public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Pavel Emelianov <xemul@openvz.org>
To: Andrew Morton <akpm@osdl.org>
Cc: Sukadev Bhattiprolu <sukadev@us.ibm.com>,
	Serge Hallyn <serue@us.ibm.com>,
	"Eric W. Biederman" <ebiederm@xmission.com>,
	Linux Containers <containers@lists.osdl.org>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	Kirill Korotaev <dev@openvz.org>
Subject: [PATCH 12/16] Reference counting of pid naspaces by pids
Date: Fri, 06 Jul 2007 12:10:06 +0400	[thread overview]
Message-ID: <468DF8DE.4000800@openvz.org> (raw)
In-Reply-To: <468DF6F7.1010906@openvz.org>

Getting and putting the pid namespace in alloc_pid() and
free_pid() is too slow. Instead this I get/put the namespace
by the pidmaps. When the pidmap allocates its first pid the
namespace is get, when the pidmap becomes empty - the namespace
is put.

Although pids may live longer than their "fingerpints" in the
pidmaps, this is ok to release the namespace with not yet freed
struct pids, as this pid is not alive and no routines will
use it.

Signed-off-by: Pavel Emelianov <xemul@openvz.org>

---

 pid.c |   19 +++++++++++++++----
 1 files changed, 15 insertions(+), 4 deletions(-)

diff -upr linux-2.6.22-rc4-mm2.orig/kernel/pid.c linux-2.6.22-rc4-mm2-2/kernel/pid.c
--- linux-2.6.22-rc4-mm2.orig/kernel/pid.c	2007-06-14 12:14:29.000000000 +0400
+++ linux-2.6.22-rc4-mm2-2/kernel/pid.c	2007-07-04 19:00:38.000000000 +0400
@@ -89,16 +95,23 @@ static  __cacheline_aligned_in_smp DEFIN
 
 static fastcall void free_pidmap(struct pid_namespace *pid_ns, int pid)
 {
-	struct pidmap *map = pid_ns->pidmap + pid / BITS_PER_PAGE;
 	int offset = pid & BITS_PER_PAGE_MASK;
+	int map_id = pid / BITS_PER_PAGE;
+	struct pidmap *map = pid_ns->pidmap + map_id;
+	int free_pids;
 
 	clear_bit(offset, map->page);
-	atomic_inc(&map->nr_free);
+	free_pids = atomic_inc_return(&map->nr_free);
+
+	if (map_id == 0)
+		free_pids++;
+	if (free_pids == BITS_PER_PAGE)
+		put_pid_ns(pid_ns);
 }
 
 static int alloc_pidmap(struct pid_namespace *pid_ns)
 {
-	int i, offset, max_scan, pid, last = pid_ns->last_pid;
+	int i, offset, max_scan, pid, last = pid_ns->last_pid, free_pids;
 	struct pidmap *map;
 
 	pid = last + 1;
@@ -126,7 +139,11 @@ static int alloc_pidmap(struct pid_names
 		if (likely(atomic_read(&map->nr_free))) {
 			do {
 				if (!test_and_set_bit(offset, map->page)) {
-					atomic_dec(&map->nr_free);
+					free_pids = atomic_dec_return(
+							&map->nr_free);
+					if (free_pids == BITS_PER_PAGE - 1)
+						get_pid_ns(pid_ns);
+
 					pid_ns->last_pid = pid;
 					return pid;
 				}


  parent reply	other threads:[~2007-07-06  8:17 UTC|newest]

Thread overview: 64+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-07-06  8:01 [PATCH 0/16] Pid namespaces Pavel Emelianov
2007-07-06  8:03 ` [PATCH 1/16] Round up the API Pavel Emelianov
2007-07-09 20:18   ` Cedric Le Goater
2007-07-10  6:40     ` Pavel Emelianov
2007-07-10  7:34       ` Andrew Morton
2007-07-06  8:03 ` [PATCH 2/16] Miscelaneous preparations for namespaces Pavel Emelianov
2007-07-09 20:22   ` Cedric Le Goater
2007-07-10  6:42     ` Pavel Emelianov
2007-07-06  8:04 ` [PATCH 3/16] Introduce MS_KERNMOUNT flag Pavel Emelianov
2007-07-06  8:05 ` [PATCH 4/16] Change data structures for pid namespaces Pavel Emelianov
2007-07-09 20:25   ` Cedric Le Goater
2007-07-10  4:32     ` sukadev
2007-07-10  7:04       ` Pavel Emelianov
2007-07-10 12:07         ` Cedric Le Goater
2007-07-06  8:05 ` [PATCH 5/16] Make proc be mountable from different " Pavel Emelianov
2007-07-06  8:06 ` [PATCH 6/16] Helpers to obtain pid numbers Pavel Emelianov
2007-07-10  5:18   ` sukadev
2007-07-10  6:49     ` Pavel Emelianov
2007-07-06  8:07 ` [PATCH 7/16] Helpers to find the task by its numerical ids Pavel Emelianov
2007-07-10  4:00   ` sukadev
2007-07-10  6:47     ` Pavel Emelianov
2007-07-06  8:07 ` [PATCH 8/16] Masquerade the siginfo when sending a pid to a foreign namespace Pavel Emelianov
2007-07-10  4:18   ` sukadev
2007-07-10  6:56     ` Pavel Emelianov
2007-07-06  8:08 ` [PATCH 9/16] Make proc_flust_task to flush entries from multiple proc trees Pavel Emelianov
2007-07-06  8:08 ` [PATCH 10/16] Changes in copy_process() to work with pid namespaces Pavel Emelianov
2007-07-12  0:21   ` sukadev
2007-07-06  8:09 ` [PATCH 11/16] Add support for multiple kmem caches for pids Pavel Emelianov
2007-07-06  8:10 ` Pavel Emelianov [this message]
2007-07-06  8:10 ` [PATCH 13/16] Switch to operating with pid_numbers instead of pids Pavel Emelianov
2007-07-25  0:36   ` sukadev
2007-07-25 10:07     ` Pavel Emelyanov
2007-07-25 19:13       ` sukadev
2007-07-26  6:42         ` Pavel Emelyanov
2007-07-06  8:11 ` [PATCH 14/16] Make pid namespaces clonnable Pavel Emelianov
2007-07-06  8:13 ` [PATCH 15/16] Changes to show virtual ids to user Pavel Emelianov
2007-07-06  8:16 ` [PATCH 16/16] Remove already unneeded memners from struct pid Pavel Emelianov
2007-07-06 16:26 ` [PATCH 0/16] Pid namespaces Dave Hansen
2007-07-09  5:58   ` Pavel Emelianov
2007-07-09 19:58     ` Dave Hansen
2007-07-09 12:02 ` Herbert Poetzl
2007-07-09 13:16   ` Pavel Emelianov
2007-07-09 19:52     ` Herbert Poetzl
2007-07-09 20:12       ` Cedric Le Goater
2007-07-10  6:59         ` Pavel Emelianov
2007-07-09 17:46 ` Badari Pulavarty
2007-07-09 20:06   ` Cedric Le Goater
2007-07-09 23:00     ` Badari Pulavarty
2007-07-10  7:05       ` Pavel Emelianov
2007-07-10 11:30     ` Pavel Emelianov
2007-07-10 12:05       ` Daniel Lezcano
2007-07-10 13:03         ` Pavel Emelianov
2007-07-10 20:34       ` Badari Pulavarty
2007-07-10 13:06   ` Pavel Emelianov
2007-07-10 20:33     ` Badari Pulavarty
2007-07-09 21:42 ` sukadev
2007-07-10  0:29 ` sukadev
2007-07-10  9:41   ` Pavel Emelianov
2007-07-10 13:08   ` Pavel Emelianov
2007-07-10  4:26 ` sukadev
2007-07-10  7:02   ` Pavel Emelianov
2007-07-11  1:16 ` Matt Mackall
2007-07-11  6:39   ` Pavel Emelianov
2007-07-11 15:14     ` Matt Mackall

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=468DF8DE.4000800@openvz.org \
    --to=xemul@openvz.org \
    --cc=akpm@osdl.org \
    --cc=containers@lists.osdl.org \
    --cc=dev@openvz.org \
    --cc=ebiederm@xmission.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=serue@us.ibm.com \
    --cc=sukadev@us.ibm.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