rhealey@umn-d-ub.D.UMN.EDU (Rob Healey) (10/20/89)
Me sez: >|> Anybody know if the SysVR3 port for the UNIX PC that was >|> "let out of the bag" at a recent USENIX BOF, is getting anywhere? >|> The rumor had it that some company was actually planning the port >|> and was starting it RSN. Anybody know if it's actually happening? >|> > Lenny sez: >At this current time, I believe were at a stand-still. To make the >SVR3 port worthwhile, one would need to keep compatibility with the rest >of the UNIX pc-specific operating system features. (ie. loadable device >drivers, the device drivers themselves, etc...) > Hmmm, sounds like the reasoning behind the stratagy of a company named Intel... Appologys to Lenny, but do we lose the possiblity of a SysVR3 UNIX because it can't support some of the fringe areas of the UNIX PC? I could live with having to mkboot/mkunix a 'la 3b2/xxx if that meant I could have Rev 3 or higher UNIX. Opinions of my fellow unix-pc'ers???? If I can add drivers by relinking and rebooting I WOULDN'T complain, how about the rest of you? What's the net.opinion here? If some company came out with SysVR3 for the UNIX PC but it DIDN'T have loadable device drivers and shared librarys a 'la 3.5x would you refuse to buy it? Lets have a survey!!! Assume you could relink the kernel a 'la 3b2/xxx and that shared librarys worked like those under 3b2/xxx's, would you still refuse to buy it? I'm well aware that there would be NASTY problems porting the UNIX PC drivers from dynamically loadable ones to statically linked ones but, the rest of the UNIX world seems to handle the restriction OK. By the way, how about making a deal for CTIX 6.2 or greater? I know some of my non-shared UNIX PC programs worked fine on a Convergent/Unisys S/80 running CTIX 6.2. Any chance CTIX could be hammered into a UNIX PC kicking and screaming all the way?? As a final thought, to get SysVR3 off the ground we UNIX PC folks gotta put some $$$ where are dreams are. I'd be willing to fork out $600.00 - $1000.00 for a full development system on the UNIX PC that was rev 3.2 or higher, what about the rest of you? What would YOU pay? How can we get the thousands of unix-pc'ers needed to make this happen? There MUST be enough of us out there to finance a project like this. Assuming there are only 1000 of us at $600.00 a head, thats 600,000 dollars. Surely $600,000 plus some contributions from the crowd could get SysVR3.2 on our 7300/3b1's!? Just some rabble rousing ideas to get unix-pc.general buzzing again... -Rob p.s. Praise to Lenny and Tom for rustlin' up those $50.00 a shot 3.51 OS's! I'm Mooooooooocho grateful guys! What patch disk sets are needed for it, just a?
lenny@icus.islp.ny.us (Lenny Tropiano) (10/21/89)
In a previous article, I said ... |>At this current time, I believe were at a stand-still. To make the |>SVR3 port worthwhile, one would need to keep compatibility with the rest |>of the UNIX pc-specific operating system features. (ie. loadable device |>drivers, the device drivers themselves, etc...) |> In article <2928@umn-d-ub.D.UMN.EDU> rhealey@ub.d.umn.edu (Rob Healey) writes: |> Hmmm, sounds like the reasoning behind the stratagy of a company |> named Intel... Appologys to Lenny, but do we lose the |> possiblity of a SysVR3 UNIX because it can't support some of |> the fringe areas of the UNIX PC? I could live with having to |> mkboot/mkunix a 'la 3b2/xxx if that meant I could have Rev 3 |> or higher UNIX. Opinions of my fellow unix-pc'ers???? If I |> can add drivers by relinking and rebooting I WOULDN'T complain, |> how about the rest of you? What's the net.opinion here? |> ... Ok, want some reasoning!! I'll give you some :-) 1. A lot of us have invested a lot of money in hardware for this machine, namely Voice Power boards, StarLAN NAU boards, DOS-73 co-processors, Tape controllers, EIA/RAM combo boards, etc. Now think about it. We'll never see the source to products like these (without paying the piper... and in a lot of cases the sources weren't even written by AT&T ... eg. Alloy/DOS-73) Should I just throw away my boards because I can't use the loadable device driver that came with it? (Note your idea about doing mkboot/mkunix is great (and nothing says it wouldn't be part of the UNIX port -- if there was one), but without object (.o) files to link to the kernel, it's pretty much useless for these things ...) 2. Writing hardware specific device drivers (... now that sounds silly, all device drivers are hardware specific ...) for this machine will be very hard. Who's going to tell us the information needed to control the on-board-modem. Shall we just forget about that? What about the WE838A Dialer/Network chip, that is slated as _AT&T Proprietary_, and no one can get specs on... how are we going to control that hardware? 3. There are some *VERY* big benefits to loadable device drivers. Relinking and Rebooting a machine for device driver development is the pits. Trust me, I've done it enough on my 6386E WGS in the office! Nothing like killing 5-10 minutes a shot. And if the device driver fails, God help you ... start the relink/reboot cycle all over again! With loadable devices, it can be bound and unbound on a running kernel! 4. Keeping existing compatibility with UNIX pc binaries ... yes, there are lots of Commercial Software packages I'd just love to throw away because I can't use them on my machine anymore... :-) (ie. Paint Power, Informix, GSS*Chart, Smart, LPI-F77, etc...) Most of these are compiled with the UNIX pc shared library. So we'd need to keep some sort of compatibility there, plus give the ability to make new shared libraries (mkshlib)... 5. A lot of programs depend on the window driver, what I'd like to see here is to write some sort of "X" based window manager, and have compatibility libraries/compatibility modes for the existing UNIX pc applications. Even AT&T kept compatibility for TAM applications in their 6386WGS SVR3.2 port (not that I'm any fan of TAM). But it does keep the transition and migration period to the SVR3.2 UNIX much easier for UNIX pc TAM applications.. 6. And one last thing ... do you feel like writing a hard disk device driver for this machine? I don't... especially since one was already written, and it's working (at least the last time I checked ;-)). What I would like to do is enhance on it, add the hooks for additional drives (for John Milton's efforts ...) and make more things "tunable parameters", much like the SVR3.2 is... (ie. NMOUNT ...) |> If some company came out with SysVR3 for the UNIX PC but it |> DIDN'T have loadable device drivers and shared librarys a 'la |> 3.5x would you refuse to buy it? Lets have a survey!!! Assume |> you could relink the kernel a 'la 3b2/xxx and that shared ... I wouldn't buy it, for all reasons above. That's why it wouldn't be worth it for a company to start something like this. If someone wanted "true" SVR3 or higher compatibility ... and scrap all the nice things about this machine (and I do consider loadable device drivers a *feature*) and scrap all the software for the machine, and scrap all the internal hardware (on-board-modem, dialer), and scrap all the extra external peripherals, when why would I want *this* particular machine. I could EASILY go sell my two 3B1 machines, the entire thing, two drives [134MB] (one machine), one on the other, Tape drive, DOS-73, StarLAN, two EIA/RAM combo boards, one with 1.0 extra RAM, and the other fully populated with 1.5), expansion box, two Voice power boards, 3.5" floppy, all my documentation (hardware reference manual, etc..), and software (Phhhhheeeew.... that's a mouth-full) for enough money to buy a UNIX box that runs SVR3. What I would like to see most of all are all the bugs worked out of this kernel, get something decent. I think the machine is great (if you haven't guessed already), and it has a lot of nice features -- its a nice working UNIX environment! |> I'm well aware that there would be NASTY problems porting the UNIX PC |> drivers from dynamically loadable ones to statically linked ones |> but, the rest of the UNIX world seems to handle the restriction |> OK. ... Without source? I doubt it... I'm not that good at reverse engineering! ... |> needed to make this happen? There MUST be enough of us out there |> to finance a project like this. Assuming there are only 1000 of |> us at $600.00 a head, thats 600,000 dollars. Surely $600,000 plus |> some contributions from the crowd could get SysVR3.2 on our |> 7300/3b1's!? |> Yes, I've thought of that. There are big *legal* and logistical problems with a group purchase of the source. I guess if someone wanted to risk $600 to a project that might not get completed (look at the IDT SCSI project that was scraped -- probably because of lack of information about the machine...), then by all means, make your checks out to "Lenny Tropiano UNIX PC Donation fund" :-) :-) |> Praise to Lenny and Tom for rustlin' up those $50.00 a shot |> 3.51 OS's! I'm Mooooooooocho grateful guys! What patch disk |> sets are needed for it, just a? ... You're welcome... Who knows, we might have some more deals down the road. Maybe Tom will find a warehouse of 3B1's or PC7300 stashed away, some Ethernet boards, or some Development sets ... I'll keep you posted! -Lenny -- | Lenny Tropiano ICUS Software Systems [w] +1 (516) 589-7930 | | lenny@icus.islp.ny.us Telex; 154232428 ICUS [h] +1 (516) 968-8576 | | {ames,pacbell,decuac,hombre,sbcs,attctc}!icus!lenny attmail!icus!lenny | +------- ICUS Software Systems -- PO Box 1; Islip Terrace, NY 11752 -------+
gil@limbic.UUCP (Gil Kloepfer Jr.) (10/22/89)
In article <990@icus.islp.ny.us> lenny@icus.islp.ny.us (Lenny Tropiano) writes: >In article <2928@umn-d-ub.D.UMN.EDU> rhealey@ub.d.umn.edu (Rob Healey) writes: >|> named Intel... Appologys to Lenny, but do we lose the >|> possiblity of a SysVR3 UNIX because it can't support some of >|> the fringe areas of the UNIX PC? I could live with having to >Ok, want some reasoning!! I'll give you some :-) I'll add some to Lenny's article and try to calm the small flames I'm starting to see eminating from this conversation (maybe I'll make some new ones ;-). I'll start by saying that this was all beaten to death at the UNIX-pc BOF at Usenix, but there is one serious consideration that we can't get around and that's the 4MB address space limitation in the memory management hardware. I'm sure that the motherboard could be hacked up to make it work, but I can imagine it would be a royal pain. After pondering with many people in many circumstances over the issues of existing software, memory management hardware limitations, device driver rewrite issues, and loss of some "neat" features in our kernel, it is *my* opinion that we'd be better off cleaning up the UNIX-pc kernel and maybe making it more complient with SVR3 than to attempt a port of SVR3 to the UNIX-pc. Again, with stress on the memory management problems, as one eats up more space with a kernel, one has less space to work in. Assuming the port could be done and drivers and software working in some kind of compatibility mode, we would still be left with a problem of a kernel of a larger size than the current one, almost guaranteed. I certainly don't say that those interested in performing a port should not do so, and I think constructive discussion about a port would be good. I am saying that for all that is gained by porting SVR3 to the UNIX-pc, you may end up losing some more important resources of the machine. Further, if you _really_ need a machine that can support SVR3, buy a 386 (choke, gag ;-). Lastly, and I think a discussion that also has merit (although there are those who would disagree) is a port of some version of BSD to the UNIX-pc. This was also discussed at the BOF, and it seemed to have more of a following than a SVR3 port. Again, followers take note-- make sure that the size of the kernel is less than or equal to the size of the current one or you'll end up with applications that don't run. ------ | Gil Kloepfer, Jr. | ICUS Software Systems/Bowne Management Systems (depending on where I am) | ...ames!limbic!gil
bdb@becker.UUCP (Bruce Becker) (10/24/89)
In article <573@limbic.UUCP> gil@limbic.UUCP (Gil Kloepfer Jr.) writes: |[...] |Lastly, and I think a discussion that also has merit (although there |are those who would disagree) is a port of some version of BSD to the |UNIX-pc. This was also discussed at the BOF, and it seemed to have |more of a following than a SVR3 port. Again, followers take note-- |make sure that the size of the kernel is less than or equal to the |size of the current one or you'll end up with applications that don't |run. I agree, but in my mumble opinion it ought to be the Mach kernel with the 4.3BSD front end... Cheers, -- .... Bruce Becker Toronto, Ont. w \**/ Internet: bdb@becker.UUCP, bruce@gpu.utcs.toronto.edu `/C/-e BitNet: BECKER@HUMBER.BITNET _/ >_ "But... but.. reality isn't *real*..." - Pippy the Zkinhead