PSA: libkscreen dependency bumped to Qt 5.4

Quick announcement: We bumped the required Qt version of libkscreen today to Qt 5.4 (it was 5.2). The reason is a regression in Qt which caused a crash on teardown of the QScreen backend. We fixed it by using API which was new in Qt 5.4. I don't expect this to be a problem, as the depen

[Differential] [Accepted] D1082: Only require Qt 5.4

bshah accepted this revision. bshah added a reviewer: bshah. bshah added a comment. This revision is now accepted and ready to land. Looks good to me. REPOSITORY rKSCREENLOCKER KScreenLocker BRANCH kcm-remove-themes REVISION DETAIL https://phabricator.kde.org/D1082 EMAIL PREFERENCES

[Differential] [Request, 34 lines] D1082: Only require Qt 5.4

graesslin created this revision. graesslin added a reviewer: Plasma. Herald added a project: Plasma. Herald added a subscriber: plasma-devel. REVISION SUMMARY We don't have any code requiring Qt 5.5, so let's go down with the requirement to make it build on CI again. SVN_SILENT made messa

Re: Review Request 122617: Require Qt 5.4

marked as submitted. Review request for kwin and Plasma. Repository: kwin Description --- Require Qt 5.4 Diffs - CMakeLists.txt 6dbfcc961fe573c5621670a8cdc08a09ac7a01d9 clients/aurorae/src/aurorae.h f6ef39ba2d4b510d3d452dba5b37e34f8a18217e clients/aurorae/src/aurorae.cpp

Re: Review Request 122617: Require Qt 5.4

b. 18, 2015, 9:06 a.m.) > > > Review request for kwin and Plasma. > > > Repository: kwin > > > Description > --- > > Require Qt 5.4 > > > Diffs > - > > CMakeLists.txt 6dbfcc961fe573c5621670a8cdc08a09ac7a01d9 > clien

Review Request 122617: Require Qt 5.4

--- Require Qt 5.4 Diffs - CMakeLists.txt 6dbfcc961fe573c5621670a8cdc08a09ac7a01d9 clients/aurorae/src/aurorae.h f6ef39ba2d4b510d3d452dba5b37e34f8a18217e clients/aurorae/src/aurorae.cpp e24687d9cd7ca690d887aba0cd7569db54cbf6cb glxbackend.cpp c97fbee4743f3af239a8d640827113fcd8d04ed5

Re: Qt 5.4

This was fixed upstream now: https://bugreports.qt.io/browse/QTBUG-43436 Cheers, Eike ___ Plasma-devel mailing list Plasma-devel@kde.org https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Re: Qt 5.4 again

On Monday 12 January 2015 15:06:23 Martin Klapetek wrote: > On Mon, Jan 12, 2015 at 2:59 PM, Marco Martin wrote: > > On Monday 12 January 2015, Jonathan Riddell wrote: > > > Last we looked we decided that Plasma 5.2 can depend on Qt 5.4. But > > > come beta time it see

Re: Qt 5.4 again

On Mon, Jan 12, 2015 at 2:52 PM, Martin Gräßlin wrote: > On Monday 12 January 2015 13:46:15 Jonathan Riddell wrote: >> Last we looked we decided that Plasma 5.2 can depend on Qt 5.4. But >> come beta time it seems nobody has felt the need to change any modules >> to require

Re: Qt 5.4 again

On Mon, Jan 12, 2015 at 2:59 PM, Marco Martin wrote: > On Monday 12 January 2015, Jonathan Riddell wrote: > > Last we looked we decided that Plasma 5.2 can depend on Qt 5.4. But > > come beta time it seems nobody has felt the need to change any modules > > to require on Q

Re: Qt 5.4 again

On Monday 12 January 2015, Jonathan Riddell wrote: > Last we looked we decided that Plasma 5.2 can depend on Qt 5.4. But > come beta time it seems nobody has felt the need to change any modules > to require on Qt 5.4. Martin G says KWin has a bunch of fixes that > work with 5.4 on

Re: Qt 5.4 again

On Mon, Jan 12, 2015 at 7:22 PM, Martin Gräßlin wrote: > +1, this should probably also go into the release announcements. Oh and please > remind distros to rebuild frameworkintegration when bringing in Qt 5.4. Also kdelibs4support IIRC -- Bhushan Shah http://bhush9.github.io IRC Nick :

Re: Qt 5.4 again

On Monday 12 January 2015 13:46:15 Jonathan Riddell wrote: > Last we looked we decided that Plasma 5.2 can depend on Qt 5.4. But > come beta time it seems nobody has felt the need to change any modules > to require on Qt 5.4. Martin G says KWin has a bunch of fixes that > work with

Qt 5.4 again

Last we looked we decided that Plasma 5.2 can depend on Qt 5.4. But come beta time it seems nobody has felt the need to change any modules to require on Qt 5.4. Martin G says KWin has a bunch of fixes that work with 5.4 only but it'll also work fine with Qt 5.3. The KDE CI system has Qt 5

Re: Qt 5.4

On Wed, Dec 10, 2014 at 4:26 PM, Eike Hein wrote: > > > On 12/10/2014 04:14 PM, Aleix Pol wrote: >> >> On Wed, Dec 10, 2014 at 4:10 PM, Eike Hein > > wrote: >> >> >> There's a bunch of unsolved new regressions (e.g. DND of files is >> broken on 5.4) but nothing out of

Re: [plasma-devel] Qt 5.4

No objections to this so I'll switch it over Jonathan ___ Plasma-devel mailing list Plasma-devel@kde.org https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Qt 5.4

On Wednesday 10 December 2014 15:58:25 Jonathan Riddell wrote: > Qt 5.4 is out and it's time to make a decision on whether Plasma 5.2 will > require it. > > Speaking to packagers from Ubuntu, opensuse and Fedora they all have RC > packages and don't expect it to b

Re: Qt 5.4

On 12/10/2014 04:14 PM, Aleix Pol wrote: On Wed, Dec 10, 2014 at 4:10 PM, Eike Hein mailto:h...@kde.org>> wrote: There's a bunch of unsolved new regressions (e.g. DND of files is broken on 5.4) but nothing out of the ordinary breakage level I guess. Cheers, Eike I wasn

Re: Qt 5.4

On Wed, Dec 10, 2014 at 4:10 PM, Eike Hein wrote: > > There's a bunch of unsolved new regressions (e.g. DND of files is > broken on 5.4) but nothing out of the ordinary breakage level I > guess. > > > Cheers, > Eike I wasn't aware, I've worked on that code, if you can provide a test case I'll g

Re: Qt 5.4

There's a bunch of unsolved new regressions (e.g. DND of files is broken on 5.4) but nothing out of the ordinary breakage level I guess. Cheers, Eike ___ Plasma-devel mailing list Plasma-devel@kde.org https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Qt 5.4

On Wed, Dec 10, 2014 at 3:58 PM, Jonathan Riddell wrote: > Qt 5.4 is out and it's time to make a decision on whether Plasma 5.2 will > require it. > > Speaking to packagers from Ubuntu, opensuse and Fedora they all have RC > packages and don't expect it to be difficu

Re: Qt 5.4

On Wednesday, December 10, 2014 15:58:25 Jonathan Riddell wrote: > Qt 5.4 is out and it's time to make a decision on whether Plasma 5.2 will > require it. > > Speaking to packagers from Ubuntu, opensuse and Fedora they all have RC > packages and don't expect it to b

Qt 5.4

Qt 5.4 is out and it's time to make a decision on whether Plasma 5.2 will require it. Speaking to packagers from Ubuntu, opensuse and Fedora they all have RC packages and don't expect it to be difficult to have final packages. I know there's a patch Martin says is needed but th

Re: Qt 5.4 issues in Ubuntu

On Tue, Dec 2, 2014 at 12:51 PM, Jonathan Riddell wrote: > > I've been pointed at the current list of Qt 5.4 issues that Ubuntu > packgers are seeing > http://is.gd/jAUhah > > I expect they're all fixable but I'm not sure it would be in time for the > Plasma

Qt 5.4 issues in Ubuntu

I've been pointed at the current list of Qt 5.4 issues that Ubuntu packgers are seeing http://is.gd/jAUhah I expect they're all fixable but I'm not sure it would be in time for the Plasma 5.2 release. Jonathan ___ Plasma-devel mail

Re: Qt 5.4 for workspace

On Sep 26, 2014 11:27 PM, "Aleix Pol" wrote: > > I'm not aware of a change of Qt dependency in KF5. Did I miss anything? No, didn't happen;) what I was saying is that perhaps we want to wait for frameworks to bump the dependency, which I have no odea when will happen

Re: Re: Qt 5.4 for workspace

On Friday 26 September 2014 20:27:07 Eike Hein wrote: > As for concrete 5.4 stuff I'm interested in - there's new API > to render QQuickItems into offscreen memory buffers, which would > be really useful to make icon DND in Folder more efficient and > a little more reliable. The current code ends u

Re: Qt 5.4 for workspace

El dia 26/09/2014 20.37, "Marco Martin" va escriure: > > On Thursday 25 September 2014 19:23:00 Eike Hein wrote: > > Hi, > > > > we're about to branch Plasma for 5.1 this week. Qt 5.4 is > > in freeze and is planning a final release on November 18th,

Re: Qt 5.4 for workspace

On Thursday 25 September 2014 19:23:00 Eike Hein wrote: > Hi, > > we're about to branch Plasma for 5.1 this week. Qt 5.4 is > in freeze and is planning a final release on November 18th, > which is six weeks shy of the 5.2 release in January 2015. > > Do we want to dep

Re: Qt 5.4 for workspace

On 26.09.2014 19:34, Aleix Pol wrote: On Fri, Sep 26, 2014 at 7:30 PM, David Edmundson mailto:da...@davidedmundson.co.uk>> wrote: Definitely a good question.​ I assume you have a specific feature in mind? I certainly think it would be OK to have branches with 5.4 stuff in them

Re: Qt 5.4 for workspace

On Fri, Sep 26, 2014 at 7:30 PM, David Edmundson wrote: > Definitely a good question.​ > > I assume you have a specific feature in mind? > I certainly think it would be OK to have branches with 5.4 stuff in them. > Or #ifdef... Aleix ___ Plasma-devel

Re: Qt 5.4 for workspace

Definitely a good question.​ I assume you have a specific feature in mind? I certainly think it would be OK to have branches with 5.4 stuff in them. ___ Plasma-devel mailing list Plasma-devel@kde.org https://mail.kde.org/mailman/listinfo/plasma-devel

Re: Qt 5.4 for workspace

On Thursday 25 September 2014 19:23:00 Eike Hein wrote: > Hi, > > we're about to branch Plasma for 5.1 this week. Qt 5.4 is > in freeze and is planning a final release on November 18th, > which is six weeks shy of the 5.2 release in January 2015. > > Do we want to dep

Qt 5.4 for workspace

Hi, we're about to branch Plasma for 5.1 this week. Qt 5.4 is in freeze and is planning a final release on November 18th, which is six weeks shy of the 5.2 release in January 2015. Do we want to depend on 5.4 in master after we branch, or is this too compressed for 5.2? We also can

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

aceFormat::setProfile. > > Alexander Richardson wrote: > By the way this causes the build to fail with Qt 5.4, > QQuickWindow::setDefaultFormat was changed to > QSurfaceFormat::setDefaultFormat() (see > https://git.reviewboard.kde.org/r/119913/ ) It's been reverted since

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

unner) or with > if'ing the cases where it doesn't work. > > Martin Gräßlin wrote: > Requesting GLES on a desktop system just feels very wrong and from KWin > experience: don't even think about ifdefing. For desktop usage I would > suggest to try

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

hmm, i see it's still trying to set opengles, that shuld be reverted as was determined wasn't a good idea On Monday 11 August 2014 09:03:23 Martin Gräßlin wrote: > > On Aug. 11, 2014, 8:35 a.m., Martin Gräßlin wrote: > > > I highly suggest to revert this change. Enforcing GLES means limiting to >

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

hat is). After some investigation, David found out we > were going through some memory-consuming path [1]. A way to workaround it is > by using only OpenGLES. Everything still works here after this patch, so it > seems to be a good change to get in. > > This change needs Qt 5.4, to get

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

> (on intel systems, that is). After some investigation, David found out we > were going through some memory-consuming path [1]. A way to workaround it is > by using only OpenGLES. Everything still works here after this patch, so it > seems to be a good change to get in. > > This

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

> (on intel systems, that is). After some investigation, David found out we > were going through some memory-consuming path [1]. A way to workaround it is > by using only OpenGLES. Everything still works here after this patch, so it > seems to be a good change to get in. > &g

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

investigation, David found out we were going through some memory-consuming path [1]. A way to workaround it is by using only OpenGLES. Everything still works here after this patch, so it seems to be a good change to get in. This change needs Qt 5.4, to get the new API that lets us enforce a

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

s a sensible thing > - wait for Martin, then decide to merge or not > > I'm cool with with either strategy. (Would be quite interesting though to > see what happens when we switch. :)) Well, Qt 5.4 won't be in for a while, so it only makes sense for wild adventurers

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

we realized that one of > the reasons we got such big memory footprint was actually the intel driver > (on intel systems, that is). After some investigation, David found out we > were going through some memory-consuming path [1]. A way to workaround it is > by using only OpenGLES. Eve

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

ms, that is). After some investigation, David found out we > were going through some memory-consuming path [1]. A way to workaround it is > by using only OpenGLES. Everything still works here after this patch, so it > seems to be a good change to get in. > > This change needs Qt 5.

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

is). After some investigation, David found out we > were going through some memory-consuming path [1]. A way to workaround it is > by using only OpenGLES. Everything still works here after this patch, so it > seems to be a good change to get in. > > This change needs Qt 5.4, t

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

er > (on intel systems, that is). After some investigation, David found out we > were going through some memory-consuming path [1]. A way to workaround it is > by using only OpenGLES. Everything still works here after this patch, so it > seems to be a good change to get in. >

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

intel systems, that is). After some investigation, David found out we > were going through some memory-consuming path [1]. A way to workaround it is > by using only OpenGLES. Everything still works here after this patch, so it > seems to be a good change to get in. > > This chan

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

uce memory usage in plasma shell, we realized that one of > the reasons we got such big memory footprint was actually the intel driver > (on intel systems, that is). After some investigation, David found out we > were going through some memory-consuming path [1]. A way to workaround it is

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

oing through some memory-consuming path [1]. A way to workaround it is > by using only OpenGLES. Everything still works here after this patch, so it > seems to be a good change to get in. > > This change needs Qt 5.4, to get the new API that lets us enforce a > QSurfaceFormat,

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

otprint was actually the intel driver > (on intel systems, that is). After some investigation, David found out we > were going through some memory-consuming path [1]. A way to workaround it is > by using only OpenGLES. Everything still works here after this patch, so it > seems to be a goo

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

going through some memory-consuming path [1]. A way to workaround it is > by using only OpenGLES. Everything still works here after this patch, so it > seems to be a good change to get in. > > This change needs Qt 5.4, to get the new API that lets us enforce a > QSurfaceForm

Re: Review Request 119524: Force OpenGLES if on Qt 5.4

> were going through some memory-consuming path [1]. A way to workaround it is > by using only OpenGLES. Everything still works here after this patch, so it > seems to be a good change to get in. > > This change needs Qt 5.4, to get the new API that lets us enforce a > QSurfac

Review Request 119524: Force OpenGLES if on Qt 5.4

workaround it is by using only OpenGLES. Everything still works here after this patch, so it seems to be a good change to get in. This change needs Qt 5.4, to get the new API that lets us enforce a QSurfaceFormat, hence having it ifdef'd. I'm quite unaware of problems we might find.

Running Plasma on Qt 5.4

Hi, Just in case somebody has the magic idea of trying Qt 5.4, bear in mind that it's currently not working (or, well, working very bad). Here's a patch if you really feel like trying it [1] which seems to work. Cheers! Aleix [1] https://codereview.qt-project.org