Re: Buildworld failure

From: Giorgos Keramidas <keramida_at_FreeBSD.org>
Date: Thu, 23 Dec 2010 10:03:44 +0100
On Wed, 22 Dec 2010 19:29:39 -0500, Derek Tattersall <dlt_at_mebtel.net> wrote:
> I blew away /usr/src and /usr/obj and re-cvsuped this morning.  I
> attempted to buildworld, only to have the process die in the mkdep step.
>
> CC='clang' mkdep -f .depend -a
> -I/usr/src/cddl/usr.bin/ctfmerge/../../../sys/cddl/compat/opensolaris
> -I/usr/src/cddl/usr.bin/ctfmerge/../../../cddl/compat/opensolaris/include
[...]
> -DNEED_SOLARIS_BOOLEAN -I/usr/obj/usr/src/tmp/legacy/usr/include
[...]
> /usr/src/cddl/usr.bin/ctfmerge/../../../cddl/contrib/opensolaris/tools/ctf/cvt/util.c
> In file included from /usr/src/cddl/usr.bin/ctfmerge/../../../cddl/contrib/opensolaris/tools/ctf/cvt/barrier.c:46:
> /usr/src/cddl/usr.bin/ctfmerge/../../../cddl/contrib/opensolaris/tools/ctf/cvt/barrier.h:39:10:
> fatal error: 'semaphore.h' file not found
> #include <semaphore.h>
>          ^
> 1 error generated.
> mkdep: compile failed

Which branch is this?  When was it last updated?  I see that you are
using CC='clang', so my guess is you are building CURRENT, right?

If that's the case, can you try:

  - building without CC='clang', to see if the breakage is clang's fault
  - if that fails, resync your CURRENT source and retry

NOTE: I did build a snapshot from /head at -r 216642 last afternoon and
it worked fine with the default CC.  So if something broke *after* that
revision try bisecting from -r 216642 to the version you built, to see
if there's something that broke the build in-between.
Received on Thu Dec 23 2010 - 08:21:15 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:40:10 UTC