It is time.
Arch: please at least test with USE="test-full", this will enable more tests in archs that supports that.
Arch: valngrind USE can be also added for extra tests.
Hi Arch, Please stabilize. Known issues: 1. for tests you should install guile-2 and rebuild autogen manually as it has no subslot operator, see bug#615008. 2. glib-networking has known test issue bug#595952. To help with testing I added USE="test-full" stable masked which will pull extra dependencies to allow verbose testing, you may also use USE="valngrind" if supported at your platform. sparc is excluded at this point due to bug#613418, waiting for a machine. Thanks!
I feel a change in the Matrix. commit 63a8c991d684c812aa38d679a2fd14c402674268 Author: Alon Bar-Lev <alonbl@gentoo.org> Date: Mon Apr 24 10:55:46 2017 +0300 net-libs/gnutls: move >=dev-scheme/guile-2 to test-full avoid block stable Package-Manager: Portage-2.3.3, Repoman-2.3.1
(In reply to Jeroen Roovers from comment #4) > I feel a change in the Matrix. I do not understand, is there any action item? recommendation?
(In reply to Alon Bar-Lev from comment #5) > (In reply to Jeroen Roovers from comment #4) > > I feel a change in the Matrix. > > I do not understand, is there any action item? recommendation? No, that change is fine as it helps the stabilisation process, especially since media-sound/lilypond depends on <dev-scheme/guile-2.0.
Most arches here aren't doing stabilization as long as there are open depend bugs that are not keyword/stabling bugs with arches CCed. So you should clear out the depends list from stuff that shouldn't block stabilization, if you want things to proceed. This is how the script used by most works - open dependent bug - no stabling with the --no-depends flag most arch devs pass to it that use it. This is also very logical even if checked manually without reading all the comments.
(In reply to Mart Raudsepp from comment #7) > Most arches here aren't doing stabilization as long as there are open depend > bugs that are not keyword/stabling bugs with arches CCed. So you should > clear out the depends list from stuff that shouldn't block stabilization, if > you want things to proceed. This is how the script used by most works - open > dependent bug - no stabling with the --no-depends flag most arch devs pass > to it that use it. This is also very logical even if checked manually > without reading all the comments. ok, done. leaving out big-endian for bug#613418 which may be nettle issue. leaving out ia64 for as pending stabilization of bug#612348 for a long time. I am hopping that we can proceed for the popular architectures.
amd64 stable
(In reply to Alon Bar-Lev from comment #8) > leaving out ia64 for as pending stabilization of bug#612348 for a long time. You didn't remove ia64 from CC though? That depend was fine though if the intention was to block ia64 until they have done it. If it's a dependent bug, for the script it will only block for architectures that are still CC'ed on the bug (if the bug is in the stabilization, keywording or appropriate security component), so for everyone but ia64 that wouldn't block anything either with the script (though if no arches are CC'ed, then it does block).
(In reply to Mart Raudsepp from comment #10) > (In reply to Alon Bar-Lev from comment #8) > > leaving out ia64 for as pending stabilization of bug#612348 for a long time. > > You didn't remove ia64 from CC though? I was sure I did. ok, I re-added dependency.
x86 stable
Stable on alpha.
ppc stable
ppc64 stable
ia64 stable
arm stable
@sparc, @hppa: ping? bug#626010 and this one should resolve the issue and allow us to drop the old gnutls.
hppa stable
sparc stable (thank to Dakon)
