Cr8-xyz/blender
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity
Files
c7e25a25b06659ed41e9f4ee101a0980135cdb24
blender/intern/cycles/kernel/geom/triangle.h

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

358 lines
13 KiB
C
Raw Normal View History

Cycles: use SPDX license headers * Replace license text in headers with SPDX identifiers. * Remove specific license info from outdated readme.txt, instead leave details to the source files. * Add list of SPDX license identifiers used, and corresponding license texts. * Update copyright dates while we're at it. Ref D14069, T95597
2022-02-11 13:53:21 +01:00
/* SPDX-License-Identifier: Apache-2.0
* Copyright 2011-2022 Blender Foundation */
Fix for attribute lookup in OSL. This uses a map in the OSL globals instead of the device texture.
2012-09-05 17:08:56 +00:00
Cycles code refactor: move more code to geom folder, add some comments.
2014-03-29 13:03:48 +01:00
/* Triangle Primitive
*
Cycles: Implement watertight ray/triangle intersection Using this paper: Sven Woop, Watertight Ray/Triangle Intersection http://jcgt.org/published/0002/01/05/paper.pdf This change is expected to address quite reasonable amount of reports from the bug tracker, plus it might help reducing the noise in some scenes. Unfortunately, it's currently about 7% slower than the previous solution with pre-computed triangle plane equations, but maybe with some smart tweaks to the code (tests reshuffle, using SIMD in a nice way or so) we can avoid the speed regression. But perhaps smartest thing to do here would be to change single triangle / ray intersection with multiple triangles / ray intersections. That's how Embree does this and it's watertight single ray intersection is not any faster that this. Currently only triangle intersection is modified accordingly to the paper, in the future we would also want to modify the node / ray intersection. Reviewers: brecht, juicyfruit Subscribers: dingto, ton Differential Revision: https://developer.blender.org/D819
2014-12-15 21:18:01 +05:00
* Basic triangle with 3 vertices is used to represent mesh surfaces. For BVH
* ray intersection we use a precomputed triangle storage to accelerate
* intersection at the cost of more memory usage */
Cycles code refactor: move more code to geom folder, add some comments.
2014-03-29 13:03:48 +01:00
Cycles: merge of cycles-x branch, a major update to the renderer This includes much improved GPU rendering performance, viewport interactivity, new shadow catcher, revamped sampling settings, subsurface scattering anisotropy, new GPU volume sampling, improved PMJ sampling pattern, and more. Some features have also been removed or changed, breaking backwards compatibility. Including the removal of the OpenCL backend, for which alternatives are under development. Release notes and code docs: https://wiki.blender.org/wiki/Reference/Release_Notes/3.0/Cycles https://wiki.blender.org/wiki/Source/Render/Cycles Credits: * Sergey Sharybin * Brecht Van Lommel * Patrick Mours (OptiX backend) * Christophe Hery (subsurface scattering anisotropy) * William Leeson (PMJ sampling pattern) * Alaska (various fixes and tweaks) * Thomas Dinges (various fixes) For the full commit history, see the cycles-x branch. This squashes together all the changes since intermediate changes would often fail building or tests. Ref T87839, T87837, T87836 Fixes T90734, T89353, T80267, T80267, T77185, T69800
2021-09-20 17:59:20 +02:00
#pragma once
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
CCL_NAMESPACE_BEGIN
Cleanup: comment blocks, trailing space in comments
2021-06-24 15:56:58 +10:00
/* Normal on triangle. */
Cycles: replace integrator state argument macros * Rename struct KernelGlobals to struct KernelGlobalsCPU * Add KernelGlobals, IntegratorState and ConstIntegratorState typedefs that every device can define in its own way. * Remove INTEGRATOR_STATE_ARGS and INTEGRATOR_STATE_PASS macros and replace with these new typedefs. * Add explicit state argument to INTEGRATOR_STATE and similar macros In preparation for decoupling main and shadow paths. Differential Revision: https://developer.blender.org/D12888
2021-10-17 16:10:10 +02:00
ccl_device_inline float3 triangle_normal(KernelGlobals kg, ccl_private ShaderData *sd)
Cycles: Calculate face normal on the fly. Instead of pre-calculation and storage, we now calculate the face normal during render. This gives a small slowdown (~1%) but decreases memory usage, which is especially important for GPUs, where you have limited VRAM. Part of my GSoC 2014.
2014-06-13 21:27:21 +02:00
{
/* load triangle vertices */
Cycles: Remove ccl_fetch and SOA
2017-02-16 06:24:13 -05:00
const uint4 tri_vindex = kernel_tex_fetch(__tri_vindex, sd->prim);
Cycles: reduce triangle memory usage with packed_float3 Depends on D13243 Differential Revision: https://developer.blender.org/D13244
2021-11-16 14:03:59 +01:00
const float3 v0 = kernel_tex_fetch(__tri_verts, tri_vindex.w + 0);
const float3 v1 = kernel_tex_fetch(__tri_verts, tri_vindex.w + 1);
const float3 v2 = kernel_tex_fetch(__tri_verts, tri_vindex.w + 2);
ClangFormat: apply to source, most of intern Apply clang format as proposed in T53211. For details on usage and instructions for migrating branches without conflicts, see: https://wiki.blender.org/wiki/Tools/ClangFormat
2019-04-17 06:17:24 +02:00
Cycles: Calculate face normal on the fly. Instead of pre-calculation and storage, we now calculate the face normal during render. This gives a small slowdown (~1%) but decreases memory usage, which is especially important for GPUs, where you have limited VRAM. Part of my GSoC 2014.
2014-06-13 21:27:21 +02:00
/* return normal */
Cycles: Remove ccl_fetch and SOA
2017-02-16 06:24:13 -05:00
if (sd->object_flag & SD_OBJECT_NEGATIVE_SCALE_APPLIED) {
Fix T41219: Cycles backface detection doesn't work properly Root of the issue goes back to the on-fly normals commit and the latest fix for it wasn't actually correct. I've mixed two fixes in there. So the idea here goes back to storing negative scaled object flag and flip runtime-calculated normal if this flag is set, which is pretty much the same as the original fix for the issue from me. The issue with motion blur wasn't caused by the rumtime normals patch and it had issues before, because it already did runtime normals calculation. Now made it so motion triangles takes the negative scale flag into account. This actually makes code more clean imo and avoids rather confusing flipping code in mesh.cpp.
2014-08-13 16:19:12 +06:00
return normalize(cross(v2 - v0, v1 - v0));
Cycles: Split ShaderData object and shader flags We started to run out of bits there, so now we separate flags which came from __object_flags and which are either runtime or coming from __shader_flags. Rule now is: SD_OBJECT_* flags are to be tested against new object_flags field of ShaderData, all the rest flags are to be tested against flags field of ShaderData. There should be no user-visible changes, and time difference should be minimal. In fact, from tests here can only see hardly measurable difference and sometimes the new code is somewhat faster (all within a noise floor, so hard to tell for sure). Reviewers: brecht, dingto, juicyfruit, lukasstockner97, maiself Differential Revision: https://developer.blender.org/D2428
2016-12-23 11:31:19 +01:00
}
else {
Fix T41219: Cycles backface detection doesn't work properly Root of the issue goes back to the on-fly normals commit and the latest fix for it wasn't actually correct. I've mixed two fixes in there. So the idea here goes back to storing negative scaled object flag and flip runtime-calculated normal if this flag is set, which is pretty much the same as the original fix for the issue from me. The issue with motion blur wasn't caused by the rumtime normals patch and it had issues before, because it already did runtime normals calculation. Now made it so motion triangles takes the negative scale flag into account. This actually makes code more clean imo and avoids rather confusing flipping code in mesh.cpp.
2014-08-13 16:19:12 +06:00
return normalize(cross(v1 - v0, v2 - v0));
Cycles: Split ShaderData object and shader flags We started to run out of bits there, so now we separate flags which came from __object_flags and which are either runtime or coming from __shader_flags. Rule now is: SD_OBJECT_* flags are to be tested against new object_flags field of ShaderData, all the rest flags are to be tested against flags field of ShaderData. There should be no user-visible changes, and time difference should be minimal. In fact, from tests here can only see hardly measurable difference and sometimes the new code is somewhat faster (all within a noise floor, so hard to tell for sure). Reviewers: brecht, dingto, juicyfruit, lukasstockner97, maiself Differential Revision: https://developer.blender.org/D2428
2016-12-23 11:31:19 +01:00
}
Cycles: Calculate face normal on the fly. Instead of pre-calculation and storage, we now calculate the face normal during render. This gives a small slowdown (~1%) but decreases memory usage, which is especially important for GPUs, where you have limited VRAM. Part of my GSoC 2014.
2014-06-13 21:27:21 +02:00
}
Cleanup: comment blocks, trailing space in comments
2021-06-24 15:56:58 +10:00
/* Point and normal on triangle. */
Cycles: replace integrator state argument macros * Rename struct KernelGlobals to struct KernelGlobalsCPU * Add KernelGlobals, IntegratorState and ConstIntegratorState typedefs that every device can define in its own way. * Remove INTEGRATOR_STATE_ARGS and INTEGRATOR_STATE_PASS macros and replace with these new typedefs. * Add explicit state argument to INTEGRATOR_STATE and similar macros In preparation for decoupling main and shadow paths. Differential Revision: https://developer.blender.org/D12888
2021-10-17 16:10:10 +02:00
ccl_device_inline void triangle_point_normal(KernelGlobals kg,
Cycles: merge of cycles-x branch, a major update to the renderer This includes much improved GPU rendering performance, viewport interactivity, new shadow catcher, revamped sampling settings, subsurface scattering anisotropy, new GPU volume sampling, improved PMJ sampling pattern, and more. Some features have also been removed or changed, breaking backwards compatibility. Including the removal of the OpenCL backend, for which alternatives are under development. Release notes and code docs: https://wiki.blender.org/wiki/Reference/Release_Notes/3.0/Cycles https://wiki.blender.org/wiki/Source/Render/Cycles Credits: * Sergey Sharybin * Brecht Van Lommel * Patrick Mours (OptiX backend) * Christophe Hery (subsurface scattering anisotropy) * William Leeson (PMJ sampling pattern) * Alaska (various fixes and tweaks) * Thomas Dinges (various fixes) For the full commit history, see the cycles-x branch. This squashes together all the changes since intermediate changes would often fail building or tests. Ref T87839, T87837, T87836 Fixes T90734, T89353, T80267, T80267, T77185, T69800
2021-09-20 17:59:20 +02:00
int object,
int prim,
float u,
float v,
Cycles: Kernel address space changes for MSL This is the first of a sequence of changes to support compiling Cycles kernels as MSL (Metal Shading Language) in preparation for a Metal GPU device implementation. MSL requires that all pointer types be declared with explicit address space attributes (device, thread, etc...). There is already precedent for this with Cycles' address space macros (ccl_global, ccl_private, etc...), therefore the first step of MSL-enablement is to apply these consistently. Line-for-line this represents the largest change required to enable MSL. Applying this change first will simplify future patches as well as offering the emergent benefit of enhanced descriptiveness. The vast majority of deltas in this patch fall into one of two cases: - Ensuring ccl_private is specified for thread-local pointer types - Ensuring ccl_global is specified for device-wide pointer types Additionally, the ccl_addr_space qualifier can be removed. Prior to Cycles X, ccl_addr_space was used as a context-dependent address space qualifier, but now it is either redundant (e.g. in struct typedefs), or can be replaced by ccl_global in the case of pointer types. Associated function variants (e.g. lcg_step_float_addrspace) are also redundant. In cases where address space qualifiers are chained with "const", this patch places the address space qualifier first. The rationale for this is that the choice of address space is likely to have the greater impact on runtime performance and overall architecture. The final part of this patch is the addition of a metal/compat.h header. This is partially complete and will be extended in future patches, paving the way for the full Metal implementation. Ref T92212 Reviewed By: brecht Maniphest Tasks: T92212 Differential Revision: https://developer.blender.org/D12864
2021-10-14 13:53:40 +01:00
ccl_private float3 *P,
ccl_private float3 *Ng,
ccl_private int *shader)
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
{
/* load triangle vertices */
Cycles: Reduce memory usage by de-duplicating triangle storage There are several internal changes for this: First idea is to make __tri_verts to behave similar to __tri_storage, meaning, __tri_verts array now contains all vertices of all triangles instead of just mesh vertices. This saves some lookup when reading triangle coordinates in functions like triangle_normal(). In order to make it efficient needed to store global triangle offset somewhere. So no __tri_vindex.w contains a global triangle index which can be used to read triangle vertices. Additionally, the order of vertices in that array is aligned with primitives from BVH. This is needed to keep cache as much coherent as possible for BVH traversal. This causes some extra tricks needed to fill the array in and deal with True Displacement but those trickery is fully required to prevent noticeable slowdown. Next idea was to use this __tri_verts instead of __tri_storage in intersection code. Unfortunately, this is quite tricky to do without noticeable speed loss. Mainly this loss is caused by extra lookup happening to access vertex coordinate. Fortunately, tricks here and there (i,e, some types changes to avoid casts which are not really coming for free) reduces those losses to an acceptable level. So now they are within couple of percent only, On a positive site we've achieved: - Few percent of memory save with triangle-only scenes. Actual save in this case is close to size of all vertices. On a more fine-subdivided scenes this benefit might become more obvious. - Huge memory save of hairy scenes. For example, on koro.blend there is about 20% memory save. Similar figure for bunny.blend. This memory save was the main goal of this commit to move forward with Hair BVH which required more memory per BVH node. So while this sounds exciting, this memory optimization will become invisible by upcoming Hair BVH work. But again on a positive side, we can add an option to NOT use Hair BVH and then we'll have same-ish render times as we've got currently but will have this 20% memory benefit on hairy scenes.
2016-06-10 16:13:50 +02:00
const uint4 tri_vindex = kernel_tex_fetch(__tri_vindex, prim);
Cycles: reduce triangle memory usage with packed_float3 Depends on D13243 Differential Revision: https://developer.blender.org/D13244
2021-11-16 14:03:59 +01:00
float3 v0 = kernel_tex_fetch(__tri_verts, tri_vindex.w + 0);
float3 v1 = kernel_tex_fetch(__tri_verts, tri_vindex.w + 1);
float3 v2 = kernel_tex_fetch(__tri_verts, tri_vindex.w + 2);
Cycles code refactor: move more geometry code into per primitive files.
2014-03-29 13:03:45 +01:00
/* compute point */
float t = 1.0f - u - v;
*P = (u * v0 + v * v1 + t * v2);
Cycles: Replace object index hack with actual checks for SD_TRANSFORM_APPLIED Using ones complement for detecting if transform has been applied was confusing and led to several bugs. With this proper checks are made. Also added a few transforms where they were missing, mostly affecting baking and displacement when `P` is used in the shader (previously `P` was in the wrong space for these shaders) Also removed `TIME_INVALID` as this may have resulted in incorrect transforms in some cases. Reviewed By: brecht Differential Revision: https://developer.blender.org/D2192
2016-09-02 21:37:17 -04:00
/* get object flags */
int object_flag = kernel_tex_fetch(__object_flag, object);
Cycles: Calculate face normal on the fly. Instead of pre-calculation and storage, we now calculate the face normal during render. This gives a small slowdown (~1%) but decreases memory usage, which is especially important for GPUs, where you have limited VRAM. Part of my GSoC 2014.
2014-06-13 21:27:21 +02:00
/* compute normal */
Cycles: Split ShaderData object and shader flags We started to run out of bits there, so now we separate flags which came from __object_flags and which are either runtime or coming from __shader_flags. Rule now is: SD_OBJECT_* flags are to be tested against new object_flags field of ShaderData, all the rest flags are to be tested against flags field of ShaderData. There should be no user-visible changes, and time difference should be minimal. In fact, from tests here can only see hardly measurable difference and sometimes the new code is somewhat faster (all within a noise floor, so hard to tell for sure). Reviewers: brecht, dingto, juicyfruit, lukasstockner97, maiself Differential Revision: https://developer.blender.org/D2428
2016-12-23 11:31:19 +01:00
if (object_flag & SD_OBJECT_NEGATIVE_SCALE_APPLIED) {
Fix T41219: Cycles backface detection doesn't work properly Root of the issue goes back to the on-fly normals commit and the latest fix for it wasn't actually correct. I've mixed two fixes in there. So the idea here goes back to storing negative scaled object flag and flip runtime-calculated normal if this flag is set, which is pretty much the same as the original fix for the issue from me. The issue with motion blur wasn't caused by the rumtime normals patch and it had issues before, because it already did runtime normals calculation. Now made it so motion triangles takes the negative scale flag into account. This actually makes code more clean imo and avoids rather confusing flipping code in mesh.cpp.
2014-08-13 16:19:12 +06:00
*Ng = normalize(cross(v2 - v0, v1 - v0));
Cycles: Split ShaderData object and shader flags We started to run out of bits there, so now we separate flags which came from __object_flags and which are either runtime or coming from __shader_flags. Rule now is: SD_OBJECT_* flags are to be tested against new object_flags field of ShaderData, all the rest flags are to be tested against flags field of ShaderData. There should be no user-visible changes, and time difference should be minimal. In fact, from tests here can only see hardly measurable difference and sometimes the new code is somewhat faster (all within a noise floor, so hard to tell for sure). Reviewers: brecht, dingto, juicyfruit, lukasstockner97, maiself Differential Revision: https://developer.blender.org/D2428
2016-12-23 11:31:19 +01:00
}
else {
Fix T41219: Cycles backface detection doesn't work properly Root of the issue goes back to the on-fly normals commit and the latest fix for it wasn't actually correct. I've mixed two fixes in there. So the idea here goes back to storing negative scaled object flag and flip runtime-calculated normal if this flag is set, which is pretty much the same as the original fix for the issue from me. The issue with motion blur wasn't caused by the rumtime normals patch and it had issues before, because it already did runtime normals calculation. Now made it so motion triangles takes the negative scale flag into account. This actually makes code more clean imo and avoids rather confusing flipping code in mesh.cpp.
2014-08-13 16:19:12 +06:00
*Ng = normalize(cross(v1 - v0, v2 - v0));
Cycles: Split ShaderData object and shader flags We started to run out of bits there, so now we separate flags which came from __object_flags and which are either runtime or coming from __shader_flags. Rule now is: SD_OBJECT_* flags are to be tested against new object_flags field of ShaderData, all the rest flags are to be tested against flags field of ShaderData. There should be no user-visible changes, and time difference should be minimal. In fact, from tests here can only see hardly measurable difference and sometimes the new code is somewhat faster (all within a noise floor, so hard to tell for sure). Reviewers: brecht, dingto, juicyfruit, lukasstockner97, maiself Differential Revision: https://developer.blender.org/D2428
2016-12-23 11:31:19 +01:00
}
Cycles: Calculate face normal on the fly. Instead of pre-calculation and storage, we now calculate the face normal during render. This gives a small slowdown (~1%) but decreases memory usage, which is especially important for GPUs, where you have limited VRAM. Part of my GSoC 2014.
2014-06-13 21:27:21 +02:00
/* shader`*/
make "tri_shader" an int instead of a float tri_shader does no longer need to a float. Reviewers: dingto, sergey Reviewed By: dingto, sergey Subscribers: dingto Projects: #cycles Differential Revision: https://developer.blender.org/D789
2014-09-24 13:34:28 +02:00
*shader = kernel_tex_fetch(__tri_shader, prim);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
Cycles code refactor: move more code to geom folder, add some comments.
2014-03-29 13:03:48 +01:00
/* Triangle vertex locations */
Cycles: replace integrator state argument macros * Rename struct KernelGlobals to struct KernelGlobalsCPU * Add KernelGlobals, IntegratorState and ConstIntegratorState typedefs that every device can define in its own way. * Remove INTEGRATOR_STATE_ARGS and INTEGRATOR_STATE_PASS macros and replace with these new typedefs. * Add explicit state argument to INTEGRATOR_STATE and similar macros In preparation for decoupling main and shadow paths. Differential Revision: https://developer.blender.org/D12888
2021-10-17 16:10:10 +02:00
ccl_device_inline void triangle_vertices(KernelGlobals kg, int prim, float3 P[3])
Cycles OSL: refactoring and fixes * Moved kernel/osl/nodes to kernel/shaders * Renamed standard attributes to use geom:, particle:, object: prefixes * Update stdosl.h to properly reflect the closures we support * Fix the wrong stdosl.h being used for building shaders * Add geom:numpolyvertices, geom:trianglevertices, geom:polyvertices attributes
2012-11-03 14:32:13 +00:00
{
Cycles: Reduce memory usage by de-duplicating triangle storage There are several internal changes for this: First idea is to make __tri_verts to behave similar to __tri_storage, meaning, __tri_verts array now contains all vertices of all triangles instead of just mesh vertices. This saves some lookup when reading triangle coordinates in functions like triangle_normal(). In order to make it efficient needed to store global triangle offset somewhere. So no __tri_vindex.w contains a global triangle index which can be used to read triangle vertices. Additionally, the order of vertices in that array is aligned with primitives from BVH. This is needed to keep cache as much coherent as possible for BVH traversal. This causes some extra tricks needed to fill the array in and deal with True Displacement but those trickery is fully required to prevent noticeable slowdown. Next idea was to use this __tri_verts instead of __tri_storage in intersection code. Unfortunately, this is quite tricky to do without noticeable speed loss. Mainly this loss is caused by extra lookup happening to access vertex coordinate. Fortunately, tricks here and there (i,e, some types changes to avoid casts which are not really coming for free) reduces those losses to an acceptable level. So now they are within couple of percent only, On a positive site we've achieved: - Few percent of memory save with triangle-only scenes. Actual save in this case is close to size of all vertices. On a more fine-subdivided scenes this benefit might become more obvious. - Huge memory save of hairy scenes. For example, on koro.blend there is about 20% memory save. Similar figure for bunny.blend. This memory save was the main goal of this commit to move forward with Hair BVH which required more memory per BVH node. So while this sounds exciting, this memory optimization will become invisible by upcoming Hair BVH work. But again on a positive side, we can add an option to NOT use Hair BVH and then we'll have same-ish render times as we've got currently but will have this 20% memory benefit on hairy scenes.
2016-06-10 16:13:50 +02:00
const uint4 tri_vindex = kernel_tex_fetch(__tri_vindex, prim);
Cycles: reduce triangle memory usage with packed_float3 Depends on D13243 Differential Revision: https://developer.blender.org/D13244
2021-11-16 14:03:59 +01:00
P[0] = kernel_tex_fetch(__tri_verts, tri_vindex.w + 0);
P[1] = kernel_tex_fetch(__tri_verts, tri_vindex.w + 1);
P[2] = kernel_tex_fetch(__tri_verts, tri_vindex.w + 2);
Cycles OSL: refactoring and fixes * Moved kernel/osl/nodes to kernel/shaders * Renamed standard attributes to use geom:, particle:, object: prefixes * Update stdosl.h to properly reflect the closures we support * Fix the wrong stdosl.h being used for building shaders * Add geom:numpolyvertices, geom:trianglevertices, geom:polyvertices attributes
2012-11-03 14:32:13 +00:00
}
Cycles: reduce shadow terminator artifacts Offset rays from the flat surface to match where they would be for a smooth surface as specified by the normals. In the shading panel there is now a Shading Offset (existing option) and Geometry Offset (new). The Geometry Offset works as follows: * 0: disabled * 0.001: only terminated triangles (normal points to the light, geometry doesn't) are affected * 0.1 (default): triangles at grazing angles are affected, and the effect fades out * 1: all triangles are affected Limitations: * The artifact is still visible in some cases, it could be that some quads require to be treated specifically as quads. * Inconsistent normals cause artifacts. * If small objects cast shadows to a big low poly surface, the shadows can appear to be in a wrong place - because the surface moved slightly above the geometry. This can be noticed only at grazing angles to light. * Approximated surfaces of two non-intersecting low-poly objects can overlap that causes off-the-wall shadows. Generally, using one or a few levels of subdivision can get rid of artifacts faster than before. Differential Revision: https://developer.blender.org/D11065
2021-06-28 13:54:18 +02:00
/* Triangle vertex locations and vertex normals */
Cycles: replace integrator state argument macros * Rename struct KernelGlobals to struct KernelGlobalsCPU * Add KernelGlobals, IntegratorState and ConstIntegratorState typedefs that every device can define in its own way. * Remove INTEGRATOR_STATE_ARGS and INTEGRATOR_STATE_PASS macros and replace with these new typedefs. * Add explicit state argument to INTEGRATOR_STATE and similar macros In preparation for decoupling main and shadow paths. Differential Revision: https://developer.blender.org/D12888
2021-10-17 16:10:10 +02:00
ccl_device_inline void triangle_vertices_and_normals(KernelGlobals kg,
Cleanup: clang-format
2021-06-29 20:13:24 +10:00
int prim,
float3 P[3],
float3 N[3])
Cycles: reduce shadow terminator artifacts Offset rays from the flat surface to match where they would be for a smooth surface as specified by the normals. In the shading panel there is now a Shading Offset (existing option) and Geometry Offset (new). The Geometry Offset works as follows: * 0: disabled * 0.001: only terminated triangles (normal points to the light, geometry doesn't) are affected * 0.1 (default): triangles at grazing angles are affected, and the effect fades out * 1: all triangles are affected Limitations: * The artifact is still visible in some cases, it could be that some quads require to be treated specifically as quads. * Inconsistent normals cause artifacts. * If small objects cast shadows to a big low poly surface, the shadows can appear to be in a wrong place - because the surface moved slightly above the geometry. This can be noticed only at grazing angles to light. * Approximated surfaces of two non-intersecting low-poly objects can overlap that causes off-the-wall shadows. Generally, using one or a few levels of subdivision can get rid of artifacts faster than before. Differential Revision: https://developer.blender.org/D11065
2021-06-28 13:54:18 +02:00
{
const uint4 tri_vindex = kernel_tex_fetch(__tri_vindex, prim);
Cycles: reduce triangle memory usage with packed_float3 Depends on D13243 Differential Revision: https://developer.blender.org/D13244
2021-11-16 14:03:59 +01:00
P[0] = kernel_tex_fetch(__tri_verts, tri_vindex.w + 0);
P[1] = kernel_tex_fetch(__tri_verts, tri_vindex.w + 1);
P[2] = kernel_tex_fetch(__tri_verts, tri_vindex.w + 2);
N[0] = kernel_tex_fetch(__tri_vnormal, tri_vindex.x);
N[1] = kernel_tex_fetch(__tri_vnormal, tri_vindex.y);
N[2] = kernel_tex_fetch(__tri_vnormal, tri_vindex.z);
Cycles: reduce shadow terminator artifacts Offset rays from the flat surface to match where they would be for a smooth surface as specified by the normals. In the shading panel there is now a Shading Offset (existing option) and Geometry Offset (new). The Geometry Offset works as follows: * 0: disabled * 0.001: only terminated triangles (normal points to the light, geometry doesn't) are affected * 0.1 (default): triangles at grazing angles are affected, and the effect fades out * 1: all triangles are affected Limitations: * The artifact is still visible in some cases, it could be that some quads require to be treated specifically as quads. * Inconsistent normals cause artifacts. * If small objects cast shadows to a big low poly surface, the shadows can appear to be in a wrong place - because the surface moved slightly above the geometry. This can be noticed only at grazing angles to light. * Approximated surfaces of two non-intersecting low-poly objects can overlap that causes off-the-wall shadows. Generally, using one or a few levels of subdivision can get rid of artifacts faster than before. Differential Revision: https://developer.blender.org/D11065
2021-06-28 13:54:18 +02:00
}
Cycles code refactor: move more code to geom folder, add some comments.
2014-03-29 13:03:48 +01:00
/* Interpolate smooth vertex normal from vertices */
Cycles: Implement denoising option for reducing noise in the rendered image This commit contains the first part of the new Cycles denoising option, which filters the resulting image using information gathered during rendering to get rid of noise while preserving visual features as well as possible. To use the option, enable it in the render layer options. The default settings fit a wide range of scenes, but the user can tweak individual settings to control the tradeoff between a noise-free image, image details, and calculation time. Note that the denoiser may still change in the future and that some features are not implemented yet. The most important missing feature is animation denoising, which uses information from multiple frames at once to produce a flicker-free and smoother result. These features will be added in the future. Finally, thanks to all the people who supported this project: - Google (through the GSoC) and Theory Studios for sponsoring the development - The authors of the papers I used for implementing the denoiser (more details on them will be included in the technical docs) - The other Cycles devs for feedback on the code, especially Sergey for mentoring the GSoC project and Brecht for the code review! - And of course the users who helped with testing, reported bugs and things that could and/or should work better!
2017-05-07 14:40:58 +02:00
ccl_device_inline float3
Cycles: replace integrator state argument macros * Rename struct KernelGlobals to struct KernelGlobalsCPU * Add KernelGlobals, IntegratorState and ConstIntegratorState typedefs that every device can define in its own way. * Remove INTEGRATOR_STATE_ARGS and INTEGRATOR_STATE_PASS macros and replace with these new typedefs. * Add explicit state argument to INTEGRATOR_STATE and similar macros In preparation for decoupling main and shadow paths. Differential Revision: https://developer.blender.org/D12888
2021-10-17 16:10:10 +02:00
triangle_smooth_normal(KernelGlobals kg, float3 Ng, int prim, float u, float v)
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
{
/* load triangle vertices */
Cycles: Reduce memory usage by de-duplicating triangle storage There are several internal changes for this: First idea is to make __tri_verts to behave similar to __tri_storage, meaning, __tri_verts array now contains all vertices of all triangles instead of just mesh vertices. This saves some lookup when reading triangle coordinates in functions like triangle_normal(). In order to make it efficient needed to store global triangle offset somewhere. So no __tri_vindex.w contains a global triangle index which can be used to read triangle vertices. Additionally, the order of vertices in that array is aligned with primitives from BVH. This is needed to keep cache as much coherent as possible for BVH traversal. This causes some extra tricks needed to fill the array in and deal with True Displacement but those trickery is fully required to prevent noticeable slowdown. Next idea was to use this __tri_verts instead of __tri_storage in intersection code. Unfortunately, this is quite tricky to do without noticeable speed loss. Mainly this loss is caused by extra lookup happening to access vertex coordinate. Fortunately, tricks here and there (i,e, some types changes to avoid casts which are not really coming for free) reduces those losses to an acceptable level. So now they are within couple of percent only, On a positive site we've achieved: - Few percent of memory save with triangle-only scenes. Actual save in this case is close to size of all vertices. On a more fine-subdivided scenes this benefit might become more obvious. - Huge memory save of hairy scenes. For example, on koro.blend there is about 20% memory save. Similar figure for bunny.blend. This memory save was the main goal of this commit to move forward with Hair BVH which required more memory per BVH node. So while this sounds exciting, this memory optimization will become invisible by upcoming Hair BVH work. But again on a positive side, we can add an option to NOT use Hair BVH and then we'll have same-ish render times as we've got currently but will have this 20% memory benefit on hairy scenes.
2016-06-10 16:13:50 +02:00
const uint4 tri_vindex = kernel_tex_fetch(__tri_vindex, prim);
Cycles: reduce triangle memory usage with packed_float3 Depends on D13243 Differential Revision: https://developer.blender.org/D13244
2021-11-16 14:03:59 +01:00
float3 n0 = kernel_tex_fetch(__tri_vnormal, tri_vindex.x);
float3 n1 = kernel_tex_fetch(__tri_vnormal, tri_vindex.y);
float3 n2 = kernel_tex_fetch(__tri_vnormal, tri_vindex.z);
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Cycles: Implement denoising option for reducing noise in the rendered image This commit contains the first part of the new Cycles denoising option, which filters the resulting image using information gathered during rendering to get rid of noise while preserving visual features as well as possible. To use the option, enable it in the render layer options. The default settings fit a wide range of scenes, but the user can tweak individual settings to control the tradeoff between a noise-free image, image details, and calculation time. Note that the denoiser may still change in the future and that some features are not implemented yet. The most important missing feature is animation denoising, which uses information from multiple frames at once to produce a flicker-free and smoother result. These features will be added in the future. Finally, thanks to all the people who supported this project: - Google (through the GSoC) and Theory Studios for sponsoring the development - The authors of the papers I used for implementing the denoiser (more details on them will be included in the technical docs) - The other Cycles devs for feedback on the code, especially Sergey for mentoring the GSoC project and Brecht for the code review! - And of course the users who helped with testing, reported bugs and things that could and/or should work better!
2017-05-07 14:40:58 +02:00
float3 N = safe_normalize((1.0f - u - v) * n2 + u * n0 + v * n1);
return is_zero(N) ? Ng : N;
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
Cycles: replace integrator state argument macros * Rename struct KernelGlobals to struct KernelGlobalsCPU * Add KernelGlobals, IntegratorState and ConstIntegratorState typedefs that every device can define in its own way. * Remove INTEGRATOR_STATE_ARGS and INTEGRATOR_STATE_PASS macros and replace with these new typedefs. * Add explicit state argument to INTEGRATOR_STATE and similar macros In preparation for decoupling main and shadow paths. Differential Revision: https://developer.blender.org/D12888
2021-10-17 16:10:10 +02:00
ccl_device_inline float3 triangle_smooth_normal_unnormalized(
KernelGlobals kg, ccl_private const ShaderData *sd, float3 Ng, int prim, float u, float v)
Cycles: avoid copying vertex normals attribute twice to the devices Vertex normals are needed for normals maps and therefore are packed and send to the device alongside the other float3 attributes. However, we already pack and send vertex normals through `DeviceScene.tri_vnormal`. This removes the packing of vertex normals from the attributes buffer, and reuses `tri_vnormal` in the kernel for normals lookup for normal maps, which reduces memory usage a bit, and speeds up device updates. This also fixes potential missing normals updates following rB12a06292af86, since the need for vertex normals for normals maps was overlooked. Reviewed By: brecht Differential Revision: https://developer.blender.org/D12237
2021-08-17 23:24:46 +02:00
{
/* load triangle vertices */
const uint4 tri_vindex = kernel_tex_fetch(__tri_vindex, prim);
Cycles: reduce triangle memory usage with packed_float3 Depends on D13243 Differential Revision: https://developer.blender.org/D13244
2021-11-16 14:03:59 +01:00
float3 n0 = kernel_tex_fetch(__tri_vnormal, tri_vindex.x);
float3 n1 = kernel_tex_fetch(__tri_vnormal, tri_vindex.y);
float3 n2 = kernel_tex_fetch(__tri_vnormal, tri_vindex.z);
Cycles: avoid copying vertex normals attribute twice to the devices Vertex normals are needed for normals maps and therefore are packed and send to the device alongside the other float3 attributes. However, we already pack and send vertex normals through `DeviceScene.tri_vnormal`. This removes the packing of vertex normals from the attributes buffer, and reuses `tri_vnormal` in the kernel for normals lookup for normal maps, which reduces memory usage a bit, and speeds up device updates. This also fixes potential missing normals updates following rB12a06292af86, since the need for vertex normals for normals maps was overlooked. Reviewed By: brecht Differential Revision: https://developer.blender.org/D12237
2021-08-17 23:24:46 +02:00
Fix T90854: Cycles, normal map fails with applied transformations Prior to rBb8ecdbcd964a normals were stored both in DeviceScene.tri_vnormal and the float3 attributes buffer. However, the normals in `DeviceScene.tri_vnormal` might have be transformed to world space if the object's transformation was applied, while the data in the float3 attributes buffer were not. This caused shading issues in cases where the objects did have transformation applied, as the math expects the normals to be in object space. To fix this, convert the normals to object space if necessary before applying the normal map. Reviewed By: brecht Maniphest Tasks: T90854 Differential Revision: https://developer.blender.org/D12294
2021-08-24 01:42:45 +02:00
/* ensure that the normals are in object space */
if (sd->object_flag & SD_OBJECT_TRANSFORM_APPLIED) {
object_inverse_normal_transform(kg, sd, &n0);
object_inverse_normal_transform(kg, sd, &n1);
object_inverse_normal_transform(kg, sd, &n2);
}
Cycles: avoid copying vertex normals attribute twice to the devices Vertex normals are needed for normals maps and therefore are packed and send to the device alongside the other float3 attributes. However, we already pack and send vertex normals through `DeviceScene.tri_vnormal`. This removes the packing of vertex normals from the attributes buffer, and reuses `tri_vnormal` in the kernel for normals lookup for normal maps, which reduces memory usage a bit, and speeds up device updates. This also fixes potential missing normals updates following rB12a06292af86, since the need for vertex normals for normals maps was overlooked. Reviewed By: brecht Differential Revision: https://developer.blender.org/D12237
2021-08-17 23:24:46 +02:00
float3 N = (1.0f - u - v) * n2 + u * n0 + v * n1;
return is_zero(N) ? Ng : N;
}
Cycles code refactor: move more code to geom folder, add some comments.
2014-03-29 13:03:48 +01:00
/* Ray differentials on triangle */
Cycles: replace integrator state argument macros * Rename struct KernelGlobals to struct KernelGlobalsCPU * Add KernelGlobals, IntegratorState and ConstIntegratorState typedefs that every device can define in its own way. * Remove INTEGRATOR_STATE_ARGS and INTEGRATOR_STATE_PASS macros and replace with these new typedefs. * Add explicit state argument to INTEGRATOR_STATE and similar macros In preparation for decoupling main and shadow paths. Differential Revision: https://developer.blender.org/D12888
2021-10-17 16:10:10 +02:00
ccl_device_inline void triangle_dPdudv(KernelGlobals kg,
Cycles: OpenCL kernel split This commit contains all the work related on the AMD megakernel split work which was mainly done by Varun Sundar, George Kyriazis and Lenny Wang, plus some help from Sergey Sharybin, Martijn Berger, Thomas Dinges and likely someone else which we're forgetting to mention. Currently only AMD cards are enabled for the new split kernel, but it is possible to force split opencl kernel to be used by setting the following environment variable: CYCLES_OPENCL_SPLIT_KERNEL_TEST=1. Not all the features are supported yet, and that being said no motion blur, camera blur, SSS and volumetrics for now. Also transparent shadows are disabled on AMD device because of some compiler bug. This kernel is also only implements regular path tracing and supporting branched one will take a bit. Branched path tracing is exposed to the interface still, which is a bit misleading and will be hidden there soon. More feature will be enabled once they're ported to the split kernel and tested. Neither regular CPU nor CUDA has any difference, they're generating the same exact code, which means no regressions/improvements there. Based on the research paper: https://research.nvidia.com/sites/default/files/publications/laine2013hpg_paper.pdf Here's the documentation: https://docs.google.com/document/d/1LuXW-CV-sVJkQaEGZlMJ86jZ8FmoPfecaMdR-oiWbUY/edit Design discussion of the patch: https://developer.blender.org/T44197 Differential Revision: https://developer.blender.org/D1200
2015-05-09 19:34:30 +05:00
int prim,
Cycles: Kernel address space changes for MSL This is the first of a sequence of changes to support compiling Cycles kernels as MSL (Metal Shading Language) in preparation for a Metal GPU device implementation. MSL requires that all pointer types be declared with explicit address space attributes (device, thread, etc...). There is already precedent for this with Cycles' address space macros (ccl_global, ccl_private, etc...), therefore the first step of MSL-enablement is to apply these consistently. Line-for-line this represents the largest change required to enable MSL. Applying this change first will simplify future patches as well as offering the emergent benefit of enhanced descriptiveness. The vast majority of deltas in this patch fall into one of two cases: - Ensuring ccl_private is specified for thread-local pointer types - Ensuring ccl_global is specified for device-wide pointer types Additionally, the ccl_addr_space qualifier can be removed. Prior to Cycles X, ccl_addr_space was used as a context-dependent address space qualifier, but now it is either redundant (e.g. in struct typedefs), or can be replaced by ccl_global in the case of pointer types. Associated function variants (e.g. lcg_step_float_addrspace) are also redundant. In cases where address space qualifiers are chained with "const", this patch places the address space qualifier first. The rationale for this is that the choice of address space is likely to have the greater impact on runtime performance and overall architecture. The final part of this patch is the addition of a metal/compat.h header. This is partially complete and will be extended in future patches, paving the way for the full Metal implementation. Ref T92212 Reviewed By: brecht Maniphest Tasks: T92212 Differential Revision: https://developer.blender.org/D12864
2021-10-14 13:53:40 +01:00
ccl_private float3 *dPdu,
ccl_private float3 *dPdv)
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
{
/* fetch triangle vertex coordinates */
Cycles: Reduce memory usage by de-duplicating triangle storage There are several internal changes for this: First idea is to make __tri_verts to behave similar to __tri_storage, meaning, __tri_verts array now contains all vertices of all triangles instead of just mesh vertices. This saves some lookup when reading triangle coordinates in functions like triangle_normal(). In order to make it efficient needed to store global triangle offset somewhere. So no __tri_vindex.w contains a global triangle index which can be used to read triangle vertices. Additionally, the order of vertices in that array is aligned with primitives from BVH. This is needed to keep cache as much coherent as possible for BVH traversal. This causes some extra tricks needed to fill the array in and deal with True Displacement but those trickery is fully required to prevent noticeable slowdown. Next idea was to use this __tri_verts instead of __tri_storage in intersection code. Unfortunately, this is quite tricky to do without noticeable speed loss. Mainly this loss is caused by extra lookup happening to access vertex coordinate. Fortunately, tricks here and there (i,e, some types changes to avoid casts which are not really coming for free) reduces those losses to an acceptable level. So now they are within couple of percent only, On a positive site we've achieved: - Few percent of memory save with triangle-only scenes. Actual save in this case is close to size of all vertices. On a more fine-subdivided scenes this benefit might become more obvious. - Huge memory save of hairy scenes. For example, on koro.blend there is about 20% memory save. Similar figure for bunny.blend. This memory save was the main goal of this commit to move forward with Hair BVH which required more memory per BVH node. So while this sounds exciting, this memory optimization will become invisible by upcoming Hair BVH work. But again on a positive side, we can add an option to NOT use Hair BVH and then we'll have same-ish render times as we've got currently but will have this 20% memory benefit on hairy scenes.
2016-06-10 16:13:50 +02:00
const uint4 tri_vindex = kernel_tex_fetch(__tri_vindex, prim);
Cycles: reduce triangle memory usage with packed_float3 Depends on D13243 Differential Revision: https://developer.blender.org/D13244
2021-11-16 14:03:59 +01:00
const float3 p0 = kernel_tex_fetch(__tri_verts, tri_vindex.w + 0);
const float3 p1 = kernel_tex_fetch(__tri_verts, tri_vindex.w + 1);
const float3 p2 = kernel_tex_fetch(__tri_verts, tri_vindex.w + 2);
ClangFormat: apply to source, most of intern Apply clang format as proposed in T53211. For details on usage and instructions for migrating branches without conflicts, see: https://wiki.blender.org/wiki/Tools/ClangFormat
2019-04-17 06:17:24 +02:00
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
/* compute derivatives of P w.r.t. uv */
*dPdu = (p0 - p2);
*dPdv = (p1 - p2);
}
Cycles code refactor: move more code to geom folder, add some comments.
2014-03-29 13:03:48 +01:00
/* Reading attributes on various triangle elements */
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
Cycles: replace integrator state argument macros * Rename struct KernelGlobals to struct KernelGlobalsCPU * Add KernelGlobals, IntegratorState and ConstIntegratorState typedefs that every device can define in its own way. * Remove INTEGRATOR_STATE_ARGS and INTEGRATOR_STATE_PASS macros and replace with these new typedefs. * Add explicit state argument to INTEGRATOR_STATE and similar macros In preparation for decoupling main and shadow paths. Differential Revision: https://developer.blender.org/D12888
2021-10-17 16:10:10 +02:00
ccl_device float triangle_attribute_float(KernelGlobals kg,
Cycles: Kernel address space changes for MSL This is the first of a sequence of changes to support compiling Cycles kernels as MSL (Metal Shading Language) in preparation for a Metal GPU device implementation. MSL requires that all pointer types be declared with explicit address space attributes (device, thread, etc...). There is already precedent for this with Cycles' address space macros (ccl_global, ccl_private, etc...), therefore the first step of MSL-enablement is to apply these consistently. Line-for-line this represents the largest change required to enable MSL. Applying this change first will simplify future patches as well as offering the emergent benefit of enhanced descriptiveness. The vast majority of deltas in this patch fall into one of two cases: - Ensuring ccl_private is specified for thread-local pointer types - Ensuring ccl_global is specified for device-wide pointer types Additionally, the ccl_addr_space qualifier can be removed. Prior to Cycles X, ccl_addr_space was used as a context-dependent address space qualifier, but now it is either redundant (e.g. in struct typedefs), or can be replaced by ccl_global in the case of pointer types. Associated function variants (e.g. lcg_step_float_addrspace) are also redundant. In cases where address space qualifiers are chained with "const", this patch places the address space qualifier first. The rationale for this is that the choice of address space is likely to have the greater impact on runtime performance and overall architecture. The final part of this patch is the addition of a metal/compat.h header. This is partially complete and will be extended in future patches, paving the way for the full Metal implementation. Ref T92212 Reviewed By: brecht Maniphest Tasks: T92212 Differential Revision: https://developer.blender.org/D12864
2021-10-14 13:53:40 +01:00
ccl_private const ShaderData *sd,
Cycles: merge of cycles-x branch, a major update to the renderer This includes much improved GPU rendering performance, viewport interactivity, new shadow catcher, revamped sampling settings, subsurface scattering anisotropy, new GPU volume sampling, improved PMJ sampling pattern, and more. Some features have also been removed or changed, breaking backwards compatibility. Including the removal of the OpenCL backend, for which alternatives are under development. Release notes and code docs: https://wiki.blender.org/wiki/Reference/Release_Notes/3.0/Cycles https://wiki.blender.org/wiki/Source/Render/Cycles Credits: * Sergey Sharybin * Brecht Van Lommel * Patrick Mours (OptiX backend) * Christophe Hery (subsurface scattering anisotropy) * William Leeson (PMJ sampling pattern) * Alaska (various fixes and tweaks) * Thomas Dinges (various fixes) For the full commit history, see the cycles-x branch. This squashes together all the changes since intermediate changes would often fail building or tests. Ref T87839, T87837, T87836 Fixes T90734, T89353, T80267, T80267, T77185, T69800
2021-09-20 17:59:20 +02:00
const AttributeDescriptor desc,
Cycles: Kernel address space changes for MSL This is the first of a sequence of changes to support compiling Cycles kernels as MSL (Metal Shading Language) in preparation for a Metal GPU device implementation. MSL requires that all pointer types be declared with explicit address space attributes (device, thread, etc...). There is already precedent for this with Cycles' address space macros (ccl_global, ccl_private, etc...), therefore the first step of MSL-enablement is to apply these consistently. Line-for-line this represents the largest change required to enable MSL. Applying this change first will simplify future patches as well as offering the emergent benefit of enhanced descriptiveness. The vast majority of deltas in this patch fall into one of two cases: - Ensuring ccl_private is specified for thread-local pointer types - Ensuring ccl_global is specified for device-wide pointer types Additionally, the ccl_addr_space qualifier can be removed. Prior to Cycles X, ccl_addr_space was used as a context-dependent address space qualifier, but now it is either redundant (e.g. in struct typedefs), or can be replaced by ccl_global in the case of pointer types. Associated function variants (e.g. lcg_step_float_addrspace) are also redundant. In cases where address space qualifiers are chained with "const", this patch places the address space qualifier first. The rationale for this is that the choice of address space is likely to have the greater impact on runtime performance and overall architecture. The final part of this patch is the addition of a metal/compat.h header. This is partially complete and will be extended in future patches, paving the way for the full Metal implementation. Ref T92212 Reviewed By: brecht Maniphest Tasks: T92212 Differential Revision: https://developer.blender.org/D12864
2021-10-14 13:53:40 +01:00
ccl_private float *dx,
ccl_private float *dy)
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
{
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
if (desc.element & (ATTR_ELEMENT_VERTEX | ATTR_ELEMENT_VERTEX_MOTION | ATTR_ELEMENT_CORNER)) {
float f0, f1, f2;
if (desc.element & (ATTR_ELEMENT_VERTEX | ATTR_ELEMENT_VERTEX_MOTION)) {
const uint4 tri_vindex = kernel_tex_fetch(__tri_vindex, sd->prim);
f0 = kernel_tex_fetch(__attributes_float, desc.offset + tri_vindex.x);
f1 = kernel_tex_fetch(__attributes_float, desc.offset + tri_vindex.y);
f2 = kernel_tex_fetch(__attributes_float, desc.offset + tri_vindex.z);
}
else {
const int tri = desc.offset + sd->prim * 3;
f0 = kernel_tex_fetch(__attributes_float, tri + 0);
f1 = kernel_tex_fetch(__attributes_float, tri + 1);
f2 = kernel_tex_fetch(__attributes_float, tri + 2);
}
Cycles Hair: refactoring to support generic attributes for hair curves. There should be no functional changes yet. UV, tangent and intercept are now stored as attributes, with the intention to add more like multiple uv's, vertex colors, generated coordinates and motion vectors later. Things got a bit messy due to having both triangle and curve data in the same mesh data structure, which also gives us two sets of attributes. This will get cleaned up when we split the mesh class.
2013-01-03 12:08:54 +00:00
#ifdef __RAY_DIFFERENTIALS__
Cycles: Remove ccl_fetch and SOA
2017-02-16 06:24:13 -05:00
if (dx)
*dx = sd->du.dx * f0 + sd->dv.dx * f1 - (sd->du.dx + sd->dv.dx) * f2;
if (dy)
*dy = sd->du.dy * f0 + sd->dv.dy * f1 - (sd->du.dy + sd->dv.dy) * f2;
Cycles Hair: refactoring to support generic attributes for hair curves. There should be no functional changes yet. UV, tangent and intercept are now stored as attributes, with the intention to add more like multiple uv's, vertex colors, generated coordinates and motion vectors later. Things got a bit messy due to having both triangle and curve data in the same mesh data structure, which also gives us two sets of attributes. This will get cleaned up when we split the mesh class.
2013-01-03 12:08:54 +00:00
#endif
Cycles: Remove ccl_fetch and SOA
2017-02-16 06:24:13 -05:00
return sd->u * f0 + sd->v * f1 + (1.0f - sd->u - sd->v) * f2;
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
else {
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
#ifdef __RAY_DIFFERENTIALS__
Cycles Hair: refactoring to support generic attributes for hair curves. There should be no functional changes yet. UV, tangent and intercept are now stored as attributes, with the intention to add more like multiple uv's, vertex colors, generated coordinates and motion vectors later. Things got a bit messy due to having both triangle and curve data in the same mesh data structure, which also gives us two sets of attributes. This will get cleaned up when we split the mesh class.
2013-01-03 12:08:54 +00:00
if (dx)
*dx = 0.0f;
if (dy)
*dy = 0.0f;
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
#endif
Cycles Hair: refactoring to support generic attributes for hair curves. There should be no functional changes yet. UV, tangent and intercept are now stored as attributes, with the intention to add more like multiple uv's, vertex colors, generated coordinates and motion vectors later. Things got a bit messy due to having both triangle and curve data in the same mesh data structure, which also gives us two sets of attributes. This will get cleaned up when we split the mesh class.
2013-01-03 12:08:54 +00:00
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
if (desc.element & (ATTR_ELEMENT_FACE | ATTR_ELEMENT_OBJECT | ATTR_ELEMENT_MESH)) {
const int offset = (desc.element == ATTR_ELEMENT_FACE) ? desc.offset + sd->prim :
desc.offset;
return kernel_tex_fetch(__attributes_float, offset);
}
else {
return 0.0f;
}
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
}
}
Cycles: replace integrator state argument macros * Rename struct KernelGlobals to struct KernelGlobalsCPU * Add KernelGlobals, IntegratorState and ConstIntegratorState typedefs that every device can define in its own way. * Remove INTEGRATOR_STATE_ARGS and INTEGRATOR_STATE_PASS macros and replace with these new typedefs. * Add explicit state argument to INTEGRATOR_STATE and similar macros In preparation for decoupling main and shadow paths. Differential Revision: https://developer.blender.org/D12888
2021-10-17 16:10:10 +02:00
ccl_device float2 triangle_attribute_float2(KernelGlobals kg,
Cycles: Kernel address space changes for MSL This is the first of a sequence of changes to support compiling Cycles kernels as MSL (Metal Shading Language) in preparation for a Metal GPU device implementation. MSL requires that all pointer types be declared with explicit address space attributes (device, thread, etc...). There is already precedent for this with Cycles' address space macros (ccl_global, ccl_private, etc...), therefore the first step of MSL-enablement is to apply these consistently. Line-for-line this represents the largest change required to enable MSL. Applying this change first will simplify future patches as well as offering the emergent benefit of enhanced descriptiveness. The vast majority of deltas in this patch fall into one of two cases: - Ensuring ccl_private is specified for thread-local pointer types - Ensuring ccl_global is specified for device-wide pointer types Additionally, the ccl_addr_space qualifier can be removed. Prior to Cycles X, ccl_addr_space was used as a context-dependent address space qualifier, but now it is either redundant (e.g. in struct typedefs), or can be replaced by ccl_global in the case of pointer types. Associated function variants (e.g. lcg_step_float_addrspace) are also redundant. In cases where address space qualifiers are chained with "const", this patch places the address space qualifier first. The rationale for this is that the choice of address space is likely to have the greater impact on runtime performance and overall architecture. The final part of this patch is the addition of a metal/compat.h header. This is partially complete and will be extended in future patches, paving the way for the full Metal implementation. Ref T92212 Reviewed By: brecht Maniphest Tasks: T92212 Differential Revision: https://developer.blender.org/D12864
2021-10-14 13:53:40 +01:00
ccl_private const ShaderData *sd,
Cycles: Added Float2 attribute type. Float2 are now a new type for attributes in Cycles. Before, the choices for attribute storage were float and float3, the latter padded to float4. This meant that UV maps were inflated to twice the size necessary. Reviewers: brecht, sergey Reviewed By: brecht Subscribers: #cycles Tags: #cycles Differential Revision: https://developer.blender.org/D4409
2019-03-05 14:54:54 +01:00
const AttributeDescriptor desc,
Cycles: Kernel address space changes for MSL This is the first of a sequence of changes to support compiling Cycles kernels as MSL (Metal Shading Language) in preparation for a Metal GPU device implementation. MSL requires that all pointer types be declared with explicit address space attributes (device, thread, etc...). There is already precedent for this with Cycles' address space macros (ccl_global, ccl_private, etc...), therefore the first step of MSL-enablement is to apply these consistently. Line-for-line this represents the largest change required to enable MSL. Applying this change first will simplify future patches as well as offering the emergent benefit of enhanced descriptiveness. The vast majority of deltas in this patch fall into one of two cases: - Ensuring ccl_private is specified for thread-local pointer types - Ensuring ccl_global is specified for device-wide pointer types Additionally, the ccl_addr_space qualifier can be removed. Prior to Cycles X, ccl_addr_space was used as a context-dependent address space qualifier, but now it is either redundant (e.g. in struct typedefs), or can be replaced by ccl_global in the case of pointer types. Associated function variants (e.g. lcg_step_float_addrspace) are also redundant. In cases where address space qualifiers are chained with "const", this patch places the address space qualifier first. The rationale for this is that the choice of address space is likely to have the greater impact on runtime performance and overall architecture. The final part of this patch is the addition of a metal/compat.h header. This is partially complete and will be extended in future patches, paving the way for the full Metal implementation. Ref T92212 Reviewed By: brecht Maniphest Tasks: T92212 Differential Revision: https://developer.blender.org/D12864
2021-10-14 13:53:40 +01:00
ccl_private float2 *dx,
ccl_private float2 *dy)
Cycles: Added Float2 attribute type. Float2 are now a new type for attributes in Cycles. Before, the choices for attribute storage were float and float3, the latter padded to float4. This meant that UV maps were inflated to twice the size necessary. Reviewers: brecht, sergey Reviewed By: brecht Subscribers: #cycles Tags: #cycles Differential Revision: https://developer.blender.org/D4409
2019-03-05 14:54:54 +01:00
{
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
if (desc.element & (ATTR_ELEMENT_VERTEX | ATTR_ELEMENT_VERTEX_MOTION | ATTR_ELEMENT_CORNER)) {
Cycles: Added Float2 attribute type. Float2 are now a new type for attributes in Cycles. Before, the choices for attribute storage were float and float3, the latter padded to float4. This meant that UV maps were inflated to twice the size necessary. Reviewers: brecht, sergey Reviewed By: brecht Subscribers: #cycles Tags: #cycles Differential Revision: https://developer.blender.org/D4409
2019-03-05 14:54:54 +01:00
float2 f0, f1, f2;
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
if (desc.element & (ATTR_ELEMENT_VERTEX | ATTR_ELEMENT_VERTEX_MOTION)) {
const uint4 tri_vindex = kernel_tex_fetch(__tri_vindex, sd->prim);
f0 = kernel_tex_fetch(__attributes_float2, desc.offset + tri_vindex.x);
f1 = kernel_tex_fetch(__attributes_float2, desc.offset + tri_vindex.y);
f2 = kernel_tex_fetch(__attributes_float2, desc.offset + tri_vindex.z);
}
else {
const int tri = desc.offset + sd->prim * 3;
Cycles: Added Float2 attribute type. Float2 are now a new type for attributes in Cycles. Before, the choices for attribute storage were float and float3, the latter padded to float4. This meant that UV maps were inflated to twice the size necessary. Reviewers: brecht, sergey Reviewed By: brecht Subscribers: #cycles Tags: #cycles Differential Revision: https://developer.blender.org/D4409
2019-03-05 14:54:54 +01:00
f0 = kernel_tex_fetch(__attributes_float2, tri + 0);
f1 = kernel_tex_fetch(__attributes_float2, tri + 1);
f2 = kernel_tex_fetch(__attributes_float2, tri + 2);
}
#ifdef __RAY_DIFFERENTIALS__
if (dx)
*dx = sd->du.dx * f0 + sd->dv.dx * f1 - (sd->du.dx + sd->dv.dx) * f2;
if (dy)
*dy = sd->du.dy * f0 + sd->dv.dy * f1 - (sd->du.dy + sd->dv.dy) * f2;
#endif
return sd->u * f0 + sd->v * f1 + (1.0f - sd->u - sd->v) * f2;
}
else {
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
#ifdef __RAY_DIFFERENTIALS__
Cycles: Added Float2 attribute type. Float2 are now a new type for attributes in Cycles. Before, the choices for attribute storage were float and float3, the latter padded to float4. This meant that UV maps were inflated to twice the size necessary. Reviewers: brecht, sergey Reviewed By: brecht Subscribers: #cycles Tags: #cycles Differential Revision: https://developer.blender.org/D4409
2019-03-05 14:54:54 +01:00
if (dx)
*dx = make_float2(0.0f, 0.0f);
if (dy)
*dy = make_float2(0.0f, 0.0f);
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
#endif
Cycles: Added Float2 attribute type. Float2 are now a new type for attributes in Cycles. Before, the choices for attribute storage were float and float3, the latter padded to float4. This meant that UV maps were inflated to twice the size necessary. Reviewers: brecht, sergey Reviewed By: brecht Subscribers: #cycles Tags: #cycles Differential Revision: https://developer.blender.org/D4409
2019-03-05 14:54:54 +01:00
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
if (desc.element & (ATTR_ELEMENT_FACE | ATTR_ELEMENT_OBJECT | ATTR_ELEMENT_MESH)) {
const int offset = (desc.element == ATTR_ELEMENT_FACE) ? desc.offset + sd->prim :
desc.offset;
return kernel_tex_fetch(__attributes_float2, offset);
}
else {
return make_float2(0.0f, 0.0f);
}
Cycles: Added Float2 attribute type. Float2 are now a new type for attributes in Cycles. Before, the choices for attribute storage were float and float3, the latter padded to float4. This meant that UV maps were inflated to twice the size necessary. Reviewers: brecht, sergey Reviewed By: brecht Subscribers: #cycles Tags: #cycles Differential Revision: https://developer.blender.org/D4409
2019-03-05 14:54:54 +01:00
}
}
Cycles: replace integrator state argument macros * Rename struct KernelGlobals to struct KernelGlobalsCPU * Add KernelGlobals, IntegratorState and ConstIntegratorState typedefs that every device can define in its own way. * Remove INTEGRATOR_STATE_ARGS and INTEGRATOR_STATE_PASS macros and replace with these new typedefs. * Add explicit state argument to INTEGRATOR_STATE and similar macros In preparation for decoupling main and shadow paths. Differential Revision: https://developer.blender.org/D12888
2021-10-17 16:10:10 +02:00
ccl_device float3 triangle_attribute_float3(KernelGlobals kg,
Cycles: Kernel address space changes for MSL This is the first of a sequence of changes to support compiling Cycles kernels as MSL (Metal Shading Language) in preparation for a Metal GPU device implementation. MSL requires that all pointer types be declared with explicit address space attributes (device, thread, etc...). There is already precedent for this with Cycles' address space macros (ccl_global, ccl_private, etc...), therefore the first step of MSL-enablement is to apply these consistently. Line-for-line this represents the largest change required to enable MSL. Applying this change first will simplify future patches as well as offering the emergent benefit of enhanced descriptiveness. The vast majority of deltas in this patch fall into one of two cases: - Ensuring ccl_private is specified for thread-local pointer types - Ensuring ccl_global is specified for device-wide pointer types Additionally, the ccl_addr_space qualifier can be removed. Prior to Cycles X, ccl_addr_space was used as a context-dependent address space qualifier, but now it is either redundant (e.g. in struct typedefs), or can be replaced by ccl_global in the case of pointer types. Associated function variants (e.g. lcg_step_float_addrspace) are also redundant. In cases where address space qualifiers are chained with "const", this patch places the address space qualifier first. The rationale for this is that the choice of address space is likely to have the greater impact on runtime performance and overall architecture. The final part of this patch is the addition of a metal/compat.h header. This is partially complete and will be extended in future patches, paving the way for the full Metal implementation. Ref T92212 Reviewed By: brecht Maniphest Tasks: T92212 Differential Revision: https://developer.blender.org/D12864
2021-10-14 13:53:40 +01:00
ccl_private const ShaderData *sd,
Cycles: Add AttributeDescriptor Adds a descriptor for attributes that can easily be passed around and extended to contain more data. Will be used for attributes on subdivision meshes. Reviewed By: brecht Differential Revision: https://developer.blender.org/D2110
2016-07-01 17:36:27 -04:00
const AttributeDescriptor desc,
Cycles: Kernel address space changes for MSL This is the first of a sequence of changes to support compiling Cycles kernels as MSL (Metal Shading Language) in preparation for a Metal GPU device implementation. MSL requires that all pointer types be declared with explicit address space attributes (device, thread, etc...). There is already precedent for this with Cycles' address space macros (ccl_global, ccl_private, etc...), therefore the first step of MSL-enablement is to apply these consistently. Line-for-line this represents the largest change required to enable MSL. Applying this change first will simplify future patches as well as offering the emergent benefit of enhanced descriptiveness. The vast majority of deltas in this patch fall into one of two cases: - Ensuring ccl_private is specified for thread-local pointer types - Ensuring ccl_global is specified for device-wide pointer types Additionally, the ccl_addr_space qualifier can be removed. Prior to Cycles X, ccl_addr_space was used as a context-dependent address space qualifier, but now it is either redundant (e.g. in struct typedefs), or can be replaced by ccl_global in the case of pointer types. Associated function variants (e.g. lcg_step_float_addrspace) are also redundant. In cases where address space qualifiers are chained with "const", this patch places the address space qualifier first. The rationale for this is that the choice of address space is likely to have the greater impact on runtime performance and overall architecture. The final part of this patch is the addition of a metal/compat.h header. This is partially complete and will be extended in future patches, paving the way for the full Metal implementation. Ref T92212 Reviewed By: brecht Maniphest Tasks: T92212 Differential Revision: https://developer.blender.org/D12864
2021-10-14 13:53:40 +01:00
ccl_private float3 *dx,
ccl_private float3 *dy)
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
{
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
if (desc.element & (ATTR_ELEMENT_VERTEX | ATTR_ELEMENT_VERTEX_MOTION | ATTR_ELEMENT_CORNER)) {
Cycles: Add support for uchar4 attributes. * Added support for uchar4 attributes to Cycles' attribute system. * This is used for Vertex Colors now, which saves some memory (4 unsigned characters, instead of 4 floats). * GPU Texture Limit on sm_20 and sm_21 decreased from 95 to 94, because we need a new texture for the uchar4 attributes. This is no problem for sm_30 or newer. Part of my GSoC 2014.
2014-06-13 23:40:39 +02:00
float3 f0, f1, f2;
ClangFormat: apply to source, most of intern Apply clang format as proposed in T53211. For details on usage and instructions for migrating branches without conflicts, see: https://wiki.blender.org/wiki/Tools/ClangFormat
2019-04-17 06:17:24 +02:00
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
if (desc.element & (ATTR_ELEMENT_VERTEX | ATTR_ELEMENT_VERTEX_MOTION)) {
const uint4 tri_vindex = kernel_tex_fetch(__tri_vindex, sd->prim);
Cycles: reduce triangle memory usage with packed_float3 Depends on D13243 Differential Revision: https://developer.blender.org/D13244
2021-11-16 14:03:59 +01:00
f0 = kernel_tex_fetch(__attributes_float3, desc.offset + tri_vindex.x);
f1 = kernel_tex_fetch(__attributes_float3, desc.offset + tri_vindex.y);
f2 = kernel_tex_fetch(__attributes_float3, desc.offset + tri_vindex.z);
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
}
else {
const int tri = desc.offset + sd->prim * 3;
Cycles: reduce triangle memory usage with packed_float3 Depends on D13243 Differential Revision: https://developer.blender.org/D13244
2021-11-16 14:03:59 +01:00
f0 = kernel_tex_fetch(__attributes_float3, tri + 0);
f1 = kernel_tex_fetch(__attributes_float3, tri + 1);
f2 = kernel_tex_fetch(__attributes_float3, tri + 2);
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
}
Cycles Hair: refactoring to support generic attributes for hair curves. There should be no functional changes yet. UV, tangent and intercept are now stored as attributes, with the intention to add more like multiple uv's, vertex colors, generated coordinates and motion vectors later. Things got a bit messy due to having both triangle and curve data in the same mesh data structure, which also gives us two sets of attributes. This will get cleaned up when we split the mesh class.
2013-01-03 12:08:54 +00:00
#ifdef __RAY_DIFFERENTIALS__
Cycles: Remove ccl_fetch and SOA
2017-02-16 06:24:13 -05:00
if (dx)
*dx = sd->du.dx * f0 + sd->dv.dx * f1 - (sd->du.dx + sd->dv.dx) * f2;
if (dy)
*dy = sd->du.dy * f0 + sd->dv.dy * f1 - (sd->du.dy + sd->dv.dy) * f2;
New feature Patch [#33445] - Experimental Cycles Hair Rendering (CPU only) This patch allows hair data to be exported to cycles and introduces a new line segment primitive to render with. The UI appears under the particle tab and there is a new hair info node available. It is only available under the experimental feature set and for cpu rendering.
2012-12-28 14:21:30 +00:00
#endif
Cycles: support for motion vector and UV passes. Most of the changes are related to adding support for motion data throughout the code. There's some code for actual camera/object motion blur raytracing but it's unfinished (it badly slows down the raytracing kernel even when the option is turned off), so that code it disabled still. Motion vector export from Blender tries to avoid computing derived meshes when the mesh does not have a deforming modifier, and it also won't store motion vectors for every vertex if only the object or camera is moving.
2012-04-30 12:49:26 +00:00
Cycles: Remove ccl_fetch and SOA
2017-02-16 06:24:13 -05:00
return sd->u * f0 + sd->v * f1 + (1.0f - sd->u - sd->v) * f2;
Fix most of #31307: cycles panorama camera not working correct with speed vectors and window texture coordinates. Only for Fisheye Equisolid it's still not working correct yet. Patch from Dalai with modifications.
2012-05-07 10:53:09 +00:00
}
else {
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
#ifdef __RAY_DIFFERENTIALS__
Cycles Hair: refactoring to support generic attributes for hair curves. There should be no functional changes yet. UV, tangent and intercept are now stored as attributes, with the intention to add more like multiple uv's, vertex colors, generated coordinates and motion vectors later. Things got a bit messy due to having both triangle and curve data in the same mesh data structure, which also gives us two sets of attributes. This will get cleaned up when we split the mesh class.
2013-01-03 12:08:54 +00:00
if (dx)
*dx = make_float3(0.0f, 0.0f, 0.0f);
if (dy)
*dy = make_float3(0.0f, 0.0f, 0.0f);
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
#endif
Cycles: support for motion vector and UV passes. Most of the changes are related to adding support for motion data throughout the code. There's some code for actual camera/object motion blur raytracing but it's unfinished (it badly slows down the raytracing kernel even when the option is turned off), so that code it disabled still. Motion vector export from Blender tries to avoid computing derived meshes when the mesh does not have a deforming modifier, and it also won't store motion vectors for every vertex if only the object or camera is moving.
2012-04-30 12:49:26 +00:00
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
if (desc.element & (ATTR_ELEMENT_FACE | ATTR_ELEMENT_OBJECT | ATTR_ELEMENT_MESH)) {
const int offset = (desc.element == ATTR_ELEMENT_FACE) ? desc.offset + sd->prim :
desc.offset;
Cycles: reduce triangle memory usage with packed_float3 Depends on D13243 Differential Revision: https://developer.blender.org/D13244
2021-11-16 14:03:59 +01:00
return kernel_tex_fetch(__attributes_float3, offset);
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
}
else {
return make_float3(0.0f, 0.0f, 0.0f);
}
Cycles Hair: refactoring to support generic attributes for hair curves. There should be no functional changes yet. UV, tangent and intercept are now stored as attributes, with the intention to add more like multiple uv's, vertex colors, generated coordinates and motion vectors later. Things got a bit messy due to having both triangle and curve data in the same mesh data structure, which also gives us two sets of attributes. This will get cleaned up when we split the mesh class.
2013-01-03 12:08:54 +00:00
}
Cycles: support for motion vector and UV passes. Most of the changes are related to adding support for motion data throughout the code. There's some code for actual camera/object motion blur raytracing but it's unfinished (it badly slows down the raytracing kernel even when the option is turned off), so that code it disabled still. Motion vector export from Blender tries to avoid computing derived meshes when the mesh does not have a deforming modifier, and it also won't store motion vectors for every vertex if only the object or camera is moving.
2012-04-30 12:49:26 +00:00
}
Cycles: replace integrator state argument macros * Rename struct KernelGlobals to struct KernelGlobalsCPU * Add KernelGlobals, IntegratorState and ConstIntegratorState typedefs that every device can define in its own way. * Remove INTEGRATOR_STATE_ARGS and INTEGRATOR_STATE_PASS macros and replace with these new typedefs. * Add explicit state argument to INTEGRATOR_STATE and similar macros In preparation for decoupling main and shadow paths. Differential Revision: https://developer.blender.org/D12888
2021-10-17 16:10:10 +02:00
ccl_device float4 triangle_attribute_float4(KernelGlobals kg,
Cycles: Kernel address space changes for MSL This is the first of a sequence of changes to support compiling Cycles kernels as MSL (Metal Shading Language) in preparation for a Metal GPU device implementation. MSL requires that all pointer types be declared with explicit address space attributes (device, thread, etc...). There is already precedent for this with Cycles' address space macros (ccl_global, ccl_private, etc...), therefore the first step of MSL-enablement is to apply these consistently. Line-for-line this represents the largest change required to enable MSL. Applying this change first will simplify future patches as well as offering the emergent benefit of enhanced descriptiveness. The vast majority of deltas in this patch fall into one of two cases: - Ensuring ccl_private is specified for thread-local pointer types - Ensuring ccl_global is specified for device-wide pointer types Additionally, the ccl_addr_space qualifier can be removed. Prior to Cycles X, ccl_addr_space was used as a context-dependent address space qualifier, but now it is either redundant (e.g. in struct typedefs), or can be replaced by ccl_global in the case of pointer types. Associated function variants (e.g. lcg_step_float_addrspace) are also redundant. In cases where address space qualifiers are chained with "const", this patch places the address space qualifier first. The rationale for this is that the choice of address space is likely to have the greater impact on runtime performance and overall architecture. The final part of this patch is the addition of a metal/compat.h header. This is partially complete and will be extended in future patches, paving the way for the full Metal implementation. Ref T92212 Reviewed By: brecht Maniphest Tasks: T92212 Differential Revision: https://developer.blender.org/D12864
2021-10-14 13:53:40 +01:00
ccl_private const ShaderData *sd,
Shading: Add Vertex Color node. This patch adds a new Vertex Color node. The node also returns the alpha of the vertex color layer as an output. Reviewers: brecht Differential Revision: https://developer.blender.org/D5767
2019-09-12 17:42:13 +02:00
const AttributeDescriptor desc,
Cycles: Kernel address space changes for MSL This is the first of a sequence of changes to support compiling Cycles kernels as MSL (Metal Shading Language) in preparation for a Metal GPU device implementation. MSL requires that all pointer types be declared with explicit address space attributes (device, thread, etc...). There is already precedent for this with Cycles' address space macros (ccl_global, ccl_private, etc...), therefore the first step of MSL-enablement is to apply these consistently. Line-for-line this represents the largest change required to enable MSL. Applying this change first will simplify future patches as well as offering the emergent benefit of enhanced descriptiveness. The vast majority of deltas in this patch fall into one of two cases: - Ensuring ccl_private is specified for thread-local pointer types - Ensuring ccl_global is specified for device-wide pointer types Additionally, the ccl_addr_space qualifier can be removed. Prior to Cycles X, ccl_addr_space was used as a context-dependent address space qualifier, but now it is either redundant (e.g. in struct typedefs), or can be replaced by ccl_global in the case of pointer types. Associated function variants (e.g. lcg_step_float_addrspace) are also redundant. In cases where address space qualifiers are chained with "const", this patch places the address space qualifier first. The rationale for this is that the choice of address space is likely to have the greater impact on runtime performance and overall architecture. The final part of this patch is the addition of a metal/compat.h header. This is partially complete and will be extended in future patches, paving the way for the full Metal implementation. Ref T92212 Reviewed By: brecht Maniphest Tasks: T92212 Differential Revision: https://developer.blender.org/D12864
2021-10-14 13:53:40 +01:00
ccl_private float4 *dx,
ccl_private float4 *dy)
Shading: Add Vertex Color node. This patch adds a new Vertex Color node. The node also returns the alpha of the vertex color layer as an output. Reviewers: brecht Differential Revision: https://developer.blender.org/D5767
2019-09-12 17:42:13 +02:00
{
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
if (desc.element & (ATTR_ELEMENT_VERTEX | ATTR_ELEMENT_VERTEX_MOTION | ATTR_ELEMENT_CORNER |
ATTR_ELEMENT_CORNER_BYTE)) {
Cycles: add support for rendering sculpt vertex colors Ref T78041
2020-06-25 13:48:44 +02:00
float4 f0, f1, f2;
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
if (desc.element & (ATTR_ELEMENT_VERTEX | ATTR_ELEMENT_VERTEX_MOTION)) {
const uint4 tri_vindex = kernel_tex_fetch(__tri_vindex, sd->prim);
Cycles: reduce triangle memory usage with packed_float3 Depends on D13243 Differential Revision: https://developer.blender.org/D13244
2021-11-16 14:03:59 +01:00
f0 = kernel_tex_fetch(__attributes_float4, desc.offset + tri_vindex.x);
f1 = kernel_tex_fetch(__attributes_float4, desc.offset + tri_vindex.y);
f2 = kernel_tex_fetch(__attributes_float4, desc.offset + tri_vindex.z);
Cycles: add support for rendering sculpt vertex colors Ref T78041
2020-06-25 13:48:44 +02:00
}
else {
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
const int tri = desc.offset + sd->prim * 3;
if (desc.element == ATTR_ELEMENT_CORNER) {
Cycles: reduce triangle memory usage with packed_float3 Depends on D13243 Differential Revision: https://developer.blender.org/D13244
2021-11-16 14:03:59 +01:00
f0 = kernel_tex_fetch(__attributes_float4, tri + 0);
f1 = kernel_tex_fetch(__attributes_float4, tri + 1);
f2 = kernel_tex_fetch(__attributes_float4, tri + 2);
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
}
else {
Merge branch 'blender-v2.91-release'
2020-11-09 17:37:21 +01:00
f0 = color_srgb_to_linear_v4(
color_uchar4_to_float4(kernel_tex_fetch(__attributes_uchar4, tri + 0)));
f1 = color_srgb_to_linear_v4(
color_uchar4_to_float4(kernel_tex_fetch(__attributes_uchar4, tri + 1)));
f2 = color_srgb_to_linear_v4(
color_uchar4_to_float4(kernel_tex_fetch(__attributes_uchar4, tri + 2)));
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
}
Cycles: add support for rendering sculpt vertex colors Ref T78041
2020-06-25 13:48:44 +02:00
}
Shading: Add Vertex Color node. This patch adds a new Vertex Color node. The node also returns the alpha of the vertex color layer as an output. Reviewers: brecht Differential Revision: https://developer.blender.org/D5767
2019-09-12 17:42:13 +02:00
#ifdef __RAY_DIFFERENTIALS__
if (dx)
*dx = sd->du.dx * f0 + sd->dv.dx * f1 - (sd->du.dx + sd->dv.dx) * f2;
if (dy)
*dy = sd->du.dy * f0 + sd->dv.dy * f1 - (sd->du.dy + sd->dv.dy) * f2;
#endif
return sd->u * f0 + sd->v * f1 + (1.0f - sd->u - sd->v) * f2;
}
else {
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
#ifdef __RAY_DIFFERENTIALS__
Shading: Add Vertex Color node. This patch adds a new Vertex Color node. The node also returns the alpha of the vertex color layer as an output. Reviewers: brecht Differential Revision: https://developer.blender.org/D5767
2019-09-12 17:42:13 +02:00
if (dx)
*dx = make_float4(0.0f, 0.0f, 0.0f, 0.0f);
if (dy)
*dy = make_float4(0.0f, 0.0f, 0.0f, 0.0f);
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
#endif
Shading: Add Vertex Color node. This patch adds a new Vertex Color node. The node also returns the alpha of the vertex color layer as an output. Reviewers: brecht Differential Revision: https://developer.blender.org/D5767
2019-09-12 17:42:13 +02:00
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
if (desc.element & (ATTR_ELEMENT_FACE | ATTR_ELEMENT_OBJECT | ATTR_ELEMENT_MESH)) {
const int offset = (desc.element == ATTR_ELEMENT_FACE) ? desc.offset + sd->prim :
desc.offset;
Cycles: reduce triangle memory usage with packed_float3 Depends on D13243 Differential Revision: https://developer.blender.org/D13244
2021-11-16 14:03:59 +01:00
return kernel_tex_fetch(__attributes_float4, offset);
Cycles: refactor to make attribute lookup slightly more efficient Ref D2057
2020-10-26 19:25:00 +01:00
}
else {
return make_float4(0.0f, 0.0f, 0.0f, 0.0f);
}
Shading: Add Vertex Color node. This patch adds a new Vertex Color node. The node also returns the alpha of the vertex color layer as an output. Reviewers: brecht Differential Revision: https://developer.blender.org/D5767
2019-09-12 17:42:13 +02:00
}
}
Cycles render engine, initial commit. This is the engine itself, blender modifications and build instructions will follow later. Cycles uses code from some great open source projects, many thanks them: * BVH building and traversal code from NVidia's "Understanding the Efficiency of Ray Traversal on GPUs": http://code.google.com/p/understanding-the-efficiency-of-ray-traversal-on-gpus/ * Open Shading Language for a large part of the shading system: http://code.google.com/p/openshadinglanguage/ * Blender for procedural textures and a few other nodes. * Approximate Catmull Clark subdivision from NVidia Mesh tools: http://code.google.com/p/nvidia-mesh-tools/ * Sobol direction vectors from: http://web.maths.unsw.edu.au/~fkuo/sobol/ * Film response functions from: http://www.cs.columbia.edu/CAVE/software/softlib/dorf.php
2011-04-27 11:58:34 +00:00
CCL_NAMESPACE_END
Reference in New Issue Copy Permalink