Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!cs.utexas.edu!oakhill!marvin From: marvin@oakhill.UUCP (Marvin Denman) Newsgroups: comp.arch Subject: Re: Question on division methods Message-ID: <3368@yogi.oakhill.UUCP> Date: 5 Jun 90 19:13:18 GMT References: <26681c80.27e8@petunia.CalPoly.EDU> Reply-To: marvin@oakhill.UUCP (Marvin Denman) Distribution: usa Organization: Motorola Inc., Austin, Texas Lines: 19 In article aglew@oberon.csg.uiuc.edu (Andy Glew) writes: >In fact, at this moment most integer processors implement divide with a shift and add, >except for the 88K which has an on-chip multiplier, and the ones which borrow >from the FPU chip. Just to clarify things, the 88k does division by shift and add. The on-chip multiplier is only used for integer and floating point multiplication. At the time of the design there were few if any IEEE 754 implementations that used the multiplier for division. Even now I am only personally aware of 2 implementations of this type that claim IEEE accuracy. There are probably several more. Marvin Denman Motorola 88000 Design cs.utexas.edu!oakhill!marvin -- Marvin Denman Motorola 88000 Design cs.utexas.edu!oakhill!marvin