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:   Thu, 7 May 2020 21:05:03 +0200
From:   Max Kellermann <mk@cm4all.com>
To:     Al Viro <viro@zeniv.linux.org.uk>
Cc:     Max Kellermann <mk@cm4all.com>, axboe@kernel.dk,
        linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
        stable@vger.kernel.org
Subject: Re: [PATCH] fs/io_uring: fix O_PATH fds in openat, openat2, statx
Message-ID: <20200507190503.GA15998@rabbit.intern.cm-ag>
References: <20200507185725.15840-1-mk@cm4all.com>
 <20200507190131.GF23230@ZenIV.linux.org.uk>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20200507190131.GF23230@ZenIV.linux.org.uk>
User-Agent: Mutt/1.10.1 (2018-07-13)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 2020/05/07 21:01, Al Viro <viro@zeniv.linux.org.uk> wrote:
> On Thu, May 07, 2020 at 08:57:25PM +0200, Max Kellermann wrote:
> > If an operation's flag `needs_file` is set, the function
> > io_req_set_file() calls io_file_get() to obtain a `struct file*`.
> > 
> > This fails for `O_PATH` file descriptors, because those have no
> > `struct file*`
> 
> O_PATH descriptors most certainly *do* have that.  What the hell
> are you talking about?

Oh, then my patch description (and my understanding of the root
problem) is wrong.  In my debugging session, io_file_get() on that fd
returned NULL, so I assumed O_PATH doesn't have that, but maybe there
are other reasons.

In any case, with a "real" fd, io_uring openat() succeeds, and my
patch makes the problem with O_PATH go away.

I guess I need to learn more about what happens inside io_file_get().

Max