From: Jan Kara <jack@suse.cz>
To: Andy Lutomirski <luto@amacapital.net>
Cc: Jan Kara <jack@suse.cz>, Andreas Dilger <adilger@dilger.ca>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux-mm@kvack.org" <linux-mm@kvack.org>,
"linux-ext4@vger.kernel.org" <linux-ext4@vger.kernel.org>
Subject: Re: Latency writing to an mlocked ext4 mapping
Date: Wed, 2 Nov 2011 21:17:07 +0100 [thread overview]
Message-ID: <20111102201707.GD31575@quack.suse.cz> (raw)
In-Reply-To: <CALCETrWNCy0VN-rQM-xPksiJ50DW-KM+w2NBprNOPhvnizZW=Q@mail.gmail.com>
On Tue 01-11-11 18:51:04, Andy Lutomirski wrote:
> On Tue, Nov 1, 2011 at 4:10 PM, Andy Lutomirski <luto@amacapital.net> wrote:
> > On Tue, Nov 1, 2011 at 4:03 PM, Jan Kara <jack@suse.cz> wrote:
> >> Avoiding IO during a minor fault would be a decent thing which might be
> >> worth pursuing. As you properly noted "stable pages during writeback"
> >> requirement is one obstacle which won't be that trivial to avoid though...
> >
> > There's an easy solution that would be good enough for me: add a mount
> > option to turn off stable pages.
> >
> > Is the other problem just a race, perhaps? __block_page_mkwrite calls
> > __block_write_begin (which calls get_block, which I think is where the
> > latency comes from) *before* wait_on_page_writeback, which means that
> > there might not be any space allocated yet.
>
> I think I'm right (other than calling it a race). If I change my code to do:
>
> - map the file (with MCL_FUTURE set)
> - fallocate
> - dirty all pages
> - fsync
> - dirty all pages again
>
> in the non-real-time thread, then a short test that was a mediocre
> reproducer seems to work.
>
> This is annoying, though -- I'm not generating twice as much write I/O
> as I used to. Is there any way to force the delalloc code to do its
> thing without triggering writeback? I don't think fallocate has this
> effect.
fallocate() will preallocate blocks on disk backing the mapped page. That
should get rid of latency in __block_write_begin(). Extents will still be
marked as uninitialized, but conversion from uninitialized to initialized
state happens during writeback / IO completion so you should not care much
about it.
Honza
--
Jan Kara <jack@suse.cz>
SUSE Labs, CR
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
WARNING: multiple messages have this Message-ID (diff)
From: Jan Kara <jack@suse.cz>
To: Andy Lutomirski <luto@amacapital.net>
Cc: Jan Kara <jack@suse.cz>, Andreas Dilger <adilger@dilger.ca>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux-mm@kvack.org" <linux-mm@kvack.org>,
"linux-ext4@vger.kernel.org" <linux-ext4@vger.kernel.org>
Subject: Re: Latency writing to an mlocked ext4 mapping
Date: Wed, 2 Nov 2011 21:17:07 +0100 [thread overview]
Message-ID: <20111102201707.GD31575@quack.suse.cz> (raw)
In-Reply-To: <CALCETrWNCy0VN-rQM-xPksiJ50DW-KM+w2NBprNOPhvnizZW=Q@mail.gmail.com>
On Tue 01-11-11 18:51:04, Andy Lutomirski wrote:
> On Tue, Nov 1, 2011 at 4:10 PM, Andy Lutomirski <luto@amacapital.net> wrote:
> > On Tue, Nov 1, 2011 at 4:03 PM, Jan Kara <jack@suse.cz> wrote:
> >> Avoiding IO during a minor fault would be a decent thing which might be
> >> worth pursuing. As you properly noted "stable pages during writeback"
> >> requirement is one obstacle which won't be that trivial to avoid though...
> >
> > There's an easy solution that would be good enough for me: add a mount
> > option to turn off stable pages.
> >
> > Is the other problem just a race, perhaps? __block_page_mkwrite calls
> > __block_write_begin (which calls get_block, which I think is where the
> > latency comes from) *before* wait_on_page_writeback, which means that
> > there might not be any space allocated yet.
>
> I think I'm right (other than calling it a race). If I change my code to do:
>
> - map the file (with MCL_FUTURE set)
> - fallocate
> - dirty all pages
> - fsync
> - dirty all pages again
>
> in the non-real-time thread, then a short test that was a mediocre
> reproducer seems to work.
>
> This is annoying, though -- I'm not generating twice as much write I/O
> as I used to. Is there any way to force the delalloc code to do its
> thing without triggering writeback? I don't think fallocate has this
> effect.
fallocate() will preallocate blocks on disk backing the mapped page. That
should get rid of latency in __block_write_begin(). Extents will still be
marked as uninitialized, but conversion from uninitialized to initialized
state happens during writeback / IO completion so you should not care much
about it.
Honza
--
Jan Kara <jack@suse.cz>
SUSE Labs, CR
WARNING: multiple messages have this Message-ID (diff)
From: Jan Kara <jack@suse.cz>
To: Andy Lutomirski <luto@amacapital.net>
Cc: Jan Kara <jack@suse.cz>, Andreas Dilger <adilger@dilger.ca>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux-mm@kvack.org" <linux-mm@kvack.org>,
"linux-ext4@vger.kernel.org" <linux-ext4@vger.kernel.org>
Subject: Re: Latency writing to an mlocked ext4 mapping
Date: Wed, 2 Nov 2011 21:17:07 +0100 [thread overview]
Message-ID: <20111102201707.GD31575@quack.suse.cz> (raw)
In-Reply-To: <CALCETrWNCy0VN-rQM-xPksiJ50DW-KM+w2NBprNOPhvnizZW=Q@mail.gmail.com>
On Tue 01-11-11 18:51:04, Andy Lutomirski wrote:
> On Tue, Nov 1, 2011 at 4:10 PM, Andy Lutomirski <luto@amacapital.net> wrote:
> > On Tue, Nov 1, 2011 at 4:03 PM, Jan Kara <jack@suse.cz> wrote:
> >> Avoiding IO during a minor fault would be a decent thing which might be
> >> worth pursuing. As you properly noted "stable pages during writeback"
> >> requirement is one obstacle which won't be that trivial to avoid though...
> >
> > There's an easy solution that would be good enough for me: add a mount
> > option to turn off stable pages.
> >
> > Is the other problem just a race, perhaps? __block_page_mkwrite calls
> > __block_write_begin (which calls get_block, which I think is where the
> > latency comes from) *before* wait_on_page_writeback, which means that
> > there might not be any space allocated yet.
>
> I think I'm right (other than calling it a race). If I change my code to do:
>
> - map the file (with MCL_FUTURE set)
> - fallocate
> - dirty all pages
> - fsync
> - dirty all pages again
>
> in the non-real-time thread, then a short test that was a mediocre
> reproducer seems to work.
>
> This is annoying, though -- I'm not generating twice as much write I/O
> as I used to. Is there any way to force the delalloc code to do its
> thing without triggering writeback? I don't think fallocate has this
> effect.
fallocate() will preallocate blocks on disk backing the mapped page. That
should get rid of latency in __block_write_begin(). Extents will still be
marked as uninitialized, but conversion from uninitialized to initialized
state happens during writeback / IO completion so you should not care much
about it.
Honza
--
Jan Kara <jack@suse.cz>
SUSE Labs, CR
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2011-11-02 20:17 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-10-20 0:39 Latency writing to an mlocked ext4 mapping Andy Lutomirski
2011-10-20 0:39 ` Andy Lutomirski
2011-10-20 1:02 ` Andreas Dilger
2011-10-20 1:02 ` Andreas Dilger
2011-10-20 1:15 ` Andy Lutomirski
2011-10-20 1:15 ` Andy Lutomirski
2011-10-20 2:17 ` Andy Lutomirski
2011-10-20 2:17 ` Andy Lutomirski
2011-10-20 2:17 ` Andy Lutomirski
2011-10-20 5:59 ` Andy Lutomirski
2011-10-20 5:59 ` Andy Lutomirski
2011-10-20 5:59 ` Andy Lutomirski
2011-10-25 12:26 ` Jan Kara
2011-10-25 12:26 ` Jan Kara
2011-10-25 12:26 ` Jan Kara
2011-10-28 23:37 ` Andy Lutomirski
2011-10-28 23:37 ` Andy Lutomirski
2011-10-28 23:39 ` [PATCH] mm: Improve cmtime update on shared writable mmaps Andy Lutomirski
2011-10-28 23:39 ` Andy Lutomirski
2011-11-01 22:53 ` Jan Kara
2011-11-01 22:53 ` Jan Kara
2011-11-01 23:02 ` Andy Lutomirski
2011-11-01 23:02 ` Andy Lutomirski
2011-11-01 23:02 ` Andy Lutomirski
2011-11-02 7:38 ` Christoph Hellwig
2011-11-02 7:38 ` Christoph Hellwig
2011-11-02 15:02 ` Jan Kara
2011-11-02 15:02 ` Jan Kara
2011-11-02 15:02 ` Jan Kara
2011-11-02 15:19 ` Ted Ts'o
2011-11-02 15:19 ` Ted Ts'o
2011-10-31 23:10 ` Latency writing to an mlocked ext4 mapping Jan Kara
2011-10-31 23:10 ` Jan Kara
2011-10-31 23:10 ` Jan Kara
2011-10-31 23:14 ` Andy Lutomirski
2011-10-31 23:14 ` Andy Lutomirski
2011-10-31 23:14 ` Andy Lutomirski
2011-11-01 23:03 ` Jan Kara
2011-11-01 23:03 ` Jan Kara
2011-11-01 23:03 ` Jan Kara
2011-11-01 23:10 ` Andy Lutomirski
2011-11-01 23:10 ` Andy Lutomirski
2011-11-01 23:10 ` Andy Lutomirski
2011-11-02 1:51 ` Andy Lutomirski
2011-11-02 1:51 ` Andy Lutomirski
2011-11-02 1:51 ` Andy Lutomirski
2011-11-02 20:17 ` Jan Kara [this message]
2011-11-02 20:17 ` Jan Kara
2011-11-02 20:17 ` Jan Kara
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=20111102201707.GD31575@quack.suse.cz \
--to=jack@suse.cz \
--cc=adilger@dilger.ca \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=luto@amacapital.net \
/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.