From: Josef Bacik <josef@toxicpanda.com>
To: linux-btrfs@vger.kernel.org, kernel-team@fb.com,
ebiggers@kernel.org, linux-fscrypt@vger.kernel.org,
ngompa13@gmail.com
Subject: [PATCH 07/35] btrfs: add infrastructure for safe em freeing
Date: Tue, 26 Sep 2023 14:01:33 -0400 [thread overview]
Message-ID: <e6b3d2767e1ca79224d00e275e9d34ebc6ffcb8f.1695750478.git.josef@toxicpanda.com> (raw)
In-Reply-To: <cover.1695750478.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 | 80 ++++++++++++++++++++++++++++++++++++++++---
fs/btrfs/extent_map.h | 10 ++++++
fs/btrfs/tree-log.c | 6 ++--
3 files changed, 89 insertions(+), 7 deletions(-)
diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index a6d8368ed0ed..af5ff6b10865 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);
}
@@ -53,9 +55,17 @@ struct extent_map *alloc_extent_map(void)
em->compress_type = BTRFS_COMPRESS_NONE;
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)
+{
+ if (test_bit(EXTENT_FLAG_FS_MAPPING, &em->flags))
+ kfree(em->map_lookup);
+ 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.
@@ -67,12 +77,69 @@ 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));
- if (test_bit(EXTENT_FLAG_FS_MAPPING, &em->flags))
- kfree(em->map_lookup);
- 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)
{
@@ -684,10 +751,12 @@ static void drop_all_extent_maps_fast(struct extent_map_tree *tree)
clear_bit(EXTENT_FLAG_PINNED, &em->flags);
clear_bit(EXTENT_FLAG_LOGGING, &em->flags);
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);
}
/*
@@ -908,13 +977,14 @@ 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 35d27c756e08..2093720271ea 100644
--- a/fs/btrfs/extent_map.h
+++ b/fs/btrfs/extent_map.h
@@ -55,11 +55,18 @@ struct extent_map {
refcount_t refs;
unsigned int compress_type;
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;
};
@@ -95,6 +102,9 @@ 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 extent_map_tree *tree, u64 start, u64 len, u64 gen);
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 6b98e0dbc0a4..c4d8900f91bb 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -4886,7 +4886,7 @@ 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;
}
@@ -4895,11 +4895,13 @@ 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)
--
2.41.0
next prev parent reply other threads:[~2023-09-26 18:03 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-26 18:01 [PATCH 00/35] btrfs: add fscrypt support Josef Bacik
2023-09-26 18:01 ` [PATCH 01/35] fscrypt: rename fscrypt_info => fscrypt_inode_info Josef Bacik
2023-09-26 18:01 ` [PATCH 02/35] fscrypt: add per-extent encryption support Josef Bacik
2023-09-26 18:01 ` [PATCH 03/35] fscrypt: disable all but standard v2 policies for extent encryption Josef Bacik
2023-09-26 18:01 ` [PATCH 04/35] blk-crypto: add a process bio callback Josef Bacik
2023-09-26 18:01 ` [PATCH 05/35] fscrypt: expose fscrypt_nokey_name Josef Bacik
2023-09-26 18:01 ` [PATCH 06/35] fscrypt: add documentation about extent encryption Josef Bacik
2023-09-26 18:01 ` Josef Bacik [this message]
2023-09-26 18:01 ` [PATCH 08/35] btrfs: disable various operations on encrypted inodes Josef Bacik
2023-09-26 18:01 ` [PATCH 09/35] btrfs: disable verity " Josef Bacik
2023-09-26 18:01 ` [PATCH 10/35] btrfs: start using fscrypt hooks Josef Bacik
2023-09-26 18:01 ` [PATCH 11/35] btrfs: add inode encryption contexts Josef Bacik
2023-09-26 18:01 ` [PATCH 12/35] btrfs: add new FEATURE_INCOMPAT_ENCRYPT flag Josef Bacik
2023-09-26 18:01 ` [PATCH 13/35] btrfs: adapt readdir for encrypted and nokey names Josef Bacik
2023-10-07 1:15 ` kernel test robot
2023-09-26 18:01 ` [PATCH 14/35] btrfs: handle " Josef Bacik
2023-09-26 18:01 ` [PATCH 15/35] btrfs: implement fscrypt ioctls Josef Bacik
2023-09-26 18:01 ` [PATCH 16/35] btrfs: add encryption to CONFIG_BTRFS_DEBUG Josef Bacik
2023-09-26 18:01 ` [PATCH 17/35] btrfs: add get_devices hook for fscrypt Josef Bacik
2023-09-26 18:01 ` [PATCH 18/35] btrfs: turn on inlinecrypt mount option for encrypt Josef Bacik
2023-09-26 18:01 ` [PATCH 19/35] btrfs: set file extent encryption excplicitly Josef Bacik
2023-09-26 18:01 ` [PATCH 20/35] btrfs: add fscrypt_info and encryption_type to extent_map Josef Bacik
2023-09-26 18:01 ` [PATCH 21/35] btrfs: add fscrypt_info and encryption_type to ordered_extent Josef Bacik
2023-09-26 18:01 ` [PATCH 22/35] btrfs: plumb through setting the fscrypt_info for ordered extents Josef Bacik
2023-09-26 18:01 ` [PATCH 23/35] btrfs: populate the ordered_extent with the fscrypt context Josef Bacik
2023-09-26 18:01 ` [PATCH 24/35] btrfs: keep track of fscrypt info and orig_start for dio reads Josef Bacik
2023-09-26 18:01 ` [PATCH 25/35] btrfs: add an optional encryption context to the end of file extents Josef Bacik
2023-09-26 18:01 ` [PATCH 26/35] btrfs: explicitly track file extent length for replace and drop Josef Bacik
2023-09-26 18:01 ` [PATCH 27/35] btrfs: pass through fscrypt_extent_info to the file extent helpers Josef Bacik
2023-09-26 18:01 ` [PATCH 28/35] btrfs: pass the fscrypt_info through the replace extent infrastructure Josef Bacik
2023-09-26 18:01 ` [PATCH 29/35] btrfs: implement the fscrypt extent encryption hooks Josef Bacik
2023-09-26 18:01 ` [PATCH 30/35] btrfs: setup fscrypt_extent_info for new extents Josef Bacik
2023-09-26 18:01 ` [PATCH 31/35] btrfs: populate ordered_extent with the orig offset Josef Bacik
2023-09-26 18:01 ` [PATCH 32/35] btrfs: set the bio fscrypt context when applicable Josef Bacik
2023-09-26 18:01 ` [PATCH 33/35] btrfs: add a bio argument to btrfs_csum_one_bio Josef Bacik
2023-09-26 18:02 ` [PATCH 34/35] btrfs: add orig_logical to btrfs_bio Josef Bacik
2023-09-26 18:02 ` [PATCH 35/35] btrfs: implement process_bio cb for fscrypt 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=e6b3d2767e1ca79224d00e275e9d34ebc6ffcb8f.1695750478.git.josef@toxicpanda.com \
--to=josef@toxicpanda.com \
--cc=ebiggers@kernel.org \
--cc=kernel-team@fb.com \
--cc=linux-btrfs@vger.kernel.org \
--cc=linux-fscrypt@vger.kernel.org \
--cc=ngompa13@gmail.com \
/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).