From: Nigel Cunningham <ncunningham@linuxmail.org>
To: Andrew Morton <akpm@osdl.org>, "Rafael J. Wysocki" <rjw@sisk.pl>,
LKML <linux-kernel@vger.kernel.org>
Subject: [PATCH] Freeze bdevs when freezing processes.
Date: Mon, 23 Oct 2006 14:12:15 +1000 [thread overview]
Message-ID: <1161576735.3466.7.camel@nigel.suspend2.net> (raw)
XFS can continue to submit I/O from a timer routine, even after
freezeable kernel and userspace threads are frozen. This doesn't seem to
be an issue for current swsusp code, but is definitely an issue for
Suspend2, where the pages being written could be overwritten by
Suspend2's atomic copy.
We can address this issue by freezing bdevs after stopping userspace
threads, and thawing them prior to thawing userspace.
Signed-off-by: Nigel Cunningham <nigel@suspend2.net>
diff --git a/kernel/power/process.c b/kernel/power/process.c
index 4a001fe..ddeeb50 100644
--- a/kernel/power/process.c
+++ b/kernel/power/process.c
@@ -13,6 +13,7 @@ #include <linux/interrupt.h>
#include <linux/suspend.h>
#include <linux/module.h>
#include <linux/syscalls.h>
+#include <linux/buffer_head.h>
#include <linux/freezer.h>
/*
@@ -20,6 +21,58 @@ #include <linux/freezer.h>
*/
#define TIMEOUT (20 * HZ)
+struct frozen_fs
+{
+ struct list_head fsb_list;
+ struct super_block *sb;
+};
+
+LIST_HEAD(frozen_fs_list);
+
+void freezer_make_fses_rw(void)
+{
+ struct frozen_fs *fs, *next_fs;
+
+ list_for_each_entry_safe(fs, next_fs, &frozen_fs_list, fsb_list) {
+ thaw_bdev(fs->sb->s_bdev, fs->sb);
+
+ list_del(&fs->fsb_list);
+ kfree(fs);
+ }
+}
+
+/*
+ * Done after userspace is frozen, so there should be no danger of
+ * fses being unmounted while we're in here.
+ */
+int freezer_make_fses_ro(void)
+{
+ struct frozen_fs *fs;
+ struct super_block *sb;
+
+ /* Generate the list */
+ list_for_each_entry(sb, &super_blocks, s_list) {
+ if (!sb->s_root || !sb->s_bdev ||
+ (sb->s_frozen == SB_FREEZE_TRANS) ||
+ (sb->s_flags & MS_RDONLY))
+ continue;
+
+ fs = kmalloc(sizeof(struct frozen_fs), GFP_ATOMIC);
+ if (!fs)
+ return 1;
+ fs->sb = sb;
+ list_add_tail(&fs->fsb_list, &frozen_fs_list);
+ };
+
+ /* Do the freezing in reverse order so filesystems dependant
+ * upon others are frozen in the right order. (Eg loopback
+ * on ext3). */
+ list_for_each_entry_reverse(fs, &frozen_fs_list, fsb_list)
+ freeze_bdev(fs->sb->s_bdev);
+
+ return 0;
+}
+
static inline int freezeable(struct task_struct * p)
{
@@ -119,7 +172,7 @@ int freeze_processes(void)
read_unlock(&tasklist_lock);
todo += nr_user;
if (!user_frozen && !nr_user) {
- sys_sync();
+ freezer_make_fses_ro();
start_time = jiffies;
}
user_frozen = !nr_user;
@@ -174,6 +227,12 @@ void thaw_some_processes(int all)
"Strange, %s not stopped\n", p->comm );
} while_each_thread(g, p);
+ if (!pass) {
+ read_unlock(&tasklist_lock);
+ freezer_make_fses_rw();
+ read_lock(&tasklist_lock);
+ }
+
pass++;
} while (pass < 2 && all);
next reply other threads:[~2006-10-23 4:12 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-10-23 4:12 Nigel Cunningham [this message]
2006-10-23 10:36 ` [PATCH] Freeze bdevs when freezing processes Rafael J. Wysocki
2006-10-23 12:09 ` Nigel Cunningham
2006-10-23 14:07 ` Rafael J. Wysocki
2006-10-23 14:15 ` Nick Piggin
2006-10-23 14:20 ` Rafael J. Wysocki
2006-10-23 23:05 ` Nigel Cunningham
2006-10-23 16:55 ` Andrew Morton
2006-10-23 17:14 ` Pavel Machek
2006-10-23 17:50 ` Andrew Morton
2006-10-23 18:06 ` Pavel Machek
2006-10-23 19:19 ` Rafael J. Wysocki
2006-10-23 22:52 ` Nigel Cunningham
2006-10-24 7:57 ` Rafael J. Wysocki
2006-10-24 8:21 ` Nigel Cunningham
2006-10-23 21:39 ` Matthew Garrett
2006-10-23 22:12 ` Rafael J. Wysocki
2006-10-24 7:58 ` Pavel Machek
2006-10-23 22:58 ` Nigel Cunningham
2006-10-24 8:01 ` Pavel Machek
2006-10-23 23:22 ` Nigel Cunningham
2006-10-24 8:37 ` Rafael J. Wysocki
2006-10-24 14:44 ` David Chinner
2006-10-24 15:29 ` Rafael J. Wysocki
2006-10-24 16:20 ` Oleg Verych
2006-10-24 16:27 ` Oleg Verych
2006-10-24 17:08 ` Christoph Hellwig
2006-10-25 8:05 ` Pavel Machek
2006-10-24 16:33 ` David Chinner
2006-10-24 21:37 ` Pavel Machek
2006-10-25 0:13 ` David Chinner
2006-10-25 8:10 ` Pavel Machek
2006-10-25 8:38 ` David Chinner
2006-10-25 8:47 ` Pavel Machek
2006-10-25 12:32 ` Rafael J. Wysocki
2006-10-25 13:23 ` Nigel Cunningham
2006-10-25 19:05 ` Rafael J. Wysocki
2006-10-26 7:30 ` David Chinner
2006-10-26 8:18 ` Nigel Cunningham
2006-10-26 8:48 ` Rafael J. Wysocki
2006-10-26 8:57 ` David Chinner
2006-10-26 9:11 ` Rafael J. Wysocki
2006-10-27 1:38 ` David Chinner
2006-10-27 14:37 ` Rafael J. Wysocki
2006-10-29 17:35 ` Pavel Machek
2006-10-29 23:29 ` Rafael J. Wysocki
2006-10-29 23:46 ` Nigel Cunningham
2006-10-26 9:18 ` Nigel Cunningham
2006-10-26 9:08 ` Rafael J. Wysocki
2006-10-25 8:12 ` Rafael J. Wysocki
2006-10-24 17:06 ` Christoph Hellwig
2006-10-24 19:09 ` Rafael J. Wysocki
2006-10-24 21:26 ` Pavel Machek
2006-10-24 21:33 ` Christoph Hellwig
2006-10-24 21:43 ` Pavel Machek
2006-10-24 22:19 ` Nigel Cunningham
2006-10-24 20:16 ` Rafael J. Wysocki
2006-10-24 22:17 ` Nigel Cunningham
2006-10-24 20:38 ` Christoph Hellwig
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=1161576735.3466.7.camel@nigel.suspend2.net \
--to=ncunningham@linuxmail.org \
--cc=akpm@osdl.org \
--cc=linux-kernel@vger.kernel.org \
--cc=rjw@sisk.pl \
/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.