Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <20190112152844.26550-1-w@1wt.eu> <CAGXu5jLGkPqA+ZY7aFizpsRj9VvOhUSfixWcLLVHXwsAo3PQyA@mail.gmail.com>
In-Reply-To: <CAGXu5jLGkPqA+ZY7aFizpsRj9VvOhUSfixWcLLVHXwsAo3PQyA@mail.gmail.com>
From:   Kees Cook <keescook@chromium.org>
Date:   Mon, 14 Jan 2019 17:07:16 -0800
Message-ID: <CAGXu5jKFoTjvr-EHUYHn7TSgEfiwn2BJ1J4Dwu3FGOLmz+TxvA@mail.gmail.com>
Subject: Re: [PATCH 1/8] lkdtm: change snprintf to scnprintf for possible overflow
To:     Willy Tarreau <w@1wt.eu>
Cc:     Silvio Cesare <silvio.cesare@gmail.com>,
        LKML <linux-kernel@vger.kernel.org>,
        Dan Carpenter <dan.carpenter@oracle.com>,
        Will Deacon <will.deacon@arm.com>, Greg KH <greg@kroah.com>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Jan 14, 2019 at 5:02 PM Kees Cook <keescook@chromium.org> wrote:
> On Sat, Jan 12, 2019 at 7:28 AM Willy Tarreau <w@1wt.eu> wrote:
> > From: Silvio Cesare <silvio.cesare@gmail.com>
> > Change snprintf to scnprintf. There are generally two cases where using
> > snprintf causes problems.
>
> (I didn't find a 0/8 cover letter, so I'm replying here...)

I forgot to mention: can we please get a Coccinelle rule added to
catch these cases in the future? (And make sure sfr is running it? :)
)

My attempt at it was:

@@
expression LEN, BUF, SIZE;
identifier FUNC;
@@

  LEN += snprintf(BUF + LEN, SIZE - LEN, ...);
  ... when != LEN &gt; SIZE
      when != LEN &gt;= SIZE
* FUNC(..., LEN, ...)

But this needs adjustment to deal with some false positives (like using min()).

--
Kees Cook