Date: Fri, 17 Nov 1995 10:35:48 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: didier@omnix.fr.org (Didier Derny) Cc: terry@lambert.org, joerg_wunsch@uriah.heep.sax.de, hackers@FreeBSD.ORG Subject: Re: pty Message-ID: <199511171735.KAA05728@phaeton.artisoft.com> In-Reply-To: <Pine.BSF.3.91.951117101009.3126A-100000@zapata.omnix.fr.org> from "Didier Derny" at Nov 17, 95 10:17:49 am
next in thread | previous in thread | raw e-mail | index | archive | help
> the data would be send and received on the pty side as small blocks. > each block would be delimited with STX an ETX > > example: > > STX CODE CHAN DATA ETX > > CODE: indicates the nature of the block > start a new connection, data, or disconnect > > CHAN: ttyp number > > It's a protocol wildely use in france for minitel servers. > (ASM protocol) HEY! I've written Minitel code before, for a contract with your ministry of defense! Big building, looks like an upside-down squared-off "U"! 8-). > either I write a program to control 32 pty and feed them with the data > from the ASM box or I control only one special pty an the driver do the > job... I think that the expensive part of the pty is going to be there whether or not you put in the endpoint. You may save on fd's doing this, but I think overall the increased code complexity in the muxed pty driver will steal back whatever savings you get that way. Something like this will probably be a false economy, unless maybe you will put *all* of your code in the kernel. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199511171735.KAA05728>