diff options
| author | Mai Lavelle <mai.lavelle@gmail.com> | 2017-03-01 09:47:08 +0300 |
|---|---|---|
| committer | Mai Lavelle <mai.lavelle@gmail.com> | 2017-03-08 09:31:07 +0300 |
| commit | cd7d5669d17070799e2d2a2b28f58a06c3417d7b (patch) | |
| tree | c5420e4cd9432c44575c1e489b9890ddd8439ba9 /intern/cycles/kernel/kernels/cuda | |
| parent | 4cf501b83557ed5d64dbd2ddb13e1e8c5add88f5 (diff) | |
Cycles: Remove sum_all_radiance kernel
This was only needed for the previous implementation of parallel samples. As
we don't have that any more it can be removed.
Real reason for removal tho is this: `per_sample_output_buffers` was being
calculated too small and artifacts resulted. The tile buffer is already
the correct size and calculating the size for `per_sample_output_buffers`
is a bit difficult with the current layout of the code. As
`per_sample_output_buffers` was only needed for `sum_all_radiance`,
removing that kernel and writing output to the tile buffer directly
fixes the artifacts.
Diffstat (limited to 'intern/cycles/kernel/kernels/cuda')
| -rw-r--r-- | intern/cycles/kernel/kernels/cuda/kernel_split.cu | 2 |
1 files changed, 0 insertions, 2 deletions
diff --git a/intern/cycles/kernel/kernels/cuda/kernel_split.cu b/intern/cycles/kernel/kernels/cuda/kernel_split.cu index 3a883265157..53a36b15e40 100644 --- a/intern/cycles/kernel/kernels/cuda/kernel_split.cu +++ b/intern/cycles/kernel/kernels/cuda/kernel_split.cu @@ -35,7 +35,6 @@ #include "../../split/kernel_direct_lighting.h" #include "../../split/kernel_shadow_blocked.h" #include "../../split/kernel_next_iteration_setup.h" -#include "../../split/kernel_sum_all_radiance.h" #include "../../kernel_film.h" @@ -92,7 +91,6 @@ DEFINE_SPLIT_KERNEL_FUNCTION(holdout_emission_blurring_pathtermination_ao) DEFINE_SPLIT_KERNEL_FUNCTION(direct_lighting) DEFINE_SPLIT_KERNEL_FUNCTION(shadow_blocked) DEFINE_SPLIT_KERNEL_FUNCTION(next_iteration_setup) -DEFINE_SPLIT_KERNEL_FUNCTION(sum_all_radiance) extern "C" __global__ void CUDA_LAUNCH_BOUNDS(CUDA_THREADS_BLOCK_WIDTH, CUDA_KERNEL_MAX_REGISTERS) |