From mboxrd@z Thu Jan 1 00:00:00 1970 From: Robin Holt Date: Sun, 21 Apr 2013 13:19:02 +0000 Subject: Re: [patch] gru: info leak in gru_get_config_info() Message-Id: <20130421131902.GT3658@sgi.com> List-Id: References: <20130421111031.GE6171@elgon.mountain> <5173D409.8010405@bfs.de> In-Reply-To: <5173D409.8010405@bfs.de> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: walter harms Cc: Dan Carpenter , Dimitri Sivanich , Robin Holt , linux-kernel@vger.kernel.org, kernel-janitors@vger.kernel.org On Sun, Apr 21, 2013 at 01:56:57PM +0200, walter harms wrote: > > > Am 21.04.2013 13:10, schrieb Dan Carpenter: > > The "info.fill" array isn't initialized so it can leak uninitialized > > stack information to user space. > > > > Signed-off-by: Dan Carpenter > > > > diff --git a/drivers/misc/sgi-gru/grufile.c b/drivers/misc/sgi-gru/grufile.c > > index 44d273c..ed5fc43 100644 > > --- a/drivers/misc/sgi-gru/grufile.c > > +++ b/drivers/misc/sgi-gru/grufile.c > > @@ -176,6 +176,7 @@ static long gru_get_config_info(unsigned long arg) > > info.nodes = num_online_nodes(); > > info.blades = info.nodes / nodesperblade; > > info.chiplets = GRU_CHIPLETS_PER_BLADE * info.blades; > > + memset(&info.fill, 0, sizeof(info.fill)); > > > > the other way around (clear first all bytes) looks more easy > in case someone will add more elements to the struct. > > memset(&info, 0, sizeof(info)); > info.nodes = num_online_nodes(); > info.blades = info.nodes / nodesperblade; That does seem more safe. Robin