[Releasing] Meeting minutes: Qt release team meeting 12.01.2016

[Knoll Lars]

Hi Thiago,

Can you explain in more detail what’s blocking 5.7 on QNX? As far as I understand, it’s mainly that std::atomic is not properly implemented in their libstdc++, right?

Do we have any other options apart from patching the QNX SDK, as I don’t believe that to be a very good option?

Thanks,
Lars




On 13/01/16 11:44, "Releasing on behalf of Heikkinen Jani" <releasing-bounces at qt-project.org on behalf of jani.heikkinen at theqtcompany.com> wrote:

>Meeting minutes from Qt Release Team meeting 12th January 2016
>
>
>Qt 5.6 status:
>- QTQAINFRA-943 should be now fixed but fixes aren't in qt5.git yet
>    * Update 13th Jan: qt5.git integration succeed,
>fixes for QTQAINFRA-943 in. Unfortunately
>changes waiting for it aren't in yet
>--> branching cannot start yet. Hoping we get missing changes in during this week and we so on can start branching at the begining of next week.
>- RC blocker
> list here: 
>https://bugreports.qt.io/issues/?filter=17225 <https://bugreports.qt.io/issues/?filter=17225>
>
>
>
>Qt 5.7 status:
>- No progress with QNX/std::atomic related issues
>--> Cannot start FF/Branching yet, hoping we can progress with those during next week
>
>
>Qt 4.8 & bugreports.qt.io:
>- Qt 4.8 branch has been closed for a while (except security fixes) and support has ended
>- bugreports.qt.io has some 1500+ bugs reported \ open \ in progress against Qt 4.8.x versions
>- Agreed to close those Qt 4.8 and older bugs with proper message
>   
>* ' 'please re-open if still valid in Qt 5' or something similar
>
>
>Next meeting 19th Jan 2016 16:00 CET
>
>
>br,
>Jani
>
>
>irc log below:
>[17:00:32] <jaheikki3> akseli: iieklund: thiago: fkleint: ZapB: tronical: vladimirM: aholza: peter-h: mapaaso: ankokko: fkleint: carewolf: fregl: ablasche: ping
>[17:00:56] <akseli> jaheikki3: pong
>[17:02:12] <fkleint> jaheikki3: pong
>[17:02:56] <jaheikki3> time to start qt release team meeting
>[17:03:04] <jaheikki3> On agenda today:
>[17:03:21] <jaheikki3> Qt 5.6 status
>[17:03:32] <jaheikki3> Qt 5.7 status
>[17:03:52] <jaheikki3> Qt 4.8 & bugreports.qt.io
>[17:04:02] <jaheikki3> Something else to the agenda?
>[17:05:21] <carewolf> pong
>[17:06:13] <jaheikki3> Ok, let's start from Qt 5.6 status:
>[17:07:11] <jaheikki3> unfortunately we cannot start branching yet. QTQAINFRA-943 should be now fixed but fixes aren't in qt5.git yet
>[17:07:38] <jaheikki3> And so on changes waiting for it aren't in yet either
>[17:08:29] <jaheikki3> qt5.git integration ongoing, let's hope it will integrate at this time.
>[17:10:07] <jaheikki3> New Qt 5.6 RC blocker list here (with Fix Version/s: 5.6.0 RC): https://bugreports.qt.io/issues/?filter=17225
>[17:10:40] <jaheikki3> Any comments/questions?
>[17:11:19] -*- thiago forgot to create his task from last week
>[17:11:40] <jaheikki3> thiago: Was that for 5.7?
>[17:11:55] <thiago> yeah
>[17:12:13] <thiago> I'm just saying stuff aloud, sorry
>[17:12:50] <jaheikki3> No problem, I planned to ask it a bit later ;)
>[17:14:06] <jaheikki3> Ok, I think it was all about 5.6 at this time. Hoping we get qt5.git integrated & those missing changes in during this week and we can start branching at the begining of next week...
>[17:14:15] <jaheikki3> Then Qt 5.7 status:
>[17:14:47] <jaheikki3> thiago: is there some progress with that QNX/std::atomic related issues?
>[17:15:14] <thiago> no reply from anyone
>[17:15:47] <jaheikki3> I shortly discussed with Simon and he said he haven't had time fo that yet
>[17:15:59] <thiago> ok, thanks
>[17:17:14] <jaheikki3> So we need still postpone the FF and branching. Let's hope we can progress with those during next week. I'll keep pushing Simon.
>[17:17:45] <jaheikki3> thiago: could you create that bug report, it will make follow up easier?
>[17:18:55] <thiago> yep
>[17:20:58] <jaheikki3> Ok, that was all about 5.7 at this time.
>[17:21:22] <jaheikki3> Then 4.8 & bug reports. Akseli
>[17:21:27] <akseli> Qt 4.8 branch has been closed for a while (except security fixes) and support has ended
>[17:21:32] <akseli> bugreports.qt.io has some 1500+ bugs reported \ open \ in progress against Qt 4.8.x versions (some of those reported against Qt5.x as well)
>[17:21:37] <akseli> Some of Qt 4.8.x bugs might have been fixed and bugreport status has not been updated correctly but it is quite obvious most of 4.8.x bugs are in reality on "won't fix" status.
>[17:21:41] <akseli> Should we take some actions for Qt 4.8.x (and older) bugs reported on bugreports.qt.io e.g. batch close Qt 4.8.x bugs with proper description?
>[17:21:45] <akseli> or just leave those bugs open for now?
>[17:22:12] <fkleint> Hm, maybe check with ablasche, 
>[17:22:12] <thiago> if they are reported only against 4.8 and no later version, they should be closed
>[17:22:19] <fkleint> JIRA-technically
>[17:22:27] <thiago> it's possible they still apply anyway, though
>[17:22:32] <fkleint> [but mgmt workshop going on ATM]
>[17:22:45] <thiago> but the close message can say we will repoen if it can be reproduced with 5.x
>[17:23:24] <fkleint> fears another mailbox flood
>[17:23:39] <akseli> batch close can be done without sending email ;)
>[17:24:12] <akseli> "reopen if it can be reproduced with 5.x" was something i was thinking
>
>[17:24:14] <jaheikki3> +1, I think we could close all 4.8 (and older) bugs with message 'please re-open if still valid in Qt 5' or something similar
>[17:25:04] <akseli> i can consult ablasche later about technicalities if closing 4.8 and older bugs is ok for everyone
>[17:26:39] <thiago> yeah, I think that's the way to go
>[17:26:59] <jaheikki3> +1
>[17:27:21] <fkleint> [Note though Symbian bugs won't be accepted in Qt 5  ;-) ]
>[17:27:31] <fkleint> and other dead platforms..
>[17:27:55] <jaheikki3> true
>[17:28:03] <The-Compiler> I'm guessing even if there's no mail when closing, the comments from people having their bug closed as one of the 1500 might be... fun?
>[17:29:14] <fkleint> yes, that's what I am afraid of
>[17:29:18] <thiago> well, by definition they can't be reproduced in Qt 5
>[17:29:57] <w00t_> realistically, i think closing things that aren't going to be looked at is better than leaving them open indefinitely
>[17:29:58] <ablasche> akseli: fkleint: important to ensure that you don't close bugs with 4.8 bugs with Qt 5.x tags
>[17:30:29] <w00t_> some people will of course be disappointed about that, but if the message is worded carefully enough there won't be many of those :)
>[17:30:39] <The-Compiler> yeah, I totally agree on that, but some people might get "upset" - not much you can do about that
>[17:31:20] <thiago> the interesting thing is that those bugs are dead already, but the act of closing them makes them live
>[17:32:41] <carewolf> that is not dead which can eternal lie
>[17:33:15] <fkleint> Hm..maybe the ones with no-longer-existing Nokia/etc-ids should be closed independently first..but that out of the scope of this meeting
>[17:36:09] <jaheikki3> Akseli will take care of closing those bugs with proper message. Let's then see what kind of discussion will arise. I'll add that plan in the meeting memo as well, at least some will notice it already from there
>[17:36:32] <thiago> if they're assigned to "Closed Nokia identity", they should be simply unassigned
>[17:37:30] <jaheikki3> I think it was all at this time. Let's end this meeting now & have new one 19th Jan at this same time. OK?
>[17:38:05] <akseli> +1
>[17:38:49] <jaheikki3> Thanks for your participation!
>[17:38:51] <jaheikki3> bye
>[17:39:01] <fkleint> bye
>[17:39:05] <carewolf> bye
>
>
>
>
>
>