog_bg = block_group->start;
+
 	ffe_ctl->found_offset = start + block_group->alloc_offset;
 	block_group->alloc_offset += num_bytes;
 	spin_lock(&ctl->tree_lock);
@@ -3887,10 +3924,13 @@ static int do_allocation_zoned(struct btrfs_block_group *block_group,
 	spin_unlock(&ctl->tree_lock);
 
 	ASSERT(IS_ALIGNED(ffe_ctl->found_offset,
-			  block_group->fs_info->stripesize));
+			  fs_info->stripesize));
 	ffe_ctl->search_start = ffe_ctl->found_offset;
 
 out:
+	if (ret && ffe_ctl->for_treelog)
+		fs_info->treelog_bg = 0;
+	spin_unlock(&fs_info->treelog_bg_lock);
 	spin_unlock(&block_group->lock);
 	spin_unlock(&space_info->lock);
 	return ret;
@@ -4140,7 +4180,12 @@ static int prepare_allocation(struct btrfs_fs_info *fs_info,
 		return prepare_allocation_clustered(fs_info, ffe_ctl,
 						    space_info, ins);
 	case BTRFS_EXTENT_ALLOC_ZONED:
-		/* nothing to do */
+		if (ffe_ctl->for_treelog) {
+			spin_lock(&fs_info->treelog_bg_lock);
+			if (fs_info->treelog_bg)
+				ffe_ctl->hint_byte = fs_info->treelog_bg;
+			spin_unlock(&fs_info->treelog_bg_lock);
+		}
 		return 0;
 	default:
 		BUG();
@@ -4184,6 +4229,7 @@ static noinline int find_free_extent(struct btrfs_root *root,
 	struct find_free_extent_ctl ffe_ctl = {0};
 	struct btrfs_space_info *space_info;
 	bool full_search = false;
+	bool for_treelog = root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID;
 
 	WARN_ON(num_bytes < fs_info->sectorsize);
 
@@ -4197,6 +4243,7 @@ static noinline int find_free_extent(struct btrfs_root *root,
 	ffe_ctl.orig_have_caching_bg = false;
 	ffe_ctl.found_offset = 0;
 	ffe_ctl.hint_byte = hint_byte_orig;
+	ffe_ctl.for_treelog = for_treelog;
 	ffe_ctl.policy = BTRFS_EXTENT_ALLOC_CLUSTERED;
 
 	/* For clustered allocation */
@@ -4271,8 +4318,15 @@ static noinline int find_free_extent(struct btrfs_root *root,
 		struct btrfs_block_group *bg_ret;
 
 		/* If the block group is read-only, we can skip it entirely. */
-		if (unlikely(block_group->ro))
+		if (unlikely(block_group->ro)) {
+			if (btrfs_fs_incompat(fs_info, ZONED) && for_treelog) {
+				spin_lock(&fs_info->treelog_bg_lock);
+				if (block_group->start == fs_info->treelog_bg)
+					fs_info->treelog_bg = 0;
+				spin_unlock(&fs_info->treelog_bg_lock);
+			}
 			continue;
+		}
 
 		btrfs_grab_block_group(block_group, delalloc);
 		ffe_ctl.search_start = block_group->start;
@@ -4460,6 +4514,7 @@ int btrfs_reserve_extent(struct btrfs_root *root, u64 ram_bytes,
 	bool final_tried = num_bytes == min_alloc_size;
 	u64 flags;
 	int ret;
+	bool for_treelog = root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID;
 
 	flags = get_alloc_profile_by_root(root, is_data);
 again:
@@ -4483,8 +4538,8 @@ int btrfs_reserve_extent(struct btrfs_root *root, u64 ram_bytes,
 
 			sinfo = btrfs_find_space_info(fs_info, flags);
 			btrfs_err(fs_info,
-				  "allocation failed flags %llu, wanted %llu",
-				  flags, num_bytes);
+			"allocation failed flags %llu, wanted %llu treelog %d",
+				  flags, num_bytes, for_treelog);
 			if (sinfo)
 				btrfs_dump_space_info(fs_info, sinfo,
 						      num_bytes, 1);
-- 
2.27.0


  parent reply	other threads:[~2020-10-01 18:39 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-01 18:36 [PATCH v8 00/41] btrfs: zoned block device support Naohiro Aota
2020-10-01 18:36 ` [PATCH v8 01/41] block: add bio_add_zone_append_page Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 02/41] btrfs: introduce ZONED feature flag Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 03/41] btrfs: Get zone information of zoned block devices Naohiro Aota
2020-10-13 15:53     ` David Sterba
2020-10-15  7:45       ` Johannes Thumshirn
2020-10-01 18:36   ` [PATCH v8 04/41] btrfs: Check and enable ZONED mode Naohiro Aota
2020-10-13 15:56     ` David Sterba
2020-10-28 14:49       ` Johannes Thumshirn
2020-10-13 16:13     ` David Sterba
2020-10-01 18:36   ` [PATCH v8 05/41] btrfs: introduce max_zone_append_size Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 06/41] btrfs: disallow space_cache in ZONED mode Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 07/41] btrfs: disallow NODATACOW " Naohiro Aota
2020-10-13 15:39     ` David Sterba
2020-10-21  8:53       ` Johannes Thumshirn
2020-10-01 18:36   ` [PATCH v8 08/41] btrfs: disable fallocate " Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 09/41] btrfs: disallow mixed-bg " Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 10/41] btrfs: disallow inode_cache " Naohiro Aota
2020-10-13 15:41     ` David Sterba
2020-10-22  6:48       ` Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 11/41] btrfs: implement log-structured superblock for " Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 12/41] btrfs: implement zoned chunk allocator Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 13/41] btrfs: verify device extent is aligned to zone Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 14/41] btrfs: load zone's alloction offset Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 15/41] btrfs: emulate write pointer for conventional zones Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 16/41] btrfs: track unusable bytes for zones Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 17/41] btrfs: do sequential extent allocation in ZONED mode Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 18/41] btrfs: reset zones of unused block groups Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 19/41] btrfs: redirty released extent buffers in ZONED mode Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 20/41] btrfs: extract page adding function Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 21/41] btrfs: use bio_add_zone_append_page for zoned btrfs Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 22/41] btrfs: handle REQ_OP_ZONE_APPEND as writing Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 23/41] btrfs: split ordered extent when bio is sent Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 24/41] btrfs: extend btrfs_rmap_block for specifying a device Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 25/41] btrfs: use ZONE_APPEND write for ZONED btrfs Naohiro Aota
2020-10-13 16:45     ` David Sterba
2020-10-01 18:36   ` [PATCH v8 26/41] btrfs: enable zone append writing for direct IO Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 27/41] btrfs: introduce dedicated data write path for ZONED mode Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 28/41] btrfs: serialize meta IOs on " Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 29/41] btrfs: wait existing extents before truncating Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 30/41] btrfs: avoid async metadata checksum on ZONED mode Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 31/41] btrfs: mark block groups to copy for device-replace Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 32/41] btrfs: implement cloning for ZONED device-replace Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 33/41] btrfs: implement copying " Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 34/41] btrfs: support dev-replace in ZONED mode Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 35/41] btrfs: enable relocation " Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 36/41] btrfs: relocate block group to repair IO failure in ZONED Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 37/41] btrfs: split alloc_log_tree() Naohiro Aota
2020-10-01 18:36   ` Naohiro Aota [this message]
2020-10-13 16:26     ` [PATCH v8 38/41] btrfs: extend zoned allocator to use dedicated tree-log block group David Sterba
2020-10-15  7:21       ` Johannes Thumshirn
2020-10-01 18:36   ` [PATCH v8 39/41] btrfs: serialize log transaction on ZONED mode Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 40/41] btrfs: reorder log node allocation Naohiro Aota
2020-10-01 18:36   ` [PATCH v8 41/41] btrfs: enable to mount ZONED incompat flag Naohiro Aota
2020-10-02 13:39   ` [PATCH v8 01/41] block: add bio_add_zone_append_page Martin K. Petersen
2020-10-05  1:46     ` Damien Le Moal
2020-10-05 13:43     ` Christoph Hellwig
2020-10-06  1:26       ` Martin K. Petersen
2020-10-06  5:12         ` Damien Le Moal
2020-10-09 15:40 ` [PATCH v8 00/41] btrfs: zoned block device support Josef Bacik
2020-10-12  9:17   ` Naohiro Aota

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=17f8b62a6fc896598378ecf88bdab5f6b3d3b9cc.1601574234.git.naohiro.aota@wdc.com \
    --to=naohiro.aota@wdc.com \
    --cc=dsterba@suse.com \
    --cc=hare@suse.com \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).