NetNews Usenet Archive 1992 #31/NN_1992_31.iso/spool/comp/std/cplus/1897

Internet Message Format | 1992-12-22 | 1.9 KB

Path: sparky!uunet!tdat!tools3!swf
From: [e-mail] (Stan Friesen)
Newsgroups: comp.std.c++
Subject: Re: Zero-length structures and pointer comparisons
Message-ID: <[e-mail]>
Date: 22 Dec 92 21:48:33 GMT
References: <[e-mail]> <[e-mail]> <[e-mail]> <[e-mail]>
Sender: [e-mail]
Distribution: world
Organization: NCR Teradata Database Business Unit
Lines: 35

In article <[e-mail]>, [e-mail] (John MAX Skaller) writes:
|>     In that case it is implementable on all systems
|>     where (opinion follows, please correct errors)
|>
|>     a) void* is big enough to hold all pointer types

This assumption, as written, is violated by compact and medium model on Intel
chips.  I suggest rewording this to:
    a) void* is big enough to hold all object pointer types
                       ^^^^^^
[function pointers are *not* guarenteed to be convertable to void *, and,
in C++, neither are member 'pointers'].

|>     b) conversions always pad systematically
|>     c) bitwise comparison is done
|>     d) the bits of a pointer cannot be magically changed by the system
|>     e) a pointer is implemented as an object in contiguous storage
|> ...
|>
|>     Certainly, (e) might cause problems on the 80x86 where part of the
|> pointer is implicit and not physically stored. Does it *actually*
|> cause this problem?

In 'small data' models, YES!  There is simply no reason for void* to contain
the segment when all data is in the same segment.

Of course, in paractice, the limitations on the 'small data' models make this lack
of little *practical* concern for ptrcmp() - that is since all data are in the
same segment a segmentless comparison will always work (as long as you use the
*stronger* form of (a) I mention above).

--
[e-mail]            (formerly tdatirv!sarima)
  or
[e-mail]