public inbox for linux-btrfs@vger.kernel.org
 help / color / mirror / Atom feed
From: Filipe David Borba Manana <fdmanana@gmail.com>
To: linux-btrfs@vger.kernel.org
Cc: Filipe David Borba Manana <fdmanana@gmail.com>
Subject: [PATCH v2] Btrfs: faster file extent item search in clone ioctl
Date: Mon, 13 Jan 2014 13:38:58 +0000	[thread overview]
Message-ID: <1389620338-17424-1-git-send-email-fdmanana@gmail.com> (raw)
In-Reply-To: <1389475828-19339-1-git-send-email-fdmanana@gmail.com>

When we are looking for file extent items that intersect the cloning
range, for each one that falls completely outside the range, don't
release the path and do another full tree search - just move on
to the next slot and copy the file extent item into our buffer only
if the item intersects the cloning range.
Also leave the btree node/leaf locks in spinning mode to not make other
tasks sleep since we process each leaf very quickly.

Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
---

V2: Set path->leave_spinning to 1.

 fs/btrfs/ioctl.c |   23 +++++++++++++----------
 1 file changed, 13 insertions(+), 10 deletions(-)

diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 3970f32..e3b72da 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -2893,7 +2893,7 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
 		vfree(buf);
 		return ret;
 	}
-
+	path->leave_spinning = 1;
 	path->reada = 2;
 	/* clone data */
 	key.objectid = btrfs_ino(src);
@@ -2911,6 +2911,7 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
 			goto out;
 
 		nritems = btrfs_header_nritems(path->nodes[0]);
+process_slot:
 		if (path->slots[0] >= nritems) {
 			ret = btrfs_next_leaf(BTRFS_I(src)->root, path);
 			if (ret < 0)
@@ -2937,11 +2938,6 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
 			u8 comp;
 			u64 endoff;
 
-			size = btrfs_item_size_nr(leaf, slot);
-			read_extent_buffer(leaf, buf,
-					   btrfs_item_ptr_offset(leaf, slot),
-					   size);
-
 			extent = btrfs_item_ptr(leaf, slot,
 						struct btrfs_file_extent_item);
 			comp = btrfs_file_extent_compression(leaf, extent);
@@ -2960,11 +2956,19 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
 				datal = btrfs_file_extent_ram_bytes(leaf,
 								    extent);
 			}
-			btrfs_release_path(path);
 
 			if (key.offset + datal <= off ||
-			    key.offset >= off + len - 1)
-				goto next;
+			    key.offset >= off + len - 1) {
+				path->slots[0]++;
+				goto process_slot;
+			}
+
+			size = btrfs_item_size_nr(leaf, slot);
+			read_extent_buffer(leaf, buf,
+					   btrfs_item_ptr_offset(leaf, slot),
+					   size);
+
+			btrfs_release_path(path);
 
 			memcpy(&new_key, &key, sizeof(new_key));
 			new_key.objectid = btrfs_ino(inode);
@@ -3135,7 +3139,6 @@ static int btrfs_clone(struct inode *src, struct inode *inode,
 			}
 			ret = btrfs_end_transaction(trans, root);
 		}
-next:
 		btrfs_release_path(path);
 		key.offset++;
 	}
-- 
1.7.9.5


  reply	other threads:[~2014-01-13 13:41 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-01-11 21:30 [PATCH] Btrfs: faster file extent item search in clone ioctl Filipe David Borba Manana
2014-01-13 13:38 ` Filipe David Borba Manana [this message]
2014-01-13 19:35 ` [PATCH v3] " Filipe David Borba Manana

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=1389620338-17424-1-git-send-email-fdmanana@gmail.com \
    --to=fdmanana@gmail.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