netdev - Re: SCTP and IP

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: <52094AEA.9090904@candelatech.com>
Date:	Mon, 12 Aug 2013 13:51:54 -0700
From:	Ben Greear <greearb@...delatech.com>
To:	Vlad Yasevich <vyasevich@...il.com>
CC:	netdev <netdev@...r.kernel.org>
Subject: Re: SCTP and IP_TOS

On 08/12/2013 01:21 PM, Vlad Yasevich wrote:
> On 08/12/2013 01:00 PM, Ben Greear wrote:
>> Hello!
>>
>> I notice that I read 0x2 as IP_TOS getsockopt on an SCTP socket configured
>> for streaming mode.
>>
>> Looks to me like the reason is that SCTP uses a socket type of
>> SCTP_SOCKET_TCP
>> and the ipv4/ip_sockglue.c code is only masking out ECN for SOCK_STREAM
>> types.
>>
>> Is this per design, or should the ip_sockglue.c mask out ECN for
>> SCTP_SOCKET_TCP
>> types as well?
>>
>> Thanks,
>> Ben
>>
>
>  From the ip_glue perspective, you would see either SOCK_STREAM or
> SOCK_SEQPACKET.
>
> So you are trying to set tos bits on a SEQPACKET socket, ip_sockglue()
> wouldn't try preserver current ECN bits as it does for STREAM sockets.

Actually, I am just reading the TOS, and should never be setting it
at all in this particular test case.  TCP connections return 0x0 in
this case, but SCTP returns 0x2.

It's a small issue, but it screwed up some sub-standard code in
my application, so that is why I noticed it.

It *could* be that my app is somehow to blame.  I'll poke around
in the kernel with printk when I get a chance.

Thanks,
Ben

>
> -vlad
>


-- 
Ben Greear <greearb@...delatech.com>
Candela Technologies Inc  http://www.candelatech.com

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html