Re: Old printers, Architectural implications of locate mode I/O

According to MitchAlsup1 <mitchalsup@aol.com>:

We students, also, figured out that TSS had dumped OS process state in
page 0 (un accessible normally) of the user's address space. We then
figured out that while we could not LD or ST that data, we could queue
up I/O (out) write the data to disk, read it back where we could diddle
with it, write it back to disk, queue up I/O (in) and read it back into
page 0.
>
All we did was to set the privilege bit !!

That was very restrained of you.  I was only a visiting high school kid
when I was using TSS at Princeton so I didn't push my luck.  But ...

We (students) used to have comment cards that cause the hammers all
fly at the same time. So, instead of a natural z z z z z of the print,
it would go BANG BANG BANG and we knew out stuff was being printed...

Princeton had high end 1403 model N1 printers on the 360/20's they
used for RJE to their 360/91. The /91 crashed a lot, and while waiting
for it to come back up, we might have done some, ah, unofficial stuff
pn the /20's, particularly once someone found a copy of the 360/20
card assembler.

On that model of printer could use different print chains with
different sequences of characters, to handle upper/lower case and
such. The printer had a Universal Character Set (UCS) buffer the
computer could load to tell the printer what characters were where on
the chain. So one day I wondered what would happen if I loaded the UCS
buffer with all the same character, then printed lines of that
character.

Let me just say, Don't Do That. I think it blew a fuse before anything
was seriously damaged.
--
Regards,
John Levine, johnl@taugh.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. https://jl.ly