Re: Cannot start GTK apps - gdk_visual_decompose_mask



On Thu, 2005-06-30 at 17:36 +0100, David Saxton wrote:
> On Thursday 30 Jun 2005 14:26, you wrote:
> > You don't really give us enough information to guess here. (My initial
> > reaction is: that can't happen!)
> >
> > But we'd appreciate a bug in bugzilla, with:
> >
> >  - Your GTK+ version
> >  - Information about your operating system (what distro and version,
> >    if Linux)
> >  - Information about your X server
> >  - The output of 'xdpyinfo' as an attachment
> >
> Before filing a bug with gtk, I want to check that it isn't either a bug 
> elsewhere (such as in X).
> 
> The output of 'xdpyinfo' gives (amongst other things) this interesting item:
> 
>   visual:
>     visual id:    0x3e
>     class:    DirectColor
>     depth:    24 planes
>     available colormap entries:    256 per subfield
>     red, green, blue masks:    0xff0000, 0x0, 0xff
>     significant bits in color specification:    8 bits
> 
> where presumably the "0x0" for the green mask is causing the problem - indeed, 
> 0 is the value in visual_list in the _gdk_visual_init function returned from 
> XGetVisualInfo for that visual.
> 
> All the other visuals (as shown by xpdyinfo) have non-zero masks.

This looks very much like an X bug ... I have no idea what this could
mean... you could have only images that were different shades of red and
blue.

On the other hand, I don't think that GTK+ should go into an infinite
loop here, and it's easy enough to catch and avoid. So a bugzilla
bug would be appreciated.

Since it's a DirectColor visual, it's almost certainly not your default
visual, so it doesn't really matter what GTK+ does with it as long as it
doesn't hang.

Regards,
							Owen

Attachment: signature.asc
Description: This is a digitally signed message part



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]