Jeff Gunter:
> FWIW, I have also had problems with SGI's C++ compiler and g++ (on the same
> SGI) being incompatible. The incompatibilities are even to the level where
> code will compile cleanly under one compiler and not compile at all under the
> other. These two compilers are definitely not interchangable.
Wildly off topic, but a fun view into `commercial quality software', let
me show you the output of CC under IRIX 64 with the following
application:
hw.cc:
------------------------------------------------------------------------
#include <iostream.h>
int main() { cout << "Hello world." << endl; return 0; }
------------------------------------------------------------------------
> CC -fullwarn -o hw hw.cc
"/usr/include/CC/iostream.h", line 236: remark(1506): implicit
conversion from
"long" to "int": rounding, sign extension, or loss of
accuracy may
result
x_blen= (eb>b) ? (eb-b) : 0 ;
^
"/usr/include/CC/iostream.h", line 265: remark(1506): implicit
conversion from
"long" to "int": rounding, sign extension, or loss of
accuracy may
result
return x_gptr<x_egptr ? x_egptr-x_gptr : 0 ;
^
"/usr/include/CC/iostream.h", line 270: remark(1506): implicit
conversion from
"long" to "int": rounding, sign extension, or loss of
accuracy may
result
if ( x_pptr ) return x_pptr-x_pbase ;
^
"/usr/include/CC/iostream.h", line 646: remark(1430): omission of
"class" is
nonstandard
friend ios ;
^
"/usr/include/CC/iostream.h", line 650: remark(1174): variable
"iostream_init"
was declared but never referenced
} iostream_init ;
^
------------------------------------------------------------------------
The 64 bit compiler doesn't seem to agree with itself!
> Suggestion: If possible take a step backwards and get g++ and SGI C++ both on
> the SGI and do a side-by-side comparison. If the both work and give the same
> output (maybe even the RIGHT output if you know what that should be), then you
> have grounds for suspecting your problem of being linux (or g++ under linux)
> related.
Probably yes. But you're never sure. That's why I'm happily awaiting
gcc 2.8.0 for some time now ... :(
I know I shouldn't ask, but has anybody a hint when this beast will hit
the, er, `market'?
Eric Meijer
--
E.L. Meijer ([EMAIL PROTECTED]) | tel. office +31 40 2472189
Eindhoven Univ. of Technology | tel. lab. +31 40 2475032
Lab. for Catalysis and Inorg. Chem. (TAK) | tel. fax +31 40 2455054
--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
[EMAIL PROTECTED] . Trouble?
e-mail to [EMAIL PROTECTED] .
