Return to BSD News archive
Xref: sserve comp.os.386bsd.questions:8665 comp.windows.x.i386unix:7197 Path: sserve!newshost.anu.edu.au!munnari.oz.au!foxhound.dsto.gov.au!fang.dsto.gov.au!yoyo.aarnet.edu.au!news.adelaide.edu.au!basser.cs.su.oz.au!news.cs.su.oz.au!metro!ultima!kralizec.zeta.org.au!godzilla.zeta.org.au!not-for-mail From: bde@kralizec.zeta.org.au (Bruce Evans) Newsgroups: comp.os.386bsd.questions,comp.windows.x.i386unix Subject: Re: Problems with NetBSD-0.9/XFree86-2.0 Date: 10 Feb 1994 01:11:33 +1100 Organization: Kralizec Dialup Unix Sydney - +61-2-837-1183, v.32bis and v.42bis Lines: 35 Message-ID: <2jaqulINN4q1@godzilla.zeta.org.au> References: <2j3r5t$34t@newsserv.cs.sunysb.edu> <MIKE.LONG.94Feb7173040@cthulhu.analog.com> <michaelv.760723129@ponderous.cc.iastate.edu> NNTP-Posting-Host: godzilla.zeta.org.au In article <michaelv.760723129@ponderous.cc.iastate.edu>, Michael L. VanLoon <michaelv@iastate.edu> wrote: >In <MIKE.LONG.94Feb7173040@cthulhu.analog.com> Mike.Long@analog.com (Michael W. Long) writes: > >>In article <2j3r5t$34t@newsserv.cs.sunysb.edu> rick@cs.sunysb.edu (Rick Spanbauer) writes: > >>> 1. com0 occasionally hangs kermit/tip in a state that kill -9 >>> will not remove them (process state SE+). > >>I've had this happen to me. The com driver is sensitive to XOFF (^S) >>even when it is told to use RTS/CTS flow control (SET FLOW RTS/CTS in >>kermit). ... > >I've never experienced this. Does the *serial port* (not the program) >know that it's supposed to be ignoring xon/xoff? Try doing a "stty -f >/dev/tty00 -ixon -ixoff crtscts" before entering kermit (assuming >you're using tty00). For kermit in interactive mode, add also for logins, ixon is normal so that you can used ^S^Q to stop and start scrolling, while crtscts helps to avoid going too fast for the modem, so it is normal to have both on at the same time. ixoff is for automatic xon-xoff flow control and you arely want that, especially in conjuction with crtscts. When a port is closed, output flow control of all types probably ought to be ignored (or output ought to be flushed if it is active), so that ttywait() doesn't hang "forever" waiting for the output to drain. I used to have problems with this on old Suns. My sio driver slill mishandles at least the crtscts case. The problem may be more serious if CLOCAL is set and/or SIGHUPs are being ignored. I think output is flushed anyway when carrier drops if SIGHUP kills the process, and output is discarded if CLOCAL is set and carrier is absent. -- Bruce Evans bde@kralizec.zeta.org.au