summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
-rw-r--r--src/mesa/main/KNOWN_BUGS25
1 files changed, 19 insertions, 6 deletions
diff --git a/src/mesa/main/KNOWN_BUGS b/src/mesa/main/KNOWN_BUGS
index 9c9076bfedc..42c52762b20 100644
--- a/src/mesa/main/KNOWN_BUGS
+++ b/src/mesa/main/KNOWN_BUGS
@@ -1,15 +1,17 @@
-$Id: KNOWN_BUGS,v 1.1 1999/08/19 00:55:41 jtg Exp $
+$Id: KNOWN_BUGS,v 1.2 2000/02/06 17:04:56 brianp Exp $
Performance issues with EXT_point_parameters & quake2
-Broken drivers:
- --> After integration of the changes in kw3, only the X and FX
-drivers are known to work. Windows and D3D are known to be broken,
-and all others are suspected to be broken. Please test your driver
-and update this entry when more is known.
+Using glPolygonMode() where the front and back modes aren't the
+same causes poor performance on 3Dfx.
+
+
+
+Broken drivers: some of the Mesa device drivers (such as BeOS, D3D,
+etc) haven't been updated for Mesa 3.3's device driver changes.
@@ -18,3 +20,14 @@ lines. Also, will have to add specular color add to pb.c (pixel buffer
code).
+
+Joe Bronkema reported a bug when drawing a series of long GL_LINE_STRIP
+primitives. Putting a glFlush() between the primitives is a work-around.
+
+
+
+glDrawRangeElements() should use vertex array locking to improve performance
+but trying to do so causes a rendering error. Reported by Scott McMillan.
+Fixed by disabling locking in glDrawRangeElements (varray.c) but that's
+really just hiding a bug in array locking.
+