From: "Marco van Hulten" Subject: Re: [maintainer update] math/hdf5 2.2.0 To: "Martin Reindl" , Date: Fri, 11 Sep 2026 14:29:23 +0200 On Fri Aug 21, 2026 at 8:38 PM CEST, Martin Reindl wrote: > Diff below updates hdf5 to 2.2.0. > > Note that upstream has dropped the autotools build, so this is > now cmake only. > > I have compile tested most of the following direct users of hdf5: > > cad/freecad > [...] > > Any additional testing (and of course OKs) is appreciated. There is a problem with patching (on arm64): ... ===> Installing gcc-16.1.0p11 from /usr/ports/packages/aarch64/all/ gcc-16.1.0p11: ok ===> Returning to build of hdf5-2.2.0 ===> hdf5-2.2.0 depends on: gcc->=16,<17 -> gcc-16.1.0p11 ===> hdf5-2.2.0 depends on: gfortran->=16,<17 - not found ===> Verifying install for gfortran->=16,<17 in lang/gcc/16 ===> gfortran-16.1.0p11 depends on: gcc->=16,<17 -> gcc-16.1.0p11 ===> Verifying specs: c gmp m mpc mpfr z ===> found c.104.0 gmp.11.1 m.10.1 mpc.1.1 mpfr.2.2 z.7.2 ===> Installing gfortran-16.1.0p11 from /usr/ports/packages/aarch64/all/ gfortran-16.1.0p11: ok ===> Returning to build of hdf5-2.2.0 ===> hdf5-2.2.0 depends on: gfortran->=16,<17 -> gfortran-16.1.0p11 ===> hdf5-2.2.0 depends on: libaec-* -> libaec-1.1.4 ===> hdf5-2.2.0 depends on: gcc-libs->=16,<17 -> gcc-libs-16.1.0p9 ===> Verifying specs: c m c++ c++abi pthread z aec sz gfortran>=11 ===> found c.104.0 m.10.1 c++.13.0 c++abi.10.0 pthread.28.1 z.7.2 aec.0.0 sz.0.0 gfortran.11.0 ===> Extracting for hdf5-2.2.0 ===> Patching for hdf5-2.2.0 ===> Applying OpenBSD patch patch-m4_aclocal_fc_f90 Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |Already fixed upstream. | |Index: m4/aclocal_fc.f90 |--- m4/aclocal_fc.f90.orig |+++ m4/aclocal_fc.f90 -------------------------- File to patch: No file found--skip this patch? [y] Should the patches be skipped/removed? Without the patches: The following tests FAILED: 30 - H5TEST-dsets (Failed) 2704 - FORTRAN_testhdf5_fortran (Failed) 2705 - FORTRAN_testhdf5_fortran_1_8 (Failed) Errors while running CTest Output from these tests are in: /usr/ports/pobj/hdf5-2.2.0/build-aarch64/Testing/Temporary/LastTest.log Use "--rerun-failed --output-on-failure" to re-run the failed cases verbosely. *** Error 8 in . (/usr/ports/devel/cmake/cmake.port.mk:53 'do-test': @cd /usr/ports/pobj/hdf5-2.2.0/build-aarch64 && exec /usr/bin/env -i LI...) *** Error 2 in . (/usr/ports/infrastructure/mk/bsd.port.mk:3108 '/usr/ports/pobj/hdf5-2.2.0/build-aarch64/.test_done': @cd /usr/ports/math/h...) *** Error 2 in /usr/ports/math/hdf5 (/usr/ports/infrastructure/mk/bsd.port.mk:2722 'test': @lock=hdf5-2.2.0; export _LOCKS_HELD=" hdf5-2.2....) Marco