DMF-32+4.3BSD+VAX8250=problems (HELP!)

[Ehud Gavron, system manager and VMS guru]

In article <1534 at atanasoff.cs.iastate.edu>, heins at atanasoff.cs.iastate.edu (Leeland Heins) writes:
> Any wizards got a quick solution or answer to this problem:
> 
>    We just tried to install a DEC DMF-32 in our VAX 8250's uba.  This DMF-32
> had previously been in an old 11/730 and had been working when it was removed.
>    We are running 4.3 BSD-Tahoe with no relevant modifications.
>    We have an Emulex CS21/F which emulates two DFM-32's installed and working
> fine in the 8250, this is the first "real" DMF-32 we've tried.  I have been
> told that it is the "official" word from DEC that the DMF-32 is questionable
> in VAX 8xxx machines, other people have reported that they seem to work.
> 
> This is its line from the config file:
> 
> device		dmf2	at uba? csr 0160440 flags 0xfc
> 	vector dmfsrint dmfsxint dmfdaint dmfdbint dmfrint dmfxint dmflint
> 
>    This config entry is exactly the same as the entries for our CS21/F,
> except for the appropriate modifications for the address and device
> number.
> 
> The DMF-32 shows up at boot as so:
> 
> dmf2: printer synch asynch.
> dmf2: at uba0 csr 160440 vec 720, ipl 15
> 
>    Things seem to be O.K. up until it prints the "Mon Sep 18 ..." message,
> but when it would normally write the login messages out to each of the
> ttys with getty enabled we get:
> 
> Mon Sep 18 ...
> trap type 8, code = 89d55865, pc = 80038128
> panic: Segmentation fault
> syncing disks... 36 35 30 19 6 done
> ....
> 
>    Getty isn't enabled on any of the ports yet, and nothing is plugged into
> any of the ports on the DMF-32.
>    Any ideas people have will be greatly appreciated.  Any comments about
> buying other multiplexers will be (temporarily at least) disregarded since we
> are currently very broke.  If we can't get it to work, we will have to do
> without temporarily or install a (ugh!) DZ-11.
>    Thanks in advance!

	Yes, reset your csr and vector properly.  We are running
	dmf-32's in our 8650s without problems.  Of course we're
	running the OS that one pays for, and that pays for itself...
		V M S...


> 
>   //|||\\   "Attack of the killer 	Lee Heins, EXNET Programmer
>  //-O-O-\\   mutant techno-hippies	Iowa State U. Extension
>  ||  v  ||   from Hell"			heins at atanasoff.cs.iastate.edu
>  || === ||				...!hplabs!hp-lsd!atanasoff!heins
>  || --- ||
> /// `|' \\\				Disclaimer: I speak for myself only.
-- 
  ________________________________________________________________________
 / Ehud Gavron, System Manager    |      SPAN:  1029::55295::GAVRON       \
|  Los Alamos National Laboratory |    HEPNET: 43115::55295::GAVRON        |
|  Meson Physics Facility         |    Bitnet: gavron at lampf                |
|  Data Analysis Center           |  Internet: gavron at dac.lanl.gov (VMS)   |
|  MS H828, Los Alamos NM 87544   |            eg at lanl.gov         (unix)  |
|  Yellnet: (505) 665-1131        |      UUCP: cmcl2!lanl!eg               |
|--------------------------------------------------------------------------|
|  How sharper than a serpent's tooth it is to have a thankless serpent.   |
 \________________________________________________________________________/