On Fri, Jun 28, 2019 at 11:17 AM Jean-Baptiste Kempf wrote:
>
> On Fri, May 31, 2019, at 17:53, Gravis wrote:
> > I'm a huge fan of POSIX but POSIX functions should not be used in the
> > MinGW-w64 CRT code as they are now. Using them in this way creates
> > mandatory dependencies on additional l
On Fri, May 31, 2019, at 17:53, Gravis wrote:
> I'm a huge fan of POSIX but POSIX functions should not be used in the
> MinGW-w64 CRT code as they are now. Using them in this way creates
> mandatory dependencies on additional libraries (e.g. MSVCRT). Since
> all libraries ultimately rely on Kerne
On Fri, May 31, 2019 at 11:54 AM Gravis wrote:
>
> I'm a huge fan of POSIX but POSIX functions should not be used in the
> MinGW-w64 CRT code as they are now. Using them in this way creates
> mandatory dependencies on additional libraries (e.g. MSVCRT). Since
> all libraries ultimately rely on K
On 01.06.2019 19:39, sotrdg sotrdg wrote:
> Aren’t these “malloc”,”strlen”,”malloc” etc standard C functions?
>
My guess is that he means that on Windows the C runtime library dependency is
not mandatory. Drivers and other kernel-level components are often written
without using it.
signature.a
gw-w64-public] CRT uses POSIX functions.
I'm a huge fan of POSIX but POSIX functions should not be used in the
MinGW-w64 CRT code as they are now. Using them in this way creates
mandatory dependencies on additional libraries (e.g. MSVCRT). Since
all libraries ultimately rely on Kernel32, it
I'm a huge fan of POSIX but POSIX functions should not be used in the
MinGW-w64 CRT code as they are now. Using them in this way creates
mandatory dependencies on additional libraries (e.g. MSVCRT). Since
all libraries ultimately rely on Kernel32, it only makes sense to
replace these POSIX functi
