Cycles: OpenCL split kernel refactor

This does a few things at once:

- Refactors host side split kernel logic into a new device
  agnostic class `DeviceSplitKernel`.
- Removes tile splitting, a new work pool implementation takes its place and
  allows as many threads as will fit in memory regardless of tile size, which
  can give performance gains.
- Refactors split state buffers into one buffer, as well as reduces the
  number of arguments passed to kernels. Means there's less code to deal
  with overall.
- Moves kernel logic out of OpenCL kernel files so they can later be used by
  other device types.
- Replaced OpenCL specific APIs with new generic versions
- Tiles can now be seen updating during rendering

1 files changed, 3 insertions, 3 deletions

diff --git a/intern/cycles/kernel/kernel_shadow.h b/intern/cycles/kernel/kernel_shadow.h
index 06a77a208cb..6abfa9c3873 100644
--- a/intern/cycles/kernel/kernel_shadow.h
+++ b/intern/cycles/kernel/kernel_shadow.h
@@ -248,7 +248,7 @@ ccl_device bool shadow_blocked_transparent_all(KernelGlobals *kg,
 }
 #  endif  /* __SHADOW_RECORD_ALL__ */
 
-#  ifdef __KERNEL_GPU__
+#  if defined(__KERNEL_GPU__) || !defined(__SHADOW_RECORD_ALL__)
 /* Shadow function to compute how much light is blocked,
  *
  * Here we raytrace from one transparent surface to the next step by step.
@@ -359,7 +359,7 @@ ccl_device bool shadow_blocked_transparent_stepped(
 	                                               shadow);
 }
 
-#  endif  /* __KERNEL_GPU__ */
+#  endif  /* __KERNEL_GPU__ || !__SHADOW_RECORD_ALL__ */
 #endif /* __TRANSPARENT_SHADOWS__ */
 
 ccl_device_inline bool shadow_blocked(KernelGlobals *kg,
@@ -374,7 +374,7 @@ ccl_device_inline bool shadow_blocked(KernelGlobals *kg,
 #ifdef __SPLIT_KERNEL__
 	Ray private_ray = *ray_input;
 	Ray *ray = &private_ray;
-	Intersection *isect = &kg->isect_shadow[SD_THREAD];
+	Intersection *isect = &kernel_split_state.isect_shadow[SD_THREAD];
 #else  /* __SPLIT_KERNEL__ */
 	Ray *ray = ray_input;
 	Intersection isect_object;