x86_64 updates

Michal Jaegermann michal at harddata.com
Tue Jan 27 16:33:19 UTC 2004


On Tue, Jan 27, 2004 at 08:59:31AM -0500, Gene C. wrote:
> 
> So far my experience in trying to port some software (which is also designed 
> to run on MS Windows) has proved to be a real learning experience.  This is 
> especially true for the "long" variable which is 4 bytes on 32 bit systems, 8 
> bytes on 64 bit *nix and Linux systems, and 4 bytes on MS Windows 64 bit 
> systems.

For some of us who were using Alpha and other 64-bit processors
this experience is some decade long or something like that.  This
detail that Red Hat had Alpha and SPARC distribution for quite a
while is not without a significance in that that 64-bit user space
on x86_64 basically "just works".  Personally I run into the issue
even earlier.  "Natural" sizes on Atari ST with MC68000 were 2
for sizeof(int) and 4 for longs and pointers.

It is indeed PITA that many were able to forget what they were
hopefuly told in an introductory C programming class.  Historically
this is known as "all the world is VAX" syndrome so the problem
is far from beeing new. :-)

   Michal





More information about the fedora-test-list mailing list