Files
blender/intern/cycles/kernel/kernel_differential.h
Michael Jones a0f269f682 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 16:14:43 +01:00

167 lines
5.1 KiB
C

/*
* Copyright 2011-2013 Blender Foundation
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#pragma once
CCL_NAMESPACE_BEGIN
/* See "Tracing Ray Differentials", Homan Igehy, 1999. */
ccl_device void differential_transfer(ccl_private differential3 *surface_dP,
const differential3 ray_dP,
float3 ray_D,
const differential3 ray_dD,
float3 surface_Ng,
float ray_t)
{
/* ray differential transfer through homogeneous medium, to
* compute dPdx/dy at a shading point from the incoming ray */
float3 tmp = ray_D / dot(ray_D, surface_Ng);
float3 tmpx = ray_dP.dx + ray_t * ray_dD.dx;
float3 tmpy = ray_dP.dy + ray_t * ray_dD.dy;
surface_dP->dx = tmpx - dot(tmpx, surface_Ng) * tmp;
surface_dP->dy = tmpy - dot(tmpy, surface_Ng) * tmp;
}
ccl_device void differential_incoming(ccl_private differential3 *dI, const differential3 dD)
{
/* compute dIdx/dy at a shading point, we just need to negate the
* differential of the ray direction */
dI->dx = -dD.dx;
dI->dy = -dD.dy;
}
ccl_device void differential_dudv(ccl_private differential *du,
ccl_private differential *dv,
float3 dPdu,
float3 dPdv,
differential3 dP,
float3 Ng)
{
/* now we have dPdx/dy from the ray differential transfer, and dPdu/dv
* from the primitive, we can compute dudx/dy and dvdx/dy. these are
* mainly used for differentials of arbitrary mesh attributes. */
/* find most stable axis to project to 2D */
float xn = fabsf(Ng.x);
float yn = fabsf(Ng.y);
float zn = fabsf(Ng.z);
if (zn < xn || zn < yn) {
if (yn < xn || yn < zn) {
dPdu.x = dPdu.y;
dPdv.x = dPdv.y;
dP.dx.x = dP.dx.y;
dP.dy.x = dP.dy.y;
}
dPdu.y = dPdu.z;
dPdv.y = dPdv.z;
dP.dx.y = dP.dx.z;
dP.dy.y = dP.dy.z;
}
/* using Cramer's rule, we solve for dudx and dvdx in a 2x2 linear system,
* and the same for dudy and dvdy. the denominator is the same for both
* solutions, so we compute it only once.
*
* dP.dx = dPdu * dudx + dPdv * dvdx;
* dP.dy = dPdu * dudy + dPdv * dvdy; */
float det = (dPdu.x * dPdv.y - dPdv.x * dPdu.y);
if (det != 0.0f)
det = 1.0f / det;
du->dx = (dP.dx.x * dPdv.y - dP.dx.y * dPdv.x) * det;
dv->dx = (dP.dx.y * dPdu.x - dP.dx.x * dPdu.y) * det;
du->dy = (dP.dy.x * dPdv.y - dP.dy.y * dPdv.x) * det;
dv->dy = (dP.dy.y * dPdu.x - dP.dy.x * dPdu.y) * det;
}
ccl_device differential differential_zero()
{
differential d;
d.dx = 0.0f;
d.dy = 0.0f;
return d;
}
ccl_device differential3 differential3_zero()
{
differential3 d;
d.dx = zero_float3();
d.dy = zero_float3();
return d;
}
/* Compact ray differentials that are just a scale to reduce memory usage and
* access cost in GPU.
*
* See above for more accurate reference implementations.
*
* TODO: also store the more compact version in ShaderData and recompute where
* needed? */
ccl_device_forceinline float differential_zero_compact()
{
return 0.0f;
}
ccl_device_forceinline float differential_make_compact(const differential3 D)
{
return 0.5f * (len(D.dx) + len(D.dy));
}
ccl_device_forceinline void differential_transfer_compact(ccl_private differential3 *surface_dP,
const float ray_dP,
const float3 /* ray_D */,
const float ray_dD,
const float3 surface_Ng,
const float ray_t)
{
/* ray differential transfer through homogeneous medium, to
* compute dPdx/dy at a shading point from the incoming ray */
float scale = ray_dP + ray_t * ray_dD;
float3 dx, dy;
make_orthonormals(surface_Ng, &dx, &dy);
surface_dP->dx = dx * scale;
surface_dP->dy = dy * scale;
}
ccl_device_forceinline void differential_incoming_compact(ccl_private differential3 *dI,
const float3 D,
const float dD)
{
/* compute dIdx/dy at a shading point, we just need to negate the
* differential of the ray direction */
float3 dx, dy;
make_orthonormals(D, &dx, &dy);
dI->dx = dD * dx;
dI->dy = dD * dy;
}
CCL_NAMESPACE_END