[Development] Question regarding commit: Fix transformation in eglfs backingstore

[laszlo.p.agocs at nokia.com]

Ok, let's just revert the change if it breaks on Pi. It wouldn't make much of a difference on eglfs anyways due to the forced-fullscreen windows.

The change originates from kms which, due to the way it is currently implemented, is potentially capable of doing non-fullscreen windows. The problem of wrong vertical positioning of top-level widgets was apparent there and the backingstore code for these two plugins is the same. There are probably some differences with regards to Y inversion so we end up with side effects on eglfs. Needs to be investigated further some day.
 
Thanks,
Laszlo

-----Original Message-----
From: development-bounces+laszlo.p.agocs=nokia.com at qt-project.org [mailto:development-bounces+laszlo.p.agocs=nokia.com at qt-project.org] On Behalf Of ext Jon Trulson
Sent: 28 August 2012 02:59
To: development at qt-project.org
Subject: [Development] Question regarding commit: Fix transformation in eglfs backingstore


Commit 12491f35bbea7008b15fb0ba0ee7ea5bf6eb6b6c from Laszlo Agocs:

     Fix transformation in eglfs backingstore

     While not visible with forced-fullscreen windows, the formula used
     to set up positioning on the Y axis is wrong. (with the GL
     coordinate system Y goes 1..-1 when moving top-to-bottom)

     Taken from kms' version of the backingstore which had the formulas
     corrected.

I am crosscompiling qt5 for a Raspberry Pi board using eglfs, and the latest raspbian (2012-08-16-wheezy-raspbian).  It has gone well with the most recent git masters, pulled today.

I noticed though that while running various qtbase widget examples (scribble, sliders, etc) that the Y axis is inverted.

Those opengl examples that work, do show up properly regardless of this patch (paintedwindow).

The commit above seems to be responsible.  If I revert it, then the apps display properly (though opening menus, etc still cause crashes, but that's another issue.)

OpenGL apps, including qml via qmlscene still look correct.

I suspect that Y inversion is already being handled elsewhere within Qt, so I am curious as to what problem was actually seen that required this change?

Should it be reverted?

--
Jon Trulson

Atoms are what make us matter.

_______________________________________________
Development mailing list
Development at qt-project.org
http://lists.qt-project.org/mailman/listinfo/development