| Re: Praise for Gfortran (finally) |
|
 |
|
 |
|
 |
|
 |
Group: comp.lang.fortran · Group Profile
Author: Phillip Helbig---remove CLOTHES to replyPhillip Helbig---remove CLOTHES to reply Date: Sep 20, 2008 03:15
In article <1inifpp.1j92xdij344rcN%%nospam@see.signature>,
nospam@see.signature (Richard Maine) writes:
> Phillip Helbig---remove CLOTHES to reply
> wrote:
>
>> Shouldn't porting (the compiler) it be trivial? That is, just compile and
>> go? Or is not everything implemented in the source code of the compiler,
>> but instead makes use of system calls etc?
>
> Um. Compile with what? Yes, after you have a compiler, you can use it to
> compile the compiler. Boostrapping cam be decidedly non-trivial. THat's
> the whole point, which it looks like you missed.
I was thinking a) it is written in C and b) I have a normal C compiler.
> Not that I've done much of that kind of thing myself, but I'd think a
> common way would be to start by making use of cross compilation.
Apropos.... Isn't the possibility of cross-compilation the reason some
things are not allowed by the Fortran standard, while there is no
"obvious" reason they should be disallowed?
| 
