Re: Wanted: 64bit Delphi

> Like Danny Thorpe said, don't underestimate the effort of identifying,
> fixing and testing critical spots, i.e. where sizes are important, but
> are simply assumed as fixed, or where pointer and integer sizes are
> assumed to be the same, or are cast to each other, etc.

Especially in times of .NET i think even a native compiler should
embrace the concept of

* translating the source lang into a meta language
* compiling the meta language into intermediate code
* generate the final byte code for the destination platform

I'm not sure if Borland has somethign similar but since they have at
least 2 different compilers (Delphi, C++) it might be worthwile to
redesign them. I'm aware that one is one-pass and the other multi-pass
but that can surely be catered for.

If the intermediate code is documented well enough i'm sure there will
be interested people who can create desintation compilers for the last
step for platforms that Borland cater for due to resource or financial
restrictions.

--
Hannes Danzl [NexusDB Developer]
Newsgroup archive at http://www.tamaracka.com/search.htm
.

Relevant Pages

  • Re: About Visual C++ 2005 Express Edition
    ... But that's not a useful comparison unless you have sizes for *the same ... program* for other compilers. ... Puts is MUCH smaller than printf OF COURSE. ... trying to compare printf implementations. ...
    (comp.lang.c)
  • Re: KIND question -- part II
    ... KIND function. ... correspond to byte sizes is simply a historical convenience. ... That they often bear similarity to byte sizes is ... This similarity is not so for all compilers. ...
    (comp.lang.fortran)
  • Re: Optimisation and INTENT
    ... arguments requiring a pointer. ... sizes. ... compilers (meaning gcc) allow that." ...
    (comp.lang.fortran)
  • Re: f2c for Opteron patch under Linux
    ... > wkpark@kldp.org writes: ... > the usual default sizes of integer and real on other machines? ... > possibly other compilers on the same platform? ...
    (comp.lang.fortran)