From: Glauber Costa <glommer@parallels.com>
To: linux-kernel@vger.kernel.org
Cc: linux-fsdevel@vger.kernel.org,
containers@lists.linux-foundation.org,
Pavel Emelyanov <xemul@parallels.com>,
Al Viro <viro@zeniv.linux.org.uk>,
Hugh Dickins <hughd@google.com>, Nick Piggin <npiggin@kernel.dk>,
Andrea Arcangeli <aarcange@redhat.com>,
Rik van Riel <riel@redhat.com>,
Dave Hansen <dave@linux.vnet.ibm.com>,
James Bottomley <JBottomley@parallels.com>,
David Chinner <david@fromorbit.com>,
Glauber Costa <glommer@parallels.com>
Subject: [PATCH v2 2/4] Keep nr_dentry per super block
Date: Fri, 5 Aug 2011 04:35:42 +0400 [thread overview]
Message-ID: <1312504544-1108-3-git-send-email-glommer@parallels.com> (raw)
In-Reply-To: <1312504544-1108-1-git-send-email-glommer@parallels.com>
Now that we have per-sb shrinkers, it makes sense to have nr_dentries
stored per sb as well. We turn them into per-cpu counters so we can
keep acessing them without locking.
Signed-off-by: Glauber Costa <glommer@parallels.com>
CC: Dave Chinner <david@fromorbit.com>
---
fs/dcache.c | 12 +++++++++++-
fs/super.c | 15 ++++++++++++++-
include/linux/fs.h | 2 ++
3 files changed, 27 insertions(+), 2 deletions(-)
diff --git a/fs/dcache.c b/fs/dcache.c
index b05aac3..ac19d24 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -151,7 +151,13 @@ static void __d_free(struct rcu_head *head)
static void d_free(struct dentry *dentry)
{
BUG_ON(dentry->d_count);
+ /*
+ * It is cheaper to keep a global counter separate
+ * then to scan through all superblocks when needed
+ */
this_cpu_dec(nr_dentry);
+ this_cpu_dec(*dentry->d_sb->s_nr_dentry);
+
if (dentry->d_op && dentry->d_op->d_release)
dentry->d_op->d_release(dentry);
@@ -1224,7 +1230,11 @@ struct dentry *__d_alloc(struct super_block *sb, const struct qstr *name)
INIT_LIST_HEAD(&dentry->d_alias);
INIT_LIST_HEAD(&dentry->d_u.d_child);
d_set_d_op(dentry, dentry->d_sb->s_d_op);
-
+ /*
+ * It is cheaper to keep a global counter separate
+ * then to scan through all superblocks when needed
+ */
+ this_cpu_inc(*dentry->d_sb->s_nr_dentry);
this_cpu_inc(nr_dentry);
return dentry;
diff --git a/fs/super.c b/fs/super.c
index 3f56a26..9345385 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -112,6 +112,7 @@ static struct super_block *alloc_super(struct file_system_type *type)
{
struct super_block *s = kzalloc(sizeof(struct super_block), GFP_USER);
static const struct super_operations default_op;
+ int i;
if (s) {
if (security_sb_alloc(s)) {
@@ -119,15 +120,26 @@ static struct super_block *alloc_super(struct file_system_type *type)
s = NULL;
goto out;
}
+
+ s->s_nr_dentry = alloc_percpu(int);
+ if (!s->s_nr_dentry) {
+ security_sb_free(s);
+ kfree(s);
+ s = NULL;
+ goto out;
+ }
+ for_each_possible_cpu(i)
+ *per_cpu_ptr(s->s_nr_dentry, i) = 0;
+
#ifdef CONFIG_SMP
s->s_files = alloc_percpu(struct list_head);
if (!s->s_files) {
+ free_percpu(s->s_nr_dentry);
security_sb_free(s);
kfree(s);
s = NULL;
goto out;
} else {
- int i;
for_each_possible_cpu(i)
INIT_LIST_HEAD(per_cpu_ptr(s->s_files, i));
@@ -198,6 +210,7 @@ static inline void destroy_super(struct super_block *s)
{
#ifdef CONFIG_SMP
free_percpu(s->s_files);
+ free_percpu(s->s_nr_dentry);
#endif
security_sb_free(s);
kfree(s->s_subtype);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index f23bcb7..35113fd 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1399,6 +1399,8 @@ struct super_block {
struct list_head s_dentry_lru; /* unused dentry lru */
int s_nr_dentry_unused; /* # of dentry on lru */
+ int __percpu *s_nr_dentry; /* # of dentry on this sb */
+
/* s_inode_lru_lock protects s_inode_lru and s_nr_inodes_unused */
spinlock_t s_inode_lru_lock ____cacheline_aligned_in_smp;
struct list_head s_inode_lru; /* unused inode lru */
--
1.7.6
next prev parent reply other threads:[~2011-08-05 0:36 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-08-05 0:35 [PATCH v2 0/4] Per-container dcache size limitation Glauber Costa
2011-08-05 0:35 ` [PATCH v2 1/4] factor out single-shrinker code Glauber Costa
2011-08-05 0:35 ` Glauber Costa [this message]
2011-08-12 13:51 ` [PATCH v2 2/4] Keep nr_dentry per super block Eric Dumazet
2011-08-05 0:35 ` [PATCH v2 3/4] limit nr_dentries per superblock Glauber Costa
2011-08-12 13:56 ` Eric Dumazet
2011-08-12 19:18 ` Glauber Costa
2011-08-12 19:20 ` Glauber Costa
2011-08-05 0:35 ` [PATCH v2 4/4] parse options in the vfs level Glauber Costa
2011-08-12 10:52 ` [PATCH v2 0/4] Per-container dcache size limitation Glauber Costa
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=1312504544-1108-3-git-send-email-glommer@parallels.com \
--to=glommer@parallels.com \
--cc=JBottomley@parallels.com \
--cc=aarcange@redhat.com \
--cc=containers@lists.linux-foundation.org \
--cc=dave@linux.vnet.ibm.com \
--cc=david@fromorbit.com \
--cc=hughd@google.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=npiggin@kernel.dk \
--cc=riel@redhat.com \
--cc=viro@zeniv.linux.org.uk \
--cc=xemul@parallels.com \
/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
Powered by JetHome