This is a discussion on Re: SIMH and real hardware - VMS ; >email@example.com (Bill Gunshannon) wrote: >I am certain I read that SIMH can use real hardware where it is available, >That being said, has anyone used a TK50Z-GA with SIMH? I wouldn't know about SIMH, but FWIW, E11 does specifically support ...
>firstname.lastname@example.org (Bill Gunshannon) wrote:
>I am certain I read that SIMH can use real hardware where it is available,
>That being said, has anyone used a TK50Z-GA with SIMH?
I wouldn't know about SIMH, but FWIW, E11 does specifically support the
TK50Z-GA (and TZ30, which has similar bugs), as well as most other SCSI
tapes that can use variable record lengths. I don't know how hard it would
be to add SCSI tape support to SIMH. In E11 I do it by generating raw SCSI
commands (which I send using ASPI8DOS.SYS/etc. on DOS, or /dev/sg* on
Linux, I assume FreeBSD has something similar).
That's definitely the hard way to do it, but it's the only way on DOS,
and on Linux it solves two problems: (1) I can support devices like the
TZ30/TK50Z-GA which are not completely SCSI-1 compatible (so might not
work so well with /dev/st*), and (2) at least at the time I was writing it
(kernel 2.0/2.2 time frame), the Linux SCSI tape driver would throw away
the residual record count on a space-by-records command that terminated
early because of a tape mark. PDP-11 OSes do that operation regularly
(e.g. during dir listings) and use the residual count to find out the #
of records in the file, so it needs to be be accurate.
Also, I forget whether the TK50Z-GA has this bug, but on the TZ30 (and
Qualstar 1260S 9-tracker), if you tell it to backspace past BOT (which
PDP-11 OSes will if they feel like it), it'll get brain-damaged. So you
have to keep track of the tape position and make sure you never ask for
Anyway it's definitely pretty neat being able to boot off of real DEC
distribution tapes. The command is e.g. "mount mu0: scsi4:" (if your tape is
SCSI ID 4), and similar for SCSI disks (which again use raw SCSI commands,
which allows E11 to do low-level formatting and dismount/unload operations).
Linux does a nice job of fitting USB and ATAPI devices which use SCSI
command sets, into the /dev/sg* system, so you can treat them like SCSI
devices from the SW side even if they aren't exactly. And Fuji DynaMOs
are *great* for interchange with real PDP-11s.
To unsubscribe (or subscribe) from (to) this list, send a message to
email@example.com, with the first line of the message
body being "unsubscribe" or "subscribe", respectively (without the quotes).