[Xcb] mesa +xcb = assertion death

[Jamey Sharp]

On Fri, 2005-04-29 at 18:06 -0400, Jeremy Kolb wrote:
> Jamey Sharp wrote:
> >>However I have no idea why the generic reply is returning 0 as a 
> >>sequence number, nor do I know what response_type 5 is.
> > 
> > Response type 5 is a ButtonRelease event. (0 is an error, 1 is a reply,
> > and everything else is an event; you can look up core events in
> > xproto.xml, xproto.h, or your friendly neighborhood X protocol
> > reference.)
> 
> I saw that but disregarded it since no button has been hit at this 
> point, indeed there isn't a window and this happens in both glxgears 
> (eventually spawns a window when this problem doesn't occur) and glxinfo 
> (no window ever).

Ethereal, run on your tcpdump trace, clearly says that a bunch of events
with sequence number 0 immediately follow the reply to the last GLX
request made. But they look like nonsense events to me (weird timestamps
and window IDs; and getting KeymapNotify, ButtonRelease, EnterNotify,
KeymapNotify, DestroyNotify, ColormapNotify, nonexistent event 39, and
then the entire same sequence again seems quite improbable). I suspect
your X server is busted, and is reporting the wrong length on the
GLXVendorPrivateWithReply reply. Dunno why it would do that, of course.

Today I saw your blog entries on this, BTW. Sorry, can't say I agree
that it's XCB's fault. ;-)

--Jamey