From: Pavel Machek <pavel-+ZI9xUNit7I@public.gmane.org>
To: Alan Jenkins <alan-jenkins-cCz0Lq7MMjm9FHfhHBbuYA@public.gmane.org>
Cc: "Rafael J. Wysocki" <rjw-KKrjLPT3xs0@public.gmane.org>,
pm list
<linux-pm-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org>,
linux-kernel
<linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
Kernel Testers List
<kernel-testers-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
Mel Gorman <mel-wPRd99KPJ+uzQB+pC5nmwQ@public.gmane.org>
Subject: Re: [PATCH] uswsusp: automatically free the in-memory image once s2disk has finished with it
Date: Wed, 2 Dec 2009 22:11:07 +0100 [thread overview]
Message-ID: <20091202211107.GA20830@elf.ucw.cz> (raw)
In-Reply-To: <4B16797C.3010304-cCz0Lq7MMjm9FHfhHBbuYA@public.gmane.org>
On Wed 2009-12-02 14:28:12, Alan Jenkins wrote:
> The original in-kernel suspend (swsusp) frees the in-memory hibernation
> image before powering off the machine. s2disk doesn't, so there is
> _much_ less free memory when it tries to power off.
>
> This is a gratuitous difference. The userspace suspend interface
> /dev/snapshot only allows the hibernation image to be read once.
> Once the s2disk program has read the last page, we can free the entire
> image.
>
> This avoids a hang after writing the hibernation image which was
> triggered by commit 5f8dcc21211a3d4e3a7a5ca366b469fb88117f61
> "page-allocator: split per-cpu list into one-list-per-migrate-type":
Yes, you work around page-allocator hang. But is it right thing to do?
I mean... Power down should not be too memory-critical. Why can't new
page allocator handle that?
(And yes, when page allocator is fixed we may want to do something
like that, but...)
Pavel
> --- a/kernel/power/user.c
> +++ b/kernel/power/user.c
> @@ -165,6 +165,10 @@ static ssize_t snapshot_read(struct file *filp, char __user *buf,
> res = -EFAULT;
> else
> *offp = data->handle.offset;
> + } else {
> + swsusp_free();
> + memset(&data->handle, 0, sizeof(struct snapshot_handle));
> + data->ready = 0;
> }
>
> Unlock:
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
WARNING: multiple messages have this Message-ID (diff)
From: Pavel Machek <pavel@ucw.cz>
To: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>,
pm list <linux-pm@lists.linux-foundation.org>,
linux-kernel <linux-kernel@vger.kernel.org>,
Kernel Testers List <kernel-testers@vger.kernel.org>,
Mel Gorman <mel@csn.ul.ie>
Subject: Re: [PATCH] uswsusp: automatically free the in-memory image once s2disk has finished with it
Date: Wed, 2 Dec 2009 22:11:07 +0100 [thread overview]
Message-ID: <20091202211107.GA20830@elf.ucw.cz> (raw)
In-Reply-To: <4B16797C.3010304@tuffmail.co.uk>
On Wed 2009-12-02 14:28:12, Alan Jenkins wrote:
> The original in-kernel suspend (swsusp) frees the in-memory hibernation
> image before powering off the machine. s2disk doesn't, so there is
> _much_ less free memory when it tries to power off.
>
> This is a gratuitous difference. The userspace suspend interface
> /dev/snapshot only allows the hibernation image to be read once.
> Once the s2disk program has read the last page, we can free the entire
> image.
>
> This avoids a hang after writing the hibernation image which was
> triggered by commit 5f8dcc21211a3d4e3a7a5ca366b469fb88117f61
> "page-allocator: split per-cpu list into one-list-per-migrate-type":
Yes, you work around page-allocator hang. But is it right thing to do?
I mean... Power down should not be too memory-critical. Why can't new
page allocator handle that?
(And yes, when page allocator is fixed we may want to do something
like that, but...)
Pavel
> --- a/kernel/power/user.c
> +++ b/kernel/power/user.c
> @@ -165,6 +165,10 @@ static ssize_t snapshot_read(struct file *filp, char __user *buf,
> res = -EFAULT;
> else
> *offp = data->handle.offset;
> + } else {
> + swsusp_free();
> + memset(&data->handle, 0, sizeof(struct snapshot_handle));
> + data->ready = 0;
> }
>
> Unlock:
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
next prev parent reply other threads:[~2009-12-02 21:11 UTC|newest]
Thread overview: 77+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-12-01 19:59 Bisected: s2disk (uswsusp only) hangs just before poweroff Alan Jenkins
2009-12-01 19:59 ` Alan Jenkins
2009-12-01 20:24 ` Justin P. Mattock
2009-12-01 21:45 ` Mel Gorman
[not found] ` <4B1575AC.6080904-cCz0Lq7MMjm9FHfhHBbuYA@public.gmane.org>
2009-12-01 20:24 ` Justin P. Mattock
2009-12-01 20:24 ` Justin P. Mattock
[not found] ` <4B157B81.9050703-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2009-12-01 20:27 ` Alan Jenkins
2009-12-01 20:27 ` Alan Jenkins
2009-12-01 21:14 ` Justin P. Mattock
2009-12-01 21:14 ` Justin P. Mattock
2009-12-01 20:27 ` Alan Jenkins
2009-12-01 21:45 ` Mel Gorman
2009-12-01 21:45 ` Mel Gorman
2009-12-01 21:53 ` Rafael J. Wysocki
2009-12-02 11:49 ` Alan Jenkins
[not found] ` <200912012253.08522.rjw-KKrjLPT3xs0@public.gmane.org>
2009-12-02 11:49 ` Alan Jenkins
2009-12-02 11:49 ` Alan Jenkins
[not found] ` <4B16545B.3090703-cCz0Lq7MMjm9FHfhHBbuYA@public.gmane.org>
2009-12-02 12:20 ` Mel Gorman
2009-12-02 12:20 ` Mel Gorman
2009-12-02 14:25 ` Alan Jenkins
2009-12-02 14:25 ` Alan Jenkins
2009-12-02 14:28 ` [PATCH] uswsusp: automatically free the in-memory image once s2disk has finished with it Alan Jenkins
[not found] ` <20091202122019.GD1457-wPRd99KPJ+uzQB+pC5nmwQ@public.gmane.org>
2009-12-02 14:28 ` Alan Jenkins
2009-12-02 14:28 ` Alan Jenkins
2009-12-02 21:11 ` Pavel Machek
[not found] ` <4B16797C.3010304-cCz0Lq7MMjm9FHfhHBbuYA@public.gmane.org>
2009-12-02 21:11 ` Pavel Machek [this message]
2009-12-02 21:11 ` Pavel Machek
[not found] ` <20091202211107.GA20830-I/5MKhXcvmPrBKCeMvbIDA@public.gmane.org>
2009-12-02 22:07 ` Mel Gorman
2009-12-02 22:07 ` Mel Gorman
2009-12-02 22:15 ` Pavel Machek
2009-12-02 22:15 ` Pavel Machek
2009-12-02 22:25 ` Mel Gorman
[not found] ` <20091202221524.GB20830-I/5MKhXcvmPrBKCeMvbIDA@public.gmane.org>
2009-12-02 22:25 ` Mel Gorman
2009-12-02 22:25 ` Mel Gorman
[not found] ` <20091202222516.GD26702-wPRd99KPJ+uzQB+pC5nmwQ@public.gmane.org>
2009-12-02 23:22 ` Rafael J. Wysocki
2009-12-02 23:22 ` Rafael J. Wysocki
2009-12-03 7:53 ` Pavel Machek
2009-12-03 7:53 ` Pavel Machek
2009-12-03 12:57 ` Alan Jenkins
2009-12-03 12:57 ` Alan Jenkins
2009-12-03 14:50 ` Mel Gorman
2009-12-03 19:50 ` Rafael J. Wysocki
2009-12-03 19:50 ` Rafael J. Wysocki
[not found] ` <4B17B5B8.1060105-cCz0Lq7MMjm9FHfhHBbuYA@public.gmane.org>
2009-12-03 14:50 ` Mel Gorman
2009-12-03 14:50 ` Mel Gorman
[not found] ` <20091203145018.GG26702-wPRd99KPJ+uzQB+pC5nmwQ@public.gmane.org>
2009-12-08 0:37 ` Alan Jenkins
2009-12-08 0:37 ` Alan Jenkins
2009-12-11 10:53 ` Mel Gorman
[not found] ` <9b2b86520912071637v6957ed24ie0f67acf6785ab08-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2009-12-11 10:53 ` Mel Gorman
2009-12-11 10:53 ` Mel Gorman
2009-12-14 11:08 ` Pavel Machek
[not found] ` <20091211105352.GB30670-wPRd99KPJ+uzQB+pC5nmwQ@public.gmane.org>
2009-12-14 11:08 ` Pavel Machek
2009-12-14 11:08 ` Pavel Machek
2009-12-08 0:37 ` Alan Jenkins
2009-12-03 20:16 ` Pavel Machek
2009-12-03 20:16 ` Pavel Machek
2009-12-03 20:16 ` Pavel Machek
2009-12-02 23:22 ` Rafael J. Wysocki
2009-12-03 7:53 ` Pavel Machek
2009-12-02 22:07 ` Mel Gorman
2009-12-02 21:47 ` Rafael J. Wysocki
2009-12-02 21:47 ` Rafael J. Wysocki
2009-12-02 21:47 ` Rafael J. Wysocki
2009-12-02 12:20 ` Bisected: s2disk (uswsusp only) hangs just before poweroff Mel Gorman
2009-12-01 21:53 ` Rafael J. Wysocki
2009-12-02 8:57 ` Alan Jenkins
[not found] ` <20091201214529.GA1457-wPRd99KPJ+uzQB+pC5nmwQ@public.gmane.org>
2009-12-02 8:57 ` Alan Jenkins
2009-12-02 8:57 ` Alan Jenkins
[not found] ` <4B162BE1.7070709-cCz0Lq7MMjm9FHfhHBbuYA@public.gmane.org>
2009-12-02 10:35 ` Mel Gorman
2009-12-02 10:35 ` Mel Gorman
2009-12-02 11:35 ` Alan Jenkins
[not found] ` <20091202103538.GB1457-wPRd99KPJ+uzQB+pC5nmwQ@public.gmane.org>
2009-12-02 11:35 ` Alan Jenkins
2009-12-02 11:35 ` Alan Jenkins
2009-12-02 10:35 ` Mel Gorman
2009-12-02 11:11 ` Alan Jenkins
2009-12-02 11:11 ` Alan Jenkins
2009-12-02 11:11 ` Alan Jenkins
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=20091202211107.GA20830@elf.ucw.cz \
--to=pavel-+zi9xunit7i@public.gmane.org \
--cc=alan-jenkins-cCz0Lq7MMjm9FHfhHBbuYA@public.gmane.org \
--cc=kernel-testers-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-pm-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \
--cc=mel-wPRd99KPJ+uzQB+pC5nmwQ@public.gmane.org \
--cc=rjw-KKrjLPT3xs0@public.gmane.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.