- Add LIBS="${LIBS}" to MAKE_ENV and CONFIGURE_ENV.
- Add an option helper for LIBS.
- Adjust all ports that already use LIBS. Also remove references to
PTHREAD_CFLAGS and PTHREAD_LIBS while here.
- Some ports did not support having a LIBS environment variable and
required additional patches.
Somewhat simplified a linker command line looks like:
${CC} ${src_LDFLAGS} ${LDFLAGS} ${src_LIBS} ${LIBS}
where src_LDFLAGS and src_LIBS are controlled by upstream and LDFLAGS and
LIBS can be controlled by us. If possible -L and -l flags need to be
added to LIBS to make sure they appear after any -L and -l flags set by
upstream. Many ports currently add -L${LOCALBASE}/lib to LDFLAGS but this
may appear too early on the command line causing installed libraries to
be linked in instead of freshly built ones.
Additional changes:
benchmarks/netio: Replace WITH_IPV6 with an IPV6 option.
comms/gnokii: Replace some patches with USES=pathfix. Also remove -fPIC.
graphics/gimageview: USES=libtool and install desktop file in DESKTOPDIR.
graphics/visionworkbench: Remove FreeBSD 7 support.
multimedia/libmovtar: New LIB_DEPENDS syntax.
multimedia/opencinematools: Use standard do-build.
net/siproxd: USES=libtool:keepla (port actually needs .la files for plugins)
net-mgmt/nagios: Remove -fPIC.
net-mgmt/nagios4: Remove -fPIC.
print/cups-base: Only add -lssp_nonshared on i386 and OSVERSION < 1000036.
security/p11-kit: Replace PTHREAD_LIBS in CONFIGURE_ENV with
ac_cv_func_pthread_mutexattr_init=no in CONFIGURE_ARGS. This skips a test
in configure that falsely detects pthread_mutexattr_init in our libc.
sysutils/dar: Fix iconv detection.
x11/rxvt-unicode: Remove -lstdc++ and patch configure to remove a FreeBSD
hack and use $CXX as linker as on other platforms.
PR: 190592
Exp-run by: antoine
Approved by: portmgr (antoine)
- support staging, strip libraries
- trim master_site
- use option helpers
- shift away from inclusion of bsd.port.pre.mk as bsd.port.options.mk
can test OSVERSION (less expensive)
- patch of Makefile was merged upstream so removed patch
PR: 185599
Submitted by: jgh@ (myself)
Approved by: maintainer timeout, jharris@widomaker.com ( 15 days )
was detected during build [1]
- Mark MAKE_JOBS_UNSAFE:
help2man: can't get '--help' info from man/stty.td/stty
gmake[2]: *** [man/stty.1] Error 127
gmake[2]: *** Waiting for unfinished jobs....
PR: ports/173218 [1]
Submitted by: Jason Harris <jharris@widomaker.com> (maintainer) [1]
Feature safe: yes
- Correct REINPLACE for permissions; \+ is a special character in a BRE
PR: ports/160060
Submitted by: Veniamin Gvozdikov <g.veniamin@googlemail.com>
Reviewed by: Jason Hellenthal, wxs, eadler
Approved by: Jason Harris <jharris@widomaker.com> (maintainer)
Add optional dependency on math/gmp and disable gmp support
properly when it's not needed.
PR: ports/146027
Reported by: Craig Leres <leres at ee.lbl.gov>
- OSVERSION check should be against 800039 or higher, as that's when
the sgtty changes were committed.
http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/sys/param.h.diff?r1=1.355;r2=1.356;f=h
- Fix stty.c patch. Previous version would result in termios bits
being overridden by #include <sys/ioctl_compat.h>, causing coreutils'
stty to modify bits different than /bin/stty. This version removes
tab1/tab2 delay support, since tab1/tab2 delays aren't available in HEAD
outside of the binary compatibility interface. Also fixes stty --help
usage to indicate only tab0 and tab3 are available.
termios bits being undefined. They've been moved to sys/ioctl_compat.h which
requires _KERNEL be defined. Somewhat of an ugly hack; possible alternative
would be to remove use of TAB1 and TAB2 in src/stty.c altogether.
Reported by: jsa @ EFnet
Specifically, newer autoconf (> 2.13) has different semantic of the
configure target. In short, one should use --build=CONFIGURE_TARGET
instead of CONFIGURE_TARGET directly. Otherwise, you will get a warning
and the old semantic may be removed in later autoconf releases.
To workaround this issue, many ports hack the CONFIGURE_TARGET variable
so that it contains the ``--build='' prefix.
To solve this issue, under the fact that some ports still have
configure script generated by the old autoconf, we use runtime detection
in the do-configure target so that the proper argument can be used.
Changes to Mk/*:
- Add runtime detection magic in bsd.port.mk
- Remove CONFIGURE_TARGET hack in various bsd.*.mk
- USE_GNOME=gnometarget is now an no-op
Changes to individual ports, other than removing the CONFIGURE_TARGET hack:
= pkg-plist changed (due to the ugly CONFIGURE_TARGET prefix in * executables)
- comms/gnuradio
- science/abinit
- science/elmer-fem
- science/elmer-matc
- science/elmer-meshgen2d
- science/elmerfront
- science/elmerpost
= use x86_64 as ARCH
- devel/g-wrap
= other changes
- print/magicfilter
GNU_CONFIGURE -> HAS_CONFIGURE since it's not generated by autoconf
Total # of ports modified: 1,027
Total # of ports affected: ~7,000 (set GNU_CONFIGURE to yes)
PR: 126524 (obsoletes 52917)
Submitted by: rafan
Tested on: two pointyhat 7-amd64 exp runs (by pav)
Approved by: portmgr (pav)
Cleanup conflicts in this port (fileutils, sh-utils and
textutils don't exist anymore)
Provide a regression-test target to determine the convenience
of updating this port. Add commented USE_BISON line as it
is recommended but not required, maybe future versions will
need it.
Update the WWW.
No functional change.
PR: ports/124623
Submitted by: "Pedro F. Giffuni" <pfgshield-freebsd@yahoo.com>
Approved by: maintainer timeout
The affected ports are the ones with gettext as a run-dependency
according to ports/INDEX-7 (5007 of them) and the ones with USE_GETTEXT
in Makefile (29 of them).
PR: ports/124340
Submitted by: edwin@
Approved by: portmgr (pav)
supports them. This is determined by running ``configure --help'' in
do-configure target and set the shell variable _LATE_CONFIGURE_ARGS
which is then passed to CONFIGURE_ARGS.
- Remove --mandir and --infodir in ports' Makefile where applicable
Few ports use REINPLACE_CMD to achieve the same effect, remove them too.
- Correct some manual pages location from PREFIX/man to MANPREFIX/man
- Define INFO_PATH where necessary
- Document that .info files are installed in a subdirectory relative to
PREFIX/INFO_PATH and slightly change add-plist-info to use INFO_PATH and
subdirectory detection.
PR: ports/111470
Approved by: portmgr
Discussed with: stas (Mk/*), gerald (info related stuffs)
Tested by: pointyhat exp run
- Update to GNU coreutils 6.9
- Eliminate configure warning
- Sort plist
While I am here:
- Remove unsupported checksum algorithm SHA1 and RMD160
PR: ports/111489
Submitted by: "Eric P. Scott" <eps+pbug0704 AT ana.com>
Approved by: Jason Harris <jharris AT widomaker.com> (maintainer)
was supposed to work is useless, because if we can't trust the distfile from
the remote machine, we can't trust the signature from the same machine either.
Our MD5 and SHA256 are good for checking both the sanity and the
trustiness of distfiles.
Approved by: portmgr (erwin), erwin (mentor)
