2022-02-11 13:53:21 +01:00
|
|
|
/* SPDX-License-Identifier: BSD-3-Clause
|
2013-05-20 15:58:37 +00:00
|
|
|
*
|
2022-02-11 13:53:21 +01:00
|
|
|
* Adapted from Open Shading Language
|
2013-05-20 15:58:37 +00:00
|
|
|
* Copyright (c) 2009-2010 Sony Pictures Imageworks Inc., et al.
|
|
|
|
* All Rights Reserved.
|
|
|
|
*
|
2022-02-11 13:53:21 +01:00
|
|
|
* Modifications Copyright 2011-2022 Blender Foundation. */
|
2013-05-20 15:58:37 +00:00
|
|
|
|
2021-10-24 14:19:19 +02:00
|
|
|
#pragma once
|
|
|
|
|
2013-05-20 15:58:37 +00:00
|
|
|
CCL_NAMESPACE_BEGIN
|
|
|
|
|
|
|
|
/* Wireframe Node */
|
|
|
|
|
2021-10-17 16:10:10 +02:00
|
|
|
ccl_device_inline float wireframe(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 ShaderData *sd,
|
|
|
|
float size,
|
|
|
|
int pixel_size,
|
|
|
|
ccl_private float3 *P)
|
2013-05-20 15:58:37 +00:00
|
|
|
{
|
2021-12-01 17:30:46 +01:00
|
|
|
#if defined(__HAIR__) || defined(__POINTCLOUD__)
|
2021-12-20 02:52:56 +01:00
|
|
|
if (sd->prim != PRIM_NONE && sd->type & PRIMITIVE_TRIANGLE)
|
2013-05-20 15:58:37 +00:00
|
|
|
#else
|
2019-04-17 06:17:24 +02:00
|
|
|
if (sd->prim != PRIM_NONE)
|
2013-05-20 15:58:37 +00:00
|
|
|
#endif
|
2019-04-17 06:17:24 +02:00
|
|
|
{
|
|
|
|
float3 Co[3];
|
|
|
|
float pixelwidth = 1.0f;
|
2013-05-20 15:58:37 +00:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
/* Triangles */
|
|
|
|
int np = 3;
|
2015-02-21 14:49:55 +05:00
|
|
|
|
2021-12-20 02:52:56 +01:00
|
|
|
if (sd->type & PRIMITIVE_MOTION) {
|
2019-04-17 06:17:24 +02:00
|
|
|
motion_triangle_vertices(kg, sd->object, sd->prim, sd->time, Co);
|
2021-12-20 02:52:56 +01:00
|
|
|
}
|
|
|
|
else {
|
|
|
|
triangle_vertices(kg, sd->prim, Co);
|
|
|
|
}
|
2013-05-20 15:58:37 +00:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
if (!(sd->object_flag & SD_OBJECT_TRANSFORM_APPLIED)) {
|
|
|
|
object_position_transform(kg, sd, &Co[0]);
|
|
|
|
object_position_transform(kg, sd, &Co[1]);
|
|
|
|
object_position_transform(kg, sd, &Co[2]);
|
|
|
|
}
|
2015-02-21 14:49:55 +05:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
if (pixel_size) {
|
|
|
|
// Project the derivatives of P to the viewing plane defined
|
|
|
|
// by I so we have a measure of how big is a pixel at this point
|
|
|
|
float pixelwidth_x = len(sd->dP.dx - dot(sd->dP.dx, sd->I) * sd->I);
|
|
|
|
float pixelwidth_y = len(sd->dP.dy - dot(sd->dP.dy, sd->I) * sd->I);
|
|
|
|
// Take the average of both axis' length
|
|
|
|
pixelwidth = (pixelwidth_x + pixelwidth_y) * 0.5f;
|
|
|
|
}
|
2015-02-21 14:49:55 +05:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
// Use half the width as the neighbor face will render the
|
|
|
|
// other half. And take the square for fast comparison
|
|
|
|
pixelwidth *= 0.5f * size;
|
|
|
|
pixelwidth *= pixelwidth;
|
|
|
|
for (int i = 0; i < np; i++) {
|
|
|
|
int i2 = i ? i - 1 : np - 1;
|
|
|
|
float3 dir = *P - Co[i];
|
|
|
|
float3 edge = Co[i] - Co[i2];
|
|
|
|
float3 crs = cross(edge, dir);
|
|
|
|
// At this point dot(crs, crs) / dot(edge, edge) is
|
|
|
|
// the square of area / length(edge) == square of the
|
|
|
|
// distance to the edge.
|
|
|
|
if (dot(crs, crs) < (dot(edge, edge) * pixelwidth))
|
|
|
|
return 1.0f;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return 0.0f;
|
2015-02-21 14:49:55 +05:00
|
|
|
}
|
|
|
|
|
2021-10-17 16:10:10 +02:00
|
|
|
ccl_device_noinline void svm_node_wireframe(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 ShaderData *sd,
|
|
|
|
ccl_private float *stack,
|
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
|
|
|
uint4 node)
|
2015-02-21 14:49:55 +05:00
|
|
|
{
|
2019-04-17 06:17:24 +02:00
|
|
|
uint in_size = node.y;
|
|
|
|
uint out_fac = node.z;
|
|
|
|
uint use_pixel_size, bump_offset;
|
2019-08-21 11:59:57 +02:00
|
|
|
svm_unpack_node_uchar2(node.w, &use_pixel_size, &bump_offset);
|
2015-02-21 14:49:55 +05:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
/* Input Data */
|
|
|
|
float size = stack_load_float(stack, in_size);
|
|
|
|
int pixel_size = (int)use_pixel_size;
|
2015-02-21 14:49:55 +05:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
/* Calculate wireframe */
|
|
|
|
float f = wireframe(kg, sd, size, pixel_size, &sd->P);
|
2015-02-21 14:49:55 +05:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
/* TODO(sergey): Think of faster way to calculate derivatives. */
|
|
|
|
if (bump_offset == NODE_BUMP_OFFSET_DX) {
|
|
|
|
float3 Px = sd->P - sd->dP.dx;
|
|
|
|
f += (f - wireframe(kg, sd, size, pixel_size, &Px)) / len(sd->dP.dx);
|
|
|
|
}
|
|
|
|
else if (bump_offset == NODE_BUMP_OFFSET_DY) {
|
|
|
|
float3 Py = sd->P - sd->dP.dy;
|
|
|
|
f += (f - wireframe(kg, sd, size, pixel_size, &Py)) / len(sd->dP.dy);
|
|
|
|
}
|
2015-02-21 14:49:55 +05:00
|
|
|
|
2019-04-17 06:17:24 +02:00
|
|
|
if (stack_valid(out_fac))
|
|
|
|
stack_store_float(stack, out_fac, f);
|
2013-05-20 15:58:37 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
CCL_NAMESPACE_END
|