lisbon@vpnet.UUCP (Gerry Swetsky) (11/05/89)
> EE TIMES, 30 Oct 1989, page 1 indicates that the generic port > of UNIX System V.4/386 and /486 will be available soon... > AT&T/Intel reportedly plan to demo it at UNIX EXPO. If it was there, I didn't see it. > BTW, > System V Release 4 is upwards compatible with V.3.2 and also adds > the following (again according to EE Times): You left one out - Standard shell is to be ksh! HOORAY! I wouldn't look for this product anytime soon. Interactive plans a new release though - Release 2.2. This WON'T be ATT V.4, but an interim release. I couldn't get anyone to tell me the reason for the release or what all it will contain. Interactive's V.4 package will be labelled 486/ix unless they change their minds. NOTE - the 486 is to refer to release V.4 and not the 486 chip! I look for them to change their minds regarding this confusing labelling! -- ============================================================================= | Help stamp out stupid .signature files! Gerry Swetsky | | | | Home (312)833-8122 Vpnet (312)833-8126 lisbon@vpnet.uucp | =============================================================================
evan@telly.on.ca (Evan Leibovitch) (11/11/89)
In article <[2553c481:416.2]comp.unix.i386;1@vpnet.UUCP> lisbon@vpnet.UUCP (Gerry Swetsky) writes: > >> EE TIMES, 30 Oct 1989, page 1 indicates that the generic port >> of UNIX System V.4/386 and /486 will be available soon... >> AT&T/Intel reportedly plan to demo it at UNIX EXPO. > > If it was there, I didn't see it. Were you there? When the lights were on? :-) It was being shown at the booths of - AT&T - Interactive - Intel - Unix International Also, though SCO and ESIX were not displaying it, staff at both companies were talking about it. At the UI booth, at least six different companies were showing off SysVr4, including Olivetti on a 486 system (with a co-processor socket for an i860!) > I wouldn't look for this product anytime soon. Intel (which after buying Bell Technologies plans to get into selling Unix itself) has announced that V.4 shipments will start in 2nd quarter 1990. Interactive and AT&T said theirs wouldn't be far behind. ESIX said 3rd quarter 1990, tentative. SCO's still trying to get out 3.2 :-) > Interactive plans a > new release though - Release 2.2. This WON'T be ATT V.4, but an > interim release. I couldn't get anyone to tell me the reason for > the release or what all it will contain. Bug fixes? ;-) > Interactive's V.4 package > will be labelled 486/ix unless they change their minds. NOTE - the > 486 is to refer to release V.4 and not the 486 chip! I was at the Intel announcements. They said that there are no plans for any software especially for the 486. Everything for V.4 will be done in 386 code and will treat the 486 as just a fast 386. > I look for them to change their minds regarding this confusing > labelling! Don't count on it. If SCO could call its Xenix product System V, anything is fair game... -- Evan Leibovitch, Telly Computing, located in beautiful Brampton, Ontario evan@telly.on.ca / uunet!attcan!telly!evan If you're smart enough to be a programmer, you're too smart to be a programmer
williams@cs.umass.edu (11/11/89)
In article <1989Nov10.180048.24410@telly.on.ca>, evan@telly.on.ca (Evan Leibovitch) writes... >I was at the Intel announcements. They said that there are no plans for >any software especially for the 486. Everything for V.4 will be done in >386 code and will treat the 486 as just a fast 386. > Huh? I thought the 486 WAS just a fast 386 with builtin cache and fpu! No new instructions, modes, etc.
stevens@hsi.UUCP (Richard Stevens) (11/11/89)
Anyone know who will ship the first binary release of SVR4
for any type of hardware ? AT&T for the 3B2 ? AT&T for their
WGS 386 ? I can understand that all the third party suppliers
have to integrate it with their drivers, etc., but if someone
(UI ?) is making such a hoopla about releasing the source code
(for a nontrivial $100K) I sure hope it has been thoroughly
tested and QC'ed on *something*.
If I, as an end-user, really can't get my hands on it for
6-12 months, then all the fanfare about its release last
week is really a nonevent.
Richard Stevens
Health Systems International, New Haven, CT
stevens@hsi.com
... { uunet | yale } ! hsi ! stevensmarc@dumbcat.UUCP (Marco S Hyman) (11/12/89)
In article <[2553c481:416.2]comp.unix.i386;1@vpnet.UUCP> lisbon@vpnet.UUCP (Gerry Swetsky) writes:
Interactive plans a
new release though - Release 2.2. This WON'T be ATT V.4, but an
interim release. I couldn't get anyone to tell me the reason for
the release or what all it will contain.
I'd guess that that would be the POSIX compliant version. The POSIX
compliant version is supposed to include job control, multiple groups, and
enhanced signals. This was supposed to be available 31august. Maybe it was
and I didn't notice. I'd sure like job control tho...
--marc
--
// Marco S. Hyman {ames,pyramid,sun}!pacbell!dumbcat!marcmadd@world.std.com (jim frost) (11/15/89)
williams@cs.umass.edu writes: >In article <1989Nov10.180048.24410@telly.on.ca>, evan@telly.on.ca (Evan Leibovitch) writes... >>Everything for V.4 will be done in >>386 code and will treat the 486 as just a fast 386. >> >Huh? I thought the 486 WAS just a fast 386 with builtin >cache and fpu! No new instructions, modes, etc. At a minimum, the fpu supports a bunch of functions which the 80387 did not. jim frost software tool & die madd@std.com
rcd@ico.isc.com (Dick Dunn) (11/15/89)
In article <824@hsi86.hsi.UUCP>, stevens@hsi.UUCP (Richard Stevens) writes: > Anyone know who will ship the first binary release of SVR4 > for any type of hardware ? AT&T for the 3B2 ? AT&T for their > WGS 386 ? I can understand that all the third party suppliers > have to integrate it with their drivers, etc.,... >...If I, as an end-user, really can't get my hands on it for > 6-12 months, then all the fanfare about its release last > week is really a nonevent. A demo at a trade show is usually NOT the event marking the introduction of the product for real end-user sales. Now, I flame as much as anyone about vaporware and non-announcements (tho I usually do it over in comp.arch:-), but I contend that as long as the status of a demo is reasonably clear, it serves a useful purpose. If you see a V.4 system up and running on real hardware, you have some reason to think that when vendors say "we'll have it available in nQ90" you might believe them. There IS a lot of work to be done to make a system which runs into a system which can be shipped as a product. However, knowing that the system really exists in some halfway-functional form still means something. Perhaps it's easier to see this with hardware. Consider, for example, that MIPS showed a very fast ECL machine at UNIX Expo. It's not shipping yet; you can't even order one yet. BUT they're showing you that all the talk you've heard about ECL RISCs is real; they've got the caches figured out; they've got the OS running, etc. The "inventing" (the part you can't schedule:-) is done; the technology exists. There may still be a lot of work to be done to make the product, but it can be done. Besides, there's a certain parity here. If people show things at the point they reach "demo quality", and you see vendor (or consortium:-) X showing something, but Y not showing anything comparable, you tend to suspect that X is ahead of Y. If X and Y are both promising something for mid-'90, and X has a demo but Y doesn't, you might have epsilon more faith that X will deliver (IF it's a believable demo). -- Dick Dunn rcd@ico.isc.com uucp: {ncar,nbires}!ico!rcd (303)449-2870 ...Keep your day job 'til your night job pays.
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (11/16/89)
In article <824@hsi86.hsi.UUCP>, stevens@hsi.UUCP (Richard Stevens) writes: | Anyone know who will ship the first binary release of SVR4 | for any type of hardware ? AT&T for the 3B2 ? My understandaing is that you can get it for the 3B2, but I haven't ordered one. You may be able to get it for Sun, as well. INteractive was demoing 486/ix (V.4 they said) at UNIX Expo, but they said 1Q90 for the ship date. Looked pretty good to me, but they didn't let me sit down and TRY to break it. | If I, as an end-user, really can't get my hands on it for | 6-12 months, then all the fanfare about its release last | week is really a nonevent. I expect to see beta and "early release" versions by the end of the year. Bear in mind there's no huge demand in the 386 market. People will let someone else buy the first one, wait for reviews, etc. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon
madd@world.std.com (jim frost) (11/16/89)
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) writes: >In article <824@hsi86.hsi.UUCP>, stevens@hsi.UUCP (Richard Stevens) writes: >| Anyone know who will ship the first binary release of SVR4 >| for any type of hardware ? >Bear in mind there's no huge demand in the 386 market. That depends on who you ask. I personally would kill for job-control and the BSD filesystem on a 386. Considering the number of times I've seen similar requests on comp.unix.questions et al, I'd say there's a lot of demand. Mach would be better, though :-). jim frost software tool & die "The World" Public Access Unix for the '90s madd@std.com +1 617-739-WRLD 24hrs {3,12,24}00bps
dyer@spdcc.COM (Steve Dyer) (11/16/89)
In article <1989Nov15.223222.11299@world.std.com> madd@world.std.com (jim frost) writes: >I personally would kill for job-control and the BSD filesystem on a 386. You can get it now for IBM PS/2 386 machines by buying AIX PS/2. Strictly speaking, the filesystem is not BSD, but it has what you need: long filenames and symbolic links. Job control is there too. >Mach would be better, though :-). Mach's already been ported to the 386 by the folks at CMU. -- Steve Dyer dyer@ursa-major.spdcc.com aka {ima,harvard,rayssd,linus,m2c}!spdcc!dyer dyer@arktouros.mit.edu, dyer@hstbme.mit.edu
plocher@sally.Sun.COM (John Plocher) (11/16/89)
+-- In <824@hsi86.hsi.UUCP> stevens@hsi.UUCP (Richard Stevens) writes | Anyone know who will ship the first binary release of SVR4 | for any type of hardware ? +---- As I have stated here before, my job at Sun is with a team that is responsible for porting Sun's X11/NeWS window server to AT&T System Vr4.0 on an AT&T WGS6386. Our development environments include 16 Mhz AT&T boxes, 16Mhz Intel boxes, a 25Mhz Compaq, and my clone 25Mhz Abacus motherboard based system. We get software releases at the same time that AT&T Internal Development gets them, in a manner similar to a "monthly snapshot" of the master development tree. The fact that we are "internal development" keeps me from saying too much more about things. :-( We are still watching all the pieces come together: Job control is wonderful, so are symbolic links. The compiler is ANSI :-) and has good transition settings for non-ANSI code. However, there are still signs that things aren't quite done. Installation scripts are fragile, 3rd party device support is minimal, .... All that aside, we all know that AT&T now resells Intel 386 boxes, right? We also know that there isn't too much about the Intel boxes that is different from the average run-of-the-mill clone box, right? Therefore it should be safe to say that the AT&T version of Vr4.0 should run on any 386 or 486 clone. It also is a good rule of thumb that it takes several months to do a final Beta/QA/QC/production cycle. Intel has said it will have a release by Q2-90; that sounds to me like the earliest it could be. Just work backwards: May Release (Q2-90) April 1 Month Production Jan-March 3 Months Beta/QA/QC/driver development December Final source shipment to Intel (maybe) Don't forget that your favorite 3rd party device driver developer won't get their own copy of Vr4 till May, so you may have to wait even longer to take advantage of all your nifty hardware .... -John Plocher
hedrick@geneva.rutgers.edu (Charles Hedrick) (11/16/89)
>That depends on who you ask. I personally would kill for job-control >and the BSD filesystem on a 386. It should not be necessary to kill. I have job control on SVr2 on my 286 at home. Ksh has code in it that does a pretty good emulation of Berkeley job control using the System V sxt device. The necessary kernel support should be present on any System V since release 2. However you may have to rebuild your kernel to include support for the sxt device. (At least with Microport, it's optional, so you have to get the linkkit and build a kernel.) If shl will run on your system, this will. The problem is getting you the appropriate ksh. Unfortunately, Korn (author of ksh) considers System V sufficiently brain damaged that he was unwilling to take back the fixes that make job control work on it. (This seems a bad attitude for an ATT employee.) So as far as I know, I've got the only copy of ksh where job control actually works. If your site has a ksh source license, I can just send you diffs. If you have or can get a binary license, I can probably give you my binary. (I'd have to check the license to make sure -- most ATT licenses allow people to give code to other sites with licenses having similar scope.) Microport SV/AT includes ksh binaries. I think their 386 product does as well. So your 386 might already have it. If you can show that your system is licensed for ksh, I think it would be legal for me to give you a copy of the version with job control. As far as I know, SV/AT executables work on all of the 386 SV's except AIX (and AIX has real job control).
pjh@mccc.uucp (Pete Holsberg) (11/17/89)
How can I get it for the 3B? My salesbeing doesn't seem to know how.
--
Pete Holsberg UUCP: {...!rutgers!}princeton!mccc!pjh
Mercer College CompuServe: 70240,334
1200 Old Trenton Road GEnie: PJHOLSBERG
Trenton, NJ 08690 Voice: 1-609-586-4800davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (11/22/89)
New information: Intel claims that they will be shipping SVR4 in February for their 386. Since that's the AT&T machine, and quite compatible with most other 386s I have hopes that we will be able to get it at that time. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) "The world is filled with fools. They blindly follow their so-called 'reason' in the face of the church and common sense. Any fool can see that the world is flat!" - anon