Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!watmath!clyde!burl!ulysses!allegra!bellcore!decvax!genrad!mit-eddie!godot!ima!johnl From: johnl@ima.UUCP Newsgroups: net.unix Subject: Re: Re: IBM to support UNIX on 4300 Message-ID: <493@ima.UUCP> Date: Sun, 24-Feb-85 23:45:43 EST Article-I.D.: ima.493 Posted: Sun Feb 24 23:45:43 1985 Date-Received: Thu, 28-Feb-85 01:21:47 EST Lines: 29 Nf-ID: #R:lsuc:-42400:ima:19700006:000:1569 Nf-From: ima!johnl Feb 24 23:06:00 1985 Actually, IX/370 is a descendant of AT&T's UNIX/370, which was built on top of the lowest level part of TSS/370, a minor modification of TSS/360, which was IBM's greatest software disaster ever. Is that all clear? TSS/360 was IBM's first attempt to build a time sharing system, on the 360/67 in the late 60's. I used it -- it was a reasonable idea but due to the Mongol Horde programming approach current at the time, it was always much too slow and much, much, too buggy, and IBM eventually withdrew it. A contemporaneous IBM research effort was CP/67 which evolved into VM/370. Except for some reason, the Bell Labs people at Indian Hill liked TSS and continued to use it, porting it along to newer 370 hardware as it came along. When they decided to port Unix to the 370 architecture, they decided to build it on top of the TSS kernel which provided virtual memory and I/O management. Up until this point TSS and UNIX/370 were standalone systems that ran directly on the 370 hardware. The point of VM is that you can run standalone systems under it, so it was no trick to bring up TSS-UNIX/370 under VM. Since few people want to run only Unix on a 370, there being existing applications that run under other quaint systems with names like MVS, I expect that the TSS-UNIX was tuned and modified to work well under VM, and that's what turned into IX/370. John Levine, ima!johnl [The above is the best of my understanding, but no warranties are expressed or implied. Others may disagree. Your milage may vary.] * - This is not a trademark acknowlegdement.