Qt 6.8 status
* Qt 6.8 feature freeze is in effect
* Branching from 'dev' to '6.8' is done
* Qt 6.8 API change review is started, see
https://lists.qt-project.org/pipermail/development/2024-June/045374.html
* Qt 6.8 Beta1 preparations are started
* Dependency update round in '6.8' succeed so Qt 6.8 Beta1 content
should be in place
* Packaging ongoing, some updates still needed to the packaging configs
for Qt 6.8.0
* Target is to release Qt 6.8 Beta1 as soon as possible, latest Thu 13th
of June
New Qt 6.8 FF Exception requests:
* QtAbstractItemModel, QtModelIndex and QtAbstactListModel:
* Exception not needed, finishing those features by writing
tests/documentation/examples does not require an exception
Qt 6.7 status
* Some quite critical regressions reported from Qt 6.7.1. Jani to check &
decide if it is still possible to release Qt 6.7.2 by the end of June (without
causing too much hassle with other releases etc.)
Improve release note & process
* Jani will sent release note review request to dev ML immediately when
note available.
Next meeting Tue 11th of June 2024 16:00 CET
br,
Jani Heikkinen
Release Manager
irc log below:
[17:00:55] <+jaheikki3> akseli: alblasch: carewolf: frkleint: lars__:mapaaso:
The-Compiler: thiago: vohi: ping
[17:01:03] <akseli> jaheikki3: pong
[17:01:05] <vohi> jaheikki3: pong
[17:02:04] <frkleint> jaheikki3: pong
[17:02:57] <+jaheikki3> time to start qt release team meeting
[17:03:06] <+jaheikki3> on agenda today:
[17:03:07] <thiago> jaheikki3: pong
[17:03:14] <+jaheikki3> Qt 6.8 status
[17:03:27] <+jaheikki3> Qt 6.8 FF exceptions
[17:03:37] <+jaheikki3> Any additional item to the agenda?
[17:04:06] <vohi> one thing to discuss, wrt the "release notes" thread on the
mailing list: what can we improve, and avoid that we claim BC if the change log
later on says it's not :)
[17:04:19] <vohi> (which are perhaps two things)
[17:04:53] <frkleint> is there any hope for a 6.7.1.1 release.. we have a kind
of facepalm bug in Qt Designer ( https://bugreports.qt.io/browse/QTBUG-125983 )
[17:05:03] <thiago> 6.7.2
[17:05:04] <frkleint> just in case some other bug shows up../
[17:05:09] <thiago> we're not running out of numbers
[17:05:16] <+jaheikki3> vohi: maybe that's something which needs a bit longer
discussion than this meeting
[17:06:03] <+jaheikki3> But we can start the discussion here
[17:06:29] <vohi> ack
[17:06:45] <+jaheikki3> and we can discuss about 6.7.2 plans as well
[17:07:01] <+jaheikki3> But let's start from Qt 6.8 status
[17:07:23] <+jaheikki3> Qt 6.8 feature freeze is in effect now (as planned)
[17:07:42] <+jaheikki3> Branching from 'dev' to '6.8' is also done
[17:08:05] <+jaheikki3> And Qt 6.8 Beta1 preparations are started
[17:08:35] <+jaheikki3> Dependency update round in '6.8' succeed already and so
on we should have Qt 6.8 Beta1 content is in place
[17:09:25] <+jaheikki3> Qt 6.8 Beta1 packaging ongoing but some updates still
needed to configs before we can start testing beta1
[17:09:50] <+jaheikki3> The target is to release Qt 6.8 Beta1 as soon as
possible, latest thu 13th of June
[17:10:18] <+jaheikki3> That's all about Qt 6.8 status at this time. Any
comments or questions?
[17:12:34] <+jaheikki3> Ok, then new Qt 6.8 FF exceptions:
[17:12:53] <+jaheikki3> In the ML there isn't any new FF exceptions, right?
[17:13:11] <vohi> There was "Feature freeze exception: QtAbstractItemModel,
QtModelIndex and QtAbstactListModel"
[17:13:40] <vohi> but I agree with Ulf that finishing those features by writing
tests/documentation/examples does not require an exception
[17:13:54] <+jaheikki3> That's true and I agree
[17:14:11] <vohi> no other exception requests that I am aware of so far
[17:14:30] <+jaheikki3> In addition to that there might be the webengine one we
shortly discussed last week, carewolf:?
[17:14:56] <+jaheikki3> (at least alblasch chatted me about that yesterday)
[17:16:41] <+jaheikki3> It seems carewolf isn't here now so I think we need to
handle that later if needed
[17:17:01] <+jaheikki3> Then we could check Qt 6.7.2 plans
[17:17:25] <+jaheikki3> Original plan was to release Qt 6.7.2 after summer
holidays, ~ mid August
[17:18:29] <carewolf> I am here now
[17:18:45] <carewolf> we have integrated all features we need in webengine now.
[17:19:05] <+jaheikki3> Ok, so no need for exception, great
[17:19:35] <vohi> I found three bugs reported as regressions 6.7.0 -> 6.7.1
(https://bugreports.qt.io/browse/QTBUG-125858,
https://bugreports.qt.io/browse/QTBUG-125497, and
https://bugreports.qt.io/browse/QTBUG-125481) in addition to the Qt Designer
bug; unclear how realistic it is to fix all of them so quickly that a 6.7.2
before summer break is realistic though
[17:20:43] <+jaheikki3> Yeah, it will be really hard to get it done before
summer break and on the other hand we need some effort to get beta1 out asap as
well
[17:20:54] <thiago> when would the deadline for a before-holidays release be?
[17:22:06] <+jaheikki3> Well, I am staring my holiday after 2 weeks and we have
done all our summertime plans (substitutions, resourcing) based on current
release plan
[17:22:42] <+jaheikki3> There is also
https://bugreports.qt.io/browse/QTBUG-124839, which seems to be quite important
for QtC 14
[17:24:13] <thiago> I can tell you I cannot reproduce that issue
[17:24:17] <+jaheikki3> Most of finns will start their holiday at the beginning
of July so I would say Qt 6.7.2 has to happen during June if we make the
decision to release it earlier
[17:25:12] <+jaheikki3> But that's something what I wouldn't like to do; it is
always a mess to release some just before holidays starts
[17:25:14] <thiago> though... I hae seen popups not appear... maybe they did
pop under and I didn't see
[17:25:37] <thiago> Anyway, that's a regressin but a normal bug
[17:25:43] <thiago> is there an emergency bug fix?
[17:25:51] <thiago> frkleint: specifically, is yours?
[17:27:14] <frkleint> thiago: Mine is already fixed in the 6.7 branch
[17:27:49] <+jaheikki3> Also https://bugreports.qt.io/browse/QTBUG-124839
should be fixed in '6.7'
[17:28:04] <thiago> yes, but is it an emergency that needs a re-release?
[17:28:16] <+jaheikki3> I don't know any at the moment
[17:28:20] <vohi> in terms of severity, I'd say that the designer issue is the
only one to justify a 6.7.2 in June, but then taking latest 6.7, not just 6.7.1
+ that fix
[17:28:52] <frkleint> thiago: The bug is embarrassing kind-of .you cannot
change int properties
[17:28:55] <thiago> but is it sufficient for a 6.7.2?
[17:29:34] <frkleint> Hm..depends
[17:29:36] <thiago> we *can* release 6.7.2 as 6.7.1+that fix alone
[17:30:38] <+jaheikki3> In my opinion if it is important enough to cause new
release it is then Qt 6.7.2 with latest from '6.7'; I don't see that one so
severe to do new release just for it
[17:31:26] <vohi> git rev-list --count shows 174 commits between v6.7.1 and
origin/6.7 in qtbase alone, so if we go through the effort, than make sure that
we release all those patches and get fast feedback to those as well
[17:31:26] <akseli> in the past when we have rushed something unplanned out
before vacation time there is usually someone(tm) who need to clean up the mess
afterwards and worst case re-release x.y.z.1 when most of the people are on
vacation.
[17:31:47] <thiago> the issue with doing 6.7 branch is that it requires more
testing
[17:32:02] <thiago> but if jaheikki3 feels confident of a quick turn-around
from there, good
[17:32:23] <+jaheikki3> That's true and I don't, I totally agree with akseli
[17:32:24] <thiago> akseli: good point
[17:32:52] <thiago> so I think we agree that 6.7 branch is not suitable for a
release in June. Anyone disagrees?
[17:33:41] <vohi> jaheikki3: let's discuss if and how a proper 6.7.2 release in
June would fit into the release testing team's schedules, and conclude next week
[17:33:54] <thiago> can we afford a week?
[17:34:09] <thiago> if we want to do a release from the 6.7 branch, we probably
need to start now
[17:34:17] <+jaheikki3> True
[17:34:47] <vohi> yes; if they confirm it can be done, then no need to wait
with starting the work; if they say it can't, then it can't
[17:34:52] <thiago> in other words, not deciding to release 6.7 branch right
now means we won't release it before the end of June. So the only opportunity
to release the fix is 6.7.1 + one patch
[17:35:16] <+jaheikki3> And my opinion is that any of these discussed issues
aren't critical enough to start hurrying with the release
[17:36:33] <+jaheikki3> I know that QTBUG-124839 is important for QtC and so on
if we make the desicion to do new release because of QTBUG-125983 it will be
hard to explain why we won't fix it also
[17:36:37] <thiago> the alternative is to tell people to apply that patch when
they build qttools
[17:37:15] <+jaheikki3> Yeah, I would just add that in qt 6.7 known issues and
keep the original plan
[17:37:27] <carewolf> we also had one P0 security patch that came in from chrome
[17:37:28] <vohi> my suggestion is basically: authorize Jani to make the
decision based on what the RTA team says
[17:37:40] <thiago> I agree
[17:37:47] <akseli> +1
[17:38:13] <thiago> security patches don't need a new release. The patches
suffice.
[17:38:13] <+jaheikki3> Ok, I'll discuss with the team tomorrow & let's see
what will be the conclusion
[17:38:19] <carewolf> great
[17:38:32] <+jaheikki3> Then the last item: improving qt release notes
[17:39:15] <+jaheikki3> It is related to this thread:
https://lists.qt-project.org/pipermail/development/2024-June/045368.html
[17:40:01] <+jaheikki3> At least one thing we can improve is to update the
release note for Qt 6.7.1
[17:40:29] <+jaheikki3> Because it is in its own repo and we can update it when
ever it is needed
[17:41:18] <+jaheikki3> vohi: did you have some improvements or ideas already
in your mind?
[17:41:25] <vohi> indeed. the main issue for now is if/how we should/can review
the release notes before release to catch issues like the one pointed out
(boiler plate header claims BC with 6.7.0, change log entry later states that
QtMultimedia is not BC with 6.7.0)
[17:42:29] <vohi> how much time is there between the generation of the release
notes, and the actual release? Is it realistic to send a heads=up to the list
for people to check and amend things in time?
[17:43:05] <vohi> (and that doesn't guarantee anything of course, but it might
be just enough)
[17:43:34] <+jaheikki3> Usually we generate the note when content is frozen
[17:43:42] <+jaheikki3> so there is time to review it
[17:44:12] <vohi> ok; so maybe a heads-up to the list when that happens is all
that's needed and the only thing that's practical anyway
[17:44:16] <+jaheikki3> Just to define the list of reviewers and I'll make sure
they will be added as a reviewers
[17:44:40] <+jaheikki3> I can also sent heads up to dev ml
[17:45:01] <+jaheikki3> if that's way we agree to proceed
[17:45:57] <vohi> +1 for mail to dev ml
[17:46:38] <+jaheikki3> Let's try that; I'll sent heads-up to dev ml next time
[17:47:24] <frkleint> [have to rush, bye]
[17:47:31] <+jaheikki3> I think it was all at this time so let's end this
meeting now & have new one tue 11th June at this same time
[17:47:45] <+jaheikki3> Thanks for your participation, bye
[17:47:47] <vohi> thanks!
--
Development mailing list
Development@qt-project.org
https://lists.qt-project.org/listinfo/development
