Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!zaphod.mps.ohio-state.edu!lavaca.uh.edu!menudo.uh.edu!sugar!ficc!peter From: peter@ficc.ferranti.com (Peter da Silva) Newsgroups: comp.lang.misc Subject: Re: Answers, Chapter 1: TeX (was C's sins... and others) Message-ID: <5N.6GH5@xds13.ferranti.com> Date: 13 Nov 90 21:08:18 GMT References: <7AZ6O15@xds13.ferranti.com> <5347@lanl.gov> <1990Nov12.194544.14504@Think.COM> Reply-To: peter@ficc.ferranti.com (Peter da Silva) Organization: Xenix Support, FICC Lines: 29 In article <1990Nov12.194544.14504@Think.COM> barmar@think.com (Barry Margolin) writes: > The 80x86 is not the only architecture with multi-level memory. There have > been quite a few systems built with segmented or object-based memory > architectures. Well, yes, but the 80x86 is the only survivor with a large enough following to make it worth while. [re: the lisp machine] > Or the C runtime could allocate a big array at program startup time, and do > all its allocation out of that (growing it when necessary). This is what the Burroughs C compiler had to do. Perhaps (here's a heretical thought) it's not worth implementing a C compiler on some machines. > You keep on talking about implementing a portable memory allocator in C. Well, Jim Giles does. I'm not sure why, but I see no great difficulty in doing it of you're willing to make the effort to carry a tag around. Not as fast as machine-dependent code, but isn't that always the case? In any case... Would it have been such a bad thing to standardise a method of determining if two pointers pointed into the same object? They did it for offsets in structures with OFFSET_OF. -- Peter da Silva. `-_-' +1 713 274 5180. 'U` peter@ferranti.com