Date: Fri, 11 Oct 1996 22:00:59 +1000 From: Bruce Evans <bde@zeta.org.au> To: alk@Think.COM, current@freebsd.org Subject: Re: muldi3.c Message-ID: <199610111200.WAA24073@godzilla.zeta.org.au>
next in thread | raw e-mail | index | archive | help
>I wonder about this redundancy. > >; diff ./lib/libc/quad/muldi3.c ./sys/libkern/muldi3.c >diff ./lib/libc/quad/muldi3.c ./sys/libkern/muldi3.c >35a36,37 >> * >> * $Id: muldi3.c,v 1.3 1995/05/30 08:06:41 rgrimes Exp $ >37,40d38 >< >< #if defined(LIBC_SCCS) && !defined(lint) >< static char sccsid[] = "@(#)muldi3.c 8.1 (Berkeley) 6/4/93"; >< #endif /* LIBC_SCCS and not lint */ libkern is supposed to be built from libc by copying, or perhaps by symlinking (see libc/Makefile). Under cvs, symlinks don't work and cvs Ids get out of sync. The sccsids aren't changed by cvs, but were nuked when Ids were added to the kernel. We could compile muldi3.c directly out of libc/quad using .PATH. However, muldi3.c isn't used for the i386, and I broke this possibility for the quad functions that are used by changing "quad.h" to <libkern/quad.h> (this was required for compiling with -I- to get unambiguous paths). Bruce
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199610111200.WAA24073>