linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Nick Piggin <npiggin@kernel.dk>
To: linux-fsdevel@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Subject: [patch 25/28] fs: dcache reduce prune_one_dentry locking
Date: Wed, 17 Nov 2010 01:09:25 +1100	[thread overview]
Message-ID: <20101116142030.612454500@kernel.dk> (raw)
In-Reply-To: 20101116140900.039761100@kernel.dk

[-- Attachment #1: dcache-prune_one_dentry-less-lock.patch --]
[-- Type: text/plain, Size: 1562 bytes --]

prune_one_dentry can avoid quite a bit of locking in the common case where
ancestors have an elevated refcount. Alternatively, we could have gone the
other way and made fewer trylocks in the case where d_count goes to zero, but
is probably less common.

Signed-off-by: Nick Piggin <npiggin@kernel.dk>

---
 fs/dcache.c |   27 +++++++++++++++------------
 1 file changed, 15 insertions(+), 12 deletions(-)

Index: linux-2.6/fs/dcache.c
===================================================================
--- linux-2.6.orig/fs/dcache.c	2010-11-17 00:52:38.000000000 +1100
+++ linux-2.6/fs/dcache.c	2010-11-17 01:05:38.000000000 +1100
@@ -547,26 +547,29 @@ static void prune_one_dentry(struct dent
 	 * Prune ancestors.
 	 */
 	while (dentry) {
-		spin_lock(&dcache_inode_lock);
-again:
+relock:
 		spin_lock(&dentry->d_lock);
+		if (dentry->d_count > 1) {
+			dentry->d_count--;
+			spin_unlock(&dentry->d_lock);
+	 		return;
+	 	}
+		if (!spin_trylock(&dcache_inode_lock)) {
+relock2:
+			spin_unlock(&dentry->d_lock);
+			cpu_relax();
+			goto relock;
+		}
+
 		if (IS_ROOT(dentry))
 			parent = NULL;
 		else
 			parent = dentry->d_parent;
 		if (parent && !spin_trylock(&parent->d_lock)) {
-			spin_unlock(&dentry->d_lock);
-			goto again;
-		}
-		dentry->d_count--;
-		if (dentry->d_count) {
-			if (parent)
-				spin_unlock(&parent->d_lock);
-			spin_unlock(&dentry->d_lock);
 			spin_unlock(&dcache_inode_lock);
-			return;
+			goto relock2;
 		}
-
+		dentry->d_count--;
 		dentry_lru_del(dentry);
 		__d_drop(dentry);
 		dentry = d_kill(dentry, parent);



  parent reply	other threads:[~2010-11-16 14:26 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-11-16 14:09 [patch 00/28] [rfc] dcache scaling part 1 Nick Piggin
2010-11-16 14:09 ` [patch 01/28] fs: d_validate fixes Nick Piggin
2010-11-17 10:44   ` Andi Kleen
2010-11-18 20:51   ` David Miller
2010-11-18 20:59     ` David Miller
2010-11-19  5:05       ` Nick Piggin
2010-11-19  5:01     ` Nick Piggin
2010-11-16 14:09 ` [patch 02/28] kernel: kmem_ptr_validate considered harmful Nick Piggin
2010-11-16 14:09 ` [patch 03/28] fs: dcache documentation cleanup Nick Piggin
2010-11-16 14:09 ` [patch 04/28] fs: change d_delete semantics Nick Piggin
2010-11-17  0:16   ` Tim Pepper
2010-11-16 14:09 ` [patch 05/28] cifs: dont overwrite dentry name in d_revalidate Nick Piggin
2010-11-16 14:09 ` [patch 06/28] jfs: " Nick Piggin
2010-11-16 14:09 ` [patch 07/28] fs: change d_compare for rcu-walk Nick Piggin
2010-11-17  0:44   ` Tim Pepper
2010-11-16 14:09 ` [patch 08/28] fs: change d_hash " Nick Piggin
2010-11-17  0:50   ` Tim Pepper
2010-11-16 14:09 ` [patch 09/28] hostfs: simplify locking Nick Piggin
2010-11-16 14:09 ` [patch 10/28] fs: dcache scale hash Nick Piggin
2010-11-16 14:09 ` [patch 11/28] fs: dcache scale lru Nick Piggin
2010-11-16 14:09 ` [patch 12/28] fs: dcache scale dentry refcount Nick Piggin
2010-11-16 14:09 ` [patch 13/28] fs: dcache scale d_unhashed Nick Piggin
2010-11-19 19:41   ` Tim Pepper
2010-11-16 14:09 ` [patch 14/28] fs: dcache scale subdirs Nick Piggin
2010-11-19 19:41   ` Tim Pepper
2010-11-16 14:09 ` [patch 15/28] fs: scale inode alias list Nick Piggin
2010-11-19 19:41   ` Tim Pepper
2010-11-16 14:09 ` [patch 16/28] fs: Use rename lock and RCU for multi-step operations Nick Piggin
2010-11-19 19:42   ` Tim Pepper
2010-11-16 14:09 ` [patch 17/28] fs: increase d_name lock coverage Nick Piggin
2010-11-16 14:09 ` [patch 18/28] fs: dcache remove dcache_lock Nick Piggin
2010-11-16 14:09 ` [patch 19/28] fs: dcache avoid starvation in dcache multi-step operations Nick Piggin
2010-11-16 14:09 ` [patch 20/28] fs: dcache reduce dput locking Nick Piggin
2010-11-16 14:09 ` [patch 21/28] fs: dcache reduce locking in d_alloc Nick Piggin
2010-11-16 14:09 ` [patch 22/28] fs: dcache reduce dcache_inode_lock Nick Piggin
2010-11-16 14:09 ` [patch 23/28] fs: dcache rationalise dget variants Nick Piggin
2010-11-16 14:09 ` [patch 24/28] fs: dcache reduce d_parent locking Nick Piggin
2010-11-16 14:09 ` Nick Piggin [this message]
2010-11-16 14:09 ` [patch 26/28] fs: reduce dcache_inode_lock width in lru scanning Nick Piggin
2010-11-16 14:09 ` [patch 27/28] fs: use RCU in shrink_dentry_list to reduce lock nesting Nick Piggin
2010-11-16 14:09 ` [patch 28/28] fs: consolidate dentry kill sequence Nick Piggin
2010-11-17  2:12 ` [patch 00/28] [rfc] dcache scaling part 1 Dave Chinner
2010-11-17 10:56 ` Andi Kleen
2010-11-17 11:19   ` Nick Piggin
2010-11-17 12:01     ` Andi Kleen
2010-11-19 19:43 ` Tim Pepper

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=20101116142030.612454500@kernel.dk \
    --to=npiggin@kernel.dk \
    --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;
as well as URLs for NNTP newsgroup(s).