This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Problem when cross-compiling trunk in libgcc2.c
2009/3/19 Ozkan Sezer <sezeroz@gmail.com>:
> On Thu, Mar 19, 2009 at 8:04 PM, Vincent R. <forumer@smartmobili.com> wrote:
>> On Thu, 19 Mar 2009 19:58:13 +0200, Ozkan Sezer <sezeroz@gmail.com> wrote:
>>> I'm a bit amazed that the prototype for VirtualProtect() is known to the
>>> compiler but the definition of DWORD is not.. In any case, it should be
>>> fixed easily by changing DWORD into unsigned int which is what a
>>> DWORD is always defined as. ?And PR 39063 is still open anyway.
>>> --
>>> Ozkan
>>
>> Yes you are right about the fact VirtualProtect is defined and not DWORD so
>> something is wrong with my includes
>
> That is my guess..
>
>> However you are wrong about DWORD definition it has always be defined
>> like this :
>>
>> typedef unsigned long DWORD, *PDWORD, *LPDWORD;
>>
>> at least windows.
>>
>
> A DWORD on windows is an unsigned 32 bit integer, therefore even
> if you use long, it won't matter because long is 32 bits for both win32
> and for win64.
>
> --
>
> Ozkan
>
Ozkan,
Not, really. Because if the function was prototyped before by DWORD
(for mingw/cygwin on x86/x64), we get a warning about different
prototypes. For the compiler unsigned long and unsigned int are
different types.
Cheers,
Kai
--
| (\_/) This is Bunny. Copy and paste
| (='.'=) Bunny into your signature to help
| (")_(") him gain world domination