On Mon, Mar 24, 2014 at 15:10:25 -0300, Lisandro Damián Nicanor Pérez Meyer 

> Hi! I'm trying to find a solution for the FTBFS of qantenna on arm* [0][1] 
> that fits the best to everyone. This issue seems to be very similar to a 
> older 
> thread "including both GL/gl.h and cogl/cogl.h fails on armel and armhf" [2], 
> but I failed to see how that was resolved.
As far as I can tell,
should have solved it.

> The FTBFSs are due to conflicting declarations for GLdouble, GLsizeiptr and 
> GLintptr.
> Basically the problem might boild down to the fact that Qt5 is built using 
> es2 
> instead of the "desktop" opengl which, to the best of my knowledge, it's 
> standard OpenGL 2.0.
> These are the possible solutions I think could be taken:
> a) Switch Qt5 to use "desktop" OpenGL on arm*. I have tested this on 
> harris.d.o and works OK. As a pro, it means that other apps will not have 
> porting issues like this. As a con, it might mean that all the OpenGL stuff 
> will be done by software though, am I correct?
Sounds about right.

> b) (supossing it's possible) provide es2 versions of mesa-common-dev, libglu1-
> mesa-dev et al. and build against that on arm*
> d) (also supossing it's possible) do not consider the FTBFS not RC (or allow 
> it's transition even if it's RC) until a better fix could be done.
> Please note that I did not include "porting the app" because, as I understand 
> it, there is no es2-enabled GLU available, or at least on Debian.
> I'm inclined for option a, but maybe you can provide alternative solutions. 
> As 
> a fallback (if it's possible at all) I would take option d, but that might be 
> needed for other apps in the no-so-distant future, when we will start to see 
> massive porting from Qt4 to Qt5.
I'd very much like to avoid b.  The GLU spec
(http://www.opengl.org/registry/doc/glu1.3.pdf) seems very much tied to
old OpenGL.

I didn't think GLU was still a thing used by any modern toolkit/app,
though, so I'm surprised "you get to choose between GLU and Qt5" is a


Attachment: signature.asc
Description: Digital signature


Reply via email to