Files
test2/intern
Hristo Gueorguiev 6bf4115c13 Cycles: Split kernel - sort shaders
Reduce thread divergence in kernel_shader_eval.

Rays are sorted in blocks of 2048 according to shader->id.

On R9 290 Classroom is ~30% faster, and Pabellon Barcelone is ~8% faster.

No sorting for CUDA split kernel.

Reviewers: sergey, maiself

Reviewed By: maiself

Differential Revision: https://developer.blender.org/D2598
2017-05-03 15:30:45 +02:00
..
2016-01-04 14:20:48 +05:00
2017-03-12 02:47:53 +11:00
2016-01-04 14:20:48 +05:00
2016-01-04 14:20:48 +05:00
2016-01-04 14:20:48 +05:00
2016-07-02 10:08:33 +10:00
2016-01-04 14:20:48 +05:00
2016-06-11 22:05:20 +02:00
2016-01-04 14:20:48 +05:00
2017-03-12 02:47:53 +11:00