Path: utzoo!mnetor!uunet!lll-winken!lll-lcc!lll-tis!ames!ucsd!sdcsvax!ucsdhub!hp-sdd!hplabs!hpda!hpcupt1!viggy
From: viggy@hpcupt1.HP.COM (Viggy Mokkarala)
Newsgroups: comp.arch
Subject: Re: Auto-shifted registers
Message-ID: <6310005@hpcupt1.HP.COM>
Date: 23 Feb 88 19:47:31 GMT
References: <1370@vaxb.calgary.UUCP>
Organization: Hewlett Packard, Cupertino
Lines: 29

radford@calgary.UUCP (Radford Neal) writes:

>It has occurred to me that this "auto-shifting" might be useful
>for ALL register references. Every reference to a register as
>a source operand would be accompanied by two bits giving a
>shift amount - equivalet to a multiplication by 1, 2, 4, or 8.

...
...

>	Radford Neal
>	The University of Calgary

The HP Precision Architecture provides for these kinds of operations by its
Shift and Add instructions.  There is a pre-shifter before one of the inputs to
the CPU.  It allows for one of the operands to be pre-shifted by upto 3 bits
before an addition happens.  These Shift and Add instructions come in two
flavors - one which traps on an overflow, and the other which doesn't.  A
general integer multiply routine takes, on the average, less than 20
instructions.

The assumptions and details of this claim can be found in the
paper titled "Integer Multiplication and Division on the HP Precision
Architecture", Proceedings of the Second International Conference on
Architectural Support for Programming Languages and Operating Systems (ASPLOS
II), pp 90 - 99.


Viggy Mokkarala, Hewlett Packard Co., Cupertino, CA.
(hpda!viggy)