On Thu, Feb 18, 2016 at 03:10:19PM +0100, Ruediger Meier wrote: > On Thursday 18 February 2016, Karel Zak wrote: > > On Thu, Feb 18, 2016 at 02:12:47PM +0100, Ruediger Meier wrote: > > > +#if defined __APPLE__ > > > +# define _DARWIN_C_SOURCE > > > +#endif > > > + > > This one is needed before including sys/mman.h. Don't know whether it > would be better to define _DARWIN_C_SOURCE globally in compiler > commandline (via configure for systems where it's needed)!? config.h is always included before anything else, so if _DARWIN_C_SOURCE has no any side effect then it seems like a good solution. For example we use config.h to define _GNU_SOURCE etc. (see AC_USE_SYSTEM_EXTENSIONS aitoconf macro). > > > #include <stdio.h> > > > #include <stdarg.h> > > > #include <stdint.h> > > > @@ -62,6 +66,10 @@ > > > #define XALLOC_EXIT_CODE FSCK_EX_ERROR > > > #include "xalloc.h" > > > > > > +#if !defined MAP_ANONYMOUS && defined MAP_ANON > > > +# define MAP_ANONYMOUS (MAP_ANON) > > > +#endif > > > > What about to move this to include/c.h where we already have some > > portability ifdefs? > > Would be nice to hide MAP_ANONYMOUS definition to include/c.h. But as > long as _DARWIN_C_SOURCE will not be globally defined we would also > need to include sys/mman.h there, right? > > BTW here is how "dateutils" does it. Using a macro CHECK_MAP_ANON to get > defines MAP_ANON_NEEDS_DARWIN_SOURCE and MAP_ANON_NEEDS_ALL_SOURCE > https://github.com/hroptatyr/dateutils/blob/master/m4/sxe-mmap.m4 > https://github.com/hroptatyr/dateutils/blob/master/src/prchunk.c seems also as possible way Karel -- Karel Zak <kzak@xxxxxxxxxx> http://karelzak.blogspot.com -- To unsubscribe from this list: send the line "unsubscribe util-linux" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
