Path: utzoo!utgpu!news-server.csri.toronto.edu!clyde.concordia.ca!uunet!overload!dillon From: dillon@overload.UUCP (Matthew Dillon) Newsgroups: comp.sys.amiga.tech Subject: Re: UUCP 1.06 and the A3000 Message-ID: Date: 10 Jul 90 03:11:47 GMT References: <1414@nyx.UUCP> Lines: 31 >In article danny@whitwiz.lonestar.org (Danny White) writes: >>In article <1414@nyx.UUCP> bscott@nyx.UUCP (Ben Scott) writes: >> ... There are indeed problems running uucico on the A3000 (UUCP 1.06D). I am attempting to find the bug (several of you are helping me trace it, thanks!). Not having an A3000 (yet) makes it difficult, though. There are apparently two bugs. One Darrin Hyrup has traced to somewhere around the close-serial-device/open-serial-device calls used to terminate the call at the end. Another bug apparently occurs just after receiving a file. I suspect both bugs may be a manifestation of a memory mash somewhere. In anycase, those of you with A3000's should avoid using UUCP 1.06D until I can find the problem! Wish somebody had a debugger on his A3000 with UUCP 1.06D setup, the uucico executable does have symbols in it. The bugs *might* be due to a bug in DICE, but there is as much evidence for as against that so at the moment it is pure conjecture. As far as the CHK instruction goes, DICE never generates it. I've verified this by using Lattice's OMD on all the uucico object files. -Matt -- Matthew Dillon dillon@Overload.Berkeley.CA.US 891 Regal Rd. uunet.uu.net!overload!dillon Berkeley, Ca. 94708 USA