[PATCH] selinux: endian fix

[PATCH] selinux: endian fix

[Alexey Dobriyan]

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
---

--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -3231,7 +3231,7 @@ static int selinux_socket_sock_rcv_skb(s
 		goto out;
 
 	/* Handle mapped IPv4 packets arriving via IPv6 sockets */
-	if (family == PF_INET6 && skb->protocol == ntohs(ETH_P_IP))
+	if (family == PF_INET6 && skb->protocol == htons(ETH_P_IP))
 		family = PF_INET;
 
  	read_lock_bh(&sk->sk_callback_lock);

Re: [PATCH] selinux: endian fix

[James Morris]

On Tue, 16 May 2006, Alexey Dobriyan wrote:

> Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>

Hmm, I'm certain this was tested (perhaps on a BE machine, though). In any 
case skb->protocol should definitely be network byte order.

Acked-by: James Morris <jmorris@namei.org>



> --- a/security/selinux/hooks.c
> +++ b/security/selinux/hooks.c
> @@ -3231,7 +3231,7 @@ static int selinux_socket_sock_rcv_skb(s
>  		goto out;
>  
>  	/* Handle mapped IPv4 packets arriving via IPv6 sockets */
> -	if (family == PF_INET6 && skb->protocol == ntohs(ETH_P_IP))
> +	if (family == PF_INET6 && skb->protocol == htons(ETH_P_IP))
>  		family = PF_INET;
>  
>   	read_lock_bh(&sk->sk_callback_lock);
> 

-- 
James Morris
<jmorris@namei.org>

Re: [PATCH] selinux: endian fix

[Andreas Schwab]

James Morris <jmorris@namei.org> writes:

> On Tue, 16 May 2006, Alexey Dobriyan wrote:
>
>> Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
>
> Hmm, I'm certain this was tested (perhaps on a BE machine, though).

ntohs and htons are identical operations.  Either you swap or you don't,
but there is only one way to swap a short.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: [PATCH] selinux: endian fix

[James Morris]

On Tue, 16 May 2006, Andreas Schwab wrote:

> ntohs and htons are identical operations.  Either you swap or you don't,
> but there is only one way to swap a short.

Indeed.


-- 
James Morris
<jmorris@namei.org>

Re: [PATCH] selinux: endian fix

[Al Viro]

On Tue, May 16, 2006 at 06:24:14PM +0200, Andreas Schwab wrote:
> James Morris <jmorris@namei.org> writes:
> 
> > On Tue, 16 May 2006, Alexey Dobriyan wrote:
> >
> >> Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
> >
> > Hmm, I'm certain this was tested (perhaps on a BE machine, though).
> 
> ntohs and htons are identical operations.  Either you swap or you don't,
> but there is only one way to swap a short.

Indeed, but that kind of crap still deserves a fix - same result, but
use that kind of misuse is begging for bugs later on.

Re: [PATCH] selinux: endian fix

[Jan Engelhardt]

>>> Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
>>
>> Hmm, I'm certain this was tested (perhaps on a BE machine, though).
>
>ntohs and htons are identical operations.  Either you swap or you don't,
>but there is only one way to swap a short.
>
...unless PDPs start to get attractive again. :>


Jan Engelhardt
-- 

Re: [PATCH] selinux: endian fix

[Andreas Schwab]

Jan Engelhardt <jengelh@linux01.gwdg.de> writes:

>>>> Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
>>>
>>> Hmm, I'm certain this was tested (perhaps on a BE machine, though).
>>
>>ntohs and htons are identical operations.  Either you swap or you don't,
>>but there is only one way to swap a short.
>>
> ...unless PDPs start to get attractive again. :>

No.  There are only two ways to order two bytes.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: [PATCH] selinux: endian fix

[Peter Chubb]

>>>>> "James" == James Morris <jmorris@namei.org> writes:

James> On Tue, 16 May 2006, Alexey Dobriyan wrote:
>> Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>

James> Hmm, I'm certain this was tested (perhaps on a BE machine,
James> though). In any case skb->protocol should definitely be network
James> byte order.

On all architectures, ntohs is the same as htons -- either the two
bytes are swapped or they're not.


-- 
Dr Peter Chubb  http://www.gelato.unsw.edu.au  peterc AT gelato.unsw.edu.au
http://www.ertos.nicta.com.au           ERTOS within National ICT Australia