Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!usc!zaphod.mps.ohio-state.edu!mips!bridge2!jarthur!nntp-server.caltech.edu!laguna.ccsf.caltech.edu!daveg From: daveg@near.cs.caltech.edu (Dave Gillespie) Newsgroups: comp.arch Subject: Re: What *should* architectural pointers point at? Message-ID: Date: 2 Sep 90 09:17:22 GMT References: <41188@mips.mips.COM> <26DE7EE3.58FC@tct.uucp> <0899@sheol.UUCP> Sender: news@laguna.ccsf.caltech.edu Organization: California Institute of Technology Lines: 49 In-Reply-To: throopw@sheol.UUCP's message of 2 Sep 90 03:27:13 GMT >>>>> On 2 Sep 90 03:27:13 GMT, throopw@sheol.UUCP (Wayne Throop) said: > Looking at it from another perspective, I think it certain that many, > many MORE programs (and mostly the same sloppy programs, BTW, IMHO) will > be broken by making pointers 64 bits long, whether or not integers are > also made 64 bits long. Good point! > From: daveg@near.cs.caltech.edu (Dave Gillespie) >> The 68000 puts a short branch in a 16-bit >> instruction. Eight bits of that instruction encode the displacement. >> Using a naive encoding, you are wasting four of eight bits, *NOT* >> four of 64. > From the rest of Dave's posting, it's not clear to me that he's really > objecting to bit-granular addresses on the above grounds. (In fact, it > seems not to be the case.) But let me respond as to a hypothetical > person who *does* think this is an objection. I hope Dave takes no > offense. No offense taken---you got my point exactly. Many existing architectures put byte offsets into the instruction word even though alignment constraints mean some of the bits of the offset will always be zero. Why not omit the zeros and put those bits on the left where they can do some good! If it's only one or two bits it's not such a big deal, but a bit-addressed machine would have to take the wastage of, say, 4-6 bits much more seriously. Some machines do this now for branch offsets, but I don't know of any that do it for all types of offsets. A couple other people have implied that 64-bit machines would have 64-bit instruction words, with plenty of bits to spare. But I really doubt this will happen: The temptation to use that wide bus to fetch two 32-bit instructions in one gulp is probably too great. Even 32-bit instruction sizes are controversial. > Sure, let the people who want to operate on entities of bit granularity > pay for the privilege. But there's no reason to require them to tie > their hands behind their backs and stand on their heads and spin > hula-hoops on their feet while they type with their noses either. Um, er, my sentiments exactly... -- Dave -- Dave Gillespie 256-80 Caltech Pasadena CA USA 91125 daveg@csvax.cs.caltech.edu, ...!cit-vax!daveg