linux-btrfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Josef Bacik <josef@toxicpanda.com>
To: linux-btrfs@vger.kernel.org, kernel-team@fb.com
Subject: [PATCH v5 09/52] btrfs: add infrastructure for safe em freeing
Date: Wed, 24 Jan 2024 12:18:31 -0500	[thread overview]
Message-ID: <6cf44f7860e94de68df242e69f4c5250bd061cff.1706116485.git.josef@toxicpanda.com> (raw)
In-Reply-To: <cover.1706116485.git.josef@toxicpanda.com>

When we add fscrypt support we're going to have fscrypt objects hanging
off of extent_maps.  This includes a block key, which if we're the last
one freeing the key we may have to unregister it from the block layer.
This requires taking a semaphore in the block layer, which means we
can't free em's under the extent map tree lock.

Thankfully we only do this in two places, one where we're dropping a
range of extent maps, and when we're freeing logged extents.  Add a
free_extent_map_safe() which will add the em to a list in the em_tree if
we free'd the object.  Currently this is unconditional but will be
changed to conditional on the fscrypt object we will add in a later
patch.

To process these delayed objects add a free_pending_extent_maps() that
is called after the lock has been dropped on the em_tree.  This will
process the extent maps on the freed list and do the appropriate freeing
work in a safe manner.

Signed-off-by: Josef Bacik <josef@toxicpanda.com>
---
 fs/btrfs/extent_map.c | 76 +++++++++++++++++++++++++++++++++++++++++--
 fs/btrfs/extent_map.h | 10 ++++++
 fs/btrfs/tree-log.c   |  6 ++--
 3 files changed, 87 insertions(+), 5 deletions(-)

diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index b61099bf97a8..f8705103819c 100644
--- a/fs/btrfs/extent_map.c
+++ b/fs/btrfs/extent_map.c
@@ -35,7 +35,9 @@ void __cold extent_map_exit(void)
 void extent_map_tree_init(struct extent_map_tree *tree)
 {
 	tree->map = RB_ROOT_CACHED;
+	tree->flags = 0;
 	INIT_LIST_HEAD(&tree->modified_extents);
+	INIT_LIST_HEAD(&tree->freed_extents);
 	rwlock_init(&tree->lock);
 }
 
@@ -52,9 +54,15 @@ struct extent_map *alloc_extent_map(void)
 	RB_CLEAR_NODE(&em->rb_node);
 	refcount_set(&em->refs, 1);
 	INIT_LIST_HEAD(&em->list);
+	INIT_LIST_HEAD(&em->free_list);
 	return em;
 }
 
+static void __free_extent_map(struct extent_map *em)
+{
+	kmem_cache_free(extent_map_cache, em);
+}
+
 /*
  * Drop the reference out on @em by one and free the structure if the reference
  * count hits zero.
@@ -66,10 +74,69 @@ void free_extent_map(struct extent_map *em)
 	if (refcount_dec_and_test(&em->refs)) {
 		WARN_ON(extent_map_in_tree(em));
 		WARN_ON(!list_empty(&em->list));
-		kmem_cache_free(extent_map_cache, em);
+		__free_extent_map(em);
 	}
 }
 
+/*
+ * Drop a ref for the extent map in the given tree.
+ *
+ * @tree:	tree that the em is a part of.
+ * @em:		the em to drop the reference to.
+ *
+ * Drop the reference count on @em by one, if the reference count hits 0 and
+ * there is an object on the em that can't be safely freed in the current
+ * context (if we are holding the extent_map_tree->lock for example), then add
+ * it to the freed_extents list on the extent_map_tree for later processing.
+ *
+ * This must be followed by a free_pending_extent_maps() to clear the pending
+ * frees.
+ */
+void free_extent_map_safe(struct extent_map_tree *tree,
+			  struct extent_map *em)
+{
+	lockdep_assert_held_write(&tree->lock);
+
+	if (!em)
+		return;
+
+	if (refcount_dec_and_test(&em->refs)) {
+		WARN_ON(extent_map_in_tree(em));
+		WARN_ON(!list_empty(&em->list));
+		list_add_tail(&em->free_list, &tree->freed_extents);
+		set_bit(EXTENT_MAP_TREE_PENDING_FREES, &tree->flags);
+	}
+}
+
+/*
+ * Free the em objects that exist on the em tree
+ *
+ * @tree:	the tree to free the objects from.
+ *
+ * If there are any objects on the em->freed_extents list go ahead and free them
+ * here in a safe way.  This is to be coupled with any uses of
+ * free_extent_map_safe().
+ */
+void free_pending_extent_maps(struct extent_map_tree *tree)
+{
+	struct extent_map *em;
+
+	/* Avoid taking the write lock if we don't have any pending frees. */
+	if (!test_and_clear_bit(EXTENT_MAP_TREE_PENDING_FREES, &tree->flags))
+		return;
+
+	write_lock(&tree->lock);
+	while ((em = list_first_entry_or_null(&tree->freed_extents,
+					      struct extent_map, free_list))) {
+		list_del_init(&em->free_list);
+		write_unlock(&tree->lock);
+		__free_extent_map(em);
+		cond_resched();
+		write_lock(&tree->lock);
+	}
+	write_unlock(&tree->lock);
+}
+
 /* Do the math around the end of an extent, handling wrapping. */
 static u64 range_end(u64 start, u64 len)
 {
@@ -654,10 +721,12 @@ static void drop_all_extent_maps_fast(struct extent_map_tree *tree)
 		em = rb_entry(node, struct extent_map, rb_node);
 		em->flags &= ~(EXTENT_FLAG_PINNED | EXTENT_FLAG_LOGGING);
 		remove_extent_mapping(tree, em);
-		free_extent_map(em);
+		free_extent_map_safe(tree, em);
 		cond_resched_rwlock_write(&tree->lock);
 	}
 	write_unlock(&tree->lock);
+
+	free_pending_extent_maps(tree);
 }
 
 /*
@@ -875,13 +944,14 @@ void btrfs_drop_extent_map_range(struct btrfs_inode *inode, u64 start, u64 end,
 		free_extent_map(em);
 next:
 		/* Once for us (for our lookup reference). */
-		free_extent_map(em);
+		free_extent_map_safe(em_tree, em);
 
 		em = next_em;
 	}
 
 	write_unlock(&em_tree->lock);
 
+	free_pending_extent_maps(em_tree);
 	free_extent_map(split);
 	free_extent_map(split2);
 }
diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h
index e380fc08bbe4..d31f2a03670e 100644
--- a/fs/btrfs/extent_map.h
+++ b/fs/btrfs/extent_map.h
@@ -55,11 +55,18 @@ struct extent_map {
 	u32 flags;
 	refcount_t refs;
 	struct list_head list;
+	struct list_head free_list;
+};
+
+enum extent_map_flags {
+	EXTENT_MAP_TREE_PENDING_FREES,
 };
 
 struct extent_map_tree {
 	struct rb_root_cached map;
+	unsigned long flags;
 	struct list_head modified_extents;
+	struct list_head freed_extents;
 	rwlock_t lock;
 };
 
@@ -122,6 +129,9 @@ int split_extent_map(struct btrfs_inode *inode, u64 start, u64 len, u64 pre,
 
 struct extent_map *alloc_extent_map(void);
 void free_extent_map(struct extent_map *em);
+void free_extent_map_safe(struct extent_map_tree *tree,
+			  struct extent_map *em);
+void free_pending_extent_maps(struct extent_map_tree *tree);
 int __init extent_map_init(void);
 void __cold extent_map_exit(void);
 int unpin_extent_cache(struct btrfs_inode *inode, u64 start, u64 len, u64 gen);
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 331fc7429952..916e54b91ecc 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -4884,7 +4884,7 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
 		 */
 		if (ret) {
 			clear_em_logging(tree, em);
-			free_extent_map(em);
+			free_extent_map_safe(tree, em);
 			continue;
 		}
 
@@ -4893,11 +4893,13 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
 		ret = log_one_extent(trans, inode, em, path, ctx);
 		write_lock(&tree->lock);
 		clear_em_logging(tree, em);
-		free_extent_map(em);
+		free_extent_map_safe(tree, em);
 	}
 	WARN_ON(!list_empty(&extents));
 	write_unlock(&tree->lock);
 
+	free_pending_extent_maps(tree);
+
 	if (!ret)
 		ret = btrfs_log_prealloc_extents(trans, inode, path);
 	if (ret)
-- 
2.43.0


  parent reply	other threads:[~2024-01-24 17:19 UTC|newest]

Thread overview: 59+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-01-24 17:18 [PATCH v5 00/52] btrfs: add fscrypt support Josef Bacik
2024-01-24 17:18 ` [PATCH v5 01/52] fscrypt: add per-extent encryption support Josef Bacik
2024-01-24 17:18 ` [PATCH v5 02/52] fscrypt: allow inline encryption for extent based encryption Josef Bacik
2024-01-24 17:18 ` [PATCH v5 03/52] fscrypt: add a fscrypt_inode_open helper Josef Bacik
2024-01-24 17:18 ` [PATCH v5 04/52] fscrypt: conditionally don't wipe mk secret until the last active user is done Josef Bacik
2024-01-24 17:18 ` [PATCH v5 05/52] blk-crypto: add a process bio callback Josef Bacik
2024-01-24 17:18 ` [PATCH v5 06/52] fscrypt: add a process_bio hook to fscrypt_operations Josef Bacik
2024-01-24 17:18 ` [PATCH v5 07/52] fscrypt: expose fscrypt_nokey_name Josef Bacik
2024-01-24 17:18 ` [PATCH v5 08/52] fscrypt: add documentation about extent encryption Josef Bacik
2024-01-24 17:18 ` Josef Bacik [this message]
2024-01-24 19:49   ` [PATCH v5 09/52] btrfs: add infrastructure for safe em freeing Boris Burkov
2024-01-24 17:18 ` [PATCH v5 10/52] btrfs: disable various operations on encrypted inodes Josef Bacik
2024-01-24 19:53   ` Boris Burkov
2024-01-24 17:18 ` [PATCH v5 11/52] btrfs: disable verity " Josef Bacik
2024-01-24 19:53   ` Boris Burkov
2024-01-24 17:18 ` [PATCH v5 12/52] btrfs: start using fscrypt hooks Josef Bacik
2024-01-24 17:18 ` [PATCH v5 13/52] btrfs: add inode encryption contexts Josef Bacik
2024-01-24 17:18 ` [PATCH v5 14/52] btrfs: add new FEATURE_INCOMPAT_ENCRYPT flag Josef Bacik
2024-01-24 17:18 ` [PATCH v5 15/52] btrfs: adapt readdir for encrypted and nokey names Josef Bacik
2024-01-24 17:18 ` [PATCH v5 16/52] btrfs: handle " Josef Bacik
2024-01-24 17:18 ` [PATCH v5 17/52] btrfs: implement fscrypt ioctls Josef Bacik
2024-01-24 17:18 ` [PATCH v5 18/52] btrfs: gate encryption behind BTRFS_DEBUG Josef Bacik
2024-01-24 17:18 ` [PATCH v5 19/52] btrfs: select encryption dependencies if FS_ENCRYPTION Josef Bacik
2024-01-24 17:18 ` [PATCH v5 20/52] btrfs: add get_devices hook for fscrypt Josef Bacik
2024-01-24 17:18 ` [PATCH v5 21/52] btrfs: set file extent encryption excplicitly Josef Bacik
2024-01-24 17:18 ` [PATCH v5 22/52] btrfs: add fscrypt_info and encryption_type to extent_map Josef Bacik
2024-01-24 17:18 ` [PATCH v5 23/52] btrfs: add fscrypt_info and encryption_type to ordered_extent Josef Bacik
2024-01-24 17:18 ` [PATCH v5 24/52] btrfs: plumb through setting the fscrypt_info for ordered extents Josef Bacik
2024-01-24 17:18 ` [PATCH v5 25/52] btrfs: plumb the fscrypt extent context through create_io_em Josef Bacik
2024-01-24 17:18 ` [PATCH v5 26/52] btrfs: populate the ordered_extent with the fscrypt context Josef Bacik
2024-01-24 17:18 ` [PATCH v5 27/52] btrfs: keep track of fscrypt info and orig_start for dio reads Josef Bacik
2024-01-24 17:18 ` [PATCH v5 28/52] btrfs: add an optional encryption context to the end of file extents Josef Bacik
2024-01-24 17:18 ` [PATCH v5 29/52] btrfs: explicitly track file extent length for replace and drop Josef Bacik
2024-01-24 17:18 ` [PATCH v5 30/52] btrfs: pass through fscrypt_extent_info to the file extent helpers Josef Bacik
2024-01-24 17:18 ` [PATCH v5 31/52] btrfs: implement the fscrypt extent encryption hooks Josef Bacik
2024-01-24 17:18 ` [PATCH v5 32/52] btrfs: setup fscrypt_extent_info for new extents Josef Bacik
2024-01-24 17:18 ` [PATCH v5 33/52] btrfs: populate ordered_extent with the orig offset Josef Bacik
2024-01-24 17:18 ` [PATCH v5 34/52] btrfs: set the bio fscrypt context when applicable Josef Bacik
2024-01-24 17:18 ` [PATCH v5 35/52] btrfs: add a bio argument to btrfs_csum_one_bio Josef Bacik
2024-01-24 17:18 ` [PATCH v5 36/52] btrfs: add orig_logical to btrfs_bio Josef Bacik
2024-01-24 17:18 ` [PATCH v5 37/52] btrfs: limit encrypted writes to 256 segments Josef Bacik
2024-01-24 17:19 ` [PATCH v5 38/52] btrfs: implement process_bio cb for fscrypt Josef Bacik
2024-01-24 17:19 ` [PATCH v5 39/52] btrfs: implement read repair for encryption Josef Bacik
2024-01-24 17:19 ` [PATCH v5 40/52] btrfs: add test_dummy_encryption support Josef Bacik
2024-01-24 17:19 ` [PATCH v5 41/52] btrfs: don't rewrite ret from inode_permission Josef Bacik
2024-01-24 17:19 ` [PATCH v5 42/52] btrfs: move inode_to_path higher in backref.c Josef Bacik
2024-01-24 17:19 ` [PATCH v5 43/52] btrfs: make btrfs_ref_to_path handle encrypted filenames Josef Bacik
2024-01-24 17:19 ` [PATCH v5 44/52] btrfs: don't search back for dir inode item in INO_LOOKUP_USER Josef Bacik
2024-01-24 17:19 ` [PATCH v5 45/52] btrfs: deal with encrypted symlinks in send Josef Bacik
2024-01-24 17:19 ` [PATCH v5 46/52] btrfs: decrypt file names for send Josef Bacik
2024-01-24 17:19 ` [PATCH v5 47/52] btrfs: load the inode context before sending writes Josef Bacik
2024-01-24 17:19 ` [PATCH v5 48/52] btrfs: set the appropriate free space settings in reconfigure Josef Bacik
2024-01-24 17:19 ` [PATCH v5 49/52] btrfs: support encryption with log replay Josef Bacik
2024-01-24 17:19 ` [PATCH v5 50/52] btrfs: disable auto defrag on encrypted files Josef Bacik
2024-01-24 17:19 ` [PATCH v5 51/52] btrfs: disable encryption on RAID5/6 Josef Bacik
2024-01-24 17:19 ` [PATCH v5 52/52] btrfs: disable send if we have encryption enabled Josef Bacik
2024-01-24 19:28   ` Boris Burkov
2024-01-24 19:18 ` [PATCH v5 00/52] btrfs: add fscrypt support Neal Gompa
2024-01-24 19:58   ` Josef Bacik

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=6cf44f7860e94de68df242e69f4c5250bd061cff.1706116485.git.josef@toxicpanda.com \
    --to=josef@toxicpanda.com \
    --cc=kernel-team@fb.com \
    --cc=linux-btrfs@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).