[Bf-blender-cvs] [c9f0c20] master: Cycle: Add a bit of explanation why we're using such a dirty hack

Sergey Sharybin noreply at git.blender.org
Tue Sep 16 08:47:00 CEST 2014


Commit: c9f0c20e4800a8179560ec678a8ab8cb48283cf2
Author: Sergey Sharybin
Date:   Tue Sep 16 12:46:31 2014 +0600
Branches: master
https://developer.blender.org/rBc9f0c20e4800a8179560ec678a8ab8cb48283cf2

Cycle: Add a bit of explanation why we're using such a dirty hack

===================================================================

M	intern/cycles/kernel/kernel_types.h

===================================================================

diff --git a/intern/cycles/kernel/kernel_types.h b/intern/cycles/kernel/kernel_types.h
index a8c24f0..1f01622 100644
--- a/intern/cycles/kernel/kernel_types.h
+++ b/intern/cycles/kernel/kernel_types.h
@@ -528,6 +528,17 @@ typedef enum AttributeStandard {
 #define MAX_CLOSURE 1
 #endif
 
+/* TODO(sergey): This is rather nasty bug happening in here, which
+ * could be simply a compilers bug for which we can't find a generic
+ * platform indipendent workaround. Also even if it's a compiler
+ * issue, it's not so simple to upgrade the compiler in the release
+ * environment for linux and doing it so closer to the release is
+ * rather a risky business.
+ *
+ * For this release it's probably safer to stick with such a rather
+ * dirty solution, and look for a cleaner fix during the next release
+ * cycle.
+ */
 typedef struct ShaderClosure {
 	ClosureType type;
 	float3 weight;




More information about the Bf-blender-cvs mailing list