fixes in Jam msvcgen -> use SSE instead of SSE2 to stay compatible with older AMD Athlon XP processors, updated build files, minor compile fixes for GPU experimental code

Demos/Box2dDemo/CMakeLists.txt

@@ -0,0 +1,32 @@
+# This is basically the overall name of the project in Visual Studio this is the name of the Solution File
+
+
+# For every executable you have with a main method you should have an add_executable line below.
+# For every add executable line you should list every .cpp and .h file you have associated with that executable.  
+
+
+# This is the variable for Windows.  I use this to define the root of my directory structure.
+SET(GLUT_ROOT ${BULLET_PHYSICS_SOURCE_DIR}/Glut)
+
+# You shouldn't have to modify anything below this line 
+########################################################
+
+INCLUDE_DIRECTORIES(
+${BULLET_PHYSICS_SOURCE_DIR}/src ${BULLET_PHYSICS_SOURCE_DIR}/Demos/OpenGL }
+)
+
+LINK_LIBRARIES(
+OpenGLSupport  BulletDynamics  BulletCollision LinearMath    ${GLUT_glut_LIBRARY} ${OPENGL_gl_LIBRARY} ${OPENGL_glu_LIBRARY}
+)
+
+ADD_EXECUTABLE(AppBox2dDemo
+	main.cpp
+	Box2dDemo.cpp
+	btBox2dBox2dCollisionAlgorithm.h
+	Box2dDemo.h
+	btBox2dShape.cpp
+	btBox2dBox2dCollisionAlgorithm.cpp	
+	btBox2dShape.h
+)
+
+

Demos/Box2dDemo/Jamfile

@@ -0,0 +1,3 @@
+SubDir TOP Demos Box2dDemo ;
+
+BulletDemo Box2dDemo : [ Wildcard *.h *.cpp  ] ;

Demos/CMakeLists.txt

@@ -1,7 +1,7 @@
 if (CMAKE_SIZEOF_VOID_P MATCHES "8")
 SUBDIRS( OpenGL AllBulletDemos  ConvexDecompositionDemo Benchmarks HelloWorld
 CcdPhysicsDemo ConstraintDemo SliderConstraintDemo GenericJointDemo Raytracer
-RagdollDemo ForkLiftDemo BasicDemo BspDemo MovingConcaveDemo VehicleDemo
+RagdollDemo ForkLiftDemo BasicDemo Box2dDemo Gpu2dDemo BspDemo MovingConcaveDemo VehicleDemo
 ColladaDemo UserCollisionAlgorithm CharacterDemo SoftDemo HeightFieldFluidDemo
 CollisionInterfaceDemo ConcaveConvexcastDemo SimplexDemo DynamicControlDemo
 DoublePrecisionDemo ConcaveDemo CollisionDemo
@@ -10,7 +10,7 @@ MultiMaterialDemo)
 else (CMAKE_SIZEOF_VOID_P MATCHES "8")
 SUBDIRS( OpenGL AllBulletDemos ConvexDecompositionDemo Benchmarks HelloWorld
 MultiThreadedDemo CcdPhysicsDemo ConstraintDemo SliderConstraintDemo Raytracer
-GenericJointDemo RagdollDemo ForkLiftDemo BasicDemo BspDemo MovingConcaveDemo
+GenericJointDemo RagdollDemo ForkLiftDemo BasicDemo Box2dDemo Gpu2dDemo BspDemo MovingConcaveDemo
 VehicleDemo ColladaDemo UserCollisionAlgorithm CharacterDemo SoftDemo
  HeightFieldFluidDemo
 CollisionInterfaceDemo ConcaveConvexcastDemo SimplexDemo DynamicControlDemo

Demos/Gpu2dDemo/CMakeLists.txt

@@ -19,8 +19,20 @@ LINK_LIBRARIES(
 OpenGLSupport  BulletDynamics  BulletCollision LinearMath    ${GLUT_glut_LIBRARY} ${OPENGL_gl_LIBRARY} ${OPENGL_glu_LIBRARY}
 )
 
-ADD_EXECUTABLE(AppBasicDemo
+ADD_EXECUTABLE(AppGpu2dDemo
 	main.cpp
 	BasicDemo.cpp
+	BasicDemo.h
+	btGpuDemoPairCache.cpp
+	btGpuDemoPairCache.h
+	btGpuDemo2dSharedTypes.h
+	btGpuDemo2dCpuFunc.cpp
+	btGpuDemoDynamicsWorld.cpp
+	btGpuDemoDynamicsWorld.h
+	oecakeLoader.cpp
+	oecakeLoader.h
+	btGpuDemo2dSharedCode.h
+	btGpuDemo2dSharedDefs.h
 )
 
+

Demos/Gpu2dDemo/Jamfile

@@ -1,3 +1,3 @@
-SubDir TOP Demos BasicDemo ;
+SubDir TOP Demos Gpu2dDemo ;
 
-BulletDemo BasicDemo : [ Wildcard *.h *.cpp  ] ;
+BulletDemo Gpu2dDemo : [ Wildcard *.h *.cpp  ] ;

Demos/Gpu2dDemo/btGpuDemo2dSharedCode.h

@@ -19,6 +19,7 @@ subject to the following restrictions:
 #define FRICTION_BOX_GROUND_FACT 0.05f
 #define FRICTION_BOX_BOX_FACT 0.05f
 #define USE_CENTERS 1
+#include "LinearMath/btMinMax.h"
 
 //------------------------------------------------------------------------------------------------
 //------------------------------------------------------------------------------------------------
@@ -58,6 +59,7 @@ BT_GPU___device__ void testSphSph(float3 aPos, float3 bPos, float radA, float ra
 	float dist = BT_GPU_dot(del, del);
 	dist = sqrtf(dist);
 	float maxD = radA + radB;
+	
 	if(dist > maxD)
 	{
 		return;
@@ -76,7 +78,9 @@ BT_GPU___device__ void testSphSph(float3 aPos, float3 bPos, float radA, float ra
 		normal = BT_GPU_make_float3(1.f, 0.f, 0.f);
 	}
 //	float3 contact = (bPos + aPos + normal * (radB - radA)) * 0.5f;
-	float3 contact = aPos - normal * radA;
+	float3 tmp = (normal * radA);
+	float3 contact = aPos - tmp;
+	
 	// now add point
 	int numPoints = 0;
 	for(int i = 0; i < MAX_VTX_PER_OBJ; i++)
@@ -143,14 +147,20 @@ BT_GPU___global__ void setConstraintDataD(int2 *constraints,
 		for(i = 0; i < numSphA; i++)
 		{	
 			float3 va = aPos;
-			va += ai * shapeA[i].x;
+			float3 tmp = ai * shapeA[i].x; 
-			va += aj * shapeA[i].y;
+			float3 tmp2 = aj * shapeA[i].y;
+			
+			va += tmp;
+			va += tmp2;
+			
 			float radA = shapeA[i].w;
 			for(j = 0; j < numSphB; j++)
 			{
 				float3 vb = bPos;
-				vb += bi * shapeB[j].x;
+				float3 tmp =bi * shapeB[j].x;
-				vb += bj * shapeB[j].y;
+				float3 tmp2 = bj * shapeB[j].y;
+				vb += tmp;
+				vb += tmp2;
 				float radB = shapeB[j].w;
 				testSphSph(va, vb, radA, radB, pOut);
 			}
@@ -174,7 +184,7 @@ BT_GPU___device__ float computeImpulse1(float3 rVel,
 	if(positionConstraint > 0)
 		return lambdaDt;
 
-	positionConstraint = min(0.0f,positionConstraint+penetrationError);
+	positionConstraint = btMin(0.0f,positionConstraint+penetrationError);
 	
 	lambdaDt	=	-(BT_GPU_dot(cNormal,rVel)*(1+collisionConstant));
 	lambdaDt	-=	(baumgarteConstant/dt*positionConstraint);
@@ -216,7 +226,8 @@ BT_GPU___global__ void collisionWithWallBoxD(float4 *pos,
 			float3 aVel = BT_GPU_make_float3(vel[idx].x, vel[idx].y, vel[idx].z);
 			float aAngVel = angVel[idx];
 			float3 rerVertex = ai * shape[iVtx].x;
-			rerVertex += aj * shape[iVtx].y;
+			float3 tmp = aj * shape[iVtx].y;
+			rerVertex += tmp;
 			float3 vPos = aPos + rerVertex;
 			float rad = shape[iVtx].w;
 			float3 vVel	=aVel+BT_GPU_cross(BT_GPU_make_float3(0.0f,0.0f,aAngVel),rerVertex);
@@ -239,11 +250,14 @@ BT_GPU___global__ void collisionWithWallBoxD(float4 *pos,
 					{
 						lat_vel_len = sqrtf(lat_vel_len);
 						lat_vel *= 1.f/lat_vel_len;	
-						impulse	-= lat_vel * BT_GPU_dot(lat_vel, vVel) * FRICTION_BOX_GROUND_FACT;
+						float3 tmp = lat_vel * BT_GPU_dot(lat_vel, vVel) * FRICTION_BOX_GROUND_FACT;
+						impulse	-= tmp;
 					}
 #endif //USE_FRICTION
-					vel[idx]	+=	BT_GPU_make_float42(impulse,0.0f);
+					float4 tmp = BT_GPU_make_float42(impulse,0.0f);
-					angVel[idx]	+=	BT_GPU_cross(rerVertex,impulse).z;
+					vel[idx]	+=	tmp;
+					float tmp2 = BT_GPU_cross(rerVertex,impulse).z;
+					angVel[idx]	+=	tmp2;
 				}
 			}
 
@@ -257,7 +271,8 @@ BT_GPU___global__ void collisionWithWallBoxD(float4 *pos,
 						BT_GPU_make_float3(1.0f,0.0f,0.0f),
 						dt);
 
-					vel[idx]	+=	BT_GPU_make_float42(impulse,0.0f);
+					float4 tmp = BT_GPU_make_float42(impulse,0.0f);
+					vel[idx]	+=	tmp;
 					angVel[idx]	+=	BT_GPU_cross(rerVertex,impulse).z;
 				}
 			}
@@ -272,7 +287,8 @@ BT_GPU___global__ void collisionWithWallBoxD(float4 *pos,
 						BT_GPU_make_float3(-1.0f,0.0f,0.0f),
 						dt);
 
-					vel[idx]	+=	BT_GPU_make_float42(impulse,0.0f);
+					float4 tmp = BT_GPU_make_float42(impulse,0.0f);
+					vel[idx]	+=	tmp;
 					angVel[idx]	+=	BT_GPU_cross(rerVertex,impulse).z;
 				}
 			}
@@ -326,7 +342,7 @@ BT_GPU___device__ void collisionResolutionBox(	int constrId,
 			{
 				float rLambdaDt=lambdaDtBox[(MAX_VTX_PER_OBJ)*(2*constrId)+iVtx];
 				float pLambdaDt=rLambdaDt;
-				rLambdaDt=max(pLambdaDt+lambdaDt,0.0f);
+				rLambdaDt=btMax(pLambdaDt+lambdaDt,0.0f);
 				lambdaDt=rLambdaDt-pLambdaDt;
 				lambdaDtBox[(MAX_VTX_PER_OBJ)*(2*constrId)+iVtx]=rLambdaDt;
 			}
@@ -340,7 +356,8 @@ BT_GPU___device__ void collisionResolutionBox(	int constrId,
 				{
 					lat_vel_len = sqrtf(lat_vel_len);
 					lat_vel *= 1.f/lat_vel_len;
-					impulse	-= lat_vel * BT_GPU_dot(lat_vel , relVel) * FRICTION_BOX_BOX_FACT;
+					float3 tmp = lat_vel * BT_GPU_dot(lat_vel , relVel) * FRICTION_BOX_BOX_FACT;
+					impulse	-= tmp;
 				}
 			}
 #endif //USE_FRICTION

Demos/Jamfile

@@ -81,6 +81,8 @@ SubInclude TOP Demos EPAPenDepthDemo ;
 SubInclude TOP Demos HelloWorld ;
 SubInclude TOP Demos BspDemo ;
 SubInclude TOP Demos BasicDemo ;
+SubInclude TOP Demos Box2dDemo ;
+SubInclude TOP Demos Gpu2dDemo ;
 SubInclude TOP Demos ConvexDecompositionDemo ;
 SubInclude TOP Demos ColladaDemo ;
 SubInclude TOP Demos CharacterDemo ;

Demos/OpenGL/GLDebugFont.cpp

@@ -16,8 +16,9 @@ subject to the following restrictions:
 #include "GLDebugFont.h"
 #include "GlutStuff.h"
 #include <stdio.h>
+#include <string>
 
-unsigned char sFontData[];
+extern unsigned char sFontData[];
 static GLuint sTexture = -1;
 static int sScreenWidth = 0;
 static int sScreenHeight = 0;

Demos/OpenGL/GL_ShapeDrawer.cpp

@@ -368,7 +368,7 @@ void renderSquareA(float x, float y, float z)
 	glEnd();
 }
 
-inline void glDrawVector(btVector3& v) { glVertex3d(v[0], v[1], v[2]); }
+inline void glDrawVector(const btVector3& v) { glVertex3d(v[0], v[1], v[2]); }
 
 
 void GL_ShapeDrawer::drawOpenGL(btScalar* m, const btCollisionShape* shape, const btVector3& color,int	debugMode,const btVector3& worldBoundsMin,const btVector3& worldBoundsMax)

mk/msvcgen/control.tlib

@@ -219,7 +219,7 @@ projtypes = {
   },
   library =>
   {
-    defines = ['_LIB', '_WINDOWS','_CRT_SECURE_NO_DEPRECATE','_CRT_NONSTDC_NO_DEPRECATE']
+    defines = ['_LIB', '_CONSOLE','_CRT_SECURE_NO_DEPRECATE','_CRT_NONSTDC_NO_DEPRECATE']
   }
 };
 

mk/msvcgen/projectx7.tlib

@@ -68,13 +68,13 @@
 
 [% MACRO compiler_release BLOCK -%]
 				EnableIntrinsicFunctions="true"
-				EnableEnhancedInstructionSet="2"
+				EnableEnhancedInstructionSet="1"
 				FloatingPointModel="2"
 				Optimization="2"
 				StringPooling="TRUE"
 				EnableFunctionLevelLinking="TRUE"
 				RuntimeLibrary="0"
-				DebugInformationFormat="3"
+				DebugInformationFormat="1"
 				BufferSecurityCheck="FALSE"
 [% END %]
 [% MACRO compiler_release_dll BLOCK -%]
@@ -88,21 +88,21 @@
 [% MACRO compiler_debug BLOCK -%]
 				Optimization="0"
 				MinimalRebuild="TRUE"
-				DebugInformationFormat="4"
+				DebugInformationFormat="3"
 				RuntimeTypeInfo="FALSE"
 				RuntimeLibrary="1"
 [% END %]
 [% MACRO compiler_debug_dll BLOCK -%]
                                 Optimization="0"
                                 MinimalRebuild="TRUE"
-                                DebugInformationFormat="4"
+                                DebugInformationFormat="3"
                                 RuntimeTypeInfo="FALSE"
                                 RuntimeLibrary="3"
 [% END %]
 [% MACRO compiler_debug_dbl BLOCK -%]
                                 Optimization="0"
                                 MinimalRebuild="TRUE"
-                                DebugInformationFormat="4"
+                                DebugInformationFormat="3"
                                 RuntimeTypeInfo="FALSE"
                                 RuntimeLibrary="1"
 [% END %]
@@ -110,7 +110,7 @@
 				LinkIncremental="1"
 				OptimizeReferences="2"
 				EnableCOMDATFolding="2"
-				GenerateDebugInformation="TRUE"
+				GenerateDebugInformation="FALSE"
 				IgnoreDefaultLibraryNames="LIBC,LIBCD"
 [% END %]
 [% MACRO linker_debug BLOCK -%]

mk/msvcgen/projectxsn71.tlib

@@ -77,7 +77,7 @@
 [% MACRO compiler_debug BLOCK -%]
 				Optimization="0"
 				MinimalRebuild="TRUE"
-				DebugInformationFormat="4"
+				DebugInformationFormat="3"
 				RuntimeTypeInfo="FALSE"
 				RuntimeLibrary="3"
 [% END %]

src/BulletMultiThreaded/btGpu3DGridBroadphase.cpp

@@ -27,6 +27,7 @@ subject to the following restrictions:
 #include "btGpu3DGridBroadphaseSharedDefs.h"
 
 #include "btGpu3DGridBroadphase.h"
+#include <string> //for memset
 
 //--------------------------------------------------------------------------
 

src/BulletMultiThreaded/btGpuDefines.h

@@ -113,11 +113,11 @@ inline int3 operator+(int3 a, int3 b)
     return bt3dGrid_make_int3(a.x + b.x, a.y + b.y, a.z + b.z);
 }
 
-inline float4 operator+(float4& a, float4& b)
+inline float4 operator+(const float4& a, const float4& b)
 {
 	float4 r; r.x = a.x+b.x; r.y = a.y+b.y; r.z = a.z+b.z; r.w = a.w+b.w; return r;
 }
-inline float4 operator*(float4& a, float fact)
+inline float4 operator*(const float4& a, float fact)
 {
 	float4 r; r.x = a.x*fact; r.y = a.y*fact; r.z = a.z*fact; r.w = a.w*fact; return r;
 }
@@ -136,15 +136,11 @@ inline float4& operator+=(float4& a, float4& b)
 	return a;
 }
 
-inline float3 operator+(float3& a, float3& b)
-{
-	float3 r; r.x = a.x+b.x; r.y = a.y+b.y; r.z = a.z+b.z; return r;
-}
 inline float3 operator+(const float3& a, const float3& b)
 {
 	float3 r; r.x = a.x+b.x; r.y = a.y+b.y; r.z = a.z+b.z; return r;
 }
-inline float3 operator-(float3& a, float3& b)
+inline float3 operator-(const float3& a, const float3& b)
 {
 	float3 r; r.x = a.x-b.x; r.y = a.y-b.y; r.z = a.z-b.z; return r;
 }
@@ -160,16 +156,13 @@ static inline float bt3dGrid_dot4(float4& a, float4& b)
 }
 #define BT_GPU_dot4(a,b) bt3dGrid_dot4(a,b)
 
-static inline float3 bt3dGrid_cross(float3& a, float3& b)
+static inline float3 bt3dGrid_cross(const float3& a, const float3& b)
 {
 	float3 r; r.x = a.y*b.z-a.z*b.y; r.y = -a.x*b.z+a.z*b.x; r.z = a.x*b.y-a.y*b.x;	return r;
 }
 #define BT_GPU_cross(a,b) bt3dGrid_cross(a,b)
 
-inline float3 operator*(float3& a, float fact)
+
-{
-	float3 r; r.x = a.x*fact; r.y = a.y*fact; r.z = a.z*fact; return r;
-}
 inline float3 operator*(const float3& a, float fact)
 {
 	float3 r; r.x = a.x*fact; r.y = a.y*fact; r.z = a.z*fact; return r;