Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Sun, 27 Dec 2020 21:25:31 +0000
From:   Al Viro <viro@zeniv.linux.org.uk>
To:     Linus Torvalds <torvalds@linux-foundation.org>
Cc:     Jussi Kivilinna <jussi.kivilinna@iki.fi>,
        Christoph Hellwig <hch@lst.de>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        stable <stable@vger.kernel.org>, lwn@lwn.net,
        Jiri Slaby <jslaby@suse.cz>
Subject: Re: LXC broken with 5.10-stable?, ok with 5.9-stable (Re: Linux
 5.10.3)
Message-ID: <20201227212531.GD3579531@ZenIV.linux.org.uk>
References: <16089960203931@kroah.com>
 <5ab86253-7703-e892-52b7-e6a8af579822@iki.fi>
 <CAHk-=wgtU5+7jPuPtDEpwhTuUUkA3CBN=V92Jg0Ag0=3LhfKqA@mail.gmail.com>
 <b45f1065-2da9-08c0-26f2-e5b69e780bc6@iki.fi>
 <CAHk-=wgy6NQrTMwiEWpHUPvW-nfgX7XrBrsxQ6TkRy6NasSFQg@mail.gmail.com>
 <CAHk-=whF=+EzrxP=3zNMH-1L2Nfs7fNoSufqDwOdRQo5qyMwfw@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAHk-=whF=+EzrxP=3zNMH-1L2Nfs7fNoSufqDwOdRQo5qyMwfw@mail.gmail.com>
Sender: Al Viro <viro@ftp.linux.org.uk>
Precedence: bulk

On Sun, Dec 27, 2020 at 12:12:00PM -0800, Linus Torvalds wrote:
> On Sun, Dec 27, 2020 at 11:05 AM Linus Torvalds
> <torvalds@linux-foundation.org> wrote:
> >
> > On Sun, Dec 27, 2020 at 10:39 AM Jussi Kivilinna <jussi.kivilinna@iki.fi> wrote:
> > >
> > > 5.10.3 with patch compiles fine, but does not solve the issue.
> >
> > Duh. adding the read_iter only fixes kernel_read(). For splice, it also needs a
> >
> >         .splice_read = generic_file_splice_read,
> >
> > in the file operations, something like this...
> 
> Ok, I verified that patch with the test-case in the bugzilla, and it
> seems trivially fine.
> 
> So it's committed as 14e3e989f6a5 ("proc mountinfo: make splice
> available again") now.

Is there any point in not doing the same (scripted, obviously) for
all instances with .read == seq_read?  IIRC, Christoph even posted
something along those lines, but it went nowhere for some reason...