can you provide more traces (all CreateSurface related traces: context errors, glx errors, ...) ?
Hi,
The reason for CreateSurface failure is that it runs out of memory... I think this used to occur before the patch, and I just didn't notice it - Oliver said it creates surfaces until it runs out of memory. If trace is enabled, it shows the memory going down until: WARN("Out of 'bogus' video memory\n");
That still doesn't explain the white screen, however - that's a pbuffer related regression. There are no FIXME messages that I haven't seen before - and the trace is 90 MB large, and I don't know what to look for. I do _not_ see a trace saying "Error in setting current context".