home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.std.c++
- Path: sparky!uunet!microsoft!hexnut!jimad
- From: jimad@microsoft.com (Jim Adcock)
- Subject: Re: Zero-length structures and pointer comparisons
- Message-ID: <1992Dec24.004534.1152@microsoft.com>
- Date: 24 Dec 92 00:45:34 GMT
- Organization: Microsoft Corporation
- References: <BzCG7K.2sG@frumious.uucp> <1992Dec18.192202.17023@ucc.su.OZ.AU> <1992Dec18.204337.3084@lth.se>
- Lines: 21
-
- In article <1992Dec18.204337.3084@lth.se> dag@bellman.control.lth.se (Dag Bruck) writes:
- |In <comp.std.c++> maxtal@extro.ucc.su.OZ.AU (John MAX Skaller) writes:
- |>
- |>In fact, p==q iff objects equal is unimplementable on some machines.
- |>That is, even the existing ARM requirement is unworkable.
- |
- |No, some machines are unworkable :-).
-
- Agreed -- we just disagree on what machines. The implementors of Smalltalk,
- Smalltalk machines, the Mac systems, The Windows systems, OODBMSs, and I,
- believe that machines with immovable objects, paging in 4K chunks,
- are "unworkable" for serious object oriented work. The basic problem is
- that objects are typically about 100 bytes, implying an average of 40
- objects per 4K page. The 40 objects do not typically consistently remain
- to the same temporal working set. Thus paging needs finer granulaty,
- or objects must be moveable. Objects cannot in general be moveable to
- an appropriate current temporal working set and simultaneously retain their
- global address space orderings in memory. Unless one had hardware paging
- in much smaller chunks than 4K -- which in fact 386/486 machines DO support
- in one addressing mode -- not commonly used in typical OS's.
-
-