From: Li Zefan <lizf@cn.fujitsu.com>
To: Chris Mason <chris.mason@oracle.com>
Cc: "linux-btrfs@vger.kernel.org" <linux-btrfs@vger.kernel.org>,
Josef Bacik <josef@redhat.com>
Subject: [PATCH 05/12] btrfs: Add a helper try_merge_free_space()
Date: Thu, 27 Jan 2011 16:44:02 +0800 [thread overview]
Message-ID: <4D413052.5060505@cn.fujitsu.com> (raw)
In-Reply-To: <4D412FE6.7050101@cn.fujitsu.com>
When adding a new extent, we'll firstly see if we can merge
this extent to the left or/and right extent. Extract this as
a helper try_merge_free_space().
As a side effect, we fix a small bug that if the new extent
has non-bitmap left entry but is unmergeble, we'll directly
link the extent without trying to drop it into bitmap.
This also prepares for the next patch.
Reviewed-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
---
fs/btrfs/free-space-cache.c | 75 ++++++++++++++++++++++++------------------
1 files changed, 43 insertions(+), 32 deletions(-)
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index 2974c47..cf67dc3 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -1363,22 +1363,14 @@ out:
return ret;
}
-int btrfs_add_free_space(struct btrfs_block_group_cache *block_group,
- u64 offset, u64 bytes)
+bool try_merge_free_space(struct btrfs_block_group_cache *block_group,
+ struct btrfs_free_space *info)
{
- struct btrfs_free_space *right_info = NULL;
- struct btrfs_free_space *left_info = NULL;
- struct btrfs_free_space *info = NULL;
- int ret = 0;
-
- info = kzalloc(sizeof(struct btrfs_free_space), GFP_NOFS);
- if (!info)
- return -ENOMEM;
-
- info->offset = offset;
- info->bytes = bytes;
-
- spin_lock(&block_group->tree_lock);
+ struct btrfs_free_space *left_info;
+ struct btrfs_free_space *right_info;
+ bool merged = false;
+ u64 offset = info->offset;
+ u64 bytes = info->bytes;
/*
* first we want to see if there is free space adjacent to the range we
@@ -1392,27 +1384,11 @@ int btrfs_add_free_space(struct btrfs_block_group_cache *block_group,
else
left_info = tree_search_offset(block_group, offset - 1, 0, 0);
- /*
- * If there was no extent directly to the left or right of this new
- * extent then we know we're going to have to allocate a new extent, so
- * before we do that see if we need to drop this into a bitmap
- */
- if ((!left_info || left_info->bitmap) &&
- (!right_info || right_info->bitmap)) {
- ret = insert_into_bitmap(block_group, info);
-
- if (ret < 0) {
- goto out;
- } else if (ret) {
- ret = 0;
- goto out;
- }
- }
-
if (right_info && !right_info->bitmap) {
unlink_free_space(block_group, right_info);
info->bytes += right_info->bytes;
kfree(right_info);
+ merged = true;
}
if (left_info && !left_info->bitmap &&
@@ -1421,8 +1397,43 @@ int btrfs_add_free_space(struct btrfs_block_group_cache *block_group,
info->offset = left_info->offset;
info->bytes += left_info->bytes;
kfree(left_info);
+ merged = true;
}
+ return merged;
+}
+
+int btrfs_add_free_space(struct btrfs_block_group_cache *block_group,
+ u64 offset, u64 bytes)
+{
+ struct btrfs_free_space *info;
+ int ret = 0;
+
+ info = kzalloc(sizeof(struct btrfs_free_space), GFP_NOFS);
+ if (!info)
+ return -ENOMEM;
+
+ info->offset = offset;
+ info->bytes = bytes;
+
+ spin_lock(&block_group->tree_lock);
+
+ if (try_merge_free_space(block_group, info))
+ goto link;
+
+ /*
+ * There was no extent directly to the left or right of this new
+ * extent then we know we're going to have to allocate a new extent, so
+ * before we do that see if we need to drop this into a bitmap
+ */
+ ret = insert_into_bitmap(block_group, info);
+ if (ret < 0) {
+ goto out;
+ } else if (ret) {
+ ret = 0;
+ goto out;
+ }
+link:
ret = link_free_space(block_group, info);
if (ret)
kfree(info);
--
1.6.3
next prev parent reply other threads:[~2011-01-27 8:44 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-01-27 8:42 [PATCH 00/12] Btrfs: Some bug fixes Li Zefan
2011-01-27 8:42 ` [PATCH 01/12] btrfs: Fix threshold calculation for block groups smaller than 1GB Li Zefan
2011-01-27 8:42 ` [PATCH 02/12] btrfs: Add helper function free_bitmap() Li Zefan
2011-01-27 8:43 ` [PATCH 03/12] btrfs: Free fully occupied bitmap in cluster Li Zefan
2011-01-27 8:43 ` [PATCH 04/12] btrfs: Update stats when allocating from a cluster Li Zefan
2011-01-27 8:44 ` Li Zefan [this message]
2011-01-27 8:44 ` [PATCH 06/12] btrfs: Check mergeable free space when removing " Li Zefan
2011-01-27 8:44 ` [PATCH 07/12] Btrfs: Fix memory leak at umount Li Zefan
2011-01-27 8:44 ` [PATCH 08/12] Btrfs: Fix memory leak on finding existing super Li Zefan
2011-01-27 8:45 ` [PATCH 09/12] Btrfs: Free correct pointer after using strsep Li Zefan
2011-01-27 8:45 ` [PATCH 10/12] Btrfs: Don't return acl info when mounting with noacl option Li Zefan
2011-01-27 8:45 ` [PATCH 11/12] Btrfs: Fix memory leak in writepage fixup work Li Zefan
2011-01-27 8:46 ` [PATCH 12/12] Btrfs: Fix file clone when source offset is not 0 Li Zefan
2012-01-26 13:52 ` Jan Schmidt
2012-01-26 16:17 ` David Sterba
2012-01-30 6:33 ` Li Zefan
2012-01-30 10:03 ` Jan Schmidt
2012-02-01 9:44 ` Li Zefan
2012-02-02 4:31 ` Yan, Zheng
2012-02-02 21:00 ` Jan Schmidt
2012-02-02 4:25 ` Yan, Zheng
2012-02-02 5:31 ` Li Zefan
2012-02-02 6:44 ` Yan, Zheng
2011-01-30 23:48 ` [PATCH 00/12] Btrfs: Some bug fixes Chris Mason
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=4D413052.5060505@cn.fujitsu.com \
--to=lizf@cn.fujitsu.com \
--cc=chris.mason@oracle.com \
--cc=josef@redhat.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).