All of lore.kernel.org
 help / color / mirror / Atom feed
From: Cedric Le Goater <clg@fr.ibm.com>
To: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	Herbert Poetzl <herbert@13thfloor.at>,
	Pavel Emelianov <xemul@openvz.org>
Subject: [PATCH -mm] add a kmem_cache for nsproxy objects
Date: Mon, 18 Jun 2007 22:53:13 +0200	[thread overview]
Message-ID: <4676F0B9.3080408@fr.ibm.com> (raw)

It should improve performance in some scenarii where a lot of 
these nsproxy objects are created by unsharing namespaces. This is
a typical use of virtual servers that are being created or entered.

This is also a good tool to find leaks and gather statistics on
namespace usage.

Signed-off-by: Cedric Le Goater <clg@fr.ibm.com>
---
 kernel/nsproxy.c |   21 +++++++++++++++++----
 1 file changed, 17 insertions(+), 4 deletions(-)

Index: 2.6.22-rc4-mm2/kernel/nsproxy.c
===================================================================
--- 2.6.22-rc4-mm2.orig/kernel/nsproxy.c
+++ 2.6.22-rc4-mm2/kernel/nsproxy.c
@@ -21,6 +21,8 @@
 #include <linux/utsname.h>
 #include <linux/pid_namespace.h>
 
+static struct kmem_cache *nsproxy_cachep;
+
 struct nsproxy init_nsproxy = INIT_NSPROXY(init_nsproxy);
 
 static inline void get_nsproxy(struct nsproxy *ns)
@@ -43,9 +45,11 @@ static inline struct nsproxy *clone_nspr
 {
 	struct nsproxy *ns;
 
-	ns = kmemdup(orig, sizeof(struct nsproxy), GFP_KERNEL);
-	if (ns)
+	ns = kmem_cache_alloc(nsproxy_cachep, GFP_KERNEL);
+	if (ns) {
+		memcpy(ns, orig, sizeof(struct nsproxy));
 		atomic_set(&ns->count, 1);
+	}
 	return ns;
 }
 
@@ -109,7 +113,7 @@ out_uts:
 	if (new_nsp->mnt_ns)
 		put_mnt_ns(new_nsp->mnt_ns);
 out_ns:
-	kfree(new_nsp);
+	kmem_cache_free(nsproxy_cachep, new_nsp);
 	return ERR_PTR(err);
 }
 
@@ -160,7 +164,7 @@ void free_nsproxy(struct nsproxy *ns)
 		put_pid_ns(ns->pid_ns);
 	if (ns->user_ns)
 		put_user_ns(ns->user_ns);
-	kfree(ns);
+	kmem_cache_free(nsproxy_cachep, ns);
 }
 
 /*
@@ -191,3 +195,12 @@ int unshare_nsproxy_namespaces(unsigned 
 	}
 	return err;
 }
+
+static int __init nsproxy_cache_init(void)
+{
+	nsproxy_cachep = kmem_cache_create("nsproxy", sizeof(struct nsproxy),
+					   0, SLAB_PANIC, NULL, NULL);
+	return 0;
+}
+
+module_init(nsproxy_cache_init);

             reply	other threads:[~2007-06-18 20:53 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-06-18 20:53 Cedric Le Goater [this message]
2007-06-19 18:35 ` [PATCH -mm] add a kmem_cache for nsproxy objects Andrew Morton
2007-06-19 19:04   ` Christoph Hellwig
2007-06-19 19:27     ` Christoph Lameter
2007-06-19 21:14       ` Pekka Enberg
2007-06-19 19:24   ` Christoph Lameter
2007-06-22  7:48     ` Cedric Le Goater
2007-06-22 16:22       ` Christoph Lameter

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=4676F0B9.3080408@fr.ibm.com \
    --to=clg@fr.ibm.com \
    --cc=akpm@linux-foundation.org \
    --cc=herbert@13thfloor.at \
    --cc=linux-kernel@vger.kernel.org \
    --cc=xemul@openvz.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.