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:   Wed, 13 Oct 2021 21:50:43 -0700
From:   Kees Cook <keescook@chromium.org>
To:     Steven Rostedt <rostedt@goodmis.org>
Cc:     Yafang Shao <laoar.shao@gmail.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Petr Mladek <pmladek@suse.com>,
        Al Viro <viro@zeniv.linux.org.uk>,
        Andrew Morton <akpm@linux-foundation.org>,
        Valentin Schneider <valentin.schneider@arm.com>,
        Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
        qiang.zhang@windriver.com, robdclark@chromium.org,
        christian@brauner.io, Dietmar Eggemann <dietmar.eggemann@arm.com>,
        Ingo Molnar <mingo@redhat.com>,
        Juri Lelli <juri.lelli@redhat.com>,
        Vincent Guittot <vincent.guittot@linaro.org>,
        Benjamin Segall <bsegall@google.com>,
        Mel Gorman <mgorman@suse.de>,
        Daniel Bristot de Oliveira <bristot@redhat.com>,
        David Miller <davem@davemloft.net>, kuba@kernel.org,
        LKML <linux-kernel@vger.kernel.org>,
        Vladimir Zapolskiy <vzapolskiy@gmail.com>,
        David Howells <dhowells@redhat.com>
Subject: Re: [PATCH v4 2/5] connector: use __get_task_comm in
 proc_comm_connector
Message-ID: <202110132148.523C3EA@keescook>
References: <20211013102346.179642-1-laoar.shao@gmail.com>
 <20211013102346.179642-3-laoar.shao@gmail.com>
 <20211013101921.0843aaf0@gandalf.local.home>
 <CALOAHbCt+rLiPE4_zZO_f5sKybKwYntqupx_L9V_J+yByoFvOw@mail.gmail.com>
 <20211013222418.7ea9727d@oasis.local.home>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20211013222418.7ea9727d@oasis.local.home>
Precedence: bulk

On Wed, Oct 13, 2021 at 10:24:18PM -0400, Steven Rostedt wrote:
> On Thu, 14 Oct 2021 09:48:09 +0800
> Yafang Shao <laoar.shao@gmail.com> wrote:
> 
> > > __get_task_comm() uses strncpy() which my understanding is, does not add
> > > the nul terminating byte when truncating. Which changes the functionality
> > > here. As all task comms have a terminating byte, the old method would copy
> > > that and include it. This won't add the terminating byte if the buffer is
> > > smaller than the comm, and that might cause issues.
> > >  
> > 
> > Right, that is a problem.
> > It seems that we should add a new helper get_task_comm_may_truncated().
> 
> Or simply change __get_task_comm() to:
> 
> char *__get_task_comm(char *buf, size_t buf_size, struct task_struct *tsk)
> {
> 	task_lock(tsk);
> 	strncpy(buf, tsk->comm, buf_size);
> 	/* The copied value is always nul terminated */
> 	buf[buf_size - 1] = '\0';
> 	task_unlock(tsk);
> 	return buf;
> }
> 
> But that should probably be a separate patch.

strscpy_pad() is the right thing here -- it'll retain the NUL-fill
properties of strncpy and terminate correctly.

The use of non-terminating issue with strncpy() wasn't a problem here
because get_task_comm() would always make sure task->comm was
terminated. (It uses strlcpy(), which I think needs to be changed to
strscpy_pad() too...)

-- 
Kees Cook