linux-hardening - Re: [PATCH] net: bonding: Use strscpy() instead of manually-truncated strncpy()

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <202106021257.F0DFED3@keescook>
Date:   Wed, 2 Jun 2021 13:02:02 -0700
From:   Kees Cook <keescook@...omium.org>
To:     Rao Shoaib <rao.shoaib@...cle.com>
Cc:     Jay Vosburgh <j.vosburgh@...il.com>,
        kernel test robot <lkp@...el.com>,
        Veaceslav Falico <vfalico@...il.com>,
        Andy Gospodarek <andy@...yhouse.net>,
        "David S. Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>, linux-kernel@...r.kernel.org,
        netdev@...r.kernel.org, linux-hardening@...r.kernel.org
Subject: Re: [PATCH] net: bonding: Use strscpy() instead of
 manually-truncated strncpy()

On Wed, Jun 02, 2021 at 12:46:46PM -0700, Rao Shoaib wrote:
> Would it make sense to also replace the other strncpy in the same file.

                strncpy(ifr.ifr_name, slave_dev->name, IFNAMSIZ);

I couldn't tell if this was a non-string, if it needed padding, etc. The
one I fixed below appears to be null-terminated? (Though now that I look
at it, perhaps it should be using strscpy_pad().)

And there are a bunch of other manual truncations in the kernel on
ifr_name:

$ git grep ifr_name | grep 'IFNAMSIZ.*=.*0'
drivers/net/tun.c:              ifr.ifr_name[IFNAMSIZ-1] = '\0';
net/core/dev_ioctl.c:   ifr->ifr_name[IFNAMSIZ-1] = 0;
net/core/dev_ioctl.c:   ifr->ifr_name[IFNAMSIZ-1] = 0;
net/decnet/dn_dev.c:    ifr->ifr_name[IFNAMSIZ-1] = 0;
net/ieee802154/socket.c:        ifr.ifr_name[IFNAMSIZ-1] = 0;
net/ipv4/devinet.c:     ifr->ifr_name[IFNAMSIZ - 1] = 0;
net/wireless/wext-core.c:       iwr.ifr_name[IFNAMSIZ-1] = 0;
tools/lib/bpf/xsk.c:    ifr.ifr_name[IFNAMSIZ - 1] = '\0';

And given the copy_to_user() that might happen, I think this should
absolutely be strscpy_pad(). I will send a v2...

-Kees

> 
> Shoaib
> 
> On 6/2/21 11:11 AM, Kees Cook wrote:
> > Silence this warning by just using strscpy() directly:
> > 
> > drivers/net/bonding/bond_main.c:4877:3: warning: 'strncpy' specified bound 16 equals destination size [-Wstringop-truncation]
> >      4877 |   strncpy(params->primary, primary, IFNAMSIZ);
> >           |   ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> > 
> > Reported-by: kernel test robot <lkp@...el.com>
> > Link: https://lore.kernel.org/lkml/202102150705.fdR6obB0-lkp@intel.com
> > Signed-off-by: Kees Cook <keescook@...omium.org>
> > ---
> >   drivers/net/bonding/bond_main.c | 6 ++----
> >   1 file changed, 2 insertions(+), 4 deletions(-)
> > 
> > diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
> > index c5a646d06102..ecfc48f2d0d0 100644
> > --- a/drivers/net/bonding/bond_main.c
> > +++ b/drivers/net/bonding/bond_main.c
> > @@ -5329,10 +5329,8 @@ static int bond_check_params(struct bond_params *params)
> >   			(struct reciprocal_value) { 0 };
> >   	}
> > -	if (primary) {
> > -		strncpy(params->primary, primary, IFNAMSIZ);
> > -		params->primary[IFNAMSIZ - 1] = 0;
> > -	}
> > +	if (primary)
> > +		strscpy(params->primary, primary, sizeof(params->primary));
> >   	memcpy(params->arp_targets, arp_target, sizeof(arp_target));

-- 
Kees Cook