Re: Size of C/C++ data type from GNU GCC/g++ compiled ELF 64-bit LSB executable, AMD x86-64 vs. ELF 32-bit LSB executable, Intel 80386

[Tim Prince <tprince at myrealbox dot com>]

Matthew Woehlke wrote:
> Brian Dessent wrote:
> > Matthew Woehlke wrote:
> > > Incidentally, does the Windows-x64 port* of gcc perpetuate this idiocy?
> >
> > That's a silly question... Of course it does, it's not gcc's decision to
> > make.  If gcc were to act in a way that is fundamentally broken from the
> > platform specification it would be utterly useless as a compiler --
> > calling any system function or anything else built by the native tools
> > would have wrong arguments, structs would have different layouts, you'd
> > get all kinds of random crashes.
>
> Hmm, true... bleh. Bummer that even gcc is stuck with Microsoft's 
> decision making.
This sub-topic is about 64-bit mingw.  Apparently, it is the only gcc 
for 64-bit Windows with any near-term prospects.  As mingw requires 
Microsoft libraries, the point about requiring data type consistency 
with those libraries has to be accepted.
32-bit cygwin already has differences in data types from mingw32, for 
example, in partial support of 80-bit long double.  As cygwin aims to 
support linux compatibility, I guess that a 64-bit long would not be out 
of the question, should a 64-bit cygwin emerge.  If 64-bit cygwin is 
totally impossible, I may be among those abandoning cygwin and Windows gcc.