Received-SPF: pass (google.com: domain of linux-kernel+bounces-56206-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Date: Wed, 7 Feb 2024 10:29:51 +0100
From: Christian Brauner <brauner@kernel.org>
To: Tycho Andersen <tycho@tycho.pizza>
Cc: Oleg Nesterov <oleg@redhat.com>, 
	"Eric W . Biederman" <ebiederm@xmission.com>, linux-kernel@vger.kernel.org, linux-api@vger.kernel.org, 
	Tycho Andersen <tandersen@netflix.com>
Subject: Re: [PATCH v2] pidfd: getfd should always report ESRCH if a task is
 exiting
Message-ID: <20240207-daran-fliesen-6039a2e36f39@brauner>
References: <20240206192357.81942-1-tycho@tycho.pizza>
 <20240207-vibrieren-waldarbeiten-30eeade05203@brauner>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <20240207-vibrieren-waldarbeiten-30eeade05203@brauner>

On Wed, Feb 07, 2024 at 10:15:25AM +0100, Christian Brauner wrote:
> On Tue, Feb 06, 2024 at 12:23:57PM -0700, Tycho Andersen wrote:
> > From: Tycho Andersen <tandersen@netflix.com>
> > 
> > We can get EBADF from __pidfd_fget() if a task is currently exiting, which
> > might be confusing. Let's check PF_EXITING, and just report ESRCH if so.
> > 
> > I chose PF_EXITING, because it is set in exit_signals(), which is called
> > before exit_files(). Since ->exit_status is mostly set after exit_files()
> > in exit_notify(), using that still leaves a window open for the race.
> > 
> > Signed-off-by: Tycho Andersen <tandersen@netflix.com>
> > v2: fix a race in the check by putting the check after __pidfd_fget()
> >     (thanks Oleg)
> > ---
> >  kernel/pid.c                                  | 17 +++++++++-
> >  .../selftests/pidfd/pidfd_getfd_test.c        | 31 ++++++++++++++++++-
> >  2 files changed, 46 insertions(+), 2 deletions(-)
> > 
> > diff --git a/kernel/pid.c b/kernel/pid.c
> > index de0bf2f8d18b..a8cd6296ed6d 100644
> > --- a/kernel/pid.c
> > +++ b/kernel/pid.c
> > @@ -693,8 +693,23 @@ static int pidfd_getfd(struct pid *pid, int fd)
> >  
> >  	file = __pidfd_fget(task, fd);
> >  	put_task_struct(task);
> > -	if (IS_ERR(file))
> > +	if (IS_ERR(file)) {
> > +		/*
> > +		 * It is possible that the target thread is exiting; it can be
> > +		 * either:
> > +		 * 1. before exit_signals(), which gives a real fd
> > +		 * 2. before exit_files() takes the task_lock() gives a real fd
> > +		 * 3. after exit_files() releases task_lock(), ->files is NULL;
> > +		 *    this has PF_EXITING, since it was set in exit_signals(),
> > +		 *    __pidfd_fget() returns EBADF.
> > +		 * In case 3 we get EBADF, but that really means ESRCH, since
> > +		 * the task is currently exiting and has freed its files
> > +		 * struct, so we fix it up.
> > +		 */
> > +		if (task->flags & PF_EXITING && PTR_ERR(file) == -EBADF)
> > +			return -ESRCH;
> 
> Isn't that a potential UAF because we called put_task_struct() above but
> this is exiting task->flags afterwards?

s/exiting/accessing/