public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: npiggin@suse.de
To: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org
Cc: Dave Hansen <haveblue@us.ibm.com>
Subject: [patch 04/27] fs: introduce mnt_clone_write
Date: Sat, 25 Apr 2009 11:20:24 +1000	[thread overview]
Message-ID: <20090425012209.513109971@suse.de> (raw)
In-Reply-To: 20090425012020.457460929@suse.de

[-- Attachment #1: mnt_clone_write.patch --]
[-- Type: text/plain, Size: 5254 bytes --]

This patch speeds up lmbench lat_mmap test by about another 2% after the
first patch.

Before:
 avg = 462.286
 std = 5.46106

After:
 avg = 453.12
 std = 9.58257

(50 runs of each, stddev gives a reasonable confidence)

It does this by introducing mnt_clone_write, which avoids some heavyweight
operations of mnt_want_write if called on a vfsmount which we know already
has a write count; and mnt_want_write_file, which can call mnt_clone_write
if the file is open for write.

After these two patches, mnt_want_write and mnt_drop_write go from 7% on
the profile down to 1.3% (including mnt_clone_write).

Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Nick Piggin <npiggin@suse.de>
---
 fs/file_table.c       |    3 +--
 fs/inode.c            |    2 +-
 fs/namespace.c        |   38 ++++++++++++++++++++++++++++++++++++++
 fs/open.c             |    4 ++--
 fs/xattr.c            |    4 ++--
 include/linux/mount.h |    4 ++++
 6 files changed, 48 insertions(+), 7 deletions(-)

Index: linux-2.6/fs/file_table.c
===================================================================
--- linux-2.6.orig/fs/file_table.c
+++ linux-2.6/fs/file_table.c
@@ -217,8 +217,7 @@ int init_file(struct file *file, struct
 	 */
 	if ((mode & FMODE_WRITE) && !special_file(dentry->d_inode->i_mode)) {
 		file_take_write(file);
-		error = mnt_want_write(mnt);
-		WARN_ON(error);
+		mnt_clone_write(mnt);
 	}
 	return error;
 }
Index: linux-2.6/fs/inode.c
===================================================================
--- linux-2.6.orig/fs/inode.c
+++ linux-2.6/fs/inode.c
@@ -1401,7 +1401,7 @@ void file_update_time(struct file *file)
 	if (IS_NOCMTIME(inode))
 		return;
 
-	err = mnt_want_write(file->f_path.mnt);
+	err = mnt_want_write_file(file->f_path.mnt, file);
 	if (err)
 		return;
 
Index: linux-2.6/fs/namespace.c
===================================================================
--- linux-2.6.orig/fs/namespace.c
+++ linux-2.6/fs/namespace.c
@@ -265,6 +265,44 @@ out:
 EXPORT_SYMBOL_GPL(mnt_want_write);
 
 /**
+ * mnt_clone_write - get write access to a mount
+ * @mnt: the mount on which to take a write
+ *
+ * This is effectively like mnt_want_write, except
+ * it must only be used to take an extra write reference
+ * on a mountpoint that we already know has a write reference
+ * on it. This allows some optimisation.
+ *
+ * After finished, mnt_drop_write must be called as usual to
+ * drop the reference.
+ */
+void mnt_clone_write(struct vfsmount *mnt)
+{
+	preempt_disable();
+	inc_mnt_writers(mnt);
+	preempt_enable();
+}
+EXPORT_SYMBOL_GPL(mnt_clone_write);
+
+/**
+ * mnt_want_write_file - get write access to a file's mount
+ * @file: the file who's mount on which to take a write
+ *
+ * This is like mnt_want_write, but it takes a file and can
+ * do some optimisations if the file is open for write already
+ */
+int mnt_want_write_file(struct vfsmount *mnt, struct file *file)
+{
+	if (!(file->f_mode & FMODE_WRITE))
+		return mnt_want_write(mnt);
+	else {
+		mnt_clone_write(mnt);
+		return 0;
+	}
+}
+EXPORT_SYMBOL_GPL(mnt_want_write_file);
+
+/**
  * mnt_drop_write - give up write access to a mount
  * @mnt: the mount on which to give up write access
  *
Index: linux-2.6/fs/open.c
===================================================================
--- linux-2.6.orig/fs/open.c
+++ linux-2.6/fs/open.c
@@ -612,7 +612,7 @@ SYSCALL_DEFINE2(fchmod, unsigned int, fd
 
 	audit_inode(NULL, dentry);
 
-	err = mnt_want_write(file->f_path.mnt);
+	err = mnt_want_write_file(file->f_path.mnt, file);
 	if (err)
 		goto out_putf;
 	mutex_lock(&inode->i_mutex);
@@ -761,7 +761,7 @@ SYSCALL_DEFINE3(fchown, unsigned int, fd
 	if (!file)
 		goto out;
 
-	error = mnt_want_write(file->f_path.mnt);
+	error = mnt_want_write_file(file->f_path.mnt, file);
 	if (error)
 		goto out_fput;
 	dentry = file->f_path.dentry;
Index: linux-2.6/fs/xattr.c
===================================================================
--- linux-2.6.orig/fs/xattr.c
+++ linux-2.6/fs/xattr.c
@@ -297,7 +297,7 @@ SYSCALL_DEFINE5(fsetxattr, int, fd, cons
 		return error;
 	dentry = f->f_path.dentry;
 	audit_inode(NULL, dentry);
-	error = mnt_want_write(f->f_path.mnt);
+	error = mnt_want_write_file(f->f_path.mnt, f);
 	if (!error) {
 		error = setxattr(dentry, name, value, size, flags);
 		mnt_drop_write(f->f_path.mnt);
@@ -524,7 +524,7 @@ SYSCALL_DEFINE2(fremovexattr, int, fd, c
 		return error;
 	dentry = f->f_path.dentry;
 	audit_inode(NULL, dentry);
-	error = mnt_want_write(f->f_path.mnt);
+	error = mnt_want_write_file(f->f_path.mnt, f);
 	if (!error) {
 		error = removexattr(dentry, name);
 		mnt_drop_write(f->f_path.mnt);
Index: linux-2.6/include/linux/mount.h
===================================================================
--- linux-2.6.orig/include/linux/mount.h
+++ linux-2.6/include/linux/mount.h
@@ -88,7 +88,11 @@ static inline struct vfsmount *mntget(st
 	return mnt;
 }
 
+struct file; /* forward dec */
+
 extern int mnt_want_write(struct vfsmount *mnt);
+extern int mnt_want_write_file(struct vfsmount *mnt, struct file *file);
+extern void mnt_clone_write(struct vfsmount *mnt);
 extern void mnt_drop_write(struct vfsmount *mnt);
 extern void mntput_no_expire(struct vfsmount *mnt);
 extern void mnt_pin(struct vfsmount *mnt);



  parent reply	other threads:[~2009-04-25  1:38 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-04-25  1:20 [patch 00/27] [rfc] vfs scalability patchset npiggin
2009-04-25  1:20 ` [patch 01/27] fs: cleanup files_lock npiggin
2009-04-25  3:20   ` Al Viro
2009-04-25  5:35   ` Eric W. Biederman
2009-04-26  6:12     ` Nick Piggin
2009-04-25  9:42   ` Alan Cox
2009-04-26  6:15     ` Nick Piggin
2009-04-25  1:20 ` [patch 02/27] fs: scale files_lock npiggin
2009-04-25  3:32   ` Al Viro
2009-04-25  1:20 ` [patch 03/27] fs: mnt_want_write speedup npiggin
2009-04-25  1:20 ` npiggin [this message]
2009-04-25  3:35   ` [patch 04/27] fs: introduce mnt_clone_write Al Viro
2009-04-25  1:20 ` [patch 05/27] fs: brlock vfsmount_lock npiggin
2009-04-25  3:50   ` Al Viro
2009-04-26  6:36     ` Nick Piggin
2009-04-25  1:20 ` [patch 06/27] fs: dcache fix LRU ordering npiggin
2009-04-25  1:20 ` [patch 07/27] fs: dcache scale hash npiggin
2009-04-25  1:20 ` [patch 08/27] fs: dcache scale lru npiggin
2009-04-25  1:20 ` [patch 09/27] fs: dcache scale nr_dentry npiggin
2009-04-25  1:20 ` [patch 10/27] fs: dcache scale dentry refcount npiggin
2009-04-25  1:20 ` [patch 11/27] fs: dcache scale d_unhashed npiggin
2009-04-25  1:20 ` [patch 12/27] fs: dcache scale subdirs npiggin
2009-04-25  1:20 ` [patch 13/27] fs: scale inode alias list npiggin
2009-04-25  1:20 ` [patch 14/27] fs: use RCU / seqlock logic for reverse and multi-step operaitons npiggin
2009-04-25  1:20 ` [patch 15/27] fs: dcache remove dcache_lock npiggin
2009-04-25  1:20 ` [patch 16/27] fs: dcache reduce dput locking npiggin
2009-04-25  1:20 ` [patch 17/27] fs: dcache per-bucket dcache hash locking npiggin
2009-04-25  1:20 ` [patch 18/27] fs: dcache reduce dcache_inode_lock npiggin
2009-04-25  1:20 ` [patch 19/27] fs: dcache per-inode inode alias locking npiggin
2009-04-25  1:20 ` [patch 20/27] fs: icache lock s_inodes list npiggin
2009-04-25  1:20 ` [patch 21/27] fs: icache lock inode hash npiggin
2009-04-25  1:20 ` [patch 22/27] fs: icache lock i_state npiggin
2009-04-25  1:20 ` [patch 23/27] fs: icache lock i_count npiggin
2009-04-25  1:20 ` [patch 24/27] fs: icache atomic inodes_stat npiggin
2009-04-25  1:20 ` [patch 25/27] fs: icache lock lru/writeback lists npiggin
2009-04-25  1:20 ` [patch 26/27] fs: icache protect inode state npiggin
2009-04-25  1:20 ` [patch 27/27] fs: icache remove inode_lock npiggin
2009-04-25  4:18 ` [patch 00/27] [rfc] vfs scalability patchset Al Viro
2009-04-25  5:02   ` Nick Piggin
2009-04-25  8:01   ` Christoph Hellwig
2009-04-25  8:06     ` Al Viro
2009-04-28  9:09       ` Christoph Hellwig
2009-04-28  9:48         ` Nick Piggin
2009-04-28 10:58         ` Peter Zijlstra
2009-04-28 11:32         ` Eric W. Biederman
2009-04-30  6:14           ` Nick Piggin
2009-04-25 19:08     ` Eric W. Biederman
2009-04-25 19:31       ` Al Viro
2009-04-25 20:29         ` Eric W. Biederman
2009-04-25 22:05           ` Theodore Tso

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=20090425012209.513109971@suse.de \
    --to=npiggin@suse.de \
    --cc=haveblue@us.ibm.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@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