From: Frederik Deweerdt <deweerdt@free.fr>
To: linux-kernel@vger.kernel.org
Cc: akpm@osdl.org, acme@mandriva.com, marcel@holtmann.org, jet@gyve.org
Subject: [02/04 mm-patch, rfc] Add lightweight rwlock (was Re: [mm-patch] bluetooth: use GFP_ATOMIC in *_sock_create's sk_alloc)
Date: Fri, 28 Jul 2006 18:23:20 +0200 [thread overview]
Message-ID: <20060728162320.GB1227@slug> (raw)
In-Reply-To: <20060728161515.GA1227@slug>
[-- Attachment #1: Type: text/plain, Size: 239 bytes --]
This patch is part of the lw_rwlock patchset, it removes the
net_family_{read,write}_{lock,unlock} functions which have been moved
to linux/lw_rwlock.h and made more generic.
Signed-off-by: Frederik Deweerdt <frederik.deweerdt@gmail.com>
[-- Attachment #2: net_socket.c-use-lw_rwlocks.patch --]
[-- Type: text/plain, Size: 2941 bytes --]
--- v2.6.18-rc2-mm1~ori/net/socket.c 2006-07-27 11:46:12.000000000 +0200
+++ v2.6.18-rc2-mm1/net/socket.c 2006-07-28 15:50:06.000000000 +0200
@@ -85,6 +85,7 @@
#include <linux/kmod.h>
#include <linux/audit.h>
#include <linux/wireless.h>
+#include <linux/lw_rwlock.h>
#include <asm/uaccess.h>
#include <asm/unistd.h>
@@ -143,50 +144,7 @@ static struct file_operations socket_fil
static struct net_proto_family *net_families[NPROTO];
-#if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT)
-static atomic_t net_family_lockct = ATOMIC_INIT(0);
-static DEFINE_SPINLOCK(net_family_lock);
-
-/* The strategy is: modifications net_family vector are short, do not
- sleep and veeery rare, but read access should be free of any exclusive
- locks.
- */
-
-static void net_family_write_lock(void)
-{
- spin_lock(&net_family_lock);
- while (atomic_read(&net_family_lockct) != 0) {
- spin_unlock(&net_family_lock);
-
- yield();
-
- spin_lock(&net_family_lock);
- }
-}
-
-static __inline__ void net_family_write_unlock(void)
-{
- spin_unlock(&net_family_lock);
-}
-
-static __inline__ void net_family_read_lock(void)
-{
- atomic_inc(&net_family_lockct);
- spin_unlock_wait(&net_family_lock);
-}
-
-static __inline__ void net_family_read_unlock(void)
-{
- atomic_dec(&net_family_lockct);
-}
-
-#else
-#define net_family_write_lock() do { } while(0)
-#define net_family_write_unlock() do { } while(0)
-#define net_family_read_lock() do { } while(0)
-#define net_family_read_unlock() do { } while(0)
-#endif
-
+static DEFINE_LW_RWLOCK(net_family_lock);
/*
* Statistics counters of the socket lists
@@ -1125,7 +1083,7 @@ static int __sock_create(int family, int
}
#endif
- net_family_read_lock();
+ lw_read_lock(&net_family_lock);
if (net_families[family] == NULL) {
err = -EAFNOSUPPORT;
goto out;
@@ -1176,7 +1134,7 @@ static int __sock_create(int family, int
security_socket_post_create(sock, family, type, protocol, kern);
out:
- net_family_read_unlock();
+ lw_read_unlock(&net_family_lock);
return err;
out_module_put:
module_put(net_families[family]->owner);
@@ -2025,13 +1983,13 @@ int sock_register(struct net_proto_famil
printk(KERN_CRIT "protocol %d >= NPROTO(%d)\n", ops->family, NPROTO);
return -ENOBUFS;
}
- net_family_write_lock();
+ lw_write_lock(&net_family_lock);
err = -EEXIST;
if (net_families[ops->family] == NULL) {
net_families[ops->family]=ops;
err = 0;
}
- net_family_write_unlock();
+ lw_write_unlock(&net_family_lock);
printk(KERN_INFO "NET: Registered protocol family %d\n",
ops->family);
return err;
@@ -2048,9 +2006,9 @@ int sock_unregister(int family)
if (family < 0 || family >= NPROTO)
return -1;
- net_family_write_lock();
+ lw_write_lock(&net_family_lock);
net_families[family]=NULL;
- net_family_write_unlock();
+ lw_write_unlock(&net_family_lock);
printk(KERN_INFO "NET: Unregistered protocol family %d\n",
family);
return 0;
next prev parent reply other threads:[~2006-07-28 16:23 UTC|newest]
Thread overview: 107+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-07-27 8:56 2.6.18-rc2-mm1 Andrew Morton
2006-07-27 10:27 ` [PATCH] Require mmap handler for a.out executables (was Re: 2.6.18-rc2-mm1) Eugene Teo
2006-07-27 11:40 ` [patch -mm] s390: remove s390 touch_nmi_watchdog() define Heiko Carstens
2006-07-27 12:26 ` 2.6.18-rc2-mm1 Frederik Deweerdt
2006-07-27 12:39 ` [patch] fix "efi_init_e820_map undefined" warning Frederik Deweerdt
2006-07-27 13:12 ` Should cpuset ABBA deadlock fix be in 2.6.18-rc2-mmx? Paul Jackson
2006-07-27 18:22 ` Andrew Morton
2006-07-27 19:32 ` Paul Jackson
2006-07-27 13:32 ` 2.6.18-rc2-mm1 Michal Piotrowski
2006-07-27 18:59 ` 2.6.18-rc2-mm1 Michal Piotrowski
2006-07-29 12:15 ` 2.6.18-rc2-mm1 Michal Piotrowski
2006-07-29 12:17 ` 2.6.18-rc2-mm1 Michal Piotrowski
2006-07-28 8:17 ` 2.6.18-rc2-mm1 Michal Piotrowski
2006-07-28 8:34 ` 2.6.18-rc2-mm1 Andrew Morton
2006-07-28 18:49 ` 2.6.18-rc2-mm1 Matt Helsley
2006-07-28 19:53 ` 2.6.18-rc2-mm1 Michal Piotrowski
2006-07-28 20:39 ` 2.6.18-rc2-mm1 Matt Helsley
2006-07-28 21:34 ` 2.6.18-rc2-mm1 Andrew Morton
2006-07-29 2:04 ` 2.6.18-rc2-mm1 Valdis.Kletnieks
2006-07-29 22:34 ` 2.6.18-rc2-mm1 Shailabh Nagar
2006-07-29 23:38 ` 2.6.18-rc2-mm1 Michal Piotrowski
2006-07-28 17:57 ` 2.6.18-rc2-mm1 Matt Helsley
2006-07-27 14:04 ` 2.6.18-rc2-mm1 Andy Whitcroft
2006-07-27 14:48 ` 2.6.18-rc2-mm1 Andy Whitcroft
2006-07-27 15:37 ` [PATCH] highmem: fixed ip27-memory.c build error Yoichi Yuasa
2006-07-27 18:16 ` [-mm patch] arch/i386/pci/mmconfig.c: fixes Adrian Bunk
2006-07-28 8:09 ` 2.6.18-rc2-mm1 Reuben Farrelly
2006-07-28 8:35 ` [mm-patch] bluetooth: use GFP_ATOMIC in *_sock_create's sk_alloc Frederik Deweerdt
2006-07-28 9:00 ` Marcel Holtmann
2006-07-28 12:36 ` Frederik Deweerdt
2006-07-28 9:17 ` Masatake YAMATO
2006-07-28 12:32 ` Frederik Deweerdt
2006-07-28 13:12 ` Masatake YAMATO
2006-07-28 16:15 ` [01/04 mm-patch, rfc] Add lightweight rwlock (was Re: [mm-patch] bluetooth: use GFP_ATOMIC in *_sock_create's sk_alloc) Frederik Deweerdt
2006-07-28 16:23 ` Frederik Deweerdt [this message]
2006-07-28 16:28 ` [03/04 mm-patch, rfc] Add lightweight rwlock to net/dccp/ccid.c " Frederik Deweerdt
2006-07-28 16:33 ` [04/04 mm-patch, rfc] Add lightweight rwlock to net/bluetooth/af_bluetooth.c " Frederik Deweerdt
2006-07-31 7:06 ` [01/04 mm-patch, rfc] Add lightweight rwlock Masatake YAMATO
2006-08-01 9:06 ` Frederik Deweerdt
2006-07-28 8:56 ` 2.6.18-rc2-mm1 Michal Piotrowski
2006-07-28 9:23 ` 2.6.18-rc2-mm1 Andrew Morton
2006-07-28 15:53 ` [PATCH] 2.6.18-rc2-mm1 i386 add_memory_region undefined Valdis.Kletnieks
2006-07-28 18:20 ` 2.6.18-rc2-mm1 - hard lockups on Dell C840 Valdis.Kletnieks
2006-07-28 18:44 ` 2.6.18-rc2-mm1 timer int 0 doesn't work Paul Fulghum
2006-07-28 21:48 ` Andrew Morton
2006-07-28 22:10 ` Paul Fulghum
2006-07-28 23:38 ` Andi Kleen
2006-07-29 0:15 ` Paul Fulghum
2006-07-29 1:16 ` Paul Fulghum
2006-07-29 1:24 ` Andrew Morton
2006-07-29 2:37 ` Paul Fulghum
2006-07-29 2:58 ` Eric W. Biederman
2006-07-29 4:03 ` Ingo Molnar
2006-07-30 23:00 ` Steven Rostedt
2006-07-29 2:36 ` Andi Kleen
2006-07-29 15:33 ` Paul Fulghum
2006-07-29 19:50 ` Eric W. Biederman
2006-07-29 22:05 ` Paul Fulghum
2006-07-31 5:31 ` Andi Kleen
2006-07-31 13:32 ` Paul Fulghum
2006-07-28 19:46 ` Kubuntu's udev broken with 2.6.18-rc2-mm1 Andrew James Wade
2006-07-27 19:56 ` Andrew Morton
2006-07-27 20:12 ` Greg KH
2006-07-28 14:33 ` Andrew James Wade
2006-07-30 14:01 ` Laurent Riffard
2006-07-31 0:03 ` Greg KH
2006-07-31 2:27 ` Andrew James Wade
2006-07-31 3:37 ` Greg KH
2006-07-31 4:22 ` Andrew Morton
2006-07-31 4:35 ` Greg KH
2006-07-31 4:50 ` Andrew Morton
2006-07-31 5:15 ` Greg KH
2006-07-31 6:00 ` Andrew Morton
2006-07-31 7:54 ` bert hubert
2006-07-31 8:30 ` Jesper Juhl
2006-07-31 11:14 ` Alan Cox
2006-07-31 8:10 ` Laurent Riffard
2006-08-01 3:01 ` Andrew James Wade
2006-07-27 21:28 ` Valdis.Kletnieks
2006-07-29 17:48 ` [-mm patch] security/selinux/hooks.c: make 4 functions static Adrian Bunk
2006-07-30 0:37 ` James Morris
2006-07-29 17:58 ` swsusp regression (s2dsk) [Was: 2.6.18-rc2-mm1] Jiri Slaby
2006-07-29 18:59 ` Rafael J. Wysocki
2006-07-29 23:06 ` Jiri Slaby
2006-07-29 23:10 ` Rafael J. Wysocki
2006-07-29 23:59 ` Jiri Slaby
2006-07-30 0:03 ` Jiri Slaby
2006-07-29 23:22 ` Pavel Machek
2006-07-29 23:58 ` Jiri Slaby
2006-07-30 0:06 ` Pavel Machek
2006-07-30 7:31 ` Rafael J. Wysocki
2006-07-30 8:08 ` Jiri Slaby
2006-07-30 9:28 ` Rafael J. Wysocki
2006-07-30 10:54 ` Jiri Slaby
2006-07-30 11:08 ` Pavel Machek
2006-07-30 11:34 ` Rafael J. Wysocki
2006-07-31 13:59 ` [Alsa-devel] " Takashi Iwai
2006-07-31 14:03 ` Pavel Machek
2006-07-30 11:36 ` James Courtier-Dutton
2006-07-30 11:35 ` 2.6.18-rc2-mm1 fails to reboot properly on Dell Latitude CPiA Christian Trefzer
2006-07-31 4:42 ` 2.6.18-rc2-mm1 Reuben Farrelly
2006-07-31 4:57 ` 2.6.18-rc2-mm1 Andrew Morton
2006-07-31 5:25 ` 2.6.18-rc2-mm1 Andi Kleen
2006-08-03 15:59 ` [2.6 patch] DVB_CORE must select I2C Adrian Bunk
2006-08-03 16:10 ` [v4l-dvb-maintainer] " Manu Abraham
2006-08-03 16:30 ` Trent Piepho
2006-08-03 19:13 ` Mauro Carvalho Chehab
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=20060728162320.GB1227@slug \
--to=deweerdt@free.fr \
--cc=acme@mandriva.com \
--cc=akpm@osdl.org \
--cc=jet@gyve.org \
--cc=linux-kernel@vger.kernel.org \
--cc=marcel@holtmann.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