Bluetooth: Prevent stack info leak from the EFS element.

From: Ben Seri <ben@armis.com>

In the function l2cap_parse_conf_rsp and in the function
l2cap_parse_conf_req the following variable is declared without
initialization:

struct l2cap_conf_efs efs;

In addition, when parsing input configuration parameters in both of
these functions, the switch case for handling EFS elements may skip the
memcpy call that will write to the efs variable:

...
case L2CAP_CONF_EFS:
if (olen == sizeof(efs))
memcpy(&efs, (void *)val, olen);
...

The olen in the above if is attacker controlled, and regardless of that
if, in both of these functions the efs variable would eventually be
added to the outgoing configuration request that is being built:

l2cap_add_conf_opt(&ptr, L2CAP_CONF_EFS, sizeof(efs), (unsigned long) &efs);

So by sending a configuration request, or response, that contains an
L2CAP_CONF_EFS element, but with an element length that is not
sizeof(efs) - the memcpy to the uninitialized efs variable can be
avoided, and the uninitialized variable would be returned to the
attacker (16 bytes).

This issue has been assigned CVE-2017-1000410

Cc: Marcel Holtmann <marcel@holtmann.org>
Cc: Gustavo Padovan <gustavo@padovan.org>
Cc: Johan Hedberg <johan.hedberg@gmail.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Ben Seri <ben@armis.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 net/bluetooth/l2cap_core.c |   20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)

Marcel, for some reason this patch never got applied, despite lots of
advance notice.  Please, someone queue it up as it resolves the above
very-well-reported issue.

On Fri, Dec 08, 2017 at 03:14:47PM +0100, gregkh@linuxfoundation.org wrote:
> From: Ben Seri <ben@armis.com>
> 
> In the function l2cap_parse_conf_rsp and in the function
> l2cap_parse_conf_req the following variable is declared without
> initialization:
> 
> struct l2cap_conf_efs efs;
> 
> In addition, when parsing input configuration parameters in both of
> these functions, the switch case for handling EFS elements may skip the
> memcpy call that will write to the efs variable:
> 
> ...
> case L2CAP_CONF_EFS:
> if (olen == sizeof(efs))
> memcpy(&efs, (void *)val, olen);
> ...
> 
> The olen in the above if is attacker controlled, and regardless of that
> if, in both of these functions the efs variable would eventually be
> added to the outgoing configuration request that is being built:
> 
> l2cap_add_conf_opt(&ptr, L2CAP_CONF_EFS, sizeof(efs), (unsigned long) &efs);
> 
> So by sending a configuration request, or response, that contains an
> L2CAP_CONF_EFS element, but with an element length that is not
> sizeof(efs) - the memcpy to the uninitialized efs variable can be
> avoided, and the uninitialized variable would be returned to the
> attacker (16 bytes).
> 
> This issue has been assigned CVE-2017-1000410
> 
> Cc: Marcel Holtmann <marcel@holtmann.org>
> Cc: Gustavo Padovan <gustavo@padovan.org>
> Cc: Johan Hedberg <johan.hedberg@gmail.com>
> Cc: stable <stable@vger.kernel.org>
> Signed-off-by: Ben Seri <ben@armis.com>
> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> ---
>  net/bluetooth/l2cap_core.c |   20 +++++++++++---------
>  1 file changed, 11 insertions(+), 9 deletions(-)
> 
> Marcel, for some reason this patch never got applied, despite lots of
> advance notice.  Please, someone queue it up as it resolves the above
> very-well-reported issue.
> 

This patch is still not upstream or in -next. Given that we (ChromeOS)
are heavy Bluetooth users, I'll go ahead and apply it without waiting
any longer. For my understanding and for tracking purposes, it would
be useful to know why it is not being applied. Does anyone know ?

Thanks,
Guenter

> diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c
> index 43ba91c440bc..fc6615d59165 100644
> --- a/net/bluetooth/l2cap_core.c
> +++ b/net/bluetooth/l2cap_core.c
> @@ -3363,9 +3363,10 @@ static int l2cap_parse_conf_req(struct l2cap_chan *chan, void *data, size_t data
>  			break;
>  
>  		case L2CAP_CONF_EFS:
> -			remote_efs = 1;
> -			if (olen == sizeof(efs))
> +			if (olen == sizeof(efs)) {
> +				remote_efs = 1;
>  				memcpy(&efs, (void *) val, olen);
> +			}
>  			break;
>  
>  		case L2CAP_CONF_EWS:
> @@ -3584,16 +3585,17 @@ static int l2cap_parse_conf_rsp(struct l2cap_chan *chan, void *rsp, int len,
>  			break;
>  
>  		case L2CAP_CONF_EFS:
> -			if (olen == sizeof(efs))
> +			if (olen == sizeof(efs)) {
>  				memcpy(&efs, (void *)val, olen);
>  
> -			if (chan->local_stype != L2CAP_SERV_NOTRAFIC &&
> -			    efs.stype != L2CAP_SERV_NOTRAFIC &&
> -			    efs.stype != chan->local_stype)
> -				return -ECONNREFUSED;
> +				if (chan->local_stype != L2CAP_SERV_NOTRAFIC &&
> +				    efs.stype != L2CAP_SERV_NOTRAFIC &&
> +				    efs.stype != chan->local_stype)
> +					return -ECONNREFUSED;
>  
> -			l2cap_add_conf_opt(&ptr, L2CAP_CONF_EFS, sizeof(efs),
> -					   (unsigned long) &efs, endptr - ptr);
> +				l2cap_add_conf_opt(&ptr, L2CAP_CONF_EFS, sizeof(efs),
> +						   (unsigned long) &efs, endptr - ptr);
> +			}
>  			break;
>  
>  		case L2CAP_CONF_FCS:

On Tue, Jan 02, 2018 at 12:26:15PM -0800, Guenter Roeck wrote:
> On Fri, Dec 08, 2017 at 03:14:47PM +0100, gregkh@linuxfoundation.org wrote:
> > From: Ben Seri <ben@armis.com>
> > 
> > In the function l2cap_parse_conf_rsp and in the function
> > l2cap_parse_conf_req the following variable is declared without
> > initialization:
> > 
> > struct l2cap_conf_efs efs;
> > 
> > In addition, when parsing input configuration parameters in both of
> > these functions, the switch case for handling EFS elements may skip the
> > memcpy call that will write to the efs variable:
> > 
> > ...
> > case L2CAP_CONF_EFS:
> > if (olen == sizeof(efs))
> > memcpy(&efs, (void *)val, olen);
> > ...
> > 
> > The olen in the above if is attacker controlled, and regardless of that
> > if, in both of these functions the efs variable would eventually be
> > added to the outgoing configuration request that is being built:
> > 
> > l2cap_add_conf_opt(&ptr, L2CAP_CONF_EFS, sizeof(efs), (unsigned long) &efs);
> > 
> > So by sending a configuration request, or response, that contains an
> > L2CAP_CONF_EFS element, but with an element length that is not
> > sizeof(efs) - the memcpy to the uninitialized efs variable can be
> > avoided, and the uninitialized variable would be returned to the
> > attacker (16 bytes).
> > 
> > This issue has been assigned CVE-2017-1000410
> > 
> > Cc: Marcel Holtmann <marcel@holtmann.org>
> > Cc: Gustavo Padovan <gustavo@padovan.org>
> > Cc: Johan Hedberg <johan.hedberg@gmail.com>
> > Cc: stable <stable@vger.kernel.org>
> > Signed-off-by: Ben Seri <ben@armis.com>
> > Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> > ---
> >  net/bluetooth/l2cap_core.c |   20 +++++++++++---------
> >  1 file changed, 11 insertions(+), 9 deletions(-)
> > 
> > Marcel, for some reason this patch never got applied, despite lots of
> > advance notice.  Please, someone queue it up as it resolves the above
> > very-well-reported issue.
> > 
> 
> This patch is still not upstream or in -next. Given that we (ChromeOS)
> are heavy Bluetooth users, I'll go ahead and apply it without waiting
> any longer. For my understanding and for tracking purposes, it would
> be useful to know why it is not being applied. Does anyone know ?

I have no idea what is going on here, I'm going to throw it in my
char-misc tree now, this really isn't acceptable at all.

greg k-h