Sorry, my mistake.

I had to choose depth=32 instead of 24 and now the gtk-demo looks fine.

What is with your patch? Is it safe to use it or is it only a quick
hack that needs more work to be done?

Regards and thanks a lot,
Matthieu.

On Fri, Jun 13, 2008 at 1:01 PM, Matthieu Poullet
<[EMAIL PROTECTED]> wrote:
> Hi,
>
> When I comment these two lines, the gtk-demo starts but the screen
> looks very strange (I'd say horizontally squeezed and colors seem
> wrong).
>
> I use an nvidia GeForce 8800GT and the nvidiafb driver, but I also had
> the same problem in VirtualBox 1.6.0 with the vesa driver. All my
> tests were done in Ubuntu 8.04.
>
> What could I do next?
>
> Regards,
> Matthieu.
>
> On Fri, Jun 13, 2008 at 10:50 AM, Denis Oliver Kropp <[EMAIL PROTECTED]> 
> wrote:
>> Matthieu Poullet wrote:
>>>
>>> (*) FBDev/Surface: Allocated 1600x1200 32 bit RGB24 buffer (index 0)
>>> at offset 0 and pitch 6400.
>>> (!) DirectFB/FBDev: Panning display failed (x=0 y=0 ywrap=0 vbl=1)!
>>>   --> Invalid argument
>>> (!) Core/LayerRegion: realize_region() in dfb_layer_region_flip_update()
>>> failed!
>>>   --> General failure!
>>> (!) [Main Thread       0.000] (21999) *** Assertion
>>> [surface->num_buffers > 0] failed *** [../../src/core/surface.h:287 in
>>> dfb_surface_get_buffer()]
>>
>> I don't know why panning fails with some drivers and it has just been
>> ignored,
>> but I think it is new that the error is returned.
>>
>> What happens if you comment out the two lines in systems/fbdev/fbdev.c as
>> indicated?
>>
>> static DFBResult
>> primarySetRegion( CoreLayer                  *layer,
>>                  void                       *driver_data,
>>                  void                       *layer_data,
>>                  void                       *region_data,
>>                  CoreLayerRegionConfig      *config,
>>                  CoreLayerRegionConfigFlags  updated,
>>                  CoreSurface                *surface,
>>                  CorePalette                *palette,
>>                  CoreSurfaceBufferLock      *lock )
>> {
>>     DFBResult    ret;
>>     FBDevShared *shared = dfb_fbdev->shared;
>>
>>     if (updated & CLRCF_SOURCE) {
>>          ret = dfb_fbdev_pan( config->source.x, lock->offset / lock->pitch +
>> config->source.y, true );
>> //          if (ret)             <======
>> //               return ret;     <======
>>     }
>>
>>     if ((updated & CLRCF_PALETTE) && palette)
>>          dfb_fbdev_set_palette( palette );
>>
>>     /* remember configuration */
>>     shared->config = *config;
>>
>>     return DFB_OK;
>> }
>>
>>
>> --
>> Best regards,
>>  Denis Oliver Kropp
>>
>> .------------------------------------------.
>> | DirectFB - Hardware accelerated graphics |
>> | http://www.directfb.org/                 |
>> "------------------------------------------"
>>
>

_______________________________________________
directfb-users mailing list
directfb-users@directfb.org
http://mail.directfb.org/cgi-bin/mailman/listinfo/directfb-users

Reply via email to