DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org C6E0520651
Date:   Mon, 22 Oct 2018 13:29:53 -0700
From:   Matthew Wilcox <willy@infradead.org>
To:     Amir Goldstein <amir73il@gmail.com>
Cc:     Olga Kornievskaia <olga.kornievskaia@gmail.com>,
        Al Viro <viro@zeniv.linux.org.uk>,
        linux-fsdevel <linux-fsdevel@vger.kernel.org>,
        Linux NFS Mailing List <linux-nfs@vger.kernel.org>,
        fweimer@redhat.com, Steve French <smfrench@gmail.com>,
        "Darrick J. Wong" <darrick.wong@oracle.com>,
        Christoph Hellwig <hch@lst.de>, linux-api@vger.kernel.org
Subject: Re: [PATCH v1 02/11] VFS permit cross device vfs_copy_file_range
Message-ID: <20181022202953.GA20085@bombadil.infradead.org>
References: <20181019153018.32507-1-olga.kornievskaia@gmail.com>
 <20181019153018.32507-2-olga.kornievskaia@gmail.com>
 <20181020040530.GG32577@ZenIV.linux.org.uk>
 <CAOQ4uxhAvypEvGap02CkH_bHtFt7tUVxTWeZUL5OV_V8xwWjDA@mail.gmail.com>
 <CAN-5tyHqtV7YhfK7DY4rY643bbfV=8MOE6=mxHYJufCsdcN9Jw@mail.gmail.com>
 <20181022190620.GA8863@bombadil.infradead.org>
 <CAN-5tyF=voE2=DBLKvJCHZc9+zMzysqD8REYYb-z5jwq2BLqGQ@mail.gmail.com>
 <CAOQ4uxjJgghh48LUsqCofS0uiuA35CQT-zEMvMESkqniLfv2fA@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAOQ4uxjJgghh48LUsqCofS0uiuA35CQT-zEMvMESkqniLfv2fA@mail.gmail.com>
User-Agent: Mutt/1.9.2 (2017-12-15)
Sender: linux-nfs-owner@vger.kernel.org
Precedence: bulk

On Mon, Oct 22, 2018 at 10:48:10PM +0300, Amir Goldstein wrote:
> On Mon, Oct 22, 2018 at 10:35 PM Olga Kornievskaia
> <olga.kornievskaia@gmail.com> wrote:
> > So remove the check all together for the VFS (that was my original
> > patch to begin with (like #1 not this one). So am I missing the point
> > again, I keep getting different corrections every time.
> 
> Because there are different opinions... although you did get the opinion
> of the VFS maintainer, which was: compare i_sb->s_type.
> 
> Jeff, Matthew, really, what's the use of "allowing" cross fs type copy inside
> filesystem code? and which method is going to be called?
> file_out->f_op->copy_file_range()?
> file_in->f_op->copy_file_range()?

The destination's method, as Olga originally had.

> Do we need to check if both are implemented? either?
> This is just confusing Olga and gives no real value to anyone.
> If we ever have a filesystem copy_file_range() method that can deal
> with cross fs type copy, we can change it then when we know the
> required semantics of that future call.

Wrong.  Go back and read my reasoning earlier this thread.

> That is not to say that we cannot relax same fs type from copy_file_range()
> syscall. That has already been done with the current patch, just not officially
> declared in commit message.
> 
> Thanks,
> Amir.