diff options
Diffstat (limited to 'intern/cycles/kernel/device/gpu')
-rw-r--r-- | intern/cycles/kernel/device/gpu/kernel.h | 9 | ||||
-rw-r--r-- | intern/cycles/kernel/device/gpu/work_stealing.h | 52 |
2 files changed, 57 insertions, 4 deletions
diff --git a/intern/cycles/kernel/device/gpu/kernel.h b/intern/cycles/kernel/device/gpu/kernel.h index 335cb1ec0c0..aa360b3016a 100644 --- a/intern/cycles/kernel/device/gpu/kernel.h +++ b/intern/cycles/kernel/device/gpu/kernel.h @@ -19,6 +19,7 @@ #include "kernel/device/gpu/parallel_active_index.h" #include "kernel/device/gpu/parallel_prefix_sum.h" #include "kernel/device/gpu/parallel_sorted_index.h" +#include "kernel/device/gpu/work_stealing.h" #include "kernel/integrator/integrator_state.h" #include "kernel/integrator/integrator_state_flow.h" @@ -36,10 +37,10 @@ #include "kernel/integrator/integrator_shade_surface.h" #include "kernel/integrator/integrator_shade_volume.h" -#include "kernel/kernel_adaptive_sampling.h" -#include "kernel/kernel_bake.h" -#include "kernel/kernel_film.h" -#include "kernel/kernel_work_stealing.h" +#include "kernel/bake/bake.h" + +#include "kernel/film/film_adaptive_sampling.h" +#include "kernel/film/film_read.h" /* -------------------------------------------------------------------- * Integrator. diff --git a/intern/cycles/kernel/device/gpu/work_stealing.h b/intern/cycles/kernel/device/gpu/work_stealing.h new file mode 100644 index 00000000000..fab0915c38e --- /dev/null +++ b/intern/cycles/kernel/device/gpu/work_stealing.h @@ -0,0 +1,52 @@ +/* + * Copyright 2011-2015 Blender Foundation + * + * Licensed under the Apache License, Version 2.0 (the "License"); + * you may not use this file except in compliance with the License. + * You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +#pragma once + +CCL_NAMESPACE_BEGIN + +/* + * Utility functions for work stealing + */ + +/* Map global work index to tile, pixel X/Y and sample. */ +ccl_device_inline void get_work_pixel(ccl_global const KernelWorkTile *tile, + uint global_work_index, + ccl_private uint *x, + ccl_private uint *y, + ccl_private uint *sample) +{ +#if 0 + /* Keep threads for the same sample together. */ + uint tile_pixels = tile->w * tile->h; + uint sample_offset = global_work_index / tile_pixels; + uint pixel_offset = global_work_index - sample_offset * tile_pixels; +#else + /* Keeping threads for the same pixel together. + * Appears to improve performance by a few % on CUDA and OptiX. */ + uint sample_offset = global_work_index % tile->num_samples; + uint pixel_offset = global_work_index / tile->num_samples; +#endif + + uint y_offset = pixel_offset / tile->w; + uint x_offset = pixel_offset - y_offset * tile->w; + + *x = tile->x + x_offset; + *y = tile->y + y_offset; + *sample = tile->start_sample + sample_offset; +} + +CCL_NAMESPACE_END |