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;
MIME-Version: 1.0
References: <20210724193449.361667-1-agruenba@redhat.com> <20210724193449.361667-2-agruenba@redhat.com>
In-Reply-To: <20210724193449.361667-2-agruenba@redhat.com>
From:   Linus Torvalds <torvalds@linux-foundation.org>
Date:   Sat, 24 Jul 2021 12:52:34 -0700
Message-ID: <CAHk-=whodi=ZPhoJy_a47VD+-aFtz385B4_GHvQp8Bp9NdTKUg@mail.gmail.com>
Subject: Re: [PATCH v4 1/8] iov_iter: Introduce iov_iter_fault_in_writeable helper
To:     Andreas Gruenbacher <agruenba@redhat.com>
Cc:     Alexander Viro <viro@zeniv.linux.org.uk>,
        Christoph Hellwig <hch@infradead.org>,
        "Darrick J. Wong" <djwong@kernel.org>, Jan Kara <jack@suse.cz>,
        Matthew Wilcox <willy@infradead.org>,
        cluster-devel <cluster-devel@redhat.com>,
        linux-fsdevel <linux-fsdevel@vger.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        ocfs2-devel@oss.oracle.com
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Sat, Jul 24, 2021 at 12:35 PM Andreas Gruenbacher
<agruenba@redhat.com> wrote:
>
> +int iov_iter_fault_in_writeable(const struct iov_iter *i, size_t bytes)
> +{
...
> +                       if (fault_in_user_pages(start, len, true) != len)
> +                               return -EFAULT;

Looking at this once more, I think this is likely wrong.

Why?

Because any user can/should only care about at least *part* of the
area being writable.

Imagine that you're doing a large read. If the *first* page is
writable, you should still return the partial read, not -EFAULT.

So I think the code needs to return 0 if _any_ fault was successful.
Or perhaps return how much it was able to fault in. Because returning
-EFAULT if any of it failed seems wrong, and doesn't allow for partial
success being reported.

The other reaction I have is that you now only do the
iov_iter_fault_in_writeable, but then you make fault_in_user_pages()
still have that "bool write" argument.

We already have 'fault_in_pages_readable()', and that one is more
efficient (well, at least if the fault isn't needed it is). So it
would make more sense to just implement fault_in_pages_writable()
instead of that "fault_in_user_pages(, bool write)".

                 Linus