2nd (minimal) set of extra dev-qt/* needed for USE=qt6 unmasking on arches where qtbase:6 is available. Will leave other packages for later, want to minimize issues at the moment. Does not count qutebrowser[qt6] and gpsbabel[qt6] due to qtwebengine. Leaving that up to calibre's bug #907080 instead given similar deps. sci-astronomy/stellarium[webengine] will get a mask for ppc64 instead, not going to consider that at the moment as it likely need patches. sparc and riscv missed the boat and are excluded for now. ppc64 should be coming soon wrt bug #914026 given issues should be resolved. assimp is currently masked on big-endian so nattka should complain, waiting to hear of what's up with that (needed by qtmultimedia->qtquick3d->assimp chain, albeit qtquick3d could be avoided by masking qtmultimedia[qml]).
Unable to check for sanity: > package masked: media-libs/assimp-5.2.5-r1, in all profiles for arch: hppa
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=7a240fe7e12286d0983888bf79bad5c3ca0f8f26 commit 7a240fe7e12286d0983888bf79bad5c3ca0f8f26 Author: Matoro Mahri <matoro_gentoo@matoro.tk> AuthorDate: 2023-11-30 15:44:27 +0000 Commit: Ionen Wolkens <ionen@gentoo.org> CommitDate: 2023-11-30 15:49:45 +0000 profiles/features/big-endian: unmask media-libs/assimp Bug: https://bugs.gentoo.org/918896 Signed-off-by: Matoro Mahri <matoro_gentoo@matoro.tk> Signed-off-by: Ionen Wolkens <ionen@gentoo.org> profiles/features/big-endian/package.mask | 1 - 1 file changed, 1 deletion(-)
All sanity-check issues have been resolved
Adding sparc
Think can go forward with this, will consolidate what's left for riscv in a later bug.
arm done
arm64 done
Adding riscv.
ppc done
ppc64 done
riscv done
x86 done
assimp test fail on hppa [----------] 3 tests from utglTFImportExport [ RUN ] utglTFImportExport.importglTFFromFileTest Info, T0: Load /var/tmp/portage/media-libs/assimp-5.2.5-r1/work/assimp-5.2.5/test/models/glTF/TwoBoxes/TwoBoxes.gltf Debug, T0: Assimp 5.2.0 <unknown architecture> gcc debug singlethreadedsingle : Info, T0: Found a possible importer: glTF Importer; trying signature-based detection Info, T0: Found a matching importer for this file format: glTF Importer. Info, T0: Import root directory is '/var/tmp/portage/media-libs/assimp-5.2.5-r1/work/assimp-5.2.5/test/models/glTF/TwoBoxes/' ai_assert failure in /var/tmp/portage/media-libs/assimp-5.2.5-r1/work/assimp-5.2.5/code/AssetLib/glTF/glTFImporter.cpp(441): validRes /var/tmp/portage/media-libs/assimp-5.2.5-r1/temp/environment: line 2136: 23 Aborted "${BUILD_DIR}/bin/unit"
(In reply to matoro from comment #13) > assimp test fail on hppa Please file separate bugs for test failures and block this one. If nobody figures out what's up with it, I could mask USE=qml on qtmultimedia for hppa as a temporary solution though (removes the need for qtquick3d->assimp).
(In reply to Ionen Wolkens from comment #14) > (In reply to matoro from comment #13) > > assimp test fail on hppa > Please file separate bugs for test failures and block this one. > > If nobody figures out what's up with it, I could mask USE=qml on > qtmultimedia for hppa as a temporary solution though (removes the need for > qtquick3d->assimp). fwiw neither qt3d nor qtquick3d (both need assimp) were ever keyworded for hppa, nor were any packages depending on qtmultimedia:5[qml], so it shouldn't be much of an issue to leave it masked too. Having a bugref for the mask be nice though.
Hm, alternatively I could do add a USE=spatialaudio and split off quick3d from qml. Guess that sounds better.
(In reply to Ionen Wolkens from comment #16) > Hm, alternatively I could do add a USE=spatialaudio and split off quick3d > from qml. Guess that sounds better. Or no.. I think that's messy, spatialaudio support is there by default but would be missing only for qml (more specifically it should be qml-spatialaudio or so).
loong done
hppa done
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=5b8f72a0333f89ac5d03dfdeaf5718aed9b0cb5a commit 5b8f72a0333f89ac5d03dfdeaf5718aed9b0cb5a Author: Ionen Wolkens <ionen@gentoo.org> AuthorDate: 2024-01-27 05:26:37 +0000 Commit: Ionen Wolkens <ionen@gentoo.org> CommitDate: 2024-01-27 06:46:41 +0000 profiles: re-arrange USE=qt6 masks, unmask where possible Follow-up to bug #918896 which keyworded requirements for unmasking on most arches (sparc is still pending, so remains masked for now). Note that still use.stable.masked where needed, stabilizations will wait a bit to see if any trouble arises in ~arch from unmasking. If add new IUSE=qt6 and missing a dependency, ideally please file (re)keywordreqs as needed rather than mask (unless need qtwebengine). Bug: https://bugs.gentoo.org/918896 Signed-off-by: Ionen Wolkens <ionen@gentoo.org> profiles/arch/alpha/use.mask | 1 + profiles/arch/amd64/package.use.mask | 4 ++++ profiles/arch/amd64/package.use.stable.mask | 4 ++++ profiles/arch/amd64/use.mask | 6 ------ profiles/arch/arm/use.stable.mask | 4 ++++ profiles/arch/arm64/package.use.mask | 4 ++++ profiles/arch/arm64/package.use.stable.mask | 4 ++++ profiles/arch/arm64/use.stable.mask | 4 ++++ profiles/arch/base/package.use.mask | 3 +++ profiles/arch/hppa/use.stable.mask | 4 ++++ profiles/arch/ia64/use.mask | 1 + profiles/arch/m68k/use.mask | 1 + profiles/arch/mips/use.mask | 1 + profiles/arch/powerpc/ppc64/64le/package.use.mask | 6 ++++++ profiles/arch/powerpc/ppc64/64le/package.use.stable.mask | 4 ---- profiles/arch/powerpc/use.stable.mask | 4 ++++ profiles/arch/s390/use.mask | 1 + profiles/arch/sparc/use.mask | 6 +++++- profiles/arch/x86/use.stable.mask | 4 ++++ profiles/base/use.mask | 7 ------- 20 files changed, 55 insertions(+), 18 deletions(-)
If more packages end up causing sparc trouble, may mask USE=qt6 on specific packages to reduce the depgraph. Main goal here is just to get rid of the USE=qt6 global masks, other needs can be revisited later.
Reducing list due to qtquicktimeline and qtpositioning apparently failing on sparc. Latter was keyworded on sparc for qt5 but is not currently needed, former was actually meant to be removed alongside qtquick3d.
I've opened https://github.com/qt/qt5compat/issues/3 upstream w.r.t. qt5compat, but I'm skeptical it will get an answer.
(In reply to matoro from comment #23) > I've opened https://github.com/qt/qt5compat/issues/3 upstream w.r.t. > qt5compat, but I'm skeptical it will get an answer. I think you may need to file a bug on their Jira.
(In reply to Sam James from comment #24) > (In reply to matoro from comment #23) > > I've opened https://github.com/qt/qt5compat/issues/3 upstream w.r.t. > > qt5compat, but I'm skeptical it will get an answer. > > I think you may need to file a bug on their Jira. Copied into https://bugreports.qt.io/browse/QTBUG-122928
self-reminder: qtdeclarative[svg] has a new dependency on qtsvg:6, masked it on sparc for now
