Problem when manipulating QT window geometry

[Marcus Nolte]

Hey Christian,

Thanks for your reply.

I built a little test case and played around a little more.
When I emit a QtSignal, which is connected to  a slot where the resizing 
is done, everything works properly (with and without your provided patch).
Still, with the patch, simply invoking a method which resizes the window 
fails.

Since I am definitely no expert on Qt, I can't really say if this is a 
genode-related problem or if I am using Qt improperly here.

If you feel, this might be a genode-related issue, I can, of course, 
send you the test case.

Cheers
Marcus


Am 15.09.2014 18:23, schrieb Christian Prochaska:
> Hello Marcus,
>
> could you please try with the following patch applied?
>
> https://github.com/cproc/genode/commit/d480771f9e9e8ede711c687ed49281bc403860b6.patch
>
> The patch is primarily about resizing Qt windows via the window manager,
> but it modifies some of the places you mentioned.
>
> If it still doesn't work, could you perhaps provide a test case which
> can be used for debugging?
>
> Christian
>
> On 09/15/2014 04:29 PM, Marcus Nolte wrote:
>> Hello Everyone,
>>
>> I am currently experimenting a little with Genode 14.08.
>> When playing around in Qt5, I encountered some problems when trying to
>> change the geometry of a QMainWindow.
>> I have an additional Genode::Thread running, which can trigger a
>> geometry-change of the application's main window by methods provided by
>> the MainWindow.
>>
>> When a change of the window's geometry is triggered, it seems as if the
>> window-decorations are properly resized, but the QtWindow underneath is
>> kept at its original size. Additionally, the thread which triggers the
>> geometry-change is blocked (this holds true for the Genode::Thread, as
>> well as for a QThread).
>>
>> Resizing the window with the mouse works properly.
>>
>> I took a look at the Nitpicker Qt-Plugin, where I noticed two things:
>> 1. - There is some unused code in qnitpickerplatformwindow.cpp in
>> QNitpickerPlatformWindow::handle_events().
>> Until now I did not dig into the Window_slave_policy, but the inactive
>> code segment looks to me, as if it should change the QtWindow's geometry.
>>
>> 2. - The Processing seems to stop somewhere before returning the
>> framebuffer of the the PlatformWindow when detaching / attaching it from
>> / to the rm_ssession.
>>
>> Since I'm not quite familiar with all the internal Nitpicker / Qt stuff,
>> it would be great if anyone with a little more experience could provide
>> some help. ;-)
>> - Or am I even trying something which is not intended?
>>
>> Cheers
>> Marcus
>>
>> ------------------------------------------------------------------------------
>> Want excitement?
>> Manually upgrade your production database.
>> When you want reliability, choose Perforce
>> Perforce version control. Predictably reliable.
>> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
>> _______________________________________________
>> genode-main mailing list
>> genode-main at lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/genode-main
>
> ------------------------------------------------------------------------------
> Want excitement?
> Manually upgrade your production database.
> When you want reliability, choose Perforce
> Perforce version control. Predictably reliable.
> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk
> _______________________________________________
> genode-main mailing list
> genode-main at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/genode-main
>

-- 
Marcus Nolte, M. Sc.

Technische Universität Braunschweig
Institut für Regelungstechnik (IfR)
Hans-Sommer-Straße 66
38106 Braunschweig

Tel.: +49 531 391 3827
Fax:  +49 531 391 7878
Email: nolte at ...254...
Internet: http://www.ifr.ing.tu-bs.de