Re: Call for tester: VirtualBox 3.1.2 for FreeBSD
John Baldwin wrote: >> I'm seeing this as well: >> >> kernel: link_elf_obj: symbol _Z6strlenPKc undefined >> >> ...which, I suppose, is the real kldload error message. > > Looks like it is trying to use a C++ named-version of strlen(): > > % echo _Z6strlenPKc | c++filt > strlen(char const*) > > Is the kernel module compiled using c++ instead of cc? Here is where the symbol is coming from: [r...@xwin /usr/ports/emulators/virtualbox-ose-additions]# find . -name \*.o | while read X; do nm "$X" | grep _Z6strlenPKc && echo $X; done U _Z6strlenPKc ./work/VirtualBox-3.1.2_OSE/out/freebsd.amd64/release/obj/RuntimeGuestR0/common/misc/thread.o U _Z6strlenPKc ./work/VirtualBox-3.1.2_OSE/out/freebsd.amd64/release/obj/RuntimeGuestR0/common/log/log.o U _Z6strlenPKc ./work/VirtualBox-3.1.2_OSE/out/freebsd.amd64/release/obj/RuntimeGuestR0/common/path/RTPathAbsEx.o U _Z6strlenPKc ./work/VirtualBox-3.1.2_OSE/out/freebsd.amd64/release/obj/RuntimeGuestR0/common/path/RTPathAppend.o U _Z6strlenPKc ./work/VirtualBox-3.1.2_OSE/out/freebsd.amd64/release/obj/RuntimeGuestR0/common/string/strformattype.o U _Z6strlenPKc ./work/VirtualBox-3.1.2_OSE/out/freebsd.amd64/release/obj/RuntimeGuestR0/common/string/string.o [r...@xwin /usr/ports/emulators/virtualbox-ose-additions]# And here are some presumed source files: [r...@xwin /usr/ports/emulators/virtualbox-ose-additions]# find . -name thread.c\* ./work/VirtualBox-3.1.2_OSE/src/VBox/Runtime/common/misc/thread.cpp ./work/VirtualBox-3.1.2_OSE/src/VBox/Additions/x11/VBoxClient/thread.cpp ./work/VirtualBox-3.1.2_OSE/src/libs/xpcom18a4/nsprpub/pr/src/cplus/tests/thread.cpp [r...@xwin /usr/ports/emulators/virtualbox-ose-additions]# src/VBox/Runtime/common/misc/thread.cpp contains: #include include/iprt/string.h has this: #elif defined(RT_OS_FREEBSD) && defined(_KERNEL) # include It seems sys/libkern.h is C namespace but thread.cpp or iprt/string.h isn't using extern "C" to make it that. ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
9.1rc1: port openssl-1.0.1_4 breaks port bind98-9.8.3.2
If I install the 9.1rc1 amd64 ISO image, then cd /usr/ports/security/openssl and make install then cd /usr/ports/dns/bind98 and make, I get this warning at the end (and likely elsewhere): /usr/bin/ld: warning: libcrypto.so.6, needed by /usr/lib/libgssapi_krb5.so, may conflict with libcrypto.so.8 The resultant named does not work. From /var/log/messages: Sep 9 03:41:45 SUPERBIG named[70507]: initializing DST: openssl failure Sep 9 03:41:45 SUPERBIG named[70507]: exiting (due to fatal error) Sep 9 03:41:45 SUPERBIG root: /etc/rc.d/named: WARNING: failed to start named ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
Re: [HEADS-UP] Announcing the end of port CVS
On 9/11/2012 8:06 AM, Frank Seltzer wrote: > I haven't seen this asked so far so I'll do it. Is there a cookbook > procedure to create a local svn server similar to the one you posted > for a single machine update? > It takes less time to just follow the instructions footnoted in the original post than to read this entire thread. Since the repository UUID doesn't change with the svnsync method "svn relocate" can change an existing svn tree (i.e., /usr/src) to use the new mirror, and change to another mirror or the baseline repository later. ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
lang/tcl-modules fails to build
This is due to the September 19 change r304506 to lang/tcl86 /usr/ports/lang/tcl-modules# make ===> License BSD accepted by the user ===> Extracting for tcl-modules-8.5.12_2 => SHA256 Checksum OK for tcl8.5.12-src.tar.gz. ===> Patching for tcl-modules-8.5.12_2 ===> Applying extra patch /usr/ports/lang/tcl86/files/patch-uk-locale cannot open /usr/ports/lang/tcl86/files/patch-uk-locale: No such file or directory *** Error code 2 Stop in /usr/ports/lang/tcl-modules. ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
Re: Splitting devel/subversion into SEVERAL ports -- how fine-grained do we want to see it?
So how does port subversion work now? I don't get mod_dav_svn installed and I don't see a knob for it. There is port www/mod_dav_svn but devel/subversion doesn't seem to reference it, and www/mod_dav_svn just gives errors when apache24 tries to start.(needs shared memory support that or some such). port devel/subversion does have the mod_dav_svn code in the work tree; it just isn't installed. ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
converters/libiconv on stable10?
port converters/libiconv doesn't compile for me. It's needed by the recent update to glib20 and hence blocks samba, etc. FreeBSD STABLE10.housenet.jrv 10.0-STABLE FreeBSD 10.0-STABLE #1 r263205M: Sun Mar 16 00:29:14 UTC 2014 r...@stable10.housenet.jrv:/usr/obj/usr/src/sys/GENERIC amd64 ... cc -c -I. -I. -I.. -I../include -I./../include -I../srclib -I./../srclib -I../lib -pipe -g -DINSTALLDIR=\"/usr/local/bin\" -DLOCALEDIR=\"/usr/local/share/locale\" ./iconv_no_i18n.c /bin/sh ../libtool --mode=link cc -pipe -g iconv_no_i18n.o ../srclib/libicrt.a ../lib/libiconv.la -o iconv_no_i18n libtool: link: cc -pipe -g iconv_no_i18n.o -o .libs/iconv_no_i18n ../srclib/libicrt.a ../lib/.libs/libiconv.so -Wl,-rpath -Wl,/usr/local/lib ../lib/.libs/libiconv.so: undefined reference to `aliases2_lookup' cc: error: linker command failed with exit code 1 (use -v to see invocation) *** Error code 1 Stop. make[3]: stopped in /usr/ports/converters/libiconv/work/libiconv-1.14/src *** Error code 1 Stop. make[2]: stopped in /usr/ports/converters/libiconv/work/libiconv-1.14 *** Error code 1 Stop. make[1]: stopped in /usr/ports/converters/libiconv *** Error code 1 Stop. make: stopped in /usr/ports/converters/libiconv STABLE10:/usr/ports/converters/libiconv# cat /etc/make.conf WITH_DEBUG=1 WITHOUT_X11=1 OPTIONS_UNSET=X11 WITHOUT_PKGNG=1 NO_WARNING_PKG_INSTALL_EOL=yes STABLE10:/usr/ports/converters/libiconv# cat /etc/src.conf DEBUG_FLAGS=-g WITH_PKGTOOLS=yes #WITHOUT_CLANG_IS_CC=yes #WITH_GCC=yes #WITH_GNUCXX=yes STABLE10:/usr/ports/converters/libiconv# uname -a FreeBSD STABLE10.housenet.jrv 10.0-STABLE FreeBSD 10.0-STABLE #1 r263205M: Sun Mar 16 00:29:14 UTC 2014 r...@stable10.housenet.jrv:/usr/obj/usr/src/sys/GENERIC amd64 STABLE10:/usr/ports/converters/libiconv# svnlite info ... Revision: 348319 Node Kind: directory Schedule: normal Last Changed Author: adamw Last Changed Rev: 346968 Last Changed Date: 2014-03-03 21:15:30 + (Mon, 03 Mar 2014) STABLE10:/usr/ports/converters/libiconv# ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
net/samba36 does not build on stable10?
FreeBSD STABLE10.housenet.jrv 10.0-STABLE FreeBSD 10.0-STABLE #0
r263107M: Sun Mar 16 17:52:42 UTC 2014
r...@stable10.housenet.jrv:/usr/obj/usr/src/sys/GENERIC amd64
Samba 3.6 doesn't compile for me on stable10, yet internet searches
don't reveal any other cases similar to this?
The symptom is that compilations fail due to a linkage error: undefined
reference to `__unsafe_string_function_usage_here__'.
These seem to be the result of this code in
work/samba-3.6.23/source3/include/safe_string.h
#ifdef HAVE_COMPILER_WILL_OPTIMIZE_OUT_FNS
...
#define safe_strcpy_fn2(fn_name, fn_line, d, s, max_len) \
(CHECK_STRING_SIZE(d, max_len+1) \
? __unsafe_string_function_usage_here__() \
: safe_strcpy_fn(fn_name, fn_line, (d), (s), (max_len)))
...
HAVE_COMPILER_WILL_OPTIMIZE_OUT_FNS is a configuration-time item set in
work/samba-3.6.23/source3/wscript:
# Check if the compiler will optimize out functions
conf.CHECK_CODE('''
if (0) {
this_function_does_not_exist();
} else {
return 1;
}''', 'HAVE_COMPILER_WILL_OPTIMIZE_OUT_FNS',
msg="Checking if the compiler will optimize out functions")
A manual #undef of HAVE_COMPILER_WILL_OPTIMIZE_OUT_FNS fixes the
problem. It appears that samba assumes such dead code removal always
happens, whereas perhaps there are cases clang does not remove. Or
perhaps the CHECK_STRING_SIZE test is finding a real bug. Either way it
appears to be a samba issue that is exposed by stable10's usage of clang.
Compilation output:
...
Linking bin/smbstatus
locking/locking.o: In function `rename_share_filename':
/usr/ports/net/samba36/work/samba-3.6.23/source3/locking/locking.c:1100:
undefined reference to `__unsafe_string_function_usage_here__'
/usr/ports/net/samba36/work/samba-3.6.23/source3/locking/locking.c:1101:
undefined reference to `__unsafe_string_function_usage_here__'
/usr/ports/net/samba36/work/samba-3.6.23/source3/locking/locking.c:1102:
undefined reference to `__unsafe_string_function_usage_here__'
locking/locking.o: In function `unparse_share_modes':
/usr/ports/net/samba36/work/samba-3.6.23/source3/locking/locking.c:851:
undefined reference to `__unsafe_string_function_usage_here__'
/usr/ports/net/samba36/work/samba-3.6.23/source3/locking/locking.c:854:
undefined reference to `__unsafe_string_function_usage_here__'
locking/locking.o:/usr/ports/net/samba36/work/samba-3.6.23/source3/locking/locking.c:857:
more undefined references to `__unsafe_string_function_usage_here__' follow
lib/util_str.o: In function `string_append':
/usr/ports/net/samba36/work/samba-3.6.23/source3/lib/util_str.c:2101:
undefined reference to `__unsafe_string_function_usage_here__'
cc: error: linker command failed with exit code 1 (use -v to see invocation)
gmake[2]: *** [bin/testparm] Error 1
gmake[2]: *** Waiting for unfinished jobs
cc: error: linker command failed with exit code 1 (use -v to see invocation)
gmake[2]: *** [bin/smbstatus] Error 1
gmake[2]: Leaving directory
`/usr/ports/net/samba36/work/samba-3.6.23/source3'
===> Compilation failed unexpectedly.
Try to set MAKE_JOBS_UNSAFE=yes and rebuild before reporting the failure to
the maintainer.
*** Error code 1
Stop.
make[1]: stopped in /usr/ports/net/samba36
*** Error code 1
Stop.
make: stopped in /usr/ports/net/samba36
STABLE10:/usr/ports/net/samba36# svnlite info
...
Repository UUID: 35697150-7ecd-e111-bb59-0022644237b5
Revision: 348319
Node Kind: directory
Schedule: normal
Last Changed Author: delphij
Last Changed Rev: 347949
Last Changed Date: 2014-03-12 01:07:30 + (Wed, 12 Mar 2014)
STABLE10:/usr/ports/net/samba36# cc -v
FreeBSD clang version 3.3 (tags/RELEASE_33/final 183502) 20130610
Target: x86_64-unknown-freebsd10.0
Thread model: posix
STABLE10:/usr/ports/net/samba36# cat /etc/make.conf
WITH_DEBUG=1
WITHOUT_X11=1
OPTIONS_UNSET=X11
WITHOUT_PKGNG=1
NO_WARNING_PKG_INSTALL_EOL=yes
STABLE10:/usr/ports/net/samba36#
___
freebsd-ports@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
Re: net/samba36 does not build on stable10?
On 3/17/2014 1:54 AM, Dimitry Andric wrote: > Try removing WITH_DEBUG=1 from your make.conf. This causes the samba > port to be compiled without any optimization (e.g. -O0). Only use the > WITH_DEBUG option for specific ports, not as a general setting. Is there a general purpose knob for adding -g and preventing stripping? DEBUG_FLAGS seems ineffective, at least in multimedia/handbrake. ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
i386 emacs run-time dependency on gcc?
I realize emacs needs to be built with gcc on i386, but should there be a run-time dependency too? The emacs package tarball requires gcc to install. ___ freebsd-ports@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
