Brian Paul wrote:
>
> Ian Romanick wrote:
> >
> > On Sat, Mar 30, 2002 at 08:11:31AM -0700, Jens Owen wrote:
> > > Ian Romanick wrote:
> > > >
> > > > On Thu, Mar 28, 2002 at 12:53:06PM -0800, Ian Romanick wrote:
> > > >
> > > > > #0 0x4309fb0a in CreateContext (dpy=0x8075708, vis=0x0, shareList=0x0,
> > > > > allowDirect=1, contextID=0) at glxcmds.c:165
> > > > > #1 0x4309fc5d in glXCreateContext (dpy=0x8075708, vis=0x0, shareList=0x0,
> > > > > allowDirect=1) at glxcmds.c:206
> > > > >
> > > > > It dies because it dereferences vis, which is NULL. The man page for
> > > > > glXCreateContext doesn't say anything about the expected behavior if vis is
> > > > > NULL. Should it return NULL or ... ?
> > > >
> > > > I worked around this problem by making CreateContext retun NULL if vis =
> > > > NULL. This allows Maya to load.
> > >
> > > Ian,
> > >
> > > Andreas has also tried loading Maya and has seen similar problems. He
> > > elluded to Maya wanting an overlay visual. If would be interesting to
> > > see if the vis == NULL problem went away if we had an overlay visual.
> > > We could hack up the visual initialization code to lie and say we had
> > > overlay visuals, just to see if this is the problem...
>
> I think that would just lead to another failure later on.
Definitely a visual failure, since the layers would actual
interfere...however, I was theorizing that there was other logic failing
at the application level when an overlay visual wasn't present. Alas, I
looked at their supported cards on there web site and they do support
non-overlay cards in some capacity. So, I wouldn't bother with this
exploration.
> > I don't know about overlay visuals, but I do know that CreateContext should
> > probably test vis and give some sort of reasonable return. SEGFAULTing is
> > probably not the best answer. The curious thing is that this seems to be in
> > the SGI GLX code. I wonder what happens on Irix (or any other OS for that
> > matter) if glXCreateContext is called with vis==NULL.
>
> I don't have my Indigo2 set up right now so I can't test. But I think
> testing the visual and Display parameters for NULL and returning NULL
> is a reasonable thing to do. I've just checked in code to do that.
Sounds good.
-- /\
Jens Owen / \/\ _
[EMAIL PROTECTED] / \ \ \ Steamboat Springs, Colorado
_______________________________________________
Dri-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/dri-devel
