Summary: | =app-emacs/ocaml-mode-4.00.1 =dev-lang/ocaml-4.02.3 =dev-ml/camldbm-1.0 =dev-ml/camlp4-4.02.1_p3 =dev-ml/camlp5-6.14 =dev-ml/labltk-8.06.0 =dev-ml/findlib-1.5.6 stable request | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Paweł Hajdan, Jr. (RETIRED) <phajdan.jr> |
Component: | Stabilization | Assignee: | Gentoo Team for the ML programming language family <ml> |
Status: | RESOLVED FIXED | ||
Severity: | enhancement | CC: | emeric.maschino, gnu-emacs, pacho, toralf |
Priority: | Normal | Keywords: | STABLEREQ |
Version: | unspecified | Flags: | kensington:
sanity-check+
|
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | last EAPI=1 ebuild! | ||
Package list: |
=app-emacs/ocaml-mode-4.00.1
=dev-lang/ocaml-4.02.3
=dev-ml/camlp4-4.02.1_p3
=dev-ml/labltk-8.06.0
=dev-ml/findlib-1.5.6
|
Runtime testing required: | --- |
Bug Depends on: | 564660 | ||
Bug Blocks: | 512122, 564404, 586914, 586916, 604934 |
Description
Paweł Hajdan, Jr. (RETIRED)
2013-03-24 20:25:44 UTC
This shares its tarball with dev-lang/ocaml, so preferably we shouldn't stabilise it at the same time. When will ocaml-4 be stabilized? (In reply to Pacho Ramos from comment #2) > When will ocaml-4 be stabilized? it could go now, but i'd like to see a tinderbox run first; ocaml breaks a lot between versions :/ (In reply to Alexis Ballier from comment #3) > (In reply to Pacho Ramos from comment #2) > > When will ocaml-4 be stabilized? > > it could go now, but i'd like to see a tinderbox run first; ocaml breaks a > lot between versions :/ For that maybe Toralf could help us: we would need a stable tinderbox with this newer ocaml to see what packages need to be fixed or stabilized with this Thanks a lot :) (In reply to Pacho Ramos from comment #4) > For that maybe Toralf could help us: we would need a stable tinderbox with > this newer ocaml to see what packages need to be fixed or stabilized with > this > > Thanks a lot :) No problem, would just need a keyword file - attached here - or is app-emacs/ocaml-mode-4.00.1 the only one ? (In reply to Toralf Förster from comment #5) > (In reply to Pacho Ramos from comment #4) > > > For that maybe Toralf could help us: we would need a stable tinderbox with > > this newer ocaml to see what packages need to be fixed or stabilized with > > this > > > > Thanks a lot :) > > No problem, would just need a keyword file - attached here - or is > app-emacs/ocaml-mode-4.00.1 the only one ? thanks dev-lang/ocaml-4.02.1 is the only one A stable chroot image is setup and now running with these 2 packages accepted : $ cat amd64-stable_20150624-154436/etc/portage/package.accept_keywords/ocaml ~app-emacs/ocaml-mode-4.00.1 ~dev-lang/ocaml-4.02.1 Are there any package you'd like to see to be emerged first ? Otherwise just the 1-liner in [1] would be run after the first world update. [1] http://www.zwiebeltoralf.de/tinderbox/index.html (In reply to Toralf Förster from comment #7) > Are there any package you'd like to see to be emerged first ? Otherwise just > the 1-liner in [1] would be run after the first world update. everything (r)depending on ocaml, but i guess this would work too :) (In reply to Alexis Ballier from comment #8) > (In reply to Toralf Förster from comment #7) > > Are there any package you'd like to see to be emerged first ? Otherwise just > > the 1-liner in [1] would be run after the first world update. > > everything (r)depending on ocaml, but i guess this would work too :) Ok, putted those packages on top of the last : find . -type f -name '*.ebuild' | xargs grep -Hl -e dev-lang/ocaml -e app-emacs/ocaml | xargs -n1 dirname | cut -f2- -d '/' | sort -u >> /tmp/packages and got the first fall out, or ? : * Package: sci-mathematics/coq-8.4_p1 * Repository: gentoo * Maintainer: ml@gentoo.org sci-mathematics@gentoo.org * USE: abi_x86_64 amd64 elibc_glibc kernel_linux ocamlopt userland_GNU * FEATURES: preserve-libs sandbox userpriv usersandbox >>> Unpacking source... >>> Unpacking coq-8.4pl1.tar.gz to /var/tmp/portage/sci-mathematics/coq-8.4_p1/work >>> Source unpacked in /var/tmp/portage/sci-mathematics/coq-8.4_p1/work >>> Preparing source in /var/tmp/portage/sci-mathematics/coq-8.4_p1/work/coq-8.4pl1 ... * Applying coq-8.4_p1-camlp4.patch ... [ ok ] >>> Source prepared. >>> Configuring source in /var/tmp/portage/sci-mathematics/coq-8.4_p1/work/coq-8.4pl1 ... GNU Make >= 3.81 is needed. Make 3.81 can be downloaded from ftp://ftp.gnu.org/gnu/make/make-3.81.tar.gz then locally installed on a Unix-style system by issuing: tar xzvf make-3.81.tar.gz cd make-3.81 ./configure make mv make .. cd .. Restart then the configure script and later use ./make instead of make. * ERROR: sci-mathematics/coq-8.4_p1::gentoo failed (configure phase): * configure failed * (In reply to Toralf Förster from comment #9) > (In reply to Alexis Ballier from comment #8) > > (In reply to Toralf Förster from comment #7) > > > Are there any package you'd like to see to be emerged first ? Otherwise just > > > the 1-liner in [1] would be run after the first world update. > > > > everything (r)depending on ocaml, but i guess this would work too :) > > Ok, putted those packages on top of the last : > find . -type f -name '*.ebuild' | xargs grep -Hl -e dev-lang/ocaml -e > app-emacs/ocaml | xargs -n1 dirname | cut -f2- -d '/' | sort -u >> > /tmp/packages this wont work as some (maybe most) packages depend on ocaml via an eclass i think > and got the first fall out, or ? : > > * Package: sci-mathematics/coq-8.4_p1 > * Repository: gentoo > * Maintainer: ml@gentoo.org sci-mathematics@gentoo.org > * USE: abi_x86_64 amd64 elibc_glibc kernel_linux ocamlopt > userland_GNU > * FEATURES: preserve-libs sandbox userpriv usersandbox > >>> Unpacking source... > >>> Unpacking coq-8.4pl1.tar.gz to /var/tmp/portage/sci-mathematics/coq-8.4_p1/work > >>> Source unpacked in /var/tmp/portage/sci-mathematics/coq-8.4_p1/work > >>> Preparing source in /var/tmp/portage/sci-mathematics/coq-8.4_p1/work/coq-8.4pl1 ... > * Applying coq-8.4_p1-camlp4.patch ... > [ ok ] > >>> Source prepared. > >>> Configuring source in /var/tmp/portage/sci-mathematics/coq-8.4_p1/work/coq-8.4pl1 ... > GNU Make >= 3.81 is needed. > Make 3.81 can be downloaded from ftp://ftp.gnu.org/gnu/make/make-3.81.tar.gz > then locally installed on a Unix-style system by issuing: > tar xzvf make-3.81.tar.gz > cd make-3.81 > ./configure > make > mv make .. > cd .. > Restart then the configure script and later use ./make instead of make. > * ERROR: sci-mathematics/coq-8.4_p1::gentoo failed (configure phase): > * configure failed > * this one probably needs to go stable, i think it is due to make-4.x; please file a new bug for it Okey, so here is the (hopefuly) complete list for ocaml 4.02.3: =app-emacs/ocaml-mode-4.00.1 =dev-lang/ocaml-4.02.3 =dev-ml/camldbm-1.0 =dev-ml/camlp4-4.02.1_p3 =dev-ml/camlp5-6.14 =dev-ml/labltk-8.06.0 =dev-ml/findlib-1.5.6 it's the usual 2 (ocaml + camlp5), those new ones that are split (camldbm, camlp4, labltk), findlib to support split packages, ocaml-mode for emacs. I have not been able to reproduce bug #544176, if anyone can trigger it and confirm that it is just a parallel make problem, feel free to add -j1 to emake calls as this is the case with current stable Stable for HPPA. Stable for PPC64. amd64 stable x86 stable ppc stable on ia64 I can't go ahead because of: dependency.bad [fatal] 14 dev-ml/camldbm/camldbm-1.0.ebuild: DEPEND: ia64(default/linux/ia64/13.0) ['>=dev-lang/ocaml-3.12:=[ocamlopt]'] (In reply to Agostino Sarubbo from comment #17) > on ia64 I can't go ahead because of: > > dependency.bad [fatal] 14 > dev-ml/camldbm/camldbm-1.0.ebuild: DEPEND: ia64(default/linux/ia64/13.0) > > ['>=dev-lang/ocaml-3.12:=[ocamlopt]'] hmm, right, it's not even keyworded there: drop camldbm from the list; nothing on ia64 should need it. arm stable This needs some work: RepoMan scours the neighborhood... dependency.bad [fatal] 28 dev-ml/camldbm/camldbm-1.0.ebuild: DEPEND: alpha(default/linux/alpha/13.0) ['>=dev-lang/ocaml-3.12:=[ocamlopt]'] The ocamlopt USE flag is masked for ocaml on alpha, ppc64ul, mips, ia64 and hppa -- but that should apply to camldbm just as it does for the ocaml ebuild. Is repoman broken here? Please advise. (In reply to Tobias Klausmann from comment #20) > This needs some work: > > RepoMan scours the neighborhood... > dependency.bad [fatal] 28 > dev-ml/camldbm/camldbm-1.0.ebuild: DEPEND: alpha(default/linux/alpha/13.0) > ['>=dev-lang/ocaml-3.12:=[ocamlopt]'] > > The ocamlopt USE flag is masked for ocaml on alpha, ppc64ul, mips, ia64 and > hppa -- but that should apply to camldbm just as it does for the ocaml > ebuild. Is repoman broken here? camldbm just doesnt properly support disabling ocamlopt. Please skip it on alpha (and other arches without ocamlopt support), nothing should need it. (In reply to Alexis Ballier from comment #21) > camldbm just doesnt properly support disabling ocamlopt. Please skip it on > alpha (and other arches without ocamlopt support), nothing should need it. Stable on alpha: =app-emacs/ocaml-mode-4.00.1 =dev-lang/ocaml-4.02.3 =dev-ml/camlp4-4.02.1_p3 =dev-ml/camlp5-6.14 =dev-ml/labltk-8.06.0 =dev-ml/findlib-1.5.6 @ia64, sparc: Friendly ping :) (In reply to Manuel Rüger from comment #23) > @ia64, sparc: Friendly ping :) What about simply use.stable.masking dev-lang/ocaml[emacs] and dropping ocaml-mode to ~sparc ~ia64 ? Ah no, it's the full ocaml ebuild that is eapi1 :S For ia64 maybe Emeric can help us for the arch testing (thanks!). For sparc I have no idea about how to fix it (I would probably opt for blindly stabilizing it and waiting them to complain... or maybe we should consider to move sparc to the status of m68k/s390/sh arches... that are stabilized by the people who care when they need it) (In reply to Pacho Ramos from comment #26) > For ia64 maybe Emeric can help us for the arch testing (thanks!). For sparc > I have no idea about how to fix it (I would probably opt for blindly > stabilizing it and waiting them to complain... or maybe we should consider > to move sparc to the status of m68k/s390/sh arches... that are stabilized by > the people who care when they need it) =app-emacs/ocaml-mode-4.00.1 =dev-lang/ocaml-4.02.3 =dev-ml/camlp4-4.02.1_p3 =dev-ml/labltk-8.06.0 =dev-ml/findlib-1.5.6 All emerged successfully, with tests enabled. But besides checking that they all build fine and pass test-suite, is there anything else I can do to ensure these would be really stable on ia64? BTW, =dev-ml/camlp5-6.14 needs ia64 keyword. Should I try this one too? Just let me know, Émeric (In reply to Émeric Maschino from comment #27) > All emerged successfully, with tests enabled. But besides checking that they > all build fine and pass test-suite, is there anything else I can do to > ensure these would be really stable on ia64? For app-emacs/ocaml-mode, we have a test plan: https://wiki.gentoo.org/wiki/Project:Emacs/Test_plans#app-emacs.2Focaml-mode (In reply to Ulrich Müller from comment #28) > For app-emacs/ocaml-mode, we have a test plan: > https://wiki.gentoo.org/wiki/Project:Emacs/Test_plans#app-emacs.2Focaml-mode No problem detected. Émeric Dear Maintainer (or who is mainly involved in this stable request), This is an auto-generated message that will move the current component to the new component Stabilization. To ensure that the stabilization will proceed correctly, please fill the fields "Atoms to stabilize" and "Runtime testing required" as described here: https://archives.gentoo.org/gentoo-dev/message/4b2ef0e9aa7588224b8ae799c5fe31fa sparc stable ia64 stable |