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, 3 Mar 2021 08:22:40 +0300
From:   Dan Carpenter <dan.carpenter@oracle.com>
To:     Stefan Richter <stefanr@s5r6.in-berlin.de>
Cc:     linux1394-devel@lists.sourceforge.net,
        linux-kernel@vger.kernel.org, kernel-janitors@vger.kernel.org
Subject: Re: [PATCH] firewire: prevent integer overflow on 32bit systems
Message-ID: <20210303052239.GY2222@kadam>
References: <YD4e9XOD8JPlJzxW@mwanda>
 <20210302221911.6c1582e2@kant>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20210302221911.6c1582e2@kant>
User-Agent: Mutt/1.9.4 (2018-02-28)
Precedence: bulk

On Tue, Mar 02, 2021 at 10:19:11PM +0100, Stefan Richter wrote:
> On Mar 02 Dan Carpenter wrote:
> > In TCODE_STREAM_DATA mode, on 32bit systems, the "sizeof(*e) +
> > request->length" operation can overflow leading to memory corruption.
> > 
> > Fixes: 18e9b10fcdc0 ("firewire: cdev: add closure to async stream ioctl")
> > Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
> > ---
> >  drivers/firewire/core-cdev.c | 3 +++
> >  1 file changed, 3 insertions(+)
> > 
> > diff --git a/drivers/firewire/core-cdev.c b/drivers/firewire/core-cdev.c
> > index fb6c651214f3..314de0384035 100644
> > --- a/drivers/firewire/core-cdev.c
> > +++ b/drivers/firewire/core-cdev.c
> > @@ -587,6 +587,9 @@ static int init_request(struct client *client,
> >  	    request->length < 4)
> >  		return -EINVAL;
> >  
> > +	if (request->length > ULONG_MAX - sizeof(*e))
> > +		return -EINVAL;
> > +
> >  	e = kmalloc(sizeof(*e) + request->length, GFP_KERNEL);
> >  	if (e == NULL)
> >  		return -ENOMEM;
> 
> There is already a length check for asynchronous stream requests.
> It happens in ioctl_send_stream_packet().

Oh, yeah.  You're right.  I should have looked more carefully.  Sorry.

regards,
dan carpenter