From: Josef Bacik <josef@redhat.com>
To: linux-btrfs@vger.kernel.org
Subject: [PATCH 2/2] Btrfs: cache ordered extent when completing io
Date: Tue, 2 Feb 2010 15:51:14 -0500 [thread overview]
Message-ID: <20100202205113.GC15957@localhost.localdomain> (raw)
In-Reply-To: <20100202205010.GB15957@localhost.localdomain>
When finishing io we run btrfs_dec_test_ordered_pending, and then immediately
run btrfs_lookup_ordered_extent, but btrfs_dec_test_ordered_pending does that
already, so we're searching twice when we don't have to. This patch lets us
pass a btrfs_ordered_extent in to btrfs_dec_test_ordered_pending so if we do
complete io on that ordered extent we can just use the one we found then instead
of having to do another btrfs_lookup_ordered_extent. This made my fio job with
the other patch go from 24 mb/s to 29 mb/s.
Signed-off-by: Josef Bacik <josef@redhat.com>
---
fs/btrfs/inode.c | 5 ++---
fs/btrfs/ordered-data.c | 7 ++++++-
fs/btrfs/ordered-data.h | 3 ++-
3 files changed, 10 insertions(+), 5 deletions(-)
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 9a1c746..26574d8 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -1692,11 +1692,10 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end)
int compressed = 0;
int ret;
- ret = btrfs_dec_test_ordered_pending(inode, start, end - start + 1);
+ ret = btrfs_dec_test_ordered_pending(inode, &ordered_extent, start,
+ end - start + 1);
if (!ret)
return 0;
-
- ordered_extent = btrfs_lookup_ordered_extent(inode, start);
BUG_ON(!ordered_extent);
if (test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags)) {
diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index 5c2a9e7..c44054b 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -232,11 +232,12 @@ int btrfs_add_ordered_sum(struct inode *inode,
* to make sure this function only returns 1 once for a given ordered extent.
*/
int btrfs_dec_test_ordered_pending(struct inode *inode,
+ struct btrfs_ordered_extent **cached,
u64 file_offset, u64 io_size)
{
struct btrfs_ordered_inode_tree *tree;
struct rb_node *node;
- struct btrfs_ordered_extent *entry;
+ struct btrfs_ordered_extent *entry = NULL;
int ret;
tree = &BTRFS_I(inode)->ordered_tree;
@@ -264,6 +265,10 @@ int btrfs_dec_test_ordered_pending(struct inode *inode,
else
ret = 1;
out:
+ if (!ret && cached && entry) {
+ *cached = entry;
+ atomic_inc(&entry->refs);
+ }
mutex_unlock(&tree->mutex);
return ret == 0;
}
diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h
index 1fe1282..398bfc6 100644
--- a/fs/btrfs/ordered-data.h
+++ b/fs/btrfs/ordered-data.h
@@ -137,7 +137,8 @@ int btrfs_put_ordered_extent(struct btrfs_ordered_extent *entry);
int btrfs_remove_ordered_extent(struct inode *inode,
struct btrfs_ordered_extent *entry);
int btrfs_dec_test_ordered_pending(struct inode *inode,
- u64 file_offset, u64 io_size);
+ struct btrfs_ordered_extent **cached,
+ u64 file_offset, u64 io_size);
int btrfs_add_ordered_extent(struct inode *inode, u64 file_offset,
u64 start, u64 len, u64 disk_len, int tyep);
int btrfs_add_ordered_sum(struct inode *inode,
--
1.6.6
prev parent reply other threads:[~2010-02-02 20:51 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-02-02 20:50 [PATCH 1/2] Btrfs: do not try and lookup the file extent when finishing ordered io Josef Bacik
2010-02-02 20:51 ` Josef Bacik [this message]
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=20100202205113.GC15957@localhost.localdomain \
--to=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