From: Marcos Paulo de Souza <mpdesouza@suse.com>
To: linux-btrfs@vger.kernel.org
Cc: dsterba@suse.com, Marcos Paulo de Souza <mpdesouza@suse.com>
Subject: [PATCH 1/8] fs: btrfs: Introduce btrfs_for_each_slot
Date: Thu, 26 Aug 2021 13:40:47 -0300 [thread overview]
Message-ID: <20210826164054.14993-2-mpdesouza@suse.com> (raw)
In-Reply-To: <20210826164054.14993-1-mpdesouza@suse.com>
There is a common pattern when search for a key in btrfs:
* Call btrfs_search_slot
* Endless loop
* If the found slot is bigger than the current items in the leaf, check the
next one
* If still not found in the next leaf, return 1
* Do something with the code
* Increment current slot, and continue
This pattern can be improved by creating an iterator macro, similar to
those for_each_X already existing in the linux kernel. Using this
approach means to reduce significantly boilerplate code, along making it
easier to newcomers to understand how to code works.
Signed-off-by: Marcos Paulo de Souza <mpdesouza@suse.com>
---
fs/btrfs/ctree.c | 28 ++++++++++++++++++++++++++++
fs/btrfs/ctree.h | 25 +++++++++++++++++++++++++
2 files changed, 53 insertions(+)
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 84627cbd5b5b..b1aa6e3987d0 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -2122,6 +2122,34 @@ int btrfs_search_backwards(struct btrfs_root *root, struct btrfs_key *key,
return ret;
}
+/* Search for a valid slot for the given path.
+ * @root: The root node of the tree.
+ * @key: Will contain a valid item if found.
+ * @path: The start point to validate the slot.
+ *
+ * Return 0 if the item is valid, 1 if not found and < 0 if error.
+ */
+int btrfs_valid_slot(struct btrfs_root *root, struct btrfs_key *key,
+ struct btrfs_path *path)
+{
+ while (1) {
+ int ret;
+ const int slot = path->slots[0];
+ const struct extent_buffer *leaf = path->nodes[0];
+
+ if (slot >= btrfs_header_nritems(leaf)) {
+ ret = btrfs_next_leaf(root, path);
+ if (ret)
+ return ret;
+ continue;
+ }
+ btrfs_item_key_to_cpu(leaf, key, slot);
+ break;
+ }
+
+ return 0;
+}
+
/*
* adjust the pointers going up the tree, starting at level
* making sure the right key of each node is points to 'key'.
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index f07c82fafa04..1e3c4a7741ca 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -2912,6 +2912,31 @@ int btrfs_next_old_leaf(struct btrfs_root *root, struct btrfs_path *path,
int btrfs_search_backwards(struct btrfs_root *root, struct btrfs_key *key,
struct btrfs_path *path);
+int btrfs_valid_slot(struct btrfs_root *root, struct btrfs_key *key,
+ struct btrfs_path *path);
+
+/* Search in @root for a given @key, and store the slot found in @found_key.
+ * @root: The root node of the tree.
+ * @key: The key we are looking for.
+ * @found_key: Will hold the found item.
+ * @path: Holds the current slot/leaf.
+ * @iter_ret: Contains the returned value from btrfs_search_slot and
+ * btrfs_valid_slot, whatever is executed later.
+ *
+ * The iter_ret is an output variable that will contain the result of the
+ * btrfs_search_slot if it returns an error, or the value returned from
+ * btrfs_valid_slot otherwise. The return value can be 0 if the something was
+ * found, 1 if there weren't bigger leaves, and <0 if error.
+ */
+#define btrfs_for_each_slot(root, key, found_key, path, iter_ret) \
+ for (iter_ret = btrfs_search_slot(NULL, root, key, path, 0, 0); \
+ ( \
+ iter_ret >= 0 && \
+ (iter_ret = btrfs_valid_slot(root, found_key, path)) == 0 \
+ ); \
+ path->slots[0]++ \
+ )
+
static inline int btrfs_next_old_item(struct btrfs_root *root,
struct btrfs_path *p, u64 time_seq)
{
--
2.31.1
next prev parent reply other threads:[~2021-08-26 16:42 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-08-26 16:40 [PATCHv2 0/8] btrfs: Create macro to iterate slots Marcos Paulo de Souza
2021-08-26 16:40 ` Marcos Paulo de Souza [this message]
2021-08-30 12:37 ` [PATCH 1/8] fs: btrfs: Introduce btrfs_for_each_slot Nikolay Borisov
2021-08-31 11:03 ` David Sterba
2021-08-31 11:06 ` David Sterba
2021-08-26 16:40 ` [PATCH 2/8] btrfs: block-group: use btrfs_for_each_slot in find_first_block_group Marcos Paulo de Souza
2021-08-26 16:40 ` [PATCH 3/8] btrfs: dev-replace: Use btrfs_for_each_slot in mark_block_group_to_copy Marcos Paulo de Souza
2021-08-26 16:40 ` [PATCH 4/8] btrfs: dir-item: use btrfs_for_each_slot in btrfs_search_dir_index_item Marcos Paulo de Souza
2021-08-26 16:40 ` [PATCH 5/8] btrfs: inode: use btrfs_for_each_slot in btrfs_read_readdir Marcos Paulo de Souza
2021-08-30 13:05 ` Nikolay Borisov
2021-08-31 11:10 ` David Sterba
2021-08-31 11:27 ` Nikolay Borisov
2021-08-26 16:40 ` [PATCH 6/8] btrfs: send: Use btrfs_for_each_slot macro Marcos Paulo de Souza
2021-08-30 13:53 ` Nikolay Borisov
2021-08-26 16:40 ` [PATCH 7/8] btrfs: volumes: use btrfs_for_each_slot in btrfs_read_chunk_tree Marcos Paulo de Souza
2021-08-26 16:40 ` [PATCH 8/8] btrfs: xattr: Use btrfs_for_each_slot macro in btrfs_listxattr Marcos Paulo de Souza
2021-08-26 18:06 ` [PATCHv2 0/8] btrfs: Create macro to iterate slots Marcos Paulo de Souza
2021-08-31 11:17 ` David Sterba
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=20210826164054.14993-2-mpdesouza@suse.com \
--to=mpdesouza@suse.com \
--cc=dsterba@suse.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).