Cycles: Add sample-based runtime profiler that measures time spent in various parts of the CPU kernel

This commit adds a sample-based profiler that runs during CPU rendering and collects statistics on time spent in different parts of the kernel (ray intersection, shader evaluation etc.) as well as time spent per material and object.

The results are currently not exposed in the user interface or per Python yet, to see the stats on the console pass the "--cycles-print-stats" argument to Cycles (e.g. "./blender -- --cycles-print-stats").

Unfortunately, there is no clear way to extend this functionality to CUDA or OpenCL, so it is CPU-only for now.

Reviewers: brecht, sergey, swerner

Reviewed By: brecht, swerner

Differential Revision: https://developer.blender.org/D3892

1 files changed, 10 insertions, 0 deletions

diff --git a/intern/cycles/kernel/bvh/bvh.h b/intern/cycles/kernel/bvh/bvh.h
index 6708a3efac1..284b1e9208c 100644
--- a/intern/cycles/kernel/bvh/bvh.h
+++ b/intern/cycles/kernel/bvh/bvh.h
@@ -186,6 +186,8 @@ ccl_device_intersect bool scene_intersect(KernelGlobals *kg,
                                           float difl,
                                           float extmax)
 {
+	PROFILING_INIT(kg, PROFILING_INTERSECT);
+
 	if(!scene_intersect_valid(&ray)) {
 		return false;
 	}
@@ -248,6 +250,8 @@ ccl_device_intersect bool scene_intersect_local(KernelGlobals *kg,
                                                 uint *lcg_state,
                                                 int max_hits)
 {
+	PROFILING_INIT(kg, PROFILING_INTERSECT_LOCAL);
+
 	if(!scene_intersect_valid(&ray)) {
 		return false;
 	}
@@ -327,6 +331,8 @@ ccl_device_intersect bool scene_intersect_shadow_all(KernelGlobals *kg,
                                                      uint max_hits,
                                                      uint *num_hits)
 {
+	PROFILING_INIT(kg, PROFILING_INTERSECT_SHADOW_ALL);
+
 	if(!scene_intersect_valid(ray)) {
 		return false;
 	}
@@ -407,6 +413,8 @@ ccl_device_intersect bool scene_intersect_volume(KernelGlobals *kg,
                                                  Intersection *isect,
                                                  const uint visibility)
 {
+	PROFILING_INIT(kg, PROFILING_INTERSECT_VOLUME);
+
 	if(!scene_intersect_valid(ray)) {
 		return false;
 	}
@@ -438,6 +446,8 @@ ccl_device_intersect uint scene_intersect_volume_all(KernelGlobals *kg,
                                                      const uint max_hits,
                                                      const uint visibility)
 {
+	PROFILING_INIT(kg, PROFILING_INTERSECT_VOLUME_ALL);
+
 	if(!scene_intersect_valid(ray)) {
 		return false;
 	}