netdev - Re: [PATCH] Packet mmap : allow the user to choose the offset of the tx payload.

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

Date:	Fri, 05 Oct 2012 21:21:22 +0200
From:	pchavent <Paul.Chavent@...ra.fr>
To:	Daniel Borkmann <danborkmann@...earbox.net>
Cc:	<davem@...emloft.net>, <edumazet@...gle.com>,
	<xemul@...allels.com>, <herbert@...dor.hengli.com.au>,
	<netdev@...r.kernel.org>, <johann.baudy@...-log.net>,
	<uaca@...mni.uv.es>
Subject: Re: [PATCH] Packet mmap : allow the user to choose the offset of
 the tx payload.

Hi

On Fri, 5 Oct 2012 16:17:12 +0200, Daniel Borkmann wrote:
> On Fri, Oct 5, 2012 at 3:10 PM, Paul Chavent <Paul.Chavent@...ra.fr> 
> wrote:
>> The tx offset of packet mmap tx ring used to be :
>> (TPACKET2_HDRLEN - sizeof(struct sockaddr_ll))
>>
>> The problem is that depending on the usage of SOCK_DGRAM or
>> SOCK_RAW, the payload could be aligned or not.
>>
>> This patch allow to let the user give an offset for it's tx
>> payload if he desires.
>>
>> Signed-off-by: Paul Chavent <paul.chavent@...ra.fr>
>
> Can you provide an example when it doesn't hit TPACKET_ALIGNMENT?

When we use tx ring, the user have to write at (TPACKET_HDRLEN - 
sizeof(struct sockaddr_ll))

This adress is aligned on TPACKET_ALIGNMENT since
TPACKET_HDRLEN = (TPACKET_ALIGN(sizeof(struct tpacket_hdr)) + 
sizeof(struct sockaddr_ll))

When we use the tx ring with SOCK_RAW option, the mac header is aligned 
on TPACKET_ALIGNMENT, but not the payload (14 bytes away).


>
> On the first look, could it be that your patch currently enforces the
> use of your tp_net's offset for *all* TX_RING users, even if they
> don't care about it? So in case off==0, you probably get a negative
> offset in case of SOCK_RAW, thus it won't hit the second 
> if-statement.
> Sure, but this does not look intuitive in my opinion. Maybe, it's
> better to only enter this path if the offset *is* used by someone.

Yes, moreover i had a problem with the signed/unsigned comparison.
I've fixed all those problems for the next submission.


>
> Also, to my knowledge tp_net is currently only applied in receive
> path. So, if for whatever reason people did not explicitly set tp_net
> to 0, it might break their code, if there's a random offset in it,
> right?

I haven't found where all frames were initialized to 
TP_STATUS_AVAILABLE, so i have supposed that the frames were initialized 
to zero.

So your are right, if tp_net is not forced to zero before sending the 
packet it could break the legacy code :(


>
> Best,
> Daniel

Thank for your review.

Paul.
--
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