glsdk-devel

[Glsdk-devel] Windows compatibility

[Branan R. <br...@gm...>]

Has anyone been building the SDK on Windows lately? Is it still compiling?

The code should all work from what I've seen, but I'd like some
confirmation of that. I can cross-compile it with MinGW and/or build
it in my VM, if a test is needed.

Branan

Re: [Glsdk-devel] Windows compatibility

[Orhun B. <orh...@gm...>]

It builds with Visual Studio 2008. I am getting the following warnings...
2>..\..\..\glsdk\src\glm\angles.c(32) : warning C4244: 'return' :
conversion from 'double' to 'GLfloat', possible loss of data
2>..\..\..\glsdk\src\glm\angles.c(38) : warning C4244: 'return' :
conversion from 'double' to 'GLfloat', possible loss of data
2>..\..\..\glsdk\src\glm\mat4f.c(535) : warning C4244: '=' :
conversion from 'double' to 'GLfloat', possible loss of data

2>..\..\..\glsdk\src\gls\gls.c(27) : warning C4068: unknown pragma

I didn't build the test, which I can try building (and running) it
tonight, and report back.

-- 
Orhun Birsoy

Re: [Glsdk-devel] Windows compatibility

[Branan R. <br...@gm...>]

Those warnings are probably just from missing 'f's on floating-point
constants. Easy enough to fix. The "unknown pragma" in gls.c got
committed by accident. I was trying to make GCC silence a warning
that's not going to actually matter anymore. That will go away with my
next commit.

You can build the tests if you want. Make sure you have CUnit
installed in a location that your vcvars.bat will pick up, otherwise
it'll fail miserably.

On Tue, Sep 2, 2008 at 2:40 PM, Orhun Birsoy <orh...@gm...> wrote:
> It builds with Visual Studio 2008. I am getting the following warnings...
> 2>..\..\..\glsdk\src\glm\angles.c(32) : warning C4244: 'return' :
> conversion from 'double' to 'GLfloat', possible loss of data
> 2>..\..\..\glsdk\src\glm\angles.c(38) : warning C4244: 'return' :
> conversion from 'double' to 'GLfloat', possible loss of data
> 2>..\..\..\glsdk\src\glm\mat4f.c(535) : warning C4244: '=' :
> conversion from 'double' to 'GLfloat', possible loss of data
>
> 2>..\..\..\glsdk\src\gls\gls.c(27) : warning C4068: unknown pragma
>
> I didn't build the test, which I can try building (and running) it
> tonight, and report back.
>
> --
> Orhun Birsoy
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Glsdk-devel mailing list
> Gls...@li...
> https://lists.sourceforge.net/lists/listinfo/glsdk-devel
>

Re: [Glsdk-devel] Windows compatibility

[Orhun B. <orh...@gm...>]

the first two glm warnings are due to M_PI usage. It has to be cast to float.
the last glm warning is due to use of tan(). tanf() should have been used.

I've already build the tests before, just not lately. I have CUnit is
already on my home machine. I am at work, I don't have CUnit here.

Note that, under windows I had to change several build settings to
build the tests. I don't exactly remember but the library name was not
what glsdk was trying to link with. I think the CUnit Visual Studio
project generated "libcunit.lib" and glsdk was trying to link with
cunit.lib. I can provide a better explanation when I have access to my
home machine.

-- 
Orhun Birsoy

Re: [Glsdk-devel] Windows compatibility

[Branan R. <br...@gm...>]

That's a difference in how CMake assumes libraries are named on
Windows and Linux.

on Linux, it's lib<name>.a
on Windows, it's <name>.lib

so on Windows, it needs to link against "libcunit" rather than just
"cunit". This is actually CUnit's fault, not ours - It doesn't follow
the standard naming conventions across platforms.

Branan

On Tue, Sep 2, 2008 at 2:58 PM, Orhun Birsoy <orh...@gm...> wrote:
> the first two glm warnings are due to M_PI usage. It has to be cast to float.
> the last glm warning is due to use of tan(). tanf() should have been used.
>
> I've already build the tests before, just not lately. I have CUnit is
> already on my home machine. I am at work, I don't have CUnit here.
>
> Note that, under windows I had to change several build settings to
> build the tests. I don't exactly remember but the library name was not
> what glsdk was trying to link with. I think the CUnit Visual Studio
> project generated "libcunit.lib" and glsdk was trying to link with
> cunit.lib. I can provide a better explanation when I have access to my
> home machine.
>
> --
> Orhun Birsoy
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Glsdk-devel mailing list
> Gls...@li...
> https://lists.sourceforge.net/lists/listinfo/glsdk-devel
>