Cr8-xyz/blender
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

Cycles / Non-Progressive integrator:

Browse Source 
* Non-Progressive integrator is now available on the GPU (CUDA, sm_20 and above). 

Implementation details:
* kernel_path_trace() has been split up into two functions:
kernel_path_trace_non_progressive() and kernel_path_trace_progressive().

* We compile two CUDA kernel entry functions (in kernel.cu) for the two integrators, they are still inside one .cubin file but due to the kernel separation there should be no performance problem. I tested with the BMW file on my Geforce 540M and the render times were the same for 100 samples (1.57 min in my case).

This is part of my GSoC project, SVN merge of r59032 + manual merge of UI changes for this from my branch.
This commit is contained in:
Thomas Dinges
2013-08-09 18:47:25 +00:00
parent 2ab9cbd208
commit a18112249d
13 changed files with 112 additions and 55 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
intern/cycles/kernel/kernel.cu
View File

@@ -26,13 +26,22 @@
#include "kernel_path.h"
#include "kernel_displace.h"
extern "C" __global__ void kernel_cuda_path_trace(float *buffer, uint *rng_state, int sample, int sx, int sy, int sw, int sh, int offset, int stride)
extern "C" __global__ void kernel_cuda_path_trace_progressive(float *buffer, uint *rng_state, int sample, int sx, int sy, int sw, int sh, int offset, int stride)
{
int x = sx + blockDim.x*blockIdx.x + threadIdx.x;
int y = sy + blockDim.y*blockIdx.y + threadIdx.y;
if(x < sx + sw && y < sy + sh)
kernel_path_trace(NULL, buffer, rng_state, sample, x, y, offset, stride);
kernel_path_trace_progressive(NULL, buffer, rng_state, sample, x, y, offset, stride);
}
extern "C" __global__ void kernel_cuda_path_trace_non_progressive(float *buffer, uint *rng_state, int sample, int sx, int sy, int sw, int sh, int offset, int stride)
{
int x = sx + blockDim.x*blockIdx.x + threadIdx.x;
int y = sy + blockDim.y*blockIdx.y + threadIdx.y;
if(x < sx + sw && y < sy + sh)
kernel_path_trace_non_progressive(NULL, buffer, rng_state, sample, x, y, offset, stride);
}
extern "C" __global__ void kernel_cuda_tonemap(uchar4 *rgba, float *buffer, int sample, int sx, int sy, int sw, int sh, int offset, int stride)