Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!mips!apple!turk
From: turk@Apple.COM (Ken "Turk" Turkowski)
Newsgroups: comp.lang.c++
Subject: Re: 64 bit architectures and C/C++
Message-ID: <13261@goofy.Apple.COM>
Date: 1 May 91 09:21:43 GMT
References: <168@shasta.Stanford.EDU> <1991Apr29.050715.22968@ux1.cso.uiuc.edu> <jac.672906119@sundance>
Organization: Advanced Technology Graphics, Apple Computer, Cupertino, CA, USA
Lines: 32

jac@gandalf.llnl.gov (James A. Crotinger) writes:


>  Anyone want to comment on experiences with Crays? I believe the C
>compilers have sizeof(int) = sizeof(short) = sizeof(long) == 46 or 64
>bits, depending on a compile time flag (Crays can do 46 bit integer
>arithmetic using the vectorizing floating point processors, so that is
>the default). 

>  Binary communications between Crays and other computers is something
>I haven't done, mostly because Cray doesn't support IEEE floating point.

Crays are a pain.  Structures are padded to 64 bits, so you waste a lot
of memory in arrays if your structures only have 16 bits.

I thought the number was 48 bits, but you might be right.  I seem to recall
that logical operations couldn't work with 64 bits, but arithmetic operations
could, or vice versa.

If you want to write out a number between 8 and 64 bits, you need to break
it into bytes.  As long as you're doing this, you may as well write the
proper endian.

It's not too bad to write a cray FP to IEEE converter.  One of the basic
rules of machine independent I/O, though, is to not write structures out
directly, but rather to go through a procedure for each element.  Same goes
for input.
-- 
Ken Turkowski @ Apple Computer, Inc., Cupertino, CA
Internet: turk@apple.com
Applelink: TURK
UUCP: sun!apple!turk