On Thu, 2007-06-28 at 15:49 +0200, Jan Engelhardt wrote: > I'll have to chime in here. > Test program: > #include <sys/socket.h> > #include <sys/stat.h> > #include <sys/types.h> > #include <arpa/inet.h> > #include <netinet/in.h> > #include <errno.h> > #include <stdio.h> > #include <stdlib.h> > #include <string.h> > #include <unistd.h> > #include <linux/in.h> /* get IP_FREEBIND */ > > Creates a lot of error messages. > (Lots of redefinitions.) > > $ rpm -q linux-kernel-headers glibc > linux-kernel-headers-2.6.21-7 > glibc-2.6-5 > (suse 10.3 factory) > > So looks like there's still something to do.
Hm, yes. But what? Is it reasonable for people to include <linux/in.h> and <netinet/in.h> at the same time? It's suboptimal that they have to include <linux/in.h> for certain definitions, but that file also provides conflicting definitions of stuff which exists elsewhere. Should we split <linux/in.h> into two parts? -- dwmw2 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/