netdev - Re: [RFC] updated Patch to option HSO driver to the kernel

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:	Tue, 22 Apr 2008 01:04:11 +0300
From:	Paulius Zaleckas <paulius.zaleckas@...tonika.lt>
To:	netdev@...r.kernel.org
Cc:	netdev@...r.kernel.org
Subject:  Re: [RFC] updated Patch to option HSO driver to the kernel

David Brownell wrote:
>> From: Paulius Zaleckas <paulius.zaleckas@...tonika.lt>
>> Date: Mon, 21 Apr 2008 12:34:39 +0300
>>
>> IMO it doesn't make sense to do multicasting since actually it is
>> point-to-point and not ethernet traffic.
>> Currently I am studying if it is possible to convert current pseudo
>> ethernet interface to point-to-point one.
> 
> For the record:  when I looked at that issue with respect to USB
> network connections for handheld devices, one issue seemed to be
> weak support for managing (lots of) point-to-point links.
> 
> The example that sticks in my memory is that of routing.  Setting
> up a point to point link between a handheld and a PC required
> the PC to act as an IP router.  Managing lots of desktop routers
> in any large configuration seemed to be a losing game ... and
> one that many network administrators would refuse to play.  And
> that's in addition to the lack of a dynamic IP address assignment
> solution for lots of point-to-point links.
> 
> The alternative was to set up the links like Ethernet, then bridge
> them.  That's a *much* more manageable solution, even though I still
> don't know of any distro that makes bridge setup work as easily as
> Windows does (sigh).  Bridging properly requires the ability to handle
> multicast and broadcast packets ... and once you've got that then
> DHCP, zeroconf, and other automatic network configuration schemes
> work easily.
> 
> QED ... that's why the "usbnet" framework doesn't try to use the
> point-to-point framework, instead allows N-casting.
> 
> - Dave

I would agree with you, but... To make this interface work you have to
send AT command with APN and optionally with PAP/CHAP username and password.
Then you call AT command witch initiates PPP with basestation and begins
to send/receive traffic on this pseudo ethernet interface... Actually it
still needs one more AT command to get your IP and DNS addresses.
Then you do: ifconfig hso0 xxx.xxx.xxx.xxx netmask 255.255.255.255 up
And then you have to write DNS addresses (got through AT interface) to
the /etc/resolv.conf. Now you have a working internet connection!

So... I don't see any chance to do multicasting or broadcasting here...
IMO the only way to distribute this kind of connection to local network
is to do NAT'ing :( Am I wrong?

P.S. Pseudo ethernet means that it receives bare IP packet from USB device,
adds dummy ethernet header and sends it for further processing to kernel.

Paulius

--
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