] enhancements on the win32 to win64 conversion
Para: mingw-w64-public@lists.sourceforge.net
Data: Segunda-feira, 23 de Maio de 2011, 12:09
Your code violates strict aliasing. This is your problem here. Use instead of
address cast to integer a union construct and your code will behave as you
expect
thanks a lot, and I will remember also to always use -Wall when converting
win32 code to win64
--- Em seg, 23/5/11, Kai Tietz escreveu:
De: Kai Tietz
Assunto: Re: [Mingw-w64-public] enhancements on the win32 to win64 conversion
Para: mingw-w64-public@lists.sourceforge.net
Data: Segunda-feira
Your code violates strict aliasing. This is your problem here. Use instead
of address cast to integer a union construct and your code will behave as
you expect.
Nevertheless why aren't you simply using here +0.5?
You can try if your code works with option -fno-strict-aliasing.
Regards,
Kai
--
believe or not it work on win32 but now for win32 and win64 I am using
int = float + 0.5
and it is also working
--- Em seg, 23/5/11, Ruben Van Boxem escreveu:
De: Ruben Van Boxem
Assunto: Re: [Mingw-w64-public] enhancements on the win32 to win64 conversion
Para: mingw-w64-public
2011/5/23 RSPsoftware
> the compiler dont show a warning about the win32 to win64 conversion in the
> following function, it would be great if the compiler report at least a
> warning about it
>
> this function fail miserably in win64, and I have only detected it after
> the compilation and tests
the compiler dont show a warning about the win32 to win64 conversion in the
following function, it would be great if the compiler report at least a warning
about it
this function fail miserably in win64, and I have only detected it after the
compilation and tests
intround_trick (float floatval
