From: "Aneesh Kumar K. V" <aneesh.kumar@linux.vnet.ibm.com>
To: Sage Weil <sage@newdream.net>, linux-fsdevel@vger.kernel.org
Cc: linux-kernel@vger.kernel.org,
Jonathan Nieder <jrnieder@gmail.com>,
akpm@linux-foundation.org, linux-api@vger.kernel.org
Subject: Re: [RFC] introduce sys_syncfs to sync a single file system (v2)
Date: Tue, 08 Mar 2011 10:57:00 +0530 [thread overview]
Message-ID: <8762ruchcr.fsf@linux.vnet.ibm.com> (raw)
In-Reply-To: <Pine.LNX.4.64.1103071515070.11152@cobra.newdream.net>
On Mon, 7 Mar 2011 15:17:57 -0800 (PST), Sage Weil <sage@newdream.net> wrote:
> Changes:
> v1->v2: Rename, simplify to just take an fd.
>
> It is frequently useful to sync a single file system, instead of all
> mounted file systems via sync(2):
>
> - On machines with many mounts, it is not at all uncommon for some of
> them to hang (e.g. unresponsive NFS server). sync(2) will get stuck on
> those and may never get to the one you do care about (e.g., /).
> - Some applications write lots of data to the file system and then
> want to make sure it is flushed to disk. Calling fsync(2) on each
> file introduces unnecessary ordering constraints that result in a large
> amount of sub-optimal writeback/flush/commit behavior by the file
> system.
>
> There are currently two ways (that I know of) to sync a single super_block:
>
> - BLKFLSBUF ioctl on the block device: That also invalidates the bdev
> mapping, which isn't usually desirable, and doesn't work for non-block
> file systems.
> - 'mount -o remount,rw' will call sync_filesystem as an artifact of the
> current implemention. Relying on this little-known side effect for
> something like data safety sounds foolish.
>
> Both of these approaches require root privileges, which some applications
> do not have (nor should they need?) given that sync(2) is an unprivileged
> operation.
>
> This patch introduces a new system call syncfs(2) that takes an fd and
> syncs only the file system it references. Maybe someday we can even
>
> $ sync /some/path
>
> and not get
>
> sync: ignoring all arguments
>
> The syscall is motivated by comments by Al and Christoph at the last LSF.
> syncfs(2) seems like an appropriate name given statfs(2).
>
> A similar ioctl was also proposed a while back, see
> http://marc.info/?l=linux-fsdevel&m=127970513829285&w=2
>
> Signed-off-by: Sage Weil <sage@newdream.net>
> ---
> arch/x86/ia32/ia32entry.S | 1 +
> arch/x86/include/asm/unistd_32.h | 3 ++-
> arch/x86/include/asm/unistd_64.h | 2 ++
> arch/x86/kernel/syscall_table_32.S | 1 +
> fs/sync.c | 24 ++++++++++++++++++++++++
> 5 files changed, 30 insertions(+), 1 deletions(-)
include/asm-generic/unistd.h may also need an update.
-aneesh
next prev parent reply other threads:[~2011-03-08 5:27 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-03-03 6:35 [RFC] introduce sys_syncat to sync a single file system Sage Weil
2011-03-03 7:22 ` Jonathan Nieder
2011-03-03 8:54 ` Aneesh Kumar K. V
2011-03-07 23:17 ` [RFC] introduce sys_syncfs to sync a single file system (v2) Sage Weil
2011-03-08 5:27 ` Aneesh Kumar K. V [this message]
2011-03-10 14:56 ` Arnd Bergmann
2011-03-10 19:28 ` Sage Weil
2011-03-10 19:31 ` [PATCH v3] introduce sys_syncfs to sync a single file system Sage Weil
2011-03-10 22:08 ` Arnd Bergmann
2011-03-11 4:44 ` Aneesh Kumar K. V
2011-03-11 11:01 ` Indan Zupancic
2011-03-11 11:55 ` Arnd Bergmann
2011-03-11 23:45 ` Indan Zupancic
2011-03-11 23:56 ` Jonathan Nieder
2011-03-12 1:53 ` Indan Zupancic
2011-03-12 2:10 ` Jonathan Nieder
2011-03-12 4:22 ` Indan Zupancic
2011-03-12 17:32 ` Greg KH
2011-03-14 1:56 ` Indan Zupancic
2011-03-14 4:29 ` Sage Weil
2011-03-14 9:27 ` Indan Zupancic
2011-03-14 10:22 ` Theodore Tso
2011-03-15 10:11 ` Dave Chinner
2011-03-15 13:00 ` Sage Weil
2011-03-15 15:56 ` Andreas Dilger
2011-03-15 16:08 ` Sage Weil
2011-03-15 20:18 ` Andrew Morton
2011-03-14 20:10 ` Andrew Morton
2011-03-14 20:29 ` Artem Bityutskiy
2011-03-14 21:11 ` Ted Ts'o
2011-03-14 21:20 ` Andrew Morton
2011-03-14 23:17 ` Ted Ts'o
2011-03-14 21:22 ` Arnd Bergmann
2011-03-12 0:40 ` Ric Wheeler
2011-03-12 1:33 ` Indan Zupancic
2011-03-12 2:52 ` Ric Wheeler
2011-03-12 3:50 ` Indan Zupancic
2011-03-12 12:41 ` Ric Wheeler
2011-03-12 18:31 ` Jeff Garzik
2011-03-14 1:31 ` Indan Zupancic
2011-03-14 1:37 ` Theodore Tso
2011-03-14 1:47 ` Indan Zupancic
2011-03-14 1:45 ` Jeff Garzik
2011-03-14 1:59 ` Indan Zupancic
2011-03-12 19:28 ` Artem Bityutskiy
2011-03-12 19:22 ` Artem Bityutskiy
2011-03-14 1:38 ` Indan Zupancic
2011-03-14 5:52 ` Artem Bityutskiy
2011-03-13 20:59 ` 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=8762ruchcr.fsf@linux.vnet.ibm.com \
--to=aneesh.kumar@linux.vnet.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=jrnieder@gmail.com \
--cc=linux-api@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=sage@newdream.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox