24 Oct
2005
24 Oct
'05
8:39 a.m.
On Mon, 24 Oct 2005 05:00:57 +0100 (BST), ard at p850ug1.demon.co.uk
(Tony Duell) wrote:
NO, just for fun... I doubt very much I could improve on the
performance
for running the VAX instruction set. I have not looked at the prints
carefully enough to determine how much of the instruction set is
hard-wired (if any), I wonder if it would be possible to run a
different
instruction set entirely.
-tony
Actually, with access to the micro instructions you can often
substantially improve the run-time of a given program. The Modcomp II
had the microbus accessible for use along with a number of unused
instructions. The communications instructions and floating point were
implemented using this bus.
I implemented an instruction store attached to this bus and which
allowed me to create and store new instructions. A friend working on
a CS Phd in pattern matching ginned up a program that would find
common instruction sequences in a program, deconstruct the microcode,
perform optimization on the sequence and then create a new
instruction which was substituted for the original sequence. We often
got increases of performance of 20% over the original code.
IIRC the Burroughs computers would load a different instruction set
depending on what language was being used. Tony could create a VAX
with Forth as the instruction set ;-)
CRC
The
11/730, of course, is mostly PALs (as I said) with some RAM as a
control store, 8 off 2901s as the ALU, and an 8085 (I think) to
load the
control store, etc. One day I am going to look into modifying the
microcode of that machine...
To what end? Increased performance?
24 Oct
24 Oct
9:35 a.m.
IIRC the Burroughs computers would load a different
instruction set
depending on what language was being used.
B1700
Lee Courtney
-----Original Message-----
From: cctalk-bounces at classiccmp.org
[mailto:cctalk-bounces at classiccmp.org]On Behalf Of CRC
Sent: Monday, October 24, 2005 8:40 AM
To: cctalk at classiccmp.org
Subject: Re: Any own an 11/725...?
On Mon, 24 Oct 2005 05:00:57 +0100 (BST), ard at p850ug1.demon.co.uk
(Tony Duell) wrote:
NO, just for fun... I doubt very much I could improve on the
performance
for running the VAX instruction set. I have not looked at the prints
carefully enough to determine how much of the instruction set is
hard-wired (if any), I wonder if it would be possible to run a
different
instruction set entirely.
-tony
Actually, with access to the micro instructions you can often
substantially improve the run-time of a given program. The Modcomp II
had the microbus accessible for use along with a number of unused
instructions. The communications instructions and floating point were
implemented using this bus.
I implemented an instruction store attached to this bus and which
allowed me to create and store new instructions. A friend working on
a CS Phd in pattern matching ginned up a program that would find
common instruction sequences in a program, deconstruct the microcode,
perform optimization on the sequence and then create a new
instruction which was substituted for the original sequence. We often
got increases of performance of 20% over the original code.
IIRC the Burroughs computers would load a different instruction set
depending on what language was being used. Tony could create a VAX
with Forth as the instruction set ;-)
CRC
The
11/730, of course, is mostly PALs (as I said) with some RAM as a
control store, 8 off 2901s as the ALU, and an 8085 (I think) to
load the
control store, etc. One day I am going to look into modifying the
microcode of that machine...
To what end? Increased performance?
24 Oct
24 Oct
4:51 p.m.
Actually, with access to the micro instructions you can often
substantially improve the run-time of a given program. The Modcomp II
I am something of a PERQ-fanatic, so I am well aware of this. The classic
PERQs have user-modifyable microcode, and the hardware is not designed
for a particular isntruction set (there is no hard-wired instruction
decoding at all, unlike, say, a PDP11/45), so you can pretty much do what
you like.
Under POS (PERQ OS or Pascal OS, depending on which book you read), you
were not expected to write any machine or assembly langage (Q-codes). The
assember was not included in the OS distribution, and the manual doesn't
really epxlain the Q-codes. But the microcode assember and placer were
included, the manual does describe user-written microcode.
For some things, like directly accessing I/O ports (say for custom
hardware -- like the tape streamer interface hack) you _have_ to use
microcode on the PERQ. But you can also use it to speed up critical
routines in your program (just don't try and improve the raster operation
microcode, you won't do it. Period.)
-tony
6997
days inactive
6999
days old
3 comments
4 participants
participants (4)
-
ard＠p850ug1.demon.co.uk -
aw288＠osfn.org -
lcourtney＠mvista.com -
technobug＠comcast.net