carlm@iplmail.orl.mmc.com (carl morris) (04/27/91)
The lab I work for recently inherited a VAX 8350 system. Looking through
the KA820/KA825 Processor Technical Manual that came with it, one of my
maintenance gurus came across a section (page 2-8) that references four
serial line units which are on the cpu board. There must be some reason
for putting those ports on the board ("Hey, Bob, we got this corner left
over on the cpu board..."), anyone have any ideas? Was it primarily for
maintenance diagnostic access, or are there some useful end-user type
uses for them?
Thanks for any suggestions -- Carlkarcher@toron.waisman.wisc.edu (04/27/91)
In a previous article, carlm@iplmail.orl.mmc.com (carl morris) wrote: >The lab I work for recently inherited a VAX 8350 system. Looking through >the KA820/KA825 Processor Technical Manual that came with it, one of my >maintenance gurus came across a section (page 2-8) that references four >serial line units which are on the cpu board. There must be some reason >for putting those ports on the board ("Hey, Bob, we got this corner left >over on the cpu board..."), anyone have any ideas? Was it primarily for >maintenance diagnostic access, or are there some useful end-user type >uses for them? These are real live serial ports you can do anything you want with. The following VMS SYSGEN commands will make them appear. SYSGEN> CONNECT SLU=1 SYSGEN> CONNECT SLU=2 SYSGEN> CONNECT SLU=3 SYSGEN> CONNECT SLU=4 They will then appear as TCA0, TCB0, TCC0, TCD0. Carl Karcher Internet: KARCHER@WAISMAN.WISC.EDU Waisman Center Bitnet: KARCHER@WISCMACC University of Wisconsin-Madison PSTnet: (608) 263-5896
alan@shodha.enet.dec.com ( Alan's Home for Wayward Notes File.) (04/27/91)
In article <1991Apr26.181122.12239@iplmail.orl.mmc.com>, carlm@iplmail.orl.mmc.com (carl morris) writes: > [ A customer has noticed the serial ports available on the KA820 > family of processors and wants to know what to do with them. ] Well, on ULTRIX they are /dev/console, /dev/ttyc1, /dev/ttyc2 and /dev/ttyc3. Since they are probably very DZ like serial lines, I would guess that each character input or output will require an interrupt. Maybe not, but lots of DEC history suggests this will be the case. It's also likely they have little if any modem control. You could probably use them for terminals or printer, but you wouldn't want to run them very fast. p.s. I don't know what the VMS name would be. Maybe CSA[1-3]? > Thanks for any suggestions -- Carl -- Alan Rollow alan@nabeth.cxn.dec.com
mike@UC780.UMD.EDU (Mike Santangelo) (04/30/91)
In article <3i3213w163w@zl2tnm.gp.co.nz>, don@zl2tnm.gp.co.nz (Don Stokes) writes: >carlm@iplmail.orl.mmc.com (carl morris) writes: > >> The lab I work for recently inherited a VAX 8350 system. Looking through >> the KA820/KA825 Processor Technical Manual that came with it, one of my >> maintenance gurus came across a section (page 2-8) that references four >> serial line units which are on the cpu board. There must be some reason >> for putting those ports on the board ("Hey, Bob, we got this corner left >> over on the cpu board..."), anyone have any ideas? Was it primarily for >> maintenance diagnostic access, or are there some useful end-user type >> uses for them? > >They're (sort of) perfectly usable serial ports. You know, those silly >things that send and receive data one bit at a time? 8-) > >Seriously, one of them is the console port -- all VAXes have at least >one console port. Under VMS, this is referred to as OPA0:. The >remaining three ports are available to you to do as you like with them. >VMS doesn't enable them by default -- you can enable them with the >following commands (from a suitably privileged account; we do this at >system startup time): > > $ MCR SYSGEN > SYSGEN> CONNECT SLU=1 > SYSGEN> CONNECT SLU=2 > SYSGEN> CONNECT SLU=3 > SYSGEN> EXIT > >You now have three devices called TCA0:, TCB0: and TCC0:. > >Note that these devices are pretty dumb, and interrupt the CPU at a >fairly high priority. Heavy use is *not* recommended. High speeds >aren't a good idea, although they do work. Your milage may vary... > > >Don Stokes, ZL2TNM / / don@zl2tnm.gp.co.nz (home) >Systems Programmer /GP/ GP PRINT LIMITED Wellington, don@gp.co.nz (work) >__________________/ / ---------------- New_Zealand__________________________ One question though, from a person who doesn't have a KA820 based VAX, but maybe they would be cheap enough to get on the 3rd party used DEC hardware market... Do these ports have connectors to go along with them? Like DB25 based RS232? Or are they something you have to fudge around with on the processor board to bring out. Inquiring minds want to know :^) +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Michael F. Santangelo + Inet: mike@uc780.umd.edu VMS / UNIX Systems + mike@socrates.umd.edu Academic Computing UMUC + Bnet: MIKE@UC780 (The University of Maryland, + MIKE@UMUC (not visited often) University College) +<Your clever net-phrase here>
cur022%cluster@ukc.ac.uk (Bob Eager) (05/01/91)
In article <26APR91.19245964@toron.waisman.wisc.edu>, karcher@toron.waisman.wisc.edu writes: > These are real live serial ports you can do anything you want with. The No modem control, though. And I believe that > 1200 bps is not supported (although I run one at 9600). I suspect that if you used all of them, you might have some slowdown problems in the CPU; I think there is a lot of microcode involved. I also believe that VMS doesn't configure them if you have other serial ports attached. -------------------------+------------------------------------------------- Bob Eager | University of Kent at Canterbury | +44 227 764000 ext 7589 -------------------------+-------------------------------------------------
don@zl2tnm.gp.co.nz (Don Stokes) (05/04/91)
cur022%cluster@ukc.ac.uk (Bob Eager) writes: > In article <26APR91.19245964@toron.waisman.wisc.edu>, karcher@toron.waisman.w > > These are real live serial ports you can do anything you want with. The > > No modem control, though. And I believe that > 1200 bps is not supported > (although I run one at 9600). I suspect that if you used all of them, you > might have some slowdown problems in the CPU; I think there is a lot of > microcode involved. I've run one at 19.2K, but it locked up on me after a bit of abuse (it may have been fixable and unrelated to the port type; I only tried it once and that was in the middle of some crisis or other). It did work, ie the UARTs and software know what you're talking about. Re microcode, yes I believe so -- these are basically the same sorts of ports as OPA0 on any other VAX, ie addressed using MTPR and MFPR (move to/from processor register) instructions rather than memory mapped I/O. As such, they're character oriented (rather like the dreaded DL-11), and tend to suck on the CPU a bit. Flow control is handled in software, so if you sling a large chunk of data at them from the attached device at a high speed, it's almost guaranteed to (a) drop half of it on the floor, and (b) choke the CPU (I'm reasonably sure this was what caused the lockup I mentioned above). Running a couple of terminals off them should be OK however. > I also believe that VMS doesn't configure them if you have other serial > ports attached. No, it doesn't. I posted the necessary SYSGEN incantations earlier. With regard to the fellow who asked if the ports have a DB25 distribution panel: I'm not sure. As I said, all VAXes have a console port, so that at least will have a DB25 present, but the other three ports may not. Without digging the manual out or opening the machine up and looking, I couldn't say -- I'm at home and the machine is at work. Don Stokes, ZL2TNM / / don@zl2tnm.gp.co.nz (home) Systems Programmer /GP/ GP PRINT LIMITED Wellington, don@gp.co.nz (work) __________________/ / ---------------- New_Zealand__________________________