The combination of a modern compiler with antique binutils isn't expected
in the ecosystem. On sparc64, LTO is enabled because the ports-gcc supports
it which then emits "e" sections which are to be excluded from executables
and shared objects by the link editor, such as
.section .gnu.lto_.profile.5712db33fb59d8ee,"e",@progbits
As reported by kmos, Fred Flintstone's gas trips over these with
Fatal error: unrecognized .section attribute: want a,w,x,M,S,G,T
no longer needed since 6.9 was dropped from quirks. (experience has shown
that having these around when not needed causes some confusion when using
another port as a template for a new one; not least by me ;)
../src/_backend_agg_wrapper.cpp:253:17: error: non-constant-expression cannot be narrowed from type 'unsigned int' to 'long' in initializer list [-Wc++11-narrowing]
numpy are no longer ABI-compatible following the update to numpy 2.x.
this may over-bump slightly, but I've seen problems with at least pandas
and scipy, and identifying others individually is too time-consuming.
- switches from previously bundled qhull to depending on math/qhull
- no more jquery-ui
- tests turned off until someone has time to make them work
ok sthen@
setuptools (it's used as a package locator but importlib.metadata in
newer Python core or the external importlib_metadata are preferred).
So drop the RDEP in that case (it's still kept for py27) and bump
revisions.
(which is not) throughout the ports Makefiles.
* Replace find|xargs with find -exec {} +
* Replace -exec {} \; with -exec {} + if applicable.
* Use the -delete operator to remove files and empty directories.
* Combine and tweak some find(1) invocations while here.
ok kn@ rsadowski@ espie@
lang/python port module. I've not yet come up with a port that
would not need this and one can always set MODPY_TESTDEP to "no"
to prevent the module from touching TEST_DEPENDS.
Idea from afresh1 who pointed out the cpan module already does this.
aja "I support this move."
OK sthen@
naddy@ and ajacoutot@, independently from each other, found that the
recent update of py-matplotlib does not build. Issue is caused by me not
noticing that matplotlib tries to download jquery-ui-1.12.1, which fails
when building in a chroot using dpb. Instead of downloading jquery-ui
use the bundled files.
REVISION is not bumped as patched file is not included in the package,
and no new files are added.
some existing COMPILER lines with arch restrictions etc. In the usual
case this is now using "COMPILER = base-clang ports-gcc base-gcc" on
ports with c++ libraries in WANTLIB.
This is basically intended to be a noop on architectures using clang
as the system compiler, but help with other architectures where we
currently have many ports knocked out due to building with an unsuitable
compiler -
- some ports require c++11/newer so the GCC version in base that is used
on these archirtectures is too old.
- some ports have conflicts where an executable is built with one compiler
(e.g. gcc from base) but a library dependency is built with a different
one (e.g. gcc from ports), resulted in mixing incompatible libraries in the
same address space.
devel/gmp is intentionally skipped as it's on the path to building gcc -
the c++ library there is unused in ports (and not built by default upstream)
so intending to disable building gmpcxx in a future commit.
tb