Cycles: Remove sum_all_radiance kernel
authorMai Lavelle <mai.lavelle@gmail.com>
Wed, 1 Mar 2017 06:47:08 +0000 (01:47 -0500)
committerMai Lavelle <mai.lavelle@gmail.com>
Wed, 8 Mar 2017 06:31:07 +0000 (01:31 -0500)
commitcd7d5669d17070799e2d2a2b28f58a06c3417d7b
treec5420e4cd9432c44575c1e489b9890ddd8439ba9
parent4cf501b83557ed5d64dbd2ddb13e1e8c5add88f5
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.
13 files changed:
intern/cycles/device/device_split_kernel.cpp
intern/cycles/device/device_split_kernel.h
intern/cycles/kernel/CMakeLists.txt
intern/cycles/kernel/kernels/cpu/kernel_cpu.h
intern/cycles/kernel/kernels/cpu/kernel_cpu_impl.h
intern/cycles/kernel/kernels/cuda/kernel_split.cu
intern/cycles/kernel/kernels/opencl/kernel_sum_all_radiance.cl [deleted file]
intern/cycles/kernel/split/kernel_background_buffer_update.h
intern/cycles/kernel/split/kernel_data_init.h
intern/cycles/kernel/split/kernel_holdout_emission_blurring_pathtermination_ao.h
intern/cycles/kernel/split/kernel_path_init.h
intern/cycles/kernel/split/kernel_split_data.h
intern/cycles/kernel/split/kernel_sum_all_radiance.h [deleted file]