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

Cycles: Split BVH nodes storage into inner and leaf nodes

Browse Source 
This way we can get rid of inefficient memory usage caused by BVH boundbox
part being unused by leaf nodes but still being allocated for them. Doing
such split allows to save 6 of float4 values for QBVH per leaf node and 3
of float4 values for regular BVH per leaf node.

This translates into following memory save using 01.01.01.G rendered
without hair:

                   Device memory size   Device memory peak   Global memory peak
Before the patch:  4957                 5051                 7668
With the patch:    4467                 4562                 7332

The measurements are done against current master. Still need to run speed tests
and it's hard to predict if it's faster or not: on the one hand leaf nodes are
now much more coherent in cache, on the other hand they're not so much coherent
with regular nodes anymore.

Reviewers: brecht, juicyfruit

Subscribers: venomgfx, eyecandy

Differential Revision: https://developer.blender.org/D1236
This commit is contained in:
Sergey Sharybin
2015-04-13 23:05:09 +05:00
parent cd44449578
commit 828abaf11c
15 changed files with 151 additions and 95 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
intern/cycles/kernel/svm/svm_image.h
View File

@@ -251,9 +251,9 @@ ccl_device float4 svm_image_texture(KernelGlobals *kg, int id, float x, float y,
case 95: r = kernel_tex_image_interp(__tex_image_095, x, y); break;
case 96: r = kernel_tex_image_interp(__tex_image_096, x, y); break;
case 97: r = kernel_tex_image_interp(__tex_image_097, x, y); break;
case 98: r = kernel_tex_image_interp(__tex_image_098, x, y); break;
#if defined(__CUDA_ARCH__) && (__CUDA_ARCH__ >= 300)
case 98: r = kernel_tex_image_interp(__tex_image_098, x, y); break;
case 99: r = kernel_tex_image_interp(__tex_image_099, x, y); break;
case 100: r = kernel_tex_image_interp(__tex_image_100, x, y); break;
case 101: r = kernel_tex_image_interp(__tex_image_101, x, y); break;