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

2 files changed, 6 insertions, 2 deletions

diff --git a/intern/cycles/blender/blender_session.cpp b/intern/cycles/blender/blender_session.cpp
index 75c7dcee05e..30ae0bc813d 100644
--- a/intern/cycles/blender/blender_session.cpp
+++ b/intern/cycles/blender/blender_session.cpp
@@ -516,7 +516,7 @@ void BlenderSession::render()
 
 		if(!b_engine.is_preview() && background && print_render_stats) {
 			RenderStats stats;
-			session->scene->collect_statistics(&stats);
+			session->collect_statistics(&stats);
 			printf("Render statistics:\n%s\n", stats.full_report().c_str());
 		}
 
diff --git a/intern/cycles/blender/blender_sync.cpp b/intern/cycles/blender/blender_sync.cpp
index 832847c179f..49b046d0a88 100644
--- a/intern/cycles/blender/blender_sync.cpp
+++ b/intern/cycles/blender/blender_sync.cpp
@@ -874,7 +874,8 @@ SessionParams BlenderSync::get_session_params(BL::RenderEngine& b_engine,
 	}
 
 	/* tiles */
-	if(params.device.type != DEVICE_CPU && !background) {
+	const bool is_cpu = (params.device.type == DEVICE_CPU);
+	if(!is_cpu && !background) {
 		/* currently GPU could be much slower than CPU when using tiles,
 		 * still need to be investigated, but meanwhile make it possible
 		 * to work in viewport smoothly
@@ -960,6 +961,9 @@ SessionParams BlenderSync::get_session_params(BL::RenderEngine& b_engine,
 		params.progressive_update_timeout = 0.1;
 	}
 
+	params.use_profiling = params.device.has_profiling && !b_engine.is_preview() &&
+	                       background && BlenderSession::print_render_stats;
+
 	return params;
 }