FGFS-4.1/OpenSceneGraph
4
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
a810b80b77
OpenSceneGraph/examples/osguniformbuffer/osguniformbuffer.cpp

205 lines
6.8 KiB
C++
Raw Normal View History

From Tim Moore, "Here is initial support for uniform buffer objects. The binding between a buffer object and an indexed target is implemented as a new StateAttribute, UniformBufferBinding. I've included an example program based on the code in the ARB_uniform_buffer_object specification. A few things remain to do: * The binding between a uniform block in a shader program and a buffer indexed target number is fixed, like a vertex attribute binding. This is too restrictive because that binding can be changed without relinking the program. This mapping should be done by name in the same way that uniform values are handled i.e., like a pseudo state attribute; * There's no direct way yet to query for the offset of uniforms in uniform block, so only the std140 layout is really usable. A helper class that implemented the std140 rules would be quite helpful for setting up uniform blocks without having to link a program first; * There's no direct support for querying parameters such as the maximum block length, minimum offset alignment, etc. Having that information available outside of the draw thread would make certain instancing techniques easier to implement."
2010-11-30 01:43:27 +08:00
#include <iostream>
#include <osg/Array>
#include <osg/BoundingSphere>
#include <osg/BufferIndexBinding>
#include <osg/BufferObject>
#include <osg/Group>
#include <osg/Math>
#include <osg/MatrixTransform>
#include <osg/Program>
#include <osg/Shader>
#include <osgDB/ReadFile>
#include <osgDB/WriteFile>
#include <osgUtil/Optimizer>
#include <osgViewer/Viewer>
#include <osgViewer/ViewerEventHandlers>
using namespace std;
using namespace osg;
// This example is based on the sample code in the
// ARB_uniform_buffer_object extension specification.
GLfloat colors1[] = {
// block
0.45,0.45,1,1,
0.45,0.45,1,1,
0.75,0.75,0.75,1,
0.0,0.0,1.0,1,
0.0,1.0,0.0,1
};
GLfloat colors2[] = {
// block
0.45,0.45,1,1,
0.45,0.45,1,1,
0.75,0.75,0.75,1,
1.0,0.0,0.0,1,
0.0,1.0,0.0,1,
};
char vertexShaderSource[] =
"// Vertex shader for Gooch shading\n"
"// Author: Randi Rost\n"
"// Copyright (c) 2002-2006 3Dlabs Inc. Ltd.\n"
"// See 3Dlabs-License.txt for license information\n"
"vec3 LightPosition = vec3(0.0, 10.0, 4.0); \n"
"varying float NdotL;\n"
"varying vec3 ReflectVec;\n"
"varying vec3 ViewVec;\n"
"void main(void)\n"
"{\n"
"vec3 ecPos = vec3 (gl_ModelViewMatrix * gl_Vertex);\n"
"vec3 tnorm = normalize(gl_NormalMatrix * gl_Normal);\n"
"vec3 lightVec = normalize(LightPosition - ecPos);\n"
"ReflectVec = normalize(reflect(-lightVec, tnorm));\n"
"ViewVec = normalize(-ecPos);\n"
"NdotL = (dot(lightVec, tnorm) + 1.0) * 0.5;\n"
"gl_Position = ftransform();\n"
"}\n";
char fragmentShaderSource[] =
"// Fragment shader for Gooch shading, adapted for ARB_uniform_buffer_object\n"
"#extension GL_ARB_uniform_buffer_object : enable\n"
"layout(std140) uniform colors0\n"
"{\n"
"float DiffuseCool;\n"
"float DiffuseWarm;\n"
"vec3 SurfaceColor;\n"
"vec3 WarmColor;\n"
"vec3 CoolColor;\n"
"};\n"
"varying float NdotL;\n"
"varying vec3 ReflectVec;\n"
"varying vec3 ViewVec;\n"
"void main (void)\n"
"{\n"
"vec3 kcool = min(CoolColor + DiffuseCool * SurfaceColor, 1.0);\n"
"vec3 kwarm = min(WarmColor + DiffuseWarm * SurfaceColor, 1.0); \n"
"vec3 kfinal = mix(kcool, kwarm, NdotL);\n"
"vec3 nreflect = normalize(ReflectVec);\n"
"vec3 nview = normalize(ViewVec);\n"
"float spec = max(dot(nreflect, nview), 0.0);\n"
"spec = pow(spec, 32.0);\n"
"gl_FragColor = vec4 (min(kfinal + spec, 1.0), 1.0);\n"
"}\n";
// Callback for animating the WarmColor
Changed name of UniformCallback to UniformBufferCallback to avoid conflict with changes to come to the osg::Uniform::Callback -> osg::UniformCallback. git-svn-id: http://svn.openscenegraph.org/osg/OpenSceneGraph/trunk@14884 16af8721-9629-0410-8352-f15c8da7e697
2015-06-02 17:18:19 +08:00
class UniformBufferCallback : public StateAttributeCallback
From Tim Moore, "Here is initial support for uniform buffer objects. The binding between a buffer object and an indexed target is implemented as a new StateAttribute, UniformBufferBinding. I've included an example program based on the code in the ARB_uniform_buffer_object specification. A few things remain to do: * The binding between a uniform block in a shader program and a buffer indexed target number is fixed, like a vertex attribute binding. This is too restrictive because that binding can be changed without relinking the program. This mapping should be done by name in the same way that uniform values are handled i.e., like a pseudo state attribute; * There's no direct way yet to query for the offset of uniforms in uniform block, so only the std140 layout is really usable. A helper class that implemented the std140 rules would be quite helpful for setting up uniform blocks without having to link a program first; * There's no direct support for querying parameters such as the maximum block length, minimum offset alignment, etc. Having that information available outside of the draw thread would make certain instancing techniques easier to implement."
2010-11-30 01:43:27 +08:00
{
public:
void operator() (StateAttribute* attr, NodeVisitor* nv)
{
UniformBufferBinding* ubb = static_cast<UniformBufferBinding*>(attr);
change the design of BufferIndexBinding to work with BufferData instead of BufferObject allow convenient BufferData abstraction + serialization of BufferIndexBinding
2017-08-24 05:42:12 +08:00
FloatArray* array = static_cast<FloatArray*>(ubb->getBufferData());
From Tim Moore, "Here is initial support for uniform buffer objects. The binding between a buffer object and an indexed target is implemented as a new StateAttribute, UniformBufferBinding. I've included an example program based on the code in the ARB_uniform_buffer_object specification. A few things remain to do: * The binding between a uniform block in a shader program and a buffer indexed target number is fixed, like a vertex attribute binding. This is too restrictive because that binding can be changed without relinking the program. This mapping should be done by name in the same way that uniform values are handled i.e., like a pseudo state attribute; * There's no direct way yet to query for the offset of uniforms in uniform block, so only the std140 layout is really usable. A helper class that implemented the std140 rules would be quite helpful for setting up uniform blocks without having to link a program first; * There's no direct support for querying parameters such as the maximum block length, minimum offset alignment, etc. Having that information available outside of the draw thread would make certain instancing techniques easier to implement."
2010-11-30 01:43:27 +08:00
double time = nv->getFrameStamp()->getSimulationTime();
double frac = fmod(time, 1.0);
Vec4f warmColor = (Vec4f(0.0, 0.0, 1.0 ,1) * frac
+ Vec4f(1.0, 0.0, 0.0, 1) * (1 - frac));
// Since we're using the std140 layout, we know where the
// warmColor variable is located in the buffer.
for (int i = 0; i < 4; ++i)
(*array)[12 + i] = warmColor[i];
array->dirty();
}
};
Introduced CMake option OSG_PROVIDE_READFILE option that defaults to ON, but when switched to OFF disables the building of the osgDB::read*File() methods, forcing users to use osgDB::readRef*File() methods. The later is preferable as it closes a potential threading bug when using paging databases in conjunction with the osgDB::Registry Object Cache. This threading bug occurs when one thread gets an object from the Cache via an osgDB::read*File() call where only a pointer to the object is passed back, so taking a reference to the object is delayed till it gets reassigned to a ref_ptr<>, but at the same time another thread calls a flush of the Object Cache deleting this object as it's referenceCount is now zero. Using osgDB::readREf*File() makes sure the a ref_ptr<> is passed back and the referenceCount never goes to zero. To ensure the OSG builds when OSG_PROVIDE_READFILE is to OFF the many cases of osgDB::read*File() usage had to be replaced with a ref_ptr<> osgDB::readRef*File() usage. The avoid this change causing lots of other client code to be rewritten to handle the use of ref_ptr<> in place of C pointer I introduced a serious of templte methods in various class to adapt ref_ptr<> to the underly C pointer to be passed to old OSG API's, example of this is found in include/osg/Group: bool addChild(Node* child); // old method which can only be used with a Node* tempalte<class T> bool addChild(const osg::ref_ptr<T>& child) { return addChild(child.get()); } // adapter template method These changes together cover 149 modified files, so it's a large submission. This extent of changes are warrent to make use of the Object Cache and multi-threaded loaded more robust. git-svn-id: http://svn.openscenegraph.org/osg/OpenSceneGraph/trunk@15164 16af8721-9629-0410-8352-f15c8da7e697
2015-10-22 21:42:19 +08:00
From Tim Moore, "Here is initial support for uniform buffer objects. The binding between a buffer object and an indexed target is implemented as a new StateAttribute, UniformBufferBinding. I've included an example program based on the code in the ARB_uniform_buffer_object specification. A few things remain to do: * The binding between a uniform block in a shader program and a buffer indexed target number is fixed, like a vertex attribute binding. This is too restrictive because that binding can be changed without relinking the program. This mapping should be done by name in the same way that uniform values are handled i.e., like a pseudo state attribute; * There's no direct way yet to query for the offset of uniforms in uniform block, so only the std140 layout is really usable. A helper class that implemented the std140 rules would be quite helpful for setting up uniform blocks without having to link a program first; * There's no direct support for querying parameters such as the maximum block length, minimum offset alignment, etc. Having that information available outside of the draw thread would make certain instancing techniques easier to implement."
2010-11-30 01:43:27 +08:00
int main(int argc, char** argv)
{
osg::ArgumentParser arguments(&argc,argv);
osgViewer::Viewer viewer(arguments);
Introduced CMake option OSG_PROVIDE_READFILE option that defaults to ON, but when switched to OFF disables the building of the osgDB::read*File() methods, forcing users to use osgDB::readRef*File() methods. The later is preferable as it closes a potential threading bug when using paging databases in conjunction with the osgDB::Registry Object Cache. This threading bug occurs when one thread gets an object from the Cache via an osgDB::read*File() call where only a pointer to the object is passed back, so taking a reference to the object is delayed till it gets reassigned to a ref_ptr<>, but at the same time another thread calls a flush of the Object Cache deleting this object as it's referenceCount is now zero. Using osgDB::readREf*File() makes sure the a ref_ptr<> is passed back and the referenceCount never goes to zero. To ensure the OSG builds when OSG_PROVIDE_READFILE is to OFF the many cases of osgDB::read*File() usage had to be replaced with a ref_ptr<> osgDB::readRef*File() usage. The avoid this change causing lots of other client code to be rewritten to handle the use of ref_ptr<> in place of C pointer I introduced a serious of templte methods in various class to adapt ref_ptr<> to the underly C pointer to be passed to old OSG API's, example of this is found in include/osg/Group: bool addChild(Node* child); // old method which can only be used with a Node* tempalte<class T> bool addChild(const osg::ref_ptr<T>& child) { return addChild(child.get()); } // adapter template method These changes together cover 149 modified files, so it's a large submission. This extent of changes are warrent to make use of the Object Cache and multi-threaded loaded more robust. git-svn-id: http://svn.openscenegraph.org/osg/OpenSceneGraph/trunk@15164 16af8721-9629-0410-8352-f15c8da7e697
2015-10-22 21:42:19 +08:00
From Tim Moore, "Here is initial support for uniform buffer objects. The binding between a buffer object and an indexed target is implemented as a new StateAttribute, UniformBufferBinding. I've included an example program based on the code in the ARB_uniform_buffer_object specification. A few things remain to do: * The binding between a uniform block in a shader program and a buffer indexed target number is fixed, like a vertex attribute binding. This is too restrictive because that binding can be changed without relinking the program. This mapping should be done by name in the same way that uniform values are handled i.e., like a pseudo state attribute; * There's no direct way yet to query for the offset of uniforms in uniform block, so only the std140 layout is really usable. A helper class that implemented the std140 rules would be quite helpful for setting up uniform blocks without having to link a program first; * There's no direct support for querying parameters such as the maximum block length, minimum offset alignment, etc. Having that information available outside of the draw thread would make certain instancing techniques easier to implement."
2010-11-30 01:43:27 +08:00
if (arguments.argc() <= 1) {
cerr << "Need a scene.\n";
return 1;
}
Introduced CMake option OSG_PROVIDE_READFILE option that defaults to ON, but when switched to OFF disables the building of the osgDB::read*File() methods, forcing users to use osgDB::readRef*File() methods. The later is preferable as it closes a potential threading bug when using paging databases in conjunction with the osgDB::Registry Object Cache. This threading bug occurs when one thread gets an object from the Cache via an osgDB::read*File() call where only a pointer to the object is passed back, so taking a reference to the object is delayed till it gets reassigned to a ref_ptr<>, but at the same time another thread calls a flush of the Object Cache deleting this object as it's referenceCount is now zero. Using osgDB::readREf*File() makes sure the a ref_ptr<> is passed back and the referenceCount never goes to zero. To ensure the OSG builds when OSG_PROVIDE_READFILE is to OFF the many cases of osgDB::read*File() usage had to be replaced with a ref_ptr<> osgDB::readRef*File() usage. The avoid this change causing lots of other client code to be rewritten to handle the use of ref_ptr<> in place of C pointer I introduced a serious of templte methods in various class to adapt ref_ptr<> to the underly C pointer to be passed to old OSG API's, example of this is found in include/osg/Group: bool addChild(Node* child); // old method which can only be used with a Node* tempalte<class T> bool addChild(const osg::ref_ptr<T>& child) { return addChild(child.get()); } // adapter template method These changes together cover 149 modified files, so it's a large submission. This extent of changes are warrent to make use of the Object Cache and multi-threaded loaded more robust. git-svn-id: http://svn.openscenegraph.org/osg/OpenSceneGraph/trunk@15164 16af8721-9629-0410-8352-f15c8da7e697
2015-10-22 21:42:19 +08:00
osg::ref_ptr<osg::Node> loadedModel = osgDB::readRefNodeFiles(arguments);
From Tim Moore, "Here is initial support for uniform buffer objects. The binding between a buffer object and an indexed target is implemented as a new StateAttribute, UniformBufferBinding. I've included an example program based on the code in the ARB_uniform_buffer_object specification. A few things remain to do: * The binding between a uniform block in a shader program and a buffer indexed target number is fixed, like a vertex attribute binding. This is too restrictive because that binding can be changed without relinking the program. This mapping should be done by name in the same way that uniform values are handled i.e., like a pseudo state attribute; * There's no direct way yet to query for the offset of uniforms in uniform block, so only the std140 layout is really usable. A helper class that implemented the std140 rules would be quite helpful for setting up uniform blocks without having to link a program first; * There's no direct support for querying parameters such as the maximum block length, minimum offset alignment, etc. Having that information available outside of the draw thread would make certain instancing techniques easier to implement."
2010-11-30 01:43:27 +08:00
if (!loadedModel) {
cerr << "couldn't load " << argv[1] << "\n";
return 1;
}
osgUtil::Optimizer optimizer;
optimizer.optimize(loadedModel.get());
const BoundingSphere bound = loadedModel->getBound();
const float displacement = 2.25 * bound.radius();
Group* scene = new Group;
StateSet* rootSS = scene->getOrCreateStateSet();
Shader* vertexShader = new Shader(Shader::VERTEX);
vertexShader->setShaderSource(vertexShaderSource);
Shader* fragmentShader = new Shader(Shader::FRAGMENT);
fragmentShader->setShaderSource(fragmentShaderSource);
Program* prog = new Program;
prog->addShader(vertexShader);
prog->addShader(fragmentShader);
prog->addBindUniformBlock("colors0", 0);
rootSS->setAttributeAndModes(prog, StateAttribute::ON);
// Place 3 instances of the loaded model with different uniform
// blocks for each.
//
// The blocksize is known because of the std140 format.
const unsigned blockSize = 20 * sizeof(GLfloat);
ref_ptr<FloatArray> colorArray
= new FloatArray(&colors1[0],
&colors1[sizeof(colors1) / sizeof(GLfloat)]);
ref_ptr<UniformBufferObject> ubo = new UniformBufferObject;
colorArray->setBufferObject(ubo.get());
Group* group1 = new Group;
StateSet* ss1 = group1->getOrCreateStateSet();
group1->addChild(loadedModel.get());
scene->addChild(group1);
ref_ptr<UniformBufferBinding> ubb1
change the design of BufferIndexBinding to work with BufferData instead of BufferObject allow convenient BufferData abstraction + serialization of BufferIndexBinding
2017-08-24 05:42:12 +08:00
= new UniformBufferBinding(0, colorArray.get(), 0, blockSize);
From Tim Moore, "Here is initial support for uniform buffer objects. The binding between a buffer object and an indexed target is implemented as a new StateAttribute, UniformBufferBinding. I've included an example program based on the code in the ARB_uniform_buffer_object specification. A few things remain to do: * The binding between a uniform block in a shader program and a buffer indexed target number is fixed, like a vertex attribute binding. This is too restrictive because that binding can be changed without relinking the program. This mapping should be done by name in the same way that uniform values are handled i.e., like a pseudo state attribute; * There's no direct way yet to query for the offset of uniforms in uniform block, so only the std140 layout is really usable. A helper class that implemented the std140 rules would be quite helpful for setting up uniform blocks without having to link a program first; * There's no direct support for querying parameters such as the maximum block length, minimum offset alignment, etc. Having that information available outside of the draw thread would make certain instancing techniques easier to implement."
2010-11-30 01:43:27 +08:00
ss1->setAttributeAndModes(ubb1.get(), StateAttribute::ON);
Introduced CMake option OSG_PROVIDE_READFILE option that defaults to ON, but when switched to OFF disables the building of the osgDB::read*File() methods, forcing users to use osgDB::readRef*File() methods. The later is preferable as it closes a potential threading bug when using paging databases in conjunction with the osgDB::Registry Object Cache. This threading bug occurs when one thread gets an object from the Cache via an osgDB::read*File() call where only a pointer to the object is passed back, so taking a reference to the object is delayed till it gets reassigned to a ref_ptr<>, but at the same time another thread calls a flush of the Object Cache deleting this object as it's referenceCount is now zero. Using osgDB::readREf*File() makes sure the a ref_ptr<> is passed back and the referenceCount never goes to zero. To ensure the OSG builds when OSG_PROVIDE_READFILE is to OFF the many cases of osgDB::read*File() usage had to be replaced with a ref_ptr<> osgDB::readRef*File() usage. The avoid this change causing lots of other client code to be rewritten to handle the use of ref_ptr<> in place of C pointer I introduced a serious of templte methods in various class to adapt ref_ptr<> to the underly C pointer to be passed to old OSG API's, example of this is found in include/osg/Group: bool addChild(Node* child); // old method which can only be used with a Node* tempalte<class T> bool addChild(const osg::ref_ptr<T>& child) { return addChild(child.get()); } // adapter template method These changes together cover 149 modified files, so it's a large submission. This extent of changes are warrent to make use of the Object Cache and multi-threaded loaded more robust. git-svn-id: http://svn.openscenegraph.org/osg/OpenSceneGraph/trunk@15164 16af8721-9629-0410-8352-f15c8da7e697
2015-10-22 21:42:19 +08:00
From Tim Moore, "Here is initial support for uniform buffer objects. The binding between a buffer object and an indexed target is implemented as a new StateAttribute, UniformBufferBinding. I've included an example program based on the code in the ARB_uniform_buffer_object specification. A few things remain to do: * The binding between a uniform block in a shader program and a buffer indexed target number is fixed, like a vertex attribute binding. This is too restrictive because that binding can be changed without relinking the program. This mapping should be done by name in the same way that uniform values are handled i.e., like a pseudo state attribute; * There's no direct way yet to query for the offset of uniforms in uniform block, so only the std140 layout is really usable. A helper class that implemented the std140 rules would be quite helpful for setting up uniform blocks without having to link a program first; * There's no direct support for querying parameters such as the maximum block length, minimum offset alignment, etc. Having that information available outside of the draw thread would make certain instancing techniques easier to implement."
2010-11-30 01:43:27 +08:00
ref_ptr<FloatArray> colorArray2
= new FloatArray(&colors2[0],
&colors2[sizeof(colors2) / sizeof(GLfloat)]);
ref_ptr<UniformBufferObject> ubo2 = new UniformBufferObject;
colorArray2->setBufferObject(ubo2.get());
MatrixTransform* group2 = new MatrixTransform;
Matrix mat2 = Matrix::translate(-displacement, 0.0, 0.0);
group2->setMatrix(mat2);
StateSet* ss2 = group2->getOrCreateStateSet();
group2->addChild(loadedModel.get());
scene->addChild(group2);
ref_ptr<UniformBufferBinding> ubb2
change the design of BufferIndexBinding to work with BufferData instead of BufferObject allow convenient BufferData abstraction + serialization of BufferIndexBinding
2017-08-24 05:42:12 +08:00
= new UniformBufferBinding(0, colorArray2.get(), 0, blockSize);
From Tim Moore, "Here is initial support for uniform buffer objects. The binding between a buffer object and an indexed target is implemented as a new StateAttribute, UniformBufferBinding. I've included an example program based on the code in the ARB_uniform_buffer_object specification. A few things remain to do: * The binding between a uniform block in a shader program and a buffer indexed target number is fixed, like a vertex attribute binding. This is too restrictive because that binding can be changed without relinking the program. This mapping should be done by name in the same way that uniform values are handled i.e., like a pseudo state attribute; * There's no direct way yet to query for the offset of uniforms in uniform block, so only the std140 layout is really usable. A helper class that implemented the std140 rules would be quite helpful for setting up uniform blocks without having to link a program first; * There's no direct support for querying parameters such as the maximum block length, minimum offset alignment, etc. Having that information available outside of the draw thread would make certain instancing techniques easier to implement."
2010-11-30 01:43:27 +08:00
ss2->setAttributeAndModes(ubb2.get(), StateAttribute::ON);
ref_ptr<FloatArray> colorArray3
= new FloatArray(&colors2[0],
&colors2[sizeof(colors2) / sizeof(GLfloat)]);
ref_ptr<UniformBufferObject> ubo3 = new UniformBufferObject;
colorArray3->setBufferObject(ubo3.get());
MatrixTransform* group3 = new MatrixTransform;
Matrix mat3 = Matrix::translate(displacement, 0.0, 0.0);
group3->setMatrix(mat3);
StateSet* ss3 = group3->getOrCreateStateSet();
group3->addChild(loadedModel.get());
Introduced CMake option OSG_PROVIDE_READFILE option that defaults to ON, but when switched to OFF disables the building of the osgDB::read*File() methods, forcing users to use osgDB::readRef*File() methods. The later is preferable as it closes a potential threading bug when using paging databases in conjunction with the osgDB::Registry Object Cache. This threading bug occurs when one thread gets an object from the Cache via an osgDB::read*File() call where only a pointer to the object is passed back, so taking a reference to the object is delayed till it gets reassigned to a ref_ptr<>, but at the same time another thread calls a flush of the Object Cache deleting this object as it's referenceCount is now zero. Using osgDB::readREf*File() makes sure the a ref_ptr<> is passed back and the referenceCount never goes to zero. To ensure the OSG builds when OSG_PROVIDE_READFILE is to OFF the many cases of osgDB::read*File() usage had to be replaced with a ref_ptr<> osgDB::readRef*File() usage. The avoid this change causing lots of other client code to be rewritten to handle the use of ref_ptr<> in place of C pointer I introduced a serious of templte methods in various class to adapt ref_ptr<> to the underly C pointer to be passed to old OSG API's, example of this is found in include/osg/Group: bool addChild(Node* child); // old method which can only be used with a Node* tempalte<class T> bool addChild(const osg::ref_ptr<T>& child) { return addChild(child.get()); } // adapter template method These changes together cover 149 modified files, so it's a large submission. This extent of changes are warrent to make use of the Object Cache and multi-threaded loaded more robust. git-svn-id: http://svn.openscenegraph.org/osg/OpenSceneGraph/trunk@15164 16af8721-9629-0410-8352-f15c8da7e697
2015-10-22 21:42:19 +08:00
scene->addChild(group3);
From Tim Moore, "Here is initial support for uniform buffer objects. The binding between a buffer object and an indexed target is implemented as a new StateAttribute, UniformBufferBinding. I've included an example program based on the code in the ARB_uniform_buffer_object specification. A few things remain to do: * The binding between a uniform block in a shader program and a buffer indexed target number is fixed, like a vertex attribute binding. This is too restrictive because that binding can be changed without relinking the program. This mapping should be done by name in the same way that uniform values are handled i.e., like a pseudo state attribute; * There's no direct way yet to query for the offset of uniforms in uniform block, so only the std140 layout is really usable. A helper class that implemented the std140 rules would be quite helpful for setting up uniform blocks without having to link a program first; * There's no direct support for querying parameters such as the maximum block length, minimum offset alignment, etc. Having that information available outside of the draw thread would make certain instancing techniques easier to implement."
2010-11-30 01:43:27 +08:00
ref_ptr<UniformBufferBinding> ubb3
change the design of BufferIndexBinding to work with BufferData instead of BufferObject allow convenient BufferData abstraction + serialization of BufferIndexBinding
2017-08-24 05:42:12 +08:00
= new UniformBufferBinding(0, colorArray3.get(), 0, blockSize);
Changed name of UniformCallback to UniformBufferCallback to avoid conflict with changes to come to the osg::Uniform::Callback -> osg::UniformCallback. git-svn-id: http://svn.openscenegraph.org/osg/OpenSceneGraph/trunk@14884 16af8721-9629-0410-8352-f15c8da7e697
2015-06-02 17:18:19 +08:00
ubb3->setUpdateCallback(new UniformBufferCallback);
From Tim Moore, "Here is initial support for uniform buffer objects. The binding between a buffer object and an indexed target is implemented as a new StateAttribute, UniformBufferBinding. I've included an example program based on the code in the ARB_uniform_buffer_object specification. A few things remain to do: * The binding between a uniform block in a shader program and a buffer indexed target number is fixed, like a vertex attribute binding. This is too restrictive because that binding can be changed without relinking the program. This mapping should be done by name in the same way that uniform values are handled i.e., like a pseudo state attribute; * There's no direct way yet to query for the offset of uniforms in uniform block, so only the std140 layout is really usable. A helper class that implemented the std140 rules would be quite helpful for setting up uniform blocks without having to link a program first; * There's no direct support for querying parameters such as the maximum block length, minimum offset alignment, etc. Having that information available outside of the draw thread would make certain instancing techniques easier to implement."
2010-11-30 01:43:27 +08:00
ubb3->setDataVariance(Object::DYNAMIC);
ss3->setAttributeAndModes(ubb3.get(), StateAttribute::ON);
Introduced CMake option OSG_PROVIDE_READFILE option that defaults to ON, but when switched to OFF disables the building of the osgDB::read*File() methods, forcing users to use osgDB::readRef*File() methods. The later is preferable as it closes a potential threading bug when using paging databases in conjunction with the osgDB::Registry Object Cache. This threading bug occurs when one thread gets an object from the Cache via an osgDB::read*File() call where only a pointer to the object is passed back, so taking a reference to the object is delayed till it gets reassigned to a ref_ptr<>, but at the same time another thread calls a flush of the Object Cache deleting this object as it's referenceCount is now zero. Using osgDB::readREf*File() makes sure the a ref_ptr<> is passed back and the referenceCount never goes to zero. To ensure the OSG builds when OSG_PROVIDE_READFILE is to OFF the many cases of osgDB::read*File() usage had to be replaced with a ref_ptr<> osgDB::readRef*File() usage. The avoid this change causing lots of other client code to be rewritten to handle the use of ref_ptr<> in place of C pointer I introduced a serious of templte methods in various class to adapt ref_ptr<> to the underly C pointer to be passed to old OSG API's, example of this is found in include/osg/Group: bool addChild(Node* child); // old method which can only be used with a Node* tempalte<class T> bool addChild(const osg::ref_ptr<T>& child) { return addChild(child.get()); } // adapter template method These changes together cover 149 modified files, so it's a large submission. This extent of changes are warrent to make use of the Object Cache and multi-threaded loaded more robust. git-svn-id: http://svn.openscenegraph.org/osg/OpenSceneGraph/trunk@15164 16af8721-9629-0410-8352-f15c8da7e697
2015-10-22 21:42:19 +08:00
From Tim Moore, "Here is initial support for uniform buffer objects. The binding between a buffer object and an indexed target is implemented as a new StateAttribute, UniformBufferBinding. I've included an example program based on the code in the ARB_uniform_buffer_object specification. A few things remain to do: * The binding between a uniform block in a shader program and a buffer indexed target number is fixed, like a vertex attribute binding. This is too restrictive because that binding can be changed without relinking the program. This mapping should be done by name in the same way that uniform values are handled i.e., like a pseudo state attribute; * There's no direct way yet to query for the offset of uniforms in uniform block, so only the std140 layout is really usable. A helper class that implemented the std140 rules would be quite helpful for setting up uniform blocks without having to link a program first; * There's no direct support for querying parameters such as the maximum block length, minimum offset alignment, etc. Having that information available outside of the draw thread would make certain instancing techniques easier to implement."
2010-11-30 01:43:27 +08:00
viewer.setSceneData(scene);
viewer.realize();
return viewer.run();
}
Reference in New Issue Copy Permalink