From owner-freebsd-hackers Sat Jun 22 12:21:31 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id MAA03845 for hackers-outgoing; Sat, 22 Jun 1996 12:21:31 -0700 (PDT) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id MAA03834 for ; Sat, 22 Jun 1996 12:21:28 -0700 (PDT) Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.12/8.6.9) id FAA09329; Sun, 23 Jun 1996 05:12:54 +1000 Date: Sun, 23 Jun 1996 05:12:54 +1000 From: Bruce Evans Message-Id: <199606221912.FAA09329@godzilla.zeta.org.au> To: FreeBSD-hackers@freebsd.org, revr@cadre.nl, wilko@yedi.iaf.nl Subject: Re: HELP: serial port grief on Asus P55TP4N Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >I'm sending this to -hackers on behalf of a friend of mine who has >just converted to FreeBSD 2.1R. His system is a P100 on a Asus >P55TP4N with 32Mb, SC200 and ATI Mach64 VGA. >He has all sorts of problems when using the internal serial ports >of the Asus in combination with UUCP. It's really weird... :-( >The uucp config files are identical to mine (I have a P55TP4XE). >Symptoms: >- when X was running -> console messages 'tty level buffer overflow' >- without X -> errors logged by uucp 'uucp checksum failed' >- he also tried ijppp and that seems to work OK... (???) Uucp checksum errors are known to be caused by bugs in the UMC i/o chip which is found in some ASUS motherboards (e.g. the P55TP4XE rev.2.4). The UARTs lose sync when their speed is set while data is arriving, and uucp sets the speed a lot as a side affect of doing tcsetattr()'s (it doesn't change the speed but it I think it toggles state bit(s)). This can be worked around in most cases by changing the driver to skip setting the speed if the speed hasn't changed. Bruce