From: Dave Chinner <david@fromorbit.com>
To: Andy Lutomirski <luto@amacapital.net>
Cc: "Darrick J. Wong" <darrick.wong@oracle.com>,
"Pádraig Brady" <P@draigbrady.com>,
"Anna Schumaker" <Anna.Schumaker@netapp.com>,
linux-nfs@vger.kernel.org,
"Linux btrfs Developers List" <linux-btrfs@vger.kernel.org>,
"Linux FS Devel" <linux-fsdevel@vger.kernel.org>,
"Linux API" <linux-api@vger.kernel.org>,
"Zach Brown" <zab@zabbo.net>, "Al Viro" <viro@zeniv.linux.org.uk>,
"Chris Mason" <clm@fb.com>,
"Michael Kerrisk-manpages" <mtk.manpages@gmail.com>,
andros@netapp.com, "Christoph Hellwig" <hch@infradead.org>,
Coreutils <coreutils@gnu.org>
Subject: Re: [PATCH v1 0/8] VFS: In-kernel copy system call
Date: Mon, 14 Sep 2015 09:25:29 +1000 [thread overview]
Message-ID: <20150913232529.GT26895@dastard> (raw)
In-Reply-To: <CALCETrVsWBdqvAgwxHcG=gbcWRNPG2ZziWUg1g=siKDrDu7S2Q@mail.gmail.com>
On Tue, Sep 08, 2015 at 04:08:43PM -0700, Andy Lutomirski wrote:
> Can we have a clean way to figure out whether two file ranges are the
> same in a way that allows false negatives? I.e. return 1 if the
> ranges are reflinks of each other and 0 if not? Pretty please? I've
> implemented that in the past on btrfs by syncing the ranges and then
> comparing FIEMAP output, but that's hideous.
That fundamentally doesn't work for userspace, because the moment
the filesystem drops it's locks on the inodes in the kernel after
doing the comparison the mappings can change. IOWs, by the time the
information gets back to userspace, it's already wrong. e.g. cp made
this mistake by trying to use FIEMAP to optimise hole detection in
files and ended up with corrupt copies.
It really doesn't matter what the syscall/ioctl interface is, trying
to make application logic decisions based on inode block mappings
from userspace is racy and not safe and will go wrong...
Cheers,
Dave.
--
Dave Chinner
david@fromorbit.com
WARNING: multiple messages have this Message-ID (diff)
From: Dave Chinner <david-FqsqvQoI3Ljby3iVrkZq2A@public.gmane.org>
To: Andy Lutomirski <luto-kltTT9wpgjJwATOyAt5JVQ@public.gmane.org>
Cc: "Darrick J. Wong"
<darrick.wong-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>,
"Pádraig Brady" <P@draigbrady.com>,
"Anna Schumaker"
<Anna.Schumaker-HgOvQuBEEgTQT0dZR+AlfA@public.gmane.org>,
linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
"Linux btrfs Developers List"
<linux-btrfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
"Linux FS Devel"
<linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
"Linux API" <linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
"Zach Brown" <zab-ugsP4Wv/S6ZeoWH0uzbU5w@public.gmane.org>,
"Al Viro"
<viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org>,
"Chris Mason" <clm-b10kYP2dOMg@public.gmane.org>,
"Michael Kerrisk-manpages"
<mtk.manpages-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>,
andros-HgOvQuBEEgTQT0dZR+AlfA@public.gmane.org,
"Christoph Hellwig" <hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>,
Coreutils <coreutils-mXXj517/zsQ@public.gmane.org>
Subject: Re: [PATCH v1 0/8] VFS: In-kernel copy system call
Date: Mon, 14 Sep 2015 09:25:29 +1000 [thread overview]
Message-ID: <20150913232529.GT26895@dastard> (raw)
In-Reply-To: <CALCETrVsWBdqvAgwxHcG=gbcWRNPG2ZziWUg1g=siKDrDu7S2Q-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
On Tue, Sep 08, 2015 at 04:08:43PM -0700, Andy Lutomirski wrote:
> Can we have a clean way to figure out whether two file ranges are the
> same in a way that allows false negatives? I.e. return 1 if the
> ranges are reflinks of each other and 0 if not? Pretty please? I've
> implemented that in the past on btrfs by syncing the ranges and then
> comparing FIEMAP output, but that's hideous.
That fundamentally doesn't work for userspace, because the moment
the filesystem drops it's locks on the inodes in the kernel after
doing the comparison the mappings can change. IOWs, by the time the
information gets back to userspace, it's already wrong. e.g. cp made
this mistake by trying to use FIEMAP to optimise hole detection in
files and ended up with corrupt copies.
It really doesn't matter what the syscall/ioctl interface is, trying
to make application logic decisions based on inode block mappings
from userspace is racy and not safe and will go wrong...
Cheers,
Dave.
--
Dave Chinner
david-FqsqvQoI3Ljby3iVrkZq2A@public.gmane.org
next prev parent reply other threads:[~2015-09-13 23:25 UTC|newest]
Thread overview: 121+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-09-04 20:16 [PATCH v1 0/8] VFS: In-kernel copy system call Anna Schumaker
2015-09-04 20:16 ` Anna Schumaker
2015-09-04 20:16 ` Anna Schumaker
2015-09-04 20:16 ` [PATCH v1 1/9] vfs: add copy_file_range syscall and vfs helper Anna Schumaker
2015-09-04 20:16 ` Anna Schumaker
2015-09-04 20:16 ` Anna Schumaker
2015-09-04 21:50 ` Darrick J. Wong
2015-09-04 20:16 ` [PATCH v1 2/8] x86: add sys_copy_file_range to syscall tables Anna Schumaker
2015-09-04 20:16 ` Anna Schumaker
2015-09-04 20:16 ` [PATCH v1 3/8] btrfs: add .copy_file_range file operation Anna Schumaker
2015-09-04 20:16 ` Anna Schumaker
2015-09-04 21:02 ` Josef Bacik
2015-09-04 21:02 ` Josef Bacik
2015-09-04 21:02 ` Josef Bacik
2015-09-09 8:39 ` David Sterba
2015-09-04 20:16 ` [PATCH v1 4/8] btrfs: Add mountpoint checking during btrfs_copy_file_range Anna Schumaker
2015-09-04 20:16 ` Anna Schumaker
2015-09-04 20:16 ` Anna Schumaker
2015-09-09 9:18 ` David Sterba
2015-09-09 15:56 ` Anna Schumaker
2015-09-09 15:56 ` Anna Schumaker
2015-09-04 20:16 ` [PATCH v1 5/8] vfs: Remove copy_file_range mountpoint checks Anna Schumaker
2015-09-04 20:16 ` Anna Schumaker
2015-09-04 20:16 ` Anna Schumaker
2015-09-04 20:17 ` [PATCH v1 6/8] vfs: Copy should check len after file open mode Anna Schumaker
2015-09-04 20:17 ` Anna Schumaker
2015-09-04 20:17 ` Anna Schumaker
2015-09-04 20:17 ` [PATCH v1 7/8] vfs: Copy should use file_out rather than file_in Anna Schumaker
2015-09-04 20:17 ` Anna Schumaker
2015-09-04 20:17 ` [PATCH v1 8/8] vfs: Fall back on splice if no copy function defined Anna Schumaker
2015-09-04 20:17 ` Anna Schumaker
2015-09-04 20:17 ` Anna Schumaker
2015-09-04 21:08 ` Darrick J. Wong
2015-09-08 14:57 ` Anna Schumaker
2015-09-08 14:57 ` Anna Schumaker
2015-09-08 14:57 ` Anna Schumaker
2015-09-04 20:17 ` [PATCH v1 9/8] copy_file_range.2: New page documenting copy_file_range() Anna Schumaker
2015-09-04 20:17 ` Anna Schumaker
2015-09-04 20:17 ` Anna Schumaker
2015-09-04 21:38 ` Darrick J. Wong
2015-09-04 22:31 ` Andreas Dilger
2015-09-04 22:31 ` Andreas Dilger
2015-09-08 15:05 ` Anna Schumaker
2015-09-08 15:05 ` Anna Schumaker
2015-09-08 15:05 ` Anna Schumaker
2015-09-08 15:04 ` Anna Schumaker
2015-09-08 15:04 ` Anna Schumaker
2015-09-08 15:04 ` Anna Schumaker
2015-09-08 20:39 ` Darrick J. Wong
2015-09-09 9:16 ` David Sterba
2015-09-09 11:38 ` Austin S Hemmelgarn
2015-09-09 11:38 ` Austin S Hemmelgarn
2015-09-09 17:17 ` Darrick J. Wong
2015-09-09 17:31 ` Anna Schumaker
2015-09-09 17:31 ` Anna Schumaker
2015-09-09 17:31 ` Anna Schumaker
2015-09-09 18:12 ` Darrick J. Wong
2015-09-09 18:12 ` Darrick J. Wong
2015-09-09 19:25 ` Anna Schumaker
2015-09-09 19:25 ` Anna Schumaker
2015-09-10 15:42 ` David Sterba
2015-09-10 15:42 ` David Sterba
2015-09-10 16:43 ` Darrick J. Wong
2015-09-10 16:43 ` Darrick J. Wong
2015-09-04 22:25 ` [PATCH v1 0/8] VFS: In-kernel copy system call Andreas Dilger
2015-09-04 22:25 ` Andreas Dilger
2015-09-05 8:33 ` Al Viro
2015-09-05 8:33 ` Al Viro
2015-09-08 15:08 ` Anna Schumaker
2015-09-08 15:08 ` Anna Schumaker
2015-09-08 15:08 ` Anna Schumaker
2015-09-08 20:45 ` Darrick J. Wong
2015-09-08 20:49 ` Anna Schumaker
2015-09-08 20:49 ` Anna Schumaker
2015-09-08 20:49 ` Anna Schumaker
2015-09-08 15:07 ` Anna Schumaker
2015-09-08 15:07 ` Anna Schumaker
2015-09-08 15:21 ` Pádraig Brady
2015-09-08 15:21 ` Pádraig Brady
2015-09-08 18:23 ` Anna Schumaker
2015-09-08 18:23 ` Anna Schumaker
2015-09-08 19:10 ` Andy Lutomirski
2015-09-08 19:10 ` Andy Lutomirski
2015-09-08 20:03 ` Pádraig Brady
2015-09-08 20:03 ` Pádraig Brady
2015-09-08 21:29 ` Darrick J. Wong
2015-09-08 21:29 ` Darrick J. Wong
2015-09-08 21:45 ` Andy Lutomirski
2015-09-08 21:45 ` Andy Lutomirski
2015-09-08 22:39 ` Darrick J. Wong
2015-09-08 22:39 ` Darrick J. Wong
2015-09-08 23:08 ` Andy Lutomirski
2015-09-08 23:08 ` Andy Lutomirski
2015-09-09 1:19 ` Darrick J. Wong
2015-09-09 1:19 ` Darrick J. Wong
2015-09-09 20:09 ` Chris Mason
2015-09-09 20:26 ` Trond Myklebust
2015-09-09 20:26 ` Trond Myklebust
2015-09-09 20:38 ` Chris Mason
2015-09-09 20:38 ` Chris Mason
2015-09-09 20:41 ` Anna Schumaker
2015-09-09 20:41 ` Anna Schumaker
2015-09-09 21:42 ` Darrick J. Wong
2015-09-09 21:42 ` Darrick J. Wong
2015-09-09 20:37 ` Andy Lutomirski
2015-09-09 20:37 ` Andy Lutomirski
2015-09-09 20:42 ` Chris Mason
2015-09-09 20:42 ` Chris Mason
2015-09-13 23:25 ` Dave Chinner [this message]
2015-09-13 23:25 ` Dave Chinner
2015-09-14 17:53 ` Andy Lutomirski
2015-09-14 17:53 ` Andy Lutomirski
2015-09-09 18:52 ` Anna Schumaker
2015-09-09 18:52 ` Anna Schumaker
2015-09-09 21:16 ` Darrick J. Wong
2015-09-09 21:16 ` Darrick J. Wong
2015-09-10 15:10 ` Anna Schumaker
2015-09-10 15:10 ` Anna Schumaker
2015-09-10 15:49 ` Austin S Hemmelgarn
2015-09-10 15:49 ` Austin S Hemmelgarn
2015-09-10 11:40 ` Austin S Hemmelgarn
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=20150913232529.GT26895@dastard \
--to=david@fromorbit.com \
--cc=Anna.Schumaker@netapp.com \
--cc=P@draigbrady.com \
--cc=andros@netapp.com \
--cc=clm@fb.com \
--cc=coreutils@gnu.org \
--cc=darrick.wong@oracle.com \
--cc=hch@infradead.org \
--cc=linux-api@vger.kernel.org \
--cc=linux-btrfs@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-nfs@vger.kernel.org \
--cc=luto@amacapital.net \
--cc=mtk.manpages@gmail.com \
--cc=viro@zeniv.linux.org.uk \
--cc=zab@zabbo.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.