From: Christoph Hellwig <hch@infradead.org> To: Austin S Hemmelgarn <ahferroin7@gmail.com> Cc: Andy Lutomirski <luto@amacapital.net>, Anna Schumaker <Anna.Schumaker@netapp.com>, "Darrick J. Wong" <darrick.wong@oracle.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 Subject: Re: [PATCH v5 8/9] vfs: Add vfs_copy_file_range() support for pagecache copies Date: Wed, 14 Oct 2015 23:36:21 -0700 [thread overview] Message-ID: <20151015063621.GA3025@infradead.org> (raw) In-Reply-To: <561EA83E.8080000@gmail.com> On Wed, Oct 14, 2015 at 03:08:46PM -0400, Austin S Hemmelgarn wrote: > Whether or not reflink is different from a copy is entirely a matter of who > is looking at it. So what? I've been trying to explain why clone semantics matter, and I've not seen a counter argument for that. I've also explained a couple times that a valid clone always is a valid copy, and I've only heard some slight disagreement, and so far none as long as we take the COPY_FALLOCATE option into account. Note that all of this also applies to storage devices - any smart array will do a clone-like operation underneath an XCOPY, but so far SCSI doesn't provide full clone _semantics_ even if you can emulate a lot of it using a lot of complexity around ROD tokens. Similar at the SCSI level you can perform a fallocate-like operation using the anchor bit in the UNMAP or WRITE SAME commands. > I agree that we shouldn't try to make a reflink by default (less than > intelligent programmers won't read the docs completely, and will make > various stupid assumptions about how this is 'supposed' to work, making the > defaults less ambiguous is a good thing), but it makes sense (at least, it > does to me) to have the ability to say 'make this block of data appear at > this location as well, I don't care how you do it as long as they are > functionally independent for userspace applications'. Yes, we absolutely should use reflink as a default implementation for copy where available. But we also need a clone or reflink interface that only gives us well specified reflink semantics, and not the much weaker copy semantics.
WARNING: multiple messages have this Message-ID (diff)
From: Christoph Hellwig <hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org> To: Austin S Hemmelgarn <ahferroin7-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> Cc: Andy Lutomirski <luto-kltTT9wpgjJwATOyAt5JVQ@public.gmane.org>, Anna Schumaker <Anna.Schumaker-HgOvQuBEEgTQT0dZR+AlfA@public.gmane.org>, "Darrick J. Wong" <darrick.wong-QHcLZuEGTsvQT0dZR+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 Subject: Re: [PATCH v5 8/9] vfs: Add vfs_copy_file_range() support for pagecache copies Date: Wed, 14 Oct 2015 23:36:21 -0700 [thread overview] Message-ID: <20151015063621.GA3025@infradead.org> (raw) In-Reply-To: <561EA83E.8080000-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> On Wed, Oct 14, 2015 at 03:08:46PM -0400, Austin S Hemmelgarn wrote: > Whether or not reflink is different from a copy is entirely a matter of who > is looking at it. So what? I've been trying to explain why clone semantics matter, and I've not seen a counter argument for that. I've also explained a couple times that a valid clone always is a valid copy, and I've only heard some slight disagreement, and so far none as long as we take the COPY_FALLOCATE option into account. Note that all of this also applies to storage devices - any smart array will do a clone-like operation underneath an XCOPY, but so far SCSI doesn't provide full clone _semantics_ even if you can emulate a lot of it using a lot of complexity around ROD tokens. Similar at the SCSI level you can perform a fallocate-like operation using the anchor bit in the UNMAP or WRITE SAME commands. > I agree that we shouldn't try to make a reflink by default (less than > intelligent programmers won't read the docs completely, and will make > various stupid assumptions about how this is 'supposed' to work, making the > defaults less ambiguous is a good thing), but it makes sense (at least, it > does to me) to have the ability to say 'make this block of data appear at > this location as well, I don't care how you do it as long as they are > functionally independent for userspace applications'. Yes, we absolutely should use reflink as a default implementation for copy where available. But we also need a clone or reflink interface that only gives us well specified reflink semantics, and not the much weaker copy semantics.
next prev parent reply other threads:[~2015-10-15 6:36 UTC|newest] Thread overview: 129+ messages / expand[flat|nested] mbox.gz Atom feed top 2015-09-30 17:26 [PATCH v5 0/9] VFS: In-kernel copy system call Anna Schumaker 2015-09-30 17:26 ` Anna Schumaker 2015-09-30 17:26 ` Anna Schumaker 2015-09-30 17:26 ` [PATCH v5 1/9] vfs: add copy_file_range syscall and vfs helper Anna Schumaker 2015-09-30 17:26 ` Anna Schumaker 2015-09-30 17:26 ` [PATCH v5 2/9] x86: add sys_copy_file_range to syscall tables Anna Schumaker 2015-09-30 17:26 ` Anna Schumaker 2015-09-30 17:26 ` [PATCH v5 3/9] btrfs: add .copy_file_range file operation Anna Schumaker 2015-09-30 17:26 ` Anna Schumaker 2015-09-30 17:26 ` [PATCH v5 4/9] vfs: Copy should check len after file open mode Anna Schumaker 2015-09-30 17:26 ` Anna Schumaker 2015-09-30 17:26 ` Anna Schumaker 2015-10-11 14:22 ` Christoph Hellwig 2015-10-11 14:22 ` Christoph Hellwig 2015-09-30 17:26 ` [PATCH v5 5/9] vfs: Copy shouldn't forbid ranges inside the same file Anna Schumaker 2015-09-30 17:26 ` Anna Schumaker 2015-10-11 14:22 ` Christoph Hellwig 2015-10-14 17:37 ` Anna Schumaker 2015-10-14 17:37 ` Anna Schumaker 2015-10-14 17:37 ` Anna Schumaker 2015-10-14 18:25 ` Christoph Hellwig 2015-10-14 18:27 ` Anna Schumaker 2015-10-14 18:27 ` Anna Schumaker 2015-10-14 18:27 ` Anna Schumaker 2015-09-30 17:26 ` [PATCH v5 6/9] vfs: Copy should use file_out rather than file_in Anna Schumaker 2015-09-30 17:26 ` Anna Schumaker 2015-10-11 14:24 ` Christoph Hellwig 2015-09-30 17:26 ` [PATCH v5 7/9] vfs: Remove copy_file_range mountpoint checks Anna Schumaker 2015-09-30 17:26 ` Anna Schumaker 2015-10-11 14:23 ` Christoph Hellwig 2015-10-14 17:41 ` Anna Schumaker 2015-10-14 17:41 ` Anna Schumaker 2015-10-14 18:25 ` Christoph Hellwig 2015-10-14 18:25 ` Christoph Hellwig 2015-09-30 17:26 ` [PATCH v5 8/9] vfs: Add vfs_copy_file_range() support for pagecache copies Anna Schumaker 2015-09-30 17:26 ` Anna Schumaker 2015-09-30 17:26 ` Anna Schumaker 2015-10-08 1:40 ` Neil Brown 2015-10-09 11:15 ` Pádraig Brady 2015-10-09 11:15 ` Pádraig Brady 2015-10-13 20:25 ` Anna Schumaker 2015-10-13 20:25 ` Anna Schumaker 2015-10-14 7:41 ` Christoph Hellwig 2015-10-14 7:41 ` Christoph Hellwig 2015-10-13 19:45 ` Anna Schumaker 2015-10-13 19:45 ` Anna Schumaker 2015-10-13 19:45 ` Anna Schumaker 2015-10-11 14:22 ` Christoph Hellwig 2015-10-11 14:22 ` Christoph Hellwig 2015-10-12 23:17 ` Darrick J. Wong 2015-10-12 23:17 ` Darrick J. Wong 2015-10-13 3:36 ` Trond Myklebust 2015-10-13 7:19 ` Darrick J. Wong 2015-10-13 7:19 ` Darrick J. Wong 2015-10-13 7:30 ` Christoph Hellwig 2015-10-13 7:30 ` Christoph Hellwig 2015-10-13 7:27 ` Christoph Hellwig 2015-10-13 7:27 ` Christoph Hellwig 2015-11-10 6:24 ` Darrick J. Wong 2015-10-14 17:59 ` Anna Schumaker 2015-10-14 17:59 ` Anna Schumaker 2015-10-14 17:59 ` Anna Schumaker 2015-10-14 18:08 ` Andy Lutomirski 2015-10-14 18:27 ` Christoph Hellwig 2015-10-14 18:38 ` Andy Lutomirski 2015-10-14 18:38 ` Andy Lutomirski 2015-10-14 18:49 ` Christoph Hellwig 2015-10-14 18:49 ` Christoph Hellwig 2015-10-14 18:53 ` Andy Lutomirski 2015-10-14 19:14 ` Austin S Hemmelgarn 2015-10-14 19:14 ` Austin S Hemmelgarn 2015-10-14 19:39 ` Pádraig Brady 2015-10-14 19:39 ` Pádraig Brady 2015-10-15 5:56 ` Christoph Hellwig 2015-10-15 5:56 ` Christoph Hellwig 2015-10-14 19:08 ` Austin S Hemmelgarn 2015-10-14 19:08 ` Austin S Hemmelgarn 2015-10-15 6:36 ` Christoph Hellwig [this message] 2015-10-15 6:36 ` Christoph Hellwig 2015-10-15 12:24 ` Austin S Hemmelgarn 2015-10-15 12:24 ` Austin S Hemmelgarn 2015-10-16 5:38 ` Christoph Hellwig 2015-10-16 5:38 ` Christoph Hellwig 2015-10-16 11:46 ` Austin S Hemmelgarn 2015-10-16 12:02 ` Pádraig Brady 2015-10-16 12:02 ` Pádraig Brady 2015-10-16 12:24 ` Christoph Hellwig 2015-10-16 12:24 ` Christoph Hellwig 2015-10-16 12:46 ` Austin S Hemmelgarn 2015-10-16 12:46 ` Austin S Hemmelgarn 2015-10-16 12:21 ` Christoph Hellwig 2015-10-16 12:21 ` Christoph Hellwig 2015-10-16 12:50 ` Austin S Hemmelgarn 2015-10-16 13:12 ` Christoph Hellwig 2015-10-16 14:11 ` Austin S Hemmelgarn 2015-10-14 18:11 ` Darrick J. Wong 2015-10-14 18:11 ` Darrick J. Wong 2015-10-14 18:26 ` Andy Lutomirski 2015-09-30 17:26 ` [PATCH v5 9/9] btrfs: btrfs_copy_file_range() only supports reflinks Anna Schumaker 2015-09-30 17:26 ` Anna Schumaker 2015-10-11 14:29 ` Christoph Hellwig 2015-10-11 14:29 ` Christoph Hellwig 2015-10-12 10:23 ` Pádraig Brady 2015-10-12 10:23 ` Pádraig Brady 2015-10-12 14:34 ` Christoph Hellwig 2015-10-12 23:41 ` Darrick J. Wong 2015-10-12 23:41 ` Darrick J. Wong 2015-10-13 7:29 ` Christoph Hellwig 2015-10-13 7:29 ` Christoph Hellwig 2015-10-14 18:46 ` Darrick J. Wong 2015-10-14 18:46 ` Darrick J. Wong 2015-10-15 6:00 ` Christoph Hellwig 2015-10-16 11:49 ` Chris Mason 2015-10-16 11:49 ` Chris Mason 2015-10-16 11:49 ` Chris Mason 2015-10-16 12:25 ` Christoph Hellwig 2015-10-16 12:25 ` Christoph Hellwig 2015-10-16 13:19 ` Chris Mason 2015-10-16 13:19 ` Chris Mason 2015-10-16 13:19 ` Chris Mason 2015-10-16 21:44 ` Dave Chinner 2015-10-17 13:44 ` Chris Mason 2015-10-17 13:44 ` Chris Mason 2015-10-17 13:44 ` Chris Mason 2015-10-15 8:35 ` Dave Chinner 2015-10-15 8:35 ` Dave Chinner 2015-09-30 17:26 ` [PATCH v5 10/9] copy_file_range.2: New page documenting copy_file_range() Anna Schumaker 2015-09-30 17:26 ` Anna Schumaker 2015-09-30 17:26 ` Anna Schumaker
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=20151015063621.GA3025@infradead.org \ --to=hch@infradead.org \ --cc=Anna.Schumaker@netapp.com \ --cc=ahferroin7@gmail.com \ --cc=andros@netapp.com \ --cc=clm@fb.com \ --cc=darrick.wong@oracle.com \ --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: linkBe 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.