Blender Git Statistics -> Branches -> split-kernel-faster-building

"Split-kernel-faster-building" branch

Total commits : 22
Total committers : 1
First Commit : November 11, 2017
Latest Commit : November 30, 2017


Commits by Date

DateNumber of Commits
November 30, 20171
November 29, 20170
November 28, 20173
November 27, 20177
November 26, 20171
November 25, 20170
November 24, 20170
November 23, 20170
November 22, 20170
November 21, 20170
November 20, 20170
November 19, 20170
November 18, 20170
November 17, 20170
November 16, 20170
November 15, 20170
November 14, 20175
November 13, 20170
November 12, 20170
November 11, 20175

Committers

AuthorNumber of Commits
Mai Lavelle22

Popular Files

FilenameTotal Edits
kernel_types.h7
kernel_emission.h6
kernel_shader.h6
kernel_shader_eval.h6
kernel_split_data_types.h6
device_split_kernel.cpp4
kernel_direct_lighting.h4
kernel_path.h4
kernel_shadow_blocked_dl.h4
kernel_cpu_impl.h3

Latest commits Feed

November 30, 2017, 01:36 (GMT)
Cycles: Fix rendering of barbershop scene
November 28, 2017, 21:58 (GMT)
Merge branch 'master' into split-kernel-faster-building
November 28, 2017, 10:49 (GMT)
Cycles: Fix issue with AO in split kernel_path_background

Watch out for changes to variables passed by value, these changes
don't persist across the split.
November 28, 2017, 05:28 (GMT)
Cycles: Fix build with OSL enabled
November 27, 2017, 07:50 (GMT)
Cycles: Split shader_eval out of the kernel_lamp_emission kernel

With this kernels for BWM and classroom scenes are building in half the
time as master. Render times are 1% faster as well.
November 27, 2017, 07:49 (GMT)
Cycles: Split indirect_lamp_emission
November 27, 2017, 07:49 (GMT)
Cycles: Split kernel_path_lamp_emission
November 27, 2017, 07:49 (GMT)
Cycles: Only use first lamp in indirect_lamp_emission

Produces different results when lamps overlap from POV of a ray, but allows
for this function to be split. The differences probably wont be noticeable
in most scenes. Old behavior could be reattained by placing lamps
into the BVH.
November 27, 2017, 05:51 (GMT)
Cycles: Fix mistake with ShaderEvalTask that could lead to crash
November 27, 2017, 02:48 (GMT)
Cycles: Merge shadow_blocked_ao and _dl kernels to improve build times

Gives another 4 seconds improvement.
November 27, 2017, 01:55 (GMT)
Cycles: Remove early out that may cause threads to diverge before barrier
November 26, 2017, 08:07 (GMT)
Cycles: Use ShaderEvalIntent more to pass flag, closures, and eval_task less

This makes it so that path_flag and max_closures are not passed to shader eval
functions, instead a limited number of ShaderEvalIntents are used. This also
removes the need for ShaderEvalTask in the mega kernel and simplifies the
code a bit.
November 14, 2017, 06:36 (GMT)
Cycles: Modify kernel_indirect_background to remove call to svm_eval_nodes

Speeds up kernel building by another 2-3 seconds. Rendering is
another 1% slower.
November 14, 2017, 06:26 (GMT)
Cycles: Add shader_eval_state to allow eval of ray states besides RAY_ACTIVE
November 14, 2017, 05:13 (GMT)
Cycles: Split kernel_path_background
November 14, 2017, 05:13 (GMT)
Cycles: Split the indirect_background function
November 14, 2017, 03:45 (GMT)
Cycles: Fix shadow catcher and branched path after recent split changes
November 11, 2017, 11:45 (GMT)
Cycles: Split apart direct_emission around shader_eval
November 11, 2017, 11:45 (GMT)
Cycles: Speed up split kernel builds by using split direct_emission functions

This removes the indirect call to `svm_eval_nodes` from the
`kernel_direct_lighting` kernel, and speeds up kernel build times
by 4-5 seconds. Render time is slower by ~1%. It should be possible
to apply this technique to other areas of the kernel to get even
better build times.

Note: this breaks branched path and shadow catcher in the split
kernel. Should be an easy fix, will do later.
November 11, 2017, 10:56 (GMT)
Cycles: Split direct_emissive_eval around shader_eval

MiikaHweb - Blender Git Statistics v1.06
By: Miika HämäläinenLast update: Nov-07-2014 14:18 MiikaHweb | 2003-2019