[Bug 85529] Surfaces not drawn in Unvanquished
bugzilla-daemon at freedesktop.org
bugzilla-daemon at freedesktop.org
Thu Dec 18 04:43:54 PST 2014
https://bugs.freedesktop.org/show_bug.cgi?id=85529
Kenneth Graunke <kenneth at whitecape.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |ASSIGNED
CC|kenneth at whitecape.org |idr at freedesktop.org
Assignee|idr at freedesktop.org |kenneth at whitecape.org
--- Comment #9 from Kenneth Graunke <kenneth at whitecape.org> ---
(In reply to vcelestialragev from comment #4)
> Created attachment 110839 [details]
> A patch that "fixes" the issue
>
> The patch just partially reverts the original commit that introduced the
> regression. It is probably not a proper fix but my help provide clues about
> where the problem actually occurs.
I'm pretty sure your patch is correct, actually.
After ~2 hours of staring at the code, and adding some assertions, I realized
that if we have multiple _mesa_prims and somehow don't flag BRW_NEW_VERTICES on
the second one, brw_try_draw_prims resets start/base_vertex_location to
prim[i].{start,basevertex}, but we don't call
brw_prepare_shader_draw_parameters, so we fail to add in
brw->vb.start_vertex_bias and brw->ib.start_vertex_offset when processing
subsequent primitives.
Your patch makes us no longer add brw->vb.start_vertex_bias to the value of
gl_BaseVertexARB. However, I think we aren't *supposed* to add it in the first
place, so the patch actually fixes two bugs.
I'd like to revert more, restoring more of the older/simpler code. I'll send
out a patch shortly.
--
You are receiving this mail because:
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/intel-3d-bugs/attachments/20141218/2d56f0e6/attachment.html>
More information about the intel-3d-bugs
mailing list