17 Mar
2023
17 Mar
'23
6:25 a.m.
On Mar 16, 2023, at 10:14 PM, Chuck Guzis via cctalk
<cctalk(a)classiccmp.org> wrote:
On 3/15/23 17:23, Jon Elson via cctalk wrote:
I'm still trying to get a good answer to "when did the first commercial (as
opposed to one-off lab) computer appear that had interrupts as a standard feature?"
It looks like there was the IBM 704, in 1958, with interrupts but some documentation made
me think it was an optional feature. The other machine from 1958 I can think of is the
Electrologica X1. That is the machine that confronted Dijkstra with the need to develop
new theory to deal with the non-sequential behavior of machines with interrupts, and his
Ph.D. thesis was the result. The X1 is also interesting in that it has what much later
would be called a BIOS -- a ROM-resident software library that included I/O services
including dealing with interrupts, an assembler/loader, and an operator interface.
Dijkstra wrote that, and the source code is in an appendix of his thesis.
paul
Yes, the IBM 709x ran in single-job fashion. I
don't think it had
interrupts, so breaking off one program to schedule another was not
possible. Also, it had no memory protection. We had a 7094 at
Washington University in the late 1960s, and it was the main computer
resource on campus. When the moved up to a 360/50, they were able to
benefit from multiprogramming, and got a boost in throughput, although
the 7094 was QUITE a bit faster than the 360/50.
...
The CDC 1604 (1959 (certainly had interrupts, both internal and
external, but I do not believe it had memory protection. It did have a
console loudspeaker driven by a 3-bit DAC, however.