From owner-freebsd-hackers Sun Aug 10 00:21:08 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id AAA07732 for hackers-outgoing; Sun, 10 Aug 1997 00:21:08 -0700 (PDT) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id AAA07724 for ; Sun, 10 Aug 1997 00:21:02 -0700 (PDT) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id JAA04718 for hackers@FreeBSD.ORG; Sun, 10 Aug 1997 09:21:01 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.7/8.8.5) id JAA00326; Sun, 10 Aug 1997 09:05:28 +0200 (MET DST) Message-ID: <19970810090528.OS10788@uriah.heep.sax.de> Date: Sun, 10 Aug 1997 09:05:28 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@FreeBSD.ORG Subject: Re: ISDN drivers/cards References: X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: ; from Tom Samplonius on Aug 9, 1997 13:15:27 -0700 Sender: owner-freebsd-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk As Tom Samplonius wrote: > > going from TA to TA or TA to Router, I'd love to see an internal card that > > doesn't use 16550's that I can put in my freebsd machine and get good > > This can be improved a lot. Most TAs support a 230400bps rate, but > FreeBSD does not. Well, that's not the first time you're spreading this misinformation around: FreeBSD _would_ support this rate (basically), but the under- lying hardware doesn't. If you've got a card where you could double the oscillator frequency, simply do it, and FreeBSD will support 230400 bps (but call it 115200 still). *shudder* (I seldom agree with dennis, but in this respect, i agree with him. Abusing an async line for feeding traffic that originally came in via a synchronuous transport is horrible.) > FreeBSD-current now detects the 16670 UART that supports 230400 (and > faster. But it doesn't seem possible to set a port to 230400. Ah, that's what you mean. So, if they support 230 kbps, they must have left the way it used to be done in a 8250-compatible UART. (The divisor 1 already yielded 115200. Are they using divisor 0 now? :-) > This is mainly > due to the extra overhead of async versus sync serial (well, sync serial > carries the "overhead" out of band). No. Sync serial (HDLC) still has the overhead in-band, but it's much less. Async serial means 10 bits per byte (plus a little more if the bytes aren't adjacent), Sync serial means slightly more than 8 bits per byte (where `slightly more' depends on the data you are sending, since it's the overhead required to escape potential Sync flagging bit sequences). > BTW, I use an 3COM Impact II extrernal ISDN TA. It works very well, > except that I'd like to drive it at 230400. At 115200 bps, I get > round-trip times of 80ms. Btw., i'm using a Teles card. :-) With one channel (i don't wanna pay for two channels anyway), i get RTTs of 35 ms. :-)) The usual data rate (FTP rate) is 7.3 KB/s, although i've already been surprised to see 8.3 KB/s once (which is impossible), and this was even across two chained ISDN lines. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-)