ivu_linear_piii.hxx

hl2 source code. Do not use it illegal.

	IVP_DOUBLE c = get_elem(0,2)*p_in->k[0] + get_elem(1,2)*p_in->k[1] + get_elem(2,2)*p_in->k[2];
	p_out->k[0] = (IVP_FLOAT)a;p_out->k[1] = (IVP_FLOAT)b; p_out->k[2] = (IVP_FLOAT)c;
    }
};

inline void IVP_U_Matrix::inline_vimult4( const IVP_U_Float_Point *p_in, IVP_U_Float_Point * p_out )const{
    IVP_IF_PREFETCH_ENABLED(IVP_TRUE){
	IVP_DOUBLE a = p_in->k[0] - vv.k[0];
	IVP_DOUBLE b = p_in->k[1] - vv.k[1];
	IVP_DOUBLE c = p_in->k[2] - vv.k[2];
	
	p_out->k[0] = get_elem(0,0)*a + get_elem(1,0)*b + get_elem(2,0)*c;
	p_out->k[1] = get_elem(0,1)*a + get_elem(1,1)*b + get_elem(2,1)*c;
	p_out->k[2] = get_elem(0,2)*a + get_elem(1,2)*b + get_elem(2,2)*c;
    }else{
	IVP_DOUBLE a = p_in->k[0] - vv.k[0];
	IVP_DOUBLE b = p_in->k[1] - vv.k[1];
	IVP_DOUBLE c = p_in->k[2] - vv.k[2];
	
	p_out->k[0] = get_elem(0,0)*a + get_elem(1,0)*b + get_elem(2,0)*c;
	p_out->k[1] = get_elem(0,1)*a + get_elem(1,1)*b + get_elem(2,1)*c;
	p_out->k[2] = get_elem(0,2)*a + get_elem(1,2)*b + get_elem(2,2)*c;
    }
};


void IVP_U_Matrix3::inline_vmult3( const IVP_U_Float_Point *p_in, IVP_U_Float_Point * p_out ) const{
    IVP_IF_PREFETCH_ENABLED(IVP_TRUE){
	IVP_DOUBLE a = get_elem(0,0)*p_in->k[0] + get_elem(0,1)*p_in->k[1] + get_elem(0,2)*p_in->k[2];
	IVP_DOUBLE b = get_elem(1,0)*p_in->k[0] + get_elem(1,1)*p_in->k[1] + get_elem(1,2)*p_in->k[2];
	IVP_DOUBLE c = get_elem(2,0)*p_in->k[0] + get_elem(2,1)*p_in->k[1] + get_elem(2,2)*p_in->k[2];
	p_out->k[0] = (IVP_FLOAT)a;p_out->k[1] = (IVP_FLOAT)b; p_out->k[2] = (IVP_FLOAT)c;
    }else{
	IVP_DOUBLE a = get_elem(0,0)*p_in->k[0] + get_elem(0,1)*p_in->k[1] + get_elem(0,2)*p_in->k[2];
	IVP_DOUBLE b = get_elem(1,0)*p_in->k[0] + get_elem(1,1)*p_in->k[1] + get_elem(1,2)*p_in->k[2];
	IVP_DOUBLE c = get_elem(2,0)*p_in->k[0] + get_elem(2,1)*p_in->k[1] + get_elem(2,2)*p_in->k[2];
	p_out->k[0] = (IVP_FLOAT)a;p_out->k[1] = (IVP_FLOAT)b; p_out->k[2] = (IVP_FLOAT)c;
    }
};


inline void IVP_U_Matrix::inline_vmult4(const IVP_U_Float_Point *p_in, IVP_U_Float_Point * p_out)const{
    IVP_IF_PREFETCH_ENABLED(IVP_TRUE){
	IVP_DOUBLE h = p_in->k[0];
	IVP_DOUBLE a = h * get_elem(0,0);
	IVP_DOUBLE b = h * get_elem(1,0);
	IVP_DOUBLE c = h * get_elem(2,0);
	IVP_DOUBLE x,y;
	h = p_in->k[1]; x = h*get_elem(0,1);	y = h*get_elem(1,1);	a += x;	x = h*get_elem(2,1);	 b+=y, c+= x;
	h = p_in->k[2]; x = h*get_elem(0,2);	y = h*get_elem(1,2);	a += x;	x = h*get_elem(2,2);	 b+=y, c+= x;
	a += vv.k[0];    b += vv.k[1];    c += vv.k[2];
	p_out->k[0] = (IVP_FLOAT)a;p_out->k[1] = (IVP_FLOAT)b; p_out->k[2] = (IVP_FLOAT)c;
    }else{
	IVP_DOUBLE h = p_in->k[0];
	IVP_DOUBLE a = h * get_elem(0,0);
	IVP_DOUBLE b = h * get_elem(1,0);
	IVP_DOUBLE c = h * get_elem(2,0);
	IVP_DOUBLE x,y;
	h = p_in->k[1]; x = h*get_elem(0,1);	y = h*get_elem(1,1);	a += x;	x = h*get_elem(2,1);	 b+=y, c+= x;
	h = p_in->k[2]; x = h*get_elem(0,2);	y = h*get_elem(1,2);	a += x;	x = h*get_elem(2,2);	 b+=y, c+= x;
	a += vv.k[0];    b += vv.k[1];    c += vv.k[2];
	p_out->k[0] = (IVP_FLOAT)a;p_out->k[1] = (IVP_FLOAT)b; p_out->k[2] = (IVP_FLOAT)c;
    }
}


void IVP_U_Float_Point::inline_calc_cross_product(const IVP_U_Float_Point *v1,const IVP_U_Float_Point *v2)
{
    IVP_IF_PREFETCH_ENABLED(IVP_TRUE){
	IVP_DOUBLE a  = v1->k[1] * v2->k[2];
	IVP_DOUBLE a2 = v2->k[1] * v1->k[2];
	IVP_DOUBLE b  = v1->k[2] * v2->k[0];	
	IVP_DOUBLE b2 = v2->k[2] * v1->k[0];	a -= a2;
	IVP_DOUBLE c  = v1->k[0] * v2->k[1];
	IVP_DOUBLE c2 = v2->k[0] * v1->k[1];	b -= b2;
	k[0] = (IVP_FLOAT)a;k[1] = (IVP_FLOAT)b;	c -= c2;
	k[2] = (IVP_FLOAT)c;
    }else{
	IVP_DOUBLE a  = v1->k[1] * v2->k[2];
	IVP_DOUBLE a2 = v2->k[1] * v1->k[2];
	IVP_DOUBLE b  = v1->k[2] * v2->k[0];	
	IVP_DOUBLE b2 = v2->k[2] * v1->k[0];	a -= a2;
	IVP_DOUBLE c  = v1->k[0] * v2->k[1];
	IVP_DOUBLE c2 = v2->k[0] * v1->k[1];	b -= b2;
	k[0] = (IVP_FLOAT)a;k[1] = (IVP_FLOAT)b;	c -= c2;
	k[2] = (IVP_FLOAT)c;
    }
}

void IVP_U_Float_Point::inline_calc_cross_product_and_normize(const IVP_U_Float_Point *v1,const IVP_U_Float_Point *v2)
{
    IVP_IF_PREFETCH_ENABLED(IVP_TRUE){
	IVP_DOUBLE a  = v1->k[1] * v2->k[2];
	IVP_DOUBLE a2 = v2->k[1] * v1->k[2];
	IVP_DOUBLE b  = v1->k[2] * v2->k[0];	
	IVP_DOUBLE b2 = v2->k[2] * v1->k[0];	a -= a2;
	IVP_DOUBLE c  = v1->k[0] * v2->k[1];
	IVP_DOUBLE c2 = v2->k[0] * v1->k[1];	b -= b2;
	k[0] = (IVP_FLOAT)a;k[1] = (IVP_FLOAT)b;	c -= c2;
	k[2] = (IVP_FLOAT)c;
	this->normize();
    }else{
	IVP_DOUBLE a  = v1->k[1] * v2->k[2];
	IVP_DOUBLE a2 = v2->k[1] * v1->k[2];
	IVP_DOUBLE b  = v1->k[2] * v2->k[0];	
	IVP_DOUBLE b2 = v2->k[2] * v1->k[0];	a -= a2;
	IVP_DOUBLE c  = v1->k[0] * v2->k[1];
	IVP_DOUBLE c2 = v2->k[0] * v1->k[1];	b -= b2;
	k[0] = (IVP_FLOAT)a;k[1] = (IVP_FLOAT)b;	c -= c2;
	k[2] = (IVP_FLOAT)c;
	this->normize();
    }
}

void IVP_U_Float_Point::inline_set_vert_to_area_defined_by_three_points(const IVP_U_Float_Point *tp0,const IVP_U_Float_Point *tp1,const IVP_U_Float_Point *tp2)
{
    IVP_IF_PREFETCH_ENABLED(IVP_TRUE){
	IVP_DOUBLE a0, a1, a2, b0, b1, b2;
	// calculate (not-normized) normal
	IVP_DOUBLE fp0 = tp0->k[0];
	IVP_DOUBLE fp1 = tp0->k[1];
	IVP_DOUBLE fp2 = tp0->k[2];
	a0 = tp1->k[0] - fp0;
	a1 = tp1->k[1] - fp1;
	a2 = tp1->k[2] - fp2;
	b0 = tp2->k[0] - fp0;
	b1 = tp2->k[1] - fp1;
	b2 = tp2->k[2] - fp2;

	IVP_DOUBLE c0 = b1*a2 - a1*b2;
	IVP_DOUBLE c1 = b2*a0 - a2*b0;
	IVP_DOUBLE c2 = b0*a1 - a0*b1;
	
	// calculate hesse-area
	this->k[0]=	c0;
	this->k[1]=	c1;
	this->k[2]=	c2;
    }else{
	IVP_DOUBLE a0, a1, a2, b0, b1, b2;
	// calculate (not-normized) normal
	IVP_DOUBLE fp0 = tp0->k[0];
	IVP_DOUBLE fp1 = tp0->k[1];
	IVP_DOUBLE fp2 = tp0->k[2];
	a0 = tp1->k[0] - fp0;
	a1 = tp1->k[1] - fp1;
	a2 = tp1->k[2] - fp2;
	b0 = tp2->k[0] - fp0;
	b1 = tp2->k[1] - fp1;
	b2 = tp2->k[2] - fp2;

	IVP_DOUBLE c0 = b1*a2 - a1*b2;
	IVP_DOUBLE c1 = b2*a0 - a2*b0;
	IVP_DOUBLE c2 = b0*a1 - a0*b1;
	
	// calculate hesse-area
	this->k[0]=	c0;
	this->k[1]=	c1;
	this->k[2]=	c2;
    }
}

inline void IVP_U_Quat::init(){
    IVP_IF_PREFETCH_ENABLED(IVP_TRUE){
	x = y= z= w= 0.0f;
    }else{
	x = y= z= w= 0.0f;
    }
}

inline IVP_DOUBLE IVP_U_Quat::acos_quat(const IVP_U_Quat* q) const{ // acosinus between two quats
    return x * q->x + y * q->y + z * q->z + w * q->w;
}

inline void IVP_U_Quat::inline_set_mult_quat(const IVP_U_Quat* q1,const  IVP_U_Quat* q2) {
    IVP_IF_PREFETCH_ENABLED(IVP_TRUE){
	IVP_U_Quat *res=this;
	IVP_DOUBLE xx,yy,zz,ww;
	xx = q1->w * q2->x + q1->x * q2->w + q1->y * q2->z - q1->z * q2->y;
	yy = q1->w * q2->y + q1->y * q2->w + q1->z * q2->x - q1->x * q2->z;
	zz = q1->w * q2->z + q1->z * q2->w + q1->x * q2->y - q1->y * q2->x;
	ww = q1->w * q2->w - q1->x * q2->x - q1->y * q2->y - q1->z * q2->z;
	res->x = xx;
	res->y = yy;
	res->z = zz;
	res->w = ww;
    }else{
	IVP_U_Quat *res=this;
	IVP_DOUBLE xx,yy,zz,ww;
	xx = q1->w * q2->x + q1->x * q2->w + q1->y * q2->z - q1->z * q2->y;
	yy = q1->w * q2->y + q1->y * q2->w + q1->z * q2->x - q1->x * q2->z;
	zz = q1->w * q2->z + q1->z * q2->w + q1->x * q2->y - q1->y * q2->x;
	ww = q1->w * q2->w - q1->x * q2->x - q1->y * q2->y - q1->z * q2->z;
	res->x = xx;
	res->y = yy;
	res->z = zz;
	res->w = ww;
    }
#if 0
  asm __volatile__
	("
		lqc2    vf4,0x0(%1)	#q1
		lqc2    vf5,0x0(%2)	#q2

	 	vmul.xyzw vf12, vf4,vf5	    #vf12 x1x2 y1y2 z1z2 w1w2
		vmulw.xyz vf8,vf4,vf5	    #vf8  x1w2 y1w2 z1w2 w1w2
		vmulw.xyz vf9,vf5,vf4	    #vf9  w1x2 w1y2 w1z2 w1w2  
		#nop 
		vsubx.w vf12,vf12,vf12	    # vf12.w = ww - xx
		vopmula.xyz	ACC,vf4,vf5
		vopmsub.xyz	vf6,vf5,vf4

		vadd.xyz vf8, vf8, vf9
		vsuby.w vf12,vf12,vf12	    # vf12.w = ww - xx - yy
		#nop nop 
		vadd.xyz vf8, vf8,vf6
		vsubz.w vf8,vf12,vf12	     # vf8.w = ww - xx - yy - zz
		#nop nop nop
		sqc2    vf8,0x0(%0)
	"
	: /*no output */
	: "r" (this) , "r" (q1) ,"r" (q2)
	: "memory"    );
#endif
}

inline void IVP_U_Quat::inline_set_mult_quat(const IVP_U_Quat* q1,const  IVP_U_Float_Quat* q2) {
    this->inline_set_mult_quat( q1, (IVP_U_Quat *)q2 );
}

void IVP_U_Quat::normize_correct_step(int steps) {
    IVP_IF_PREFETCH_ENABLED(IVP_TRUE){
	IVP_U_Quat *quat = this;
	IVP_DOUBLE square = quat->x * quat->x + quat->y * quat->y + quat->z * quat->z + quat->w * quat->w;
	square *= 0.5f;
	IVP_DOUBLE factor = 1.5f - square;
	if (steps > 1)    factor += 0.5f - ( factor * factor * square );
	if (steps > 2)    factor += 0.5f - ( factor * factor * square );
	if (steps > 3)    factor += 0.5f - ( factor * factor * square );
	if (steps > 4)    factor += 0.5f - ( factor * factor * square );
	if (steps > 5)    factor += 0.5f - ( factor * factor * square );
	quat->x *= factor;
	quat->y *= factor;
	quat->z *= factor;
	quat->w *= factor;
    }else{
	IVP_U_Quat *quat = this;
	IVP_DOUBLE square = quat->x * quat->x + quat->y * quat->y + quat->z * quat->z + quat->w * quat->w;
	square *= 0.5f;
	IVP_DOUBLE factor = 1.5f - square;
	if (steps > 1)    factor += 0.5f - ( factor * factor * square );
	if (steps > 2)    factor += 0.5f - ( factor * factor * square );
	if (steps > 3)    factor += 0.5f - ( factor * factor * square );
	if (steps > 4)    factor += 0.5f - ( factor * factor * square );
	if (steps > 5)    factor += 0.5f - ( factor * factor * square );
	quat->x *= factor;
	quat->y *= factor;
	quat->z *= factor;
	quat->w *= factor;
    }
}

void IVP_U_Float_Quat::set(const IVP_U_Quat *source){
    x = (IVP_FLOAT)source->x;
    y = (IVP_FLOAT)source->y;
    z = (IVP_FLOAT)source->z;
    w = (IVP_FLOAT)source->w;
}