blender/intern/cycles/kernel/osl/osl_shader.cpp
Brecht Van Lommel 9ebfcea8f3 Cycles: revert commit that joined surface/volume socket into a single shader socket,
on second thought this makes it a bit too difficult to understand what's going on.
2011-10-12 23:03:12 +00:00

478 lines
15 KiB
C++

/*
* Copyright 2011, Blender Foundation.
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software Foundation,
* Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
*/
#include "kernel_compat_cpu.h"
#include "kernel_types.h"
#include "kernel_globals.h"
#include "kernel_object.h"
#include "osl_services.h"
#include "osl_shader.h"
#include "util_foreach.h"
#include <OSL/oslexec.h>
#include <oslexec_pvt.h>
CCL_NAMESPACE_BEGIN
tls_ptr(OSLGlobals::ThreadData, OSLGlobals::thread_data);
/* Threads */
void OSLShader::thread_init(KernelGlobals *kg)
{
OSL::pvt::ShadingSystemImpl *ssi = (OSL::pvt::ShadingSystemImpl*)kg->osl.ss;
OSLGlobals::ThreadData *tdata = new OSLGlobals::ThreadData();
memset(&tdata->globals, 0, sizeof(OSL::ShaderGlobals));
tdata->thread_info = ssi->create_thread_info();
tls_set(kg->osl.thread_data, tdata);
((OSLRenderServices*)ssi->renderer())->thread_init(kg);
}
void OSLShader::thread_free(KernelGlobals *kg)
{
OSL::pvt::ShadingSystemImpl *ssi = (OSL::pvt::ShadingSystemImpl*)kg->osl.ss;
OSLGlobals::ThreadData *tdata = tls_get(OSLGlobals::ThreadData, kg->osl.thread_data);
ssi->destroy_thread_info(tdata->thread_info);
delete tdata;
}
/* Globals */
#define TO_VEC3(v) (*(OSL::Vec3*)&(v))
#define TO_COLOR3(v) (*(OSL::Color3*)&(v))
#define TO_FLOAT3(v) make_float3(v[0], v[1], v[2])
static void shaderdata_to_shaderglobals(KernelGlobals *kg, ShaderData *sd,
int path_flag, OSL::ShaderGlobals *globals)
{
/* copy from shader data to shader globals */
globals->P = TO_VEC3(sd->P);
globals->dPdx = TO_VEC3(sd->dP.dx);
globals->dPdy = TO_VEC3(sd->dP.dy);
globals->I = TO_VEC3(sd->I);
globals->dIdx = TO_VEC3(sd->dI.dx);
globals->dIdy = TO_VEC3(sd->dI.dy);
globals->N = TO_VEC3(sd->N);
globals->Ng = TO_VEC3(sd->Ng);
globals->u = sd->u;
globals->dudx = sd->du.dx;
globals->dudy = sd->du.dy;
globals->v = sd->v;
globals->dvdx = sd->dv.dx;
globals->dvdy = sd->dv.dy;
globals->dPdu = TO_VEC3(sd->dPdu);
globals->dPdv = TO_VEC3(sd->dPdv);
globals->surfacearea = (sd->object == ~0)? 1.0f: object_surface_area(kg, sd->object);
/* booleans */
globals->raytype = path_flag; /* todo: add our own ray types */
globals->backfacing = (sd->flag & SD_BACKFACING);
/* don't know yet if we need this */
globals->flipHandedness = false;
/* shader data to be used in services callbacks */
globals->renderstate = sd;
/* hacky, we leave it to services to fetch actual object matrix */
globals->shader2common = sd;
globals->object2common = sd;
/* must be set to NULL before execute */
globals->Ci = NULL;
}
/* Surface */
static void flatten_surface_closure_tree(ShaderData *sd, bool no_glossy,
const OSL::ClosureColor *closure, float3 weight = make_float3(1.0f, 1.0f, 1.0f))
{
/* OSL gives use a closure tree, we flatten it into arrays per
* closure type, for evaluation, sampling, etc later on. */
if(closure->type == OSL::ClosureColor::COMPONENT) {
OSL::ClosureComponent *comp = (OSL::ClosureComponent*)closure;
OSL::ClosurePrimitive *prim = (OSL::ClosurePrimitive*)comp->data();
if(prim) {
ShaderClosure sc;
sc.prim = prim;
sc.weight = weight;
switch(prim->category()) {
case ClosurePrimitive::BSDF: {
if(sd->num_closure == MAX_CLOSURE)
return;
OSL::BSDFClosure *bsdf = (OSL::BSDFClosure*)prim;
ustring scattering = bsdf->scattering();
/* no caustics option */
if(no_glossy && scattering == OSL::Labels::GLOSSY)
return;
/* sample weight */
float albedo = bsdf->albedo(TO_VEC3(sd->I));
float sample_weight = fabsf(average(weight))*albedo;
float sample_sum = sd->osl_closure.bsdf_sample_sum + sample_weight;
sc.sample_weight = sample_weight;
sc.type = CLOSURE_BSDF_ID;
sd->osl_closure.bsdf_sample_sum = sample_sum;
/* scattering flags */
if(scattering == OSL::Labels::DIFFUSE)
sd->flag |= SD_BSDF|SD_BSDF_HAS_EVAL;
else if(scattering == OSL::Labels::GLOSSY)
sd->flag |= SD_BSDF|SD_BSDF_HAS_EVAL|SD_BSDF_GLOSSY;
else
sd->flag |= SD_BSDF;
/* add */
sd->closure[sd->num_closure++] = sc;
break;
}
case ClosurePrimitive::Emissive: {
if(sd->num_closure == MAX_CLOSURE)
return;
/* sample weight */
float sample_weight = fabsf(average(weight));
float sample_sum = sd->osl_closure.emissive_sample_sum + sample_weight;
sc.sample_weight = sample_weight;
sc.type = CLOSURE_EMISSION_ID;
sd->osl_closure.emissive_sample_sum = sample_sum;
/* flag */
sd->flag |= SD_EMISSION;
sd->closure[sd->num_closure++] = sc;
break;
}
case ClosurePrimitive::Holdout:
if(sd->num_closure == MAX_CLOSURE)
return;
sc.sample_weight = 0.0f;
sc.type = CLOSURE_HOLDOUT_ID;
sd->flag |= SD_HOLDOUT;
sd->closure[sd->num_closure++] = sc;
break;
case ClosurePrimitive::BSSRDF:
case ClosurePrimitive::Debug:
break; /* not implemented */
case ClosurePrimitive::Background:
case ClosurePrimitive::Volume:
break; /* not relevant */
}
}
}
else if(closure->type == OSL::ClosureColor::MUL) {
OSL::ClosureMul *mul = (OSL::ClosureMul*)closure;
flatten_surface_closure_tree(sd, no_glossy, mul->closure, TO_FLOAT3(mul->weight) * weight);
}
else if(closure->type == OSL::ClosureColor::ADD) {
OSL::ClosureAdd *add = (OSL::ClosureAdd*)closure;
flatten_surface_closure_tree(sd, no_glossy, add->closureA, weight);
flatten_surface_closure_tree(sd, no_glossy, add->closureB, weight);
}
}
void OSLShader::eval_surface(KernelGlobals *kg, ShaderData *sd, float randb, int path_flag)
{
/* gather pointers */
OSL::pvt::ShadingSystemImpl *ssi = (OSL::pvt::ShadingSystemImpl*)kg->osl.ss;
OSLGlobals::ThreadData *tdata = tls_get(OSLGlobals::ThreadData, kg->osl.thread_data);
OSL::ShaderGlobals *globals = &tdata->globals;
OSL::pvt::ShadingContext *ctx = ssi->get_context(tdata->thread_info);
/* setup shader globals from shader data */
sd->osl_ctx = ctx;
shaderdata_to_shaderglobals(kg, sd, path_flag, globals);
/* execute shader for this point */
int shader = sd->shader & SHADER_MASK;
if(kg->osl.surface_state[shader])
ctx->execute(OSL::pvt::ShadUseSurface, *(kg->osl.surface_state[shader]), *globals);
/* flatten closure tree */
sd->num_closure = 0;
sd->randb_closure = randb;
if(globals->Ci) {
bool no_glossy = (path_flag & PATH_RAY_DIFFUSE) && kernel_data.integrator.no_caustics;
flatten_surface_closure_tree(sd, no_glossy, globals->Ci);
}
}
/* Background */
static float3 flatten_background_closure_tree(const OSL::ClosureColor *closure)
{
/* OSL gives use a closure tree, if we are shading for background there
* is only one supported closure type at the moment, which has no evaluation
* functions, so we just sum the weights */
if(closure->type == OSL::ClosureColor::COMPONENT) {
OSL::ClosureComponent *comp = (OSL::ClosureComponent*)closure;
OSL::ClosurePrimitive *prim = (OSL::ClosurePrimitive*)comp->data();
if(prim && prim->category() == OSL::ClosurePrimitive::Background)
return make_float3(1.0f, 1.0f, 1.0f);
}
else if(closure->type == OSL::ClosureColor::MUL) {
OSL::ClosureMul *mul = (OSL::ClosureMul*)closure;
return TO_FLOAT3(mul->weight) * flatten_background_closure_tree(mul->closure);
}
else if(closure->type == OSL::ClosureColor::ADD) {
OSL::ClosureAdd *add = (OSL::ClosureAdd*)closure;
return flatten_background_closure_tree(add->closureA) +
flatten_background_closure_tree(add->closureB);
}
return make_float3(0.0f, 0.0f, 0.0f);
}
float3 OSLShader::eval_background(KernelGlobals *kg, ShaderData *sd, int path_flag)
{
/* gather pointers */
OSL::pvt::ShadingSystemImpl *ssi = (OSL::pvt::ShadingSystemImpl*)kg->osl.ss;
OSLGlobals::ThreadData *tdata = tls_get(OSLGlobals::ThreadData, kg->osl.thread_data);
OSL::ShaderGlobals *globals = &tdata->globals;
OSL::pvt::ShadingContext *ctx = ssi->get_context(tdata->thread_info);
/* setup shader globals from shader data */
sd->osl_ctx = ctx;
shaderdata_to_shaderglobals(kg, sd, path_flag, globals);
/* execute shader for this point */
if(kg->osl.background_state)
ctx->execute(OSL::pvt::ShadUseSurface, *kg->osl.background_state, *globals);
/* return background color immediately */
if(globals->Ci)
return flatten_background_closure_tree(globals->Ci);
return make_float3(0.0f, 0.0f, 0.0f);
}
/* Volume */
static void flatten_volume_closure_tree(ShaderData *sd,
const OSL::ClosureColor *closure, float3 weight = make_float3(1.0f, 1.0f, 1.0f))
{
/* OSL gives use a closure tree, we flatten it into arrays per
* closure type, for evaluation, sampling, etc later on. */
if(closure->type == OSL::ClosureColor::COMPONENT) {
OSL::ClosureComponent *comp = (OSL::ClosureComponent*)closure;
OSL::ClosurePrimitive *prim = (OSL::ClosurePrimitive*)comp->data();
if(prim) {
ShaderClosure sc;
sc.prim = prim;
sc.weight = weight;
switch(prim->category()) {
case ClosurePrimitive::Volume: {
if(sd->num_closure == MAX_CLOSURE)
return;
/* sample weight */
float sample_weight = fabsf(average(weight));
float sample_sum = sd->osl_closure.volume_sample_sum + sample_weight;
sc.sample_weight = sample_weight;
sc.type = CLOSURE_VOLUME_ID;
sd->osl_closure.volume_sample_sum = sample_sum;
/* add */
sd->closure[sd->num_closure++] = sc;
break;
}
case ClosurePrimitive::Holdout:
case ClosurePrimitive::Debug:
break; /* not implemented */
case ClosurePrimitive::Background:
case ClosurePrimitive::BSDF:
case ClosurePrimitive::Emissive:
case ClosurePrimitive::BSSRDF:
break; /* not relevant */
}
}
}
else if(closure->type == OSL::ClosureColor::MUL) {
OSL::ClosureMul *mul = (OSL::ClosureMul*)closure;
flatten_volume_closure_tree(sd, mul->closure, TO_FLOAT3(mul->weight) * weight);
}
else if(closure->type == OSL::ClosureColor::ADD) {
OSL::ClosureAdd *add = (OSL::ClosureAdd*)closure;
flatten_volume_closure_tree(sd, add->closureA, weight);
flatten_volume_closure_tree(sd, add->closureB, weight);
}
}
void OSLShader::eval_volume(KernelGlobals *kg, ShaderData *sd, float randb, int path_flag)
{
/* gather pointers */
OSL::pvt::ShadingSystemImpl *ssi = (OSL::pvt::ShadingSystemImpl*)kg->osl.ss;
OSLGlobals::ThreadData *tdata = tls_get(OSLGlobals::ThreadData, kg->osl.thread_data);
OSL::ShaderGlobals *globals = &tdata->globals;
OSL::pvt::ShadingContext *ctx = ssi->get_context(tdata->thread_info);
/* setup shader globals from shader data */
sd->osl_ctx = ctx;
shaderdata_to_shaderglobals(kg, sd, path_flag, globals);
/* execute shader */
int shader = sd->shader & SHADER_MASK;
if(kg->osl.volume_state[shader])
ctx->execute(OSL::pvt::ShadUseSurface, *(kg->osl.volume_state[shader]), *globals);
/* retrieve resulting closures */
sd->osl_closure.volume_sample_sum = 0.0f;
sd->osl_closure.num_volume = 0;
sd->osl_closure.randb = randb;
if(globals->Ci)
flatten_volume_closure_tree(sd, globals->Ci);
}
/* Displacement */
void OSLShader::eval_displacement(KernelGlobals *kg, ShaderData *sd)
{
/* gather pointers */
OSL::pvt::ShadingSystemImpl *ssi = (OSL::pvt::ShadingSystemImpl*)kg->osl.ss;
OSLGlobals::ThreadData *tdata = tls_get(OSLGlobals::ThreadData, kg->osl.thread_data);
OSL::ShaderGlobals *globals = &tdata->globals;
OSL::pvt::ShadingContext *ctx = ssi->get_context(tdata->thread_info);
/* setup shader globals from shader data */
sd->osl_ctx = ctx;
shaderdata_to_shaderglobals(kg, sd, 0, globals);
/* execute shader */
int shader = sd->shader & SHADER_MASK;
if(kg->osl.displacement_state[shader])
ctx->execute(OSL::pvt::ShadUseSurface, *(kg->osl.displacement_state[shader]), *globals);
/* get back position */
sd->P = TO_FLOAT3(globals->P);
}
void OSLShader::release(KernelGlobals *kg, const ShaderData *sd)
{
OSL::pvt::ShadingSystemImpl *ssi = (OSL::pvt::ShadingSystemImpl*)kg->osl.ss;
OSLGlobals::ThreadData *tdata = tls_get(OSLGlobals::ThreadData, kg->osl.thread_data);
ssi->release_context((OSL::pvt::ShadingContext*)sd->osl_ctx, tdata->thread_info);
}
/* BSDF Closure */
int OSLShader::bsdf_sample(const ShaderData *sd, const ShaderClosure *sc, float randu, float randv, float3& eval, float3& omega_in, differential3& domega_in, float& pdf)
{
OSL::BSDFClosure *sample_bsdf = (OSL::BSDFClosure*)sc->prim;
int label = LABEL_NONE;
pdf = 0.0f;
/* sample BSDF closure */
ustring ulabel;
ulabel = sample_bsdf->sample(TO_VEC3(sd->Ng),
TO_VEC3(sd->I), TO_VEC3(sd->dI.dx), TO_VEC3(sd->dI.dy),
randu, randv,
TO_VEC3(omega_in), TO_VEC3(domega_in.dx), TO_VEC3(domega_in.dy),
pdf, TO_COLOR3(eval));
/* convert OSL label */
if(ulabel == OSL::Labels::REFLECT)
label = LABEL_REFLECT;
else if(ulabel == OSL::Labels::TRANSMIT)
label = LABEL_TRANSMIT;
else
return LABEL_NONE; /* sampling failed */
/* convert scattering to our bitflag label */
ustring uscattering = sample_bsdf->scattering();
if(uscattering == OSL::Labels::DIFFUSE)
label |= LABEL_DIFFUSE;
else if(uscattering == OSL::Labels::GLOSSY)
label |= LABEL_GLOSSY;
else if(uscattering == OSL::Labels::SINGULAR)
label |= LABEL_SINGULAR;
else
label |= LABEL_TRANSPARENT;
return label;
}
float3 OSLShader::bsdf_eval(const ShaderData *sd, const ShaderClosure *sc, const float3& omega_in, float& pdf)
{
OSL::BSDFClosure *bsdf = (OSL::BSDFClosure*)sc->prim;
OSL::Color3 bsdf_eval;
if(dot(sd->Ng, omega_in) >= 0.0f)
bsdf_eval = bsdf->eval_reflect(TO_VEC3(sd->I), TO_VEC3(omega_in), pdf);
else
bsdf_eval = bsdf->eval_transmit(TO_VEC3(sd->I), TO_VEC3(omega_in), pdf);
return TO_FLOAT3(bsdf_eval);
}
/* Emissive Closure */
float3 OSLShader::emissive_eval(const ShaderData *sd, const ShaderClosure *sc)
{
OSL::EmissiveClosure *emissive = (OSL::EmissiveClosure*)sc->prim;
OSL::Color3 emissive_eval = emissive->eval(TO_VEC3(sd->Ng), TO_VEC3(sd->I));
eval += TO_FLOAT3(emissive_eval);
return eval;
}
/* Volume Closure */
float3 OSLShader::volume_eval_phase(const ShaderData *sd, const ShaderClosure *sc, const float3 omega_in, const float3 omega_out)
{
OSL::VolumeClosure *volume = (OSL::VolumeClosure*)sc->prim;
OSL::Color3 volume_eval = volume->eval_phase(TO_VEC3(omega_in), TO_VEC3(omega_out));
return TO_FLOAT3(volume_eval)*sc->weight;
}
CCL_NAMESPACE_END