TFP test

 Hello,

 could people who _don't_ have nvidia and can run KWin in
TFP mode please run 
the attached testapp and give me the output (plus what card
etc.)? Thanks

-- 
Lubos Lunak
KDE developer
------------------------------------------------------------
--
SUSE LINUX, s.r.o.   e-mail: l.lunaksuse.cz , l.lunakkde.org
Lihovarska 1060/12   tel: +420 284 028 972
190 00 Prague 9      fax: +420 284 028 951
Czech Republic       http//www.suse.cz

_______________________________________________
Kwin mailing list
Kwinkde.org
https://ma
il.kde.org/mailman/listinfo/kwin

pre: I'm having an nvidia card, but I was curious 

On Tuesday 19 June 2007 15:31:11 Lubos Lunak wrote:
>  Hello,
>
>  could people who _don't_ have nvidia and can run KWin
in TFP mode please
> run the attached testapp and give me the output (plus
what card etc.)?
> Thanks

I were not able to compile the test anyway. So here I made a

little KDE-independance-patch, and added a (obviousely!) 
missing include to glxtokens:

compile with: g++ -o m m.cpp -lX11 -lGL

--- g.cpp       2007-06-20 08:55:51.000000000 +0200
+++ m.cpp       2007-06-20 09:00:09.000000000 +0200
 -1,7
+1,9 
 #include <X11/Xlib.h>
 #include <GL/glx.h>
+#include <GL/glxtokens.h>
 #include <limits.h>
-#include <kdebug.h>
+#include <iostream>
+#include <iomanip>

 bool tfp_mode = true;
 Display* dpy;
 -108,7
+110,7 
 //            fbcdrawableinfo[ i ].y_inverted = value;
 //            fbcdrawableinfo[ i ].mipmap = mipmap;
             glXGetFBConfigAttrib( display(), fbconfigs[ j
], GLX_BIND_TO_TEXTURE_TARGETS_EXT, &value );
-            kdDebug() << "XXX:" << i
<< ":" << QString::number( value, 16 )
<< ":" << (value &
GLX_TEXTURE_2D_BIT_EXT) << endl;
+                       std::cout << "XXX:"
<< i << ":" << std::setw(16)
<< value << ":" << (value &
GLX_TEXTURE_2D_BIT_EXT) << std::endl;
             }
         }
     if( cnt )

_______________________________________________
Kwin mailing list
Kwinkde.org
https://ma
il.kde.org/mailman/listinfo/kwin

On Tuesday 19 June 2007 15:31:11 Lubos Lunak wrote:
>  Hello,
>
>  could people who _don't_ have nvidia and can run KWin
in TFP mode please
> run the attached testapp and give me the output (plus
what card etc.)?
> Thanks

here, I'm running an intel mac mini:

00:02.0 VGA compatible controller: Intel Corporation Mobile
945GM/GMS/940GML 
Express Integrated Graphics Controller (rev 03)

trapnihasute ~ $ ./g
XXX:24:               0:0
XXX:24:               0:0
XXX:24:               0:0

funny, as I know I definitely got a much bigger and
number-rich result  home 
(using GFX 8500 GT)

Regards,
Christian Parpart.

_______________________________________________
Kwin mailing list
Kwinkde.org
https://ma
il.kde.org/mailman/listinfo/kwin

On Wednesday 20 of June 2007, Christian Parpart wrote:
> On Tuesday 19 June 2007 15:31:11 Lubos Lunak wrote:
> >  Hello,
> >
> >  could people who _don't_ have nvidia and can run
KWin in TFP mode please
> > run the attached testapp and give me the output
(plus what card etc.)?
> > Thanks
>
> here, I'm running an intel mac mini:
>
> 00:02.0 VGA compatible controller: Intel Corporation
Mobile
> 945GM/GMS/940GML Express Integrated Graphics Controller
(rev 03)
>
> trapnihasute ~ $ ./g
> XXX:24:               0:0
> XXX:24:               0:0
> XXX:24:               0:0

 Hmm, ok, it seems here GLX_BIND_TO_TEXTURE_TARGETS_EXT is
not supported at 
all ...

> funny, as I know I definitely got a much bigger and
number-rich result 
> home (using GFX 8500 GT)

 ... in contrast to nVidia where
GLX_BIND_TO_TEXTURE_TARGETS_EXT returns 
nonsense.

 Ok, I'll need a different way of working around the nvidia
bug where TFP POT 
textures fail :-/.

-- 
Lubos Lunak
KDE developer
------------------------------------------------------------
--
SUSE LINUX, s.r.o.   e-mail: l.lunaksuse.cz , l.lunakkde.org
Lihovarska 1060/12   tel: +420 284 028 972
190 00 Prague 9      fax: +420 284 028 951
Czech Republic       http//www.suse.cz
_______________________________________________
Kwin mailing list
Kwinkde.org
https://ma
il.kde.org/mailman/listinfo/kwin

On Monday 25 of June 2007, Christian Parpart wrote:
> On Thursday 21 June 2007 17:02:06 Lubos Lunak wrote:
> >  Ok, I'll need a different way of working around
the nvidia bug where TFP
> > POT textures fail :-/.
>
> did you hug nvidia about this potential "bug"
already? might be more
> helpful in getting a fix from upstream than working
around the bug 

 It's been already fixed in the latest nvidia drivers. The
ones with which two 
out of my three gfx cards will never work because of being
"legacy".

-- 
Lubos Lunak
KDE developer
------------------------------------------------------------
--
SUSE LINUX, s.r.o.   e-mail: l.lunaksuse.cz , l.lunakkde.org
Lihovarska 1060/12   tel: +420 284 028 972
190 00 Prague 9      fax: +420 284 028 951
Czech Republic       http//www.suse.cz
_______________________________________________
Kwin mailing list
Kwinkde.org
https://ma
il.kde.org/mailman/listinfo/kwin

On Tuesday 26 of June 2007, Ignacio Castaño wrote:
> On Lunes 25 Junio 2007, Lubos Lunak wrote:
> > On Monday 25 of June 2007, Christian Parpart
wrote:
> > > On Thursday 21 June 2007 17:02:06 Lubos Lunak
wrote:
> > > >  Ok, I'll need a different way of
working around the nvidia bug where
> > > > TFP POT textures fail :-/.
> > >
> > > did you hug nvidia about this potential
"bug" already? might be more
> > > helpful in getting a fix from upstream than
working around the bug 
> >
> >  It's been already fixed in the latest nvidia
drivers. The ones with
> > which two out of my three gfx cards will never
work because of being
> > "legacy".
>
> Haven't followed the whole thread, but if you send me
more details about
> the issue to: icastanonvidia.com I'll contact the
appropiate people and
> push to get it fixed.

 There's nothing to fix, technically. It's been fixed in the
latest drivers. 
My problem is that those don't support many older cards (two
out of my three, 
among others) and the fix is not going to be backported 
(ht
tps://bugzilla.novell.com/show_bug.cgi?id=252951). I'll
be just a simple 
workaround somewhere in KWin anyway, I hope.

-- 
Lubos Lunak
KDE developer
------------------------------------------------------------
--
SUSE LINUX, s.r.o.   e-mail: l.lunaksuse.cz , l.lunakkde.org
Lihovarska 1060/12   tel: +420 284 028 972
190 00 Prague 9      fax: +420 284 028 951
Czech Republic       http//www.suse.cz
_______________________________________________
Kwin mailing list
Kwinkde.org
https://ma
il.kde.org/mailman/listinfo/kwin

On Martes 26 Junio 2007, Lubos Lunak wrote:
>  There's nothing to fix, technically. It's been fixed
in the latest
> drivers. My problem is that those don't support many
older cards (two out
> of my three, among others) and the fix is not going to
be backported
> (ht
tps://bugzilla.novell.com/show_bug.cgi?id=252951). I'll
be just a simple
> workaround somewhere in KWin anyway, I hope.

Lubos,

the following legacy driver appears to have fixed the bug
you refer to:

http://www.nvidia.com/object/linux_display_ia32_1.0
-9639.html

at least your test app works fine on it. Can you confirm
that it solves your 
problem?

-- 
Ignacio Castaño
castanyoyahoo.es
_______________________________________________
Kwin mailing list
Kwinkde.org
https://ma
il.kde.org/mailman/listinfo/kwin

On Wednesday 27 of June 2007, Ignacio Castaño wrote:
> On Martes 26 Junio 2007, Lubos Lunak wrote:
> >  There's nothing to fix, technically. It's been
fixed in the latest
> > drivers. My problem is that those don't support
many older cards (two out
> > of my three, among others) and the fix is not
going to be backported
> > (ht
tps://bugzilla.novell.com/show_bug.cgi?id=252951). I'll
be just a
> > simple workaround somewhere in KWin anyway, I
hope.
>
> Lubos,
>
> the following legacy driver appears to have fixed the
bug you refer to:
>
> http://www.nvidia.com/object/linux_display_ia32_1.0
-9639.html
>
> at least your test app works fine on it. Can you
confirm that it solves
> your problem?

 Yes, that seems to have fixed the problem, great, thanks. I
think I didn't 
notice this driver because of getting confused by the
versioning of different 
driver series (9755,9639,100.14.09).

-- 
Lubos Lunak
KDE developer
------------------------------------------------------------
--
SUSE LINUX, s.r.o.   e-mail: l.lunaksuse.cz , l.lunakkde.org
Lihovarska 1060/12   tel: +420 284 028 972
190 00 Prague 9      fax: +420 284 028 951
Czech Republic       http//www.suse.cz
_______________________________________________
Kwin mailing list
Kwinkde.org
https://ma
il.kde.org/mailman/listinfo/kwin

On Miércoles 27 Junio 2007, Lubos Lunak wrote:
>  Yes, that seems to have fixed the problem, great,
thanks. I think I didn't
> notice this driver because of getting confused by the
versioning of
> different driver series (9755,9639,100.14.09).

No problem, if you ever encounter problems like that feel
free to contact me.

-- 
Ignacio Castaño
castanyoyahoo.es
_______________________________________________
Kwin mailing list
Kwinkde.org
https://ma
il.kde.org/mailman/listinfo/kwin