TCP loopback socket fusing

[Vadim Goncharov]

Hi Andre Oppermann! 

On Wed, 15 Sep 2010 17:48:07 +0200; Andre Oppermann wrote about 'Re: TCP loopback socket fusing':

>> 3 If properly doing this for TCP, we should probably also do it for
>> other protocols.
> UNIX domain sockets already do this.  This implementation is particular
> for TCP and only touches the protocol specific parts.  It's not done at
> the socket layer.  For UDP it's not that easy to do as most UDP connections
> are one-off packets and no permanent binding between two sockets exists.
> For SCTP I don't know.  From glancing over the code it seems they have,
> at least partially, their own socket buffer code.  How difficult a fused
> socket there would be I can't say.

This hack is for TCP only, if an application author chooses a non-TCP protocol
for loopback, then he is already aware of Unix domain sockets and have reasons
for protocol he have chosen to be not PF_LOCAL. And hacking those protocols
adds another piece of complexity to maintenance. In fact, care should be taken
to consider non-fusing way of operation as primary because packetization is
useful not just for filtering SYNs.

BTW, I have heard once man needing something like tcpdump for Unix socket. Is
something convenient like this is possible?..

-- 
WBR, Vadim Goncharov. ICQ#166852181       mailto:vadim_nuclight at mail.ru
[Moderator of RU.ANTI-ECOLOGY][FreeBSD][http://antigreen.org][LJ:/nuclight]