/**
* Compute point size for each vertex from the vertex eye-space Z
* coordinate and the point size attenuation factors.
* Only done when point size attenuation is enabled and vertex program is
* disabled.
*/
static GLboolean
run_point_stage(struct gl_context *ctx, struct tnl_pipeline_stage *stage)
{
if (ctx->Point._Attenuated && !ctx->VertexProgram._Current) {
struct point_stage_data *store = POINT_STAGE_DATA(stage);
struct vertex_buffer *VB = &TNL_CONTEXT(ctx)->vb;
const GLfloat *eyeCoord = (GLfloat *) VB->EyePtr->data + 2;
const GLint eyeCoordStride = VB->EyePtr->stride / sizeof(GLfloat);
const GLfloat p0 = ctx->Point.Params[0];
const GLfloat p1 = ctx->Point.Params[1];
const GLfloat p2 = ctx->Point.Params[2];
const GLfloat pointSize = ctx->Point.Size;
GLfloat (*size)[4] = store->PointSize.data;
GLuint i;
for (i = 0; i < VB->Count; i++) {
const GLfloat dist = FABSF(*eyeCoord);
const GLfloat q = p0 + dist * (p1 + dist * p2);
const GLfloat atten = (q != 0.0F) ? INV_SQRTF(q) : 1.0F;
size[i][0] = pointSize * atten; /* clamping done in rasterization */
eyeCoord += eyeCoordStride;
}
VB->AttribPtr[_TNL_ATTRIB_POINTSIZE] = &store->PointSize;
}
return GL_TRUE;
}
/**
* Compute per-vertex fog blend factors from fog coordinates by
* evaluating the GL_LINEAR, GL_EXP or GL_EXP2 fog function.
* Fog coordinates are distances from the eye (typically between the
* near and far clip plane distances).
* Note the fog (eye Z) coords may be negative so we use ABS(z) below.
* Fog blend factors are in the range [0,1].
*/
float
radeonComputeFogBlendFactor( struct gl_context *ctx, GLfloat fogcoord )
{
GLfloat end = ctx->Fog.End;
GLfloat d, temp;
const GLfloat z = FABSF(fogcoord);
switch (ctx->Fog.Mode) {
case GL_LINEAR:
if (ctx->Fog.Start == ctx->Fog.End)
d = 1.0F;
else
d = 1.0F / (ctx->Fog.End - ctx->Fog.Start);
temp = (end - z) * d;
return CLAMP(temp, 0.0F, 1.0F);
break;
case GL_EXP:
d = ctx->Fog.Density;
NEG_EXP( temp, d * z );
return temp;
break;
case GL_EXP2:
d = ctx->Fog.Density*ctx->Fog.Density;
NEG_EXP( temp, d * z * z );
return temp;
break;
default:
_mesa_problem(ctx, "Bad fog mode in make_fog_coord");
return 0;
}
}
/**
* Multiply a matrix with a general scaling matrix.
*
* \param mat matrix.
* \param x x axis scale factor.
* \param y y axis scale factor.
* \param z z axis scale factor.
*
* Multiplies in-place the elements of \p mat by the scale factors. Checks if
* the scales factors are roughly the same, marking the MAT_FLAG_UNIFORM_SCALE
* flag, or MAT_FLAG_GENERAL_SCALE. Marks the MAT_DIRTY_TYPE and
* MAT_DIRTY_INVERSE dirty flags.
*/
void
_math_matrix_scale( GLmatrix *mat, GLfloat x, GLfloat y, GLfloat z )
{
GLfloat *m = mat->m;
m[0] *= x; m[4] *= y; m[8] *= z;
m[1] *= x; m[5] *= y; m[9] *= z;
m[2] *= x; m[6] *= y; m[10] *= z;
m[3] *= x; m[7] *= y; m[11] *= z;
if (FABSF(x - y) < 1e-8 && FABSF(x - z) < 1e-8)
mat->flags |= MAT_FLAG_UNIFORM_SCALE;
else
mat->flags |= MAT_FLAG_GENERAL_SCALE;
mat->flags |= (MAT_DIRTY_TYPE |
MAT_DIRTY_INVERSE);
}
/**
* Compute inverse of a general 3d transformation matrix.
*
* \param mat pointer to a GLmatrix structure. The matrix inverse will be
* stored in the GLmatrix::inv attribute.
*
* \return GL_TRUE for success, GL_FALSE for failure (\p singular matrix).
*
* \author Adapted from graphics gems II.
*
* Calculates the inverse of the upper left by first calculating its
* determinant and multiplying it to the symmetric adjust matrix of each
* element. Finally deals with the translation part by transforming the
* original translation vector using by the calculated submatrix inverse.
*/
static GLboolean invert_matrix_3d_general( GLmatrix *mat )
{
const GLfloat *in = mat->m;
GLfloat *out = mat->inv;
GLfloat pos, neg, t;
GLfloat det;
/* Calculate the determinant of upper left 3x3 submatrix and
* determine if the matrix is singular.
*/
pos = neg = 0.0;
t = MAT(in,0,0) * MAT(in,1,1) * MAT(in,2,2);
if (t >= 0.0) pos += t; else neg += t;
t = MAT(in,1,0) * MAT(in,2,1) * MAT(in,0,2);
if (t >= 0.0) pos += t; else neg += t;
t = MAT(in,2,0) * MAT(in,0,1) * MAT(in,1,2);
if (t >= 0.0) pos += t; else neg += t;
t = -MAT(in,2,0) * MAT(in,1,1) * MAT(in,0,2);
if (t >= 0.0) pos += t; else neg += t;
t = -MAT(in,1,0) * MAT(in,0,1) * MAT(in,2,2);
if (t >= 0.0) pos += t; else neg += t;
t = -MAT(in,0,0) * MAT(in,2,1) * MAT(in,1,2);
if (t >= 0.0) pos += t; else neg += t;
det = pos + neg;
if (FABSF(det) < 1e-25)
return GL_FALSE;
det = 1.0F / det;
MAT(out,0,0) = ( (MAT(in,1,1)*MAT(in,2,2) - MAT(in,2,1)*MAT(in,1,2) )*det);
MAT(out,0,1) = (- (MAT(in,0,1)*MAT(in,2,2) - MAT(in,2,1)*MAT(in,0,2) )*det);
MAT(out,0,2) = ( (MAT(in,0,1)*MAT(in,1,2) - MAT(in,1,1)*MAT(in,0,2) )*det);
MAT(out,1,0) = (- (MAT(in,1,0)*MAT(in,2,2) - MAT(in,2,0)*MAT(in,1,2) )*det);
MAT(out,1,1) = ( (MAT(in,0,0)*MAT(in,2,2) - MAT(in,2,0)*MAT(in,0,2) )*det);
MAT(out,1,2) = (- (MAT(in,0,0)*MAT(in,1,2) - MAT(in,1,0)*MAT(in,0,2) )*det);
MAT(out,2,0) = ( (MAT(in,1,0)*MAT(in,2,1) - MAT(in,2,0)*MAT(in,1,1) )*det);
MAT(out,2,1) = (- (MAT(in,0,0)*MAT(in,2,1) - MAT(in,2,0)*MAT(in,0,1) )*det);
MAT(out,2,2) = ( (MAT(in,0,0)*MAT(in,1,1) - MAT(in,1,0)*MAT(in,0,1) )*det);
/* Do the translation part */
MAT(out,0,3) = - (MAT(in,0,3) * MAT(out,0,0) +
MAT(in,1,3) * MAT(out,0,1) +
MAT(in,2,3) * MAT(out,0,2) );
MAT(out,1,3) = - (MAT(in,0,3) * MAT(out,1,0) +
MAT(in,1,3) * MAT(out,1,1) +
MAT(in,2,3) * MAT(out,1,2) );
MAT(out,2,3) = - (MAT(in,0,3) * MAT(out,2,0) +
MAT(in,1,3) * MAT(out,2,1) +
MAT(in,2,3) * MAT(out,2,2) );
return GL_TRUE;
}
/**
* Compute per-vertex fog blend factors from fog coordinates by
* evaluating the GL_LINEAR, GL_EXP or GL_EXP2 fog function.
* Fog coordinates are distances from the eye (typically between the
* near and far clip plane distances).
* Note the fog (eye Z) coords may be negative so we use ABS(z) below.
* Fog blend factors are in the range [0,1].
*/
static void
compute_fog_blend_factors(GLcontext *ctx, GLvector4f *out, const GLvector4f *in)
{
GLfloat end = ctx->Fog.End;
GLfloat *v = in->start;
GLuint stride = in->stride;
GLuint n = in->count;
GLfloat (*data)[4] = out->data;
GLfloat d;
GLuint i;
out->count = in->count;
switch (ctx->Fog.Mode) {
case GL_LINEAR:
if (ctx->Fog.Start == ctx->Fog.End)
d = 1.0F;
else
d = 1.0F / (ctx->Fog.End - ctx->Fog.Start);
for ( i = 0 ; i < n ; i++, STRIDE_F(v, stride)) {
const GLfloat z = FABSF(*v);
GLfloat f = (end - z) * d;
data[i][0] = CLAMP(f, 0.0F, 1.0F);
}
break;
case GL_EXP:
d = ctx->Fog.Density;
for ( i = 0 ; i < n ; i++, STRIDE_F(v,stride)) {
const GLfloat z = FABSF(*v);
NEG_EXP( data[i][0], d * z );
}
break;
case GL_EXP2:
d = ctx->Fog.Density*ctx->Fog.Density;
for ( i = 0 ; i < n ; i++, STRIDE_F(v, stride)) {
const GLfloat z = FABSF(*v);
NEG_EXP( data[i][0], d * z * z );
}
break;
default:
_mesa_problem(ctx, "Bad fog mode in make_fog_coord");
return;
}
}
GLfloat
_swrast_compute_lambda(GLfloat dsdx, GLfloat dsdy, GLfloat dtdx, GLfloat dtdy,
GLfloat dqdx, GLfloat dqdy, GLfloat texW, GLfloat texH,
GLfloat s, GLfloat t, GLfloat q, GLfloat invQ)
{
GLfloat dsdx2 = (s + dsdx) / (q + dqdx) - s * invQ;
GLfloat dtdx2 = (t + dtdx) / (q + dqdx) - t * invQ;
GLfloat dsdy2 = (s + dsdy) / (q + dqdy) - s * invQ;
GLfloat dtdy2 = (t + dtdy) / (q + dqdy) - t * invQ;
GLfloat maxU, maxV, rho, lambda;
dsdx2 = FABSF(dsdx2);
dsdy2 = FABSF(dsdy2);
dtdx2 = FABSF(dtdx2);
dtdy2 = FABSF(dtdy2);
maxU = MAX2(dsdx2, dsdy2) * texW;
maxV = MAX2(dtdx2, dtdy2) * texH;
rho = MAX2(maxU, maxV);
lambda = LOG2(rho);
return lambda;
}
DLLEXPORT float rootnf(float x, int y)
{
int y_odd = (y&1);
int y_even = !y_odd;
int y_neg = (y<0);
int x_neg = (x<0);
if (y == 0) return NANF;
if (EXPF(x) == 0) /* Flush denormalized input to zero */
return (y > 0) ? 0.0f : (y_even ? INFF : copysign(INFF, x));
if (y == 1) return x;
if (ISNEGINFF(x) && y_even) return NANF;
if (ISINFF(x)) return y_neg ? copysign(0.0f, x) : x;
if (ISANYZEROF(x) && y_odd && y_neg) return copysign(INFF, x);
int negresult = 0;
if (x_neg && y_odd) { negresult = 1; x = FABSF(x); }
float result = powf(x, reciprocalf(y));
if (negresult) result = -result;
return result;
}
ENTRYPOINT void
draw_braid(ModeInfo * mi)
{
Display *display = MI_DISPLAY(mi);
Window window = MI_WINDOW(mi);
int num_points = 500;
float t_inc;
float theta, psi;
float t, r_diff;
int i, s;
float x_1, y_1, x_2, y_2, r1, r2;
float color, color_use = 0.0, color_inc;
braidtype *braid;
if (braids == NULL)
return;
braid = &braids[MI_SCREEN(mi)];
#ifdef STANDALONE
if (braid->eraser) {
braid->eraser = erase_window (MI_DISPLAY(mi), MI_WINDOW(mi), braid->eraser);
return;
}
#endif
MI_IS_DRAWN(mi) = True;
XSetLineAttributes(display, MI_GC(mi), braid->linewidth,
LineSolid,
(braid->linewidth <= 3 ? CapButt : CapRound),
JoinMiter);
theta = (2.0 * M_PI) / (float) (braid->braidlength);
t_inc = (2.0 * M_PI) / (float) num_points;
color_inc = (float) MI_NPIXELS(mi) * braid->color_direction /
(float) num_points;
braid->startcolor += SPINRATE * color_inc;
if (((int) braid->startcolor) >= MI_NPIXELS(mi))
braid->startcolor = 0.0;
r_diff = (braid->max_radius - braid->min_radius) / (float) (braid->nstrands);
color = braid->startcolor;
psi = 0.0;
for (i = 0; i < braid->braidlength; i++) {
psi += theta;
for (t = 0.0; t < theta; t += t_inc) {
#ifdef COLORROUND
color += color_inc;
if (((int) color) >= MI_NPIXELS(mi))
color = 0.0;
color_use = color;
#endif
for (s = 0; s < braid->nstrands; s++) {
if (ABS(braid->braidword[i]) == s)
continue;
if (ABS(braid->braidword[i]) - 1 == s) {
/* crosSINFg */
#ifdef COLORCOMP
if (MI_NPIXELS(mi) > 2) {
color_use = color + SPINRATE *
braid->components[applywordbackto(braid, s, i)] +
(psi + t) / 2.0 / M_PI * (float) MI_NPIXELS(mi);
while (((int) color_use) >= MI_NPIXELS(mi))
color_use -= (float) MI_NPIXELS(mi);
while (((int) color_use) < 0)
color_use += (float) MI_NPIXELS(mi);
}
#endif
#ifdef COLORROUND
if (MI_NPIXELS(mi) > 2) {
color_use += SPINRATE * color_inc;
while (((int) color_use) >= MI_NPIXELS(mi))
color_use -= (float) MI_NPIXELS(mi);
}
#endif
r1 = braid->min_radius + r_diff * (float) (s);
r2 = braid->min_radius + r_diff * (float) (s + 1);
if (braid->braidword[i] > 0 ||
(FABSF(t - theta / 2.0) > theta / 7.0)) {
x_1 = ((0.5 * (1.0 + SINF(t / theta * M_PI - M_PI_2)) * r2 +
0.5 * (1.0 + SINF((theta - t) / theta * M_PI - M_PI_2)) * r1)) *
COSF(t + psi) + braid->center_x;
y_1 = ((0.5 * (1.0 + SINF(t / theta * M_PI - M_PI_2)) * r2 +
0.5 * (1.0 + SINF((theta - t) / theta * M_PI - M_PI_2)) * r1)) *
SINF(t + psi) + braid->center_y;
x_2 = ((0.5 * (1.0 + SINF((t + t_inc) / theta * M_PI - M_PI_2)) * r2 +
0.5 * (1.0 + SINF((theta - t - t_inc) / theta * M_PI - M_PI_2)) * r1)) *
COSF(t + t_inc + psi) + braid->center_x;
y_2 = ((0.5 * (1.0 + SINF((t + t_inc) / theta * M_PI - M_PI_2)) * r2 +
0.5 * (1.0 + SINF((theta - t - t_inc) / theta * M_PI - M_PI_2)) * r1)) *
SINF(t + t_inc + psi) + braid->center_y;
if (MI_NPIXELS(mi) > 2)
XSetForeground(display, MI_GC(mi), MI_PIXEL(mi, (int) color_use));
else
XSetForeground(display, MI_GC(mi), MI_WHITE_PIXEL(mi));
XDrawLine(display, window, MI_GC(mi),
(int) (x_1), (int) (y_1), (int) (x_2), (int) (y_2));
}
#ifdef COLORCOMP
if (MI_NPIXELS(mi) > 2) {
color_use = color + SPINRATE *
braid->components[applywordbackto(braid, s + 1, i)] +
(psi + t) / 2.0 / M_PI * (float) MI_NPIXELS(mi);
while (((int) color_use) >= MI_NPIXELS(mi))
color_use -= (float) MI_NPIXELS(mi);
while (((int) color_use) < 0)
color_use += (float) MI_NPIXELS(mi);
}
#endif
if (braid->braidword[i] < 0 ||
(FABSF(t - theta / 2.0) > theta / 7.0)) {
x_1 = ((0.5 * (1.0 + SINF(t / theta * M_PI - M_PI_2)) * r1 +
0.5 * (1.0 + SINF((theta - t) / theta * M_PI - M_PI_2)) * r2)) *
COSF(t + psi) + braid->center_x;
y_1 = ((0.5 * (1.0 + SINF(t / theta * M_PI - M_PI_2)) * r1 +
0.5 * (1.0 + SINF((theta - t) / theta * M_PI - M_PI_2)) * r2)) *
SINF(t + psi) + braid->center_y;
x_2 = ((0.5 * (1.0 + SINF((t + t_inc) / theta * M_PI - M_PI_2)) * r1 +
0.5 * (1.0 + SINF((theta - t - t_inc) / theta * M_PI - M_PI_2)) * r2)) *
COSF(t + t_inc + psi) + braid->center_x;
y_2 = ((0.5 * (1.0 + SINF((t + t_inc) / theta * M_PI - M_PI_2)) * r1 +
0.5 * (1.0 + SINF((theta - t - t_inc) / theta * M_PI - M_PI_2)) * r2)) *
SINF(t + t_inc + psi) + braid->center_y;
if (MI_NPIXELS(mi) > 2)
XSetForeground(display, MI_GC(mi), MI_PIXEL(mi, (int) color_use));
else
XSetForeground(display, MI_GC(mi), MI_WHITE_PIXEL(mi));
XDrawLine(display, window, MI_GC(mi),
(int) (x_1), (int) (y_1), (int) (x_2), (int) (y_2));
}
} else {
/* no crosSINFg */
#ifdef COLORCOMP
if (MI_NPIXELS(mi) > 2) {
color_use = color + SPINRATE *
braid->components[applywordbackto(braid, s, i)] +
(psi + t) / 2.0 / M_PI * (float) MI_NPIXELS(mi);
while (((int) color_use) >= MI_NPIXELS(mi))
color_use -= (float) MI_NPIXELS(mi);
while (((int) color_use) < 0)
color_use += (float) MI_NPIXELS(mi);
}
#endif
#ifdef COLORROUND
if (MI_NPIXELS(mi) > 2) {
color_use += SPINRATE * color_inc;
while (((int) color_use) >= MI_NPIXELS(mi))
color_use -= (float) MI_NPIXELS(mi);
}
#endif
r1 = braid->min_radius + r_diff * (float) (s);
x_1 = r1 * COSF(t + psi) + braid->center_x;
y_1 = r1 * SINF(t + psi) + braid->center_y;
x_2 = r1 * COSF(t + t_inc + psi) + braid->center_x;
y_2 = r1 * SINF(t + t_inc + psi) + braid->center_y;
if (MI_NPIXELS(mi) > 2)
XSetForeground(display, MI_GC(mi), MI_PIXEL(mi, (int) color_use));
else
XSetForeground(display, MI_GC(mi), MI_WHITE_PIXEL(mi));
XDrawLine(display, window, MI_GC(mi),
(int) (x_1), (int) (y_1), (int) (x_2), (int) (y_2));
}
}
}
}
XSetLineAttributes(display, MI_GC(mi), 1, LineSolid, CapNotLast, JoinRound);
if (++braid->age > MI_CYCLES(mi)) {
#ifdef STANDALONE
braid->eraser = erase_window (MI_DISPLAY(mi), MI_WINDOW(mi), braid->eraser);
#endif
init_braid(mi);
}
}
/** * Compute inverse of 4x4 transformation matrix. * * \param mat pointer to a GLmatrix structure. The matrix inverse will be * stored in the GLmatrix::inv attribute. * * \return GL_TRUE for success, GL_FALSE for failure (\p singular matrix). * * \author * Code contributed by Jacques Leroy [email protected] * * Calculates the inverse matrix by performing the gaussian matrix reduction * with partial pivoting followed by back/substitution with the loops manually * unrolled. */ static GLboolean invert_matrix_general( GLmatrix *mat ) { const GLfloat *m = mat->m; GLfloat *out = mat->inv; GLfloat wtmp[4][8]; GLfloat m0, m1, m2, m3, s; GLfloat *r0, *r1, *r2, *r3; r0 = wtmp[0], r1 = wtmp[1], r2 = wtmp[2], r3 = wtmp[3]; r0[0] = MAT(m,0,0), r0[1] = MAT(m,0,1), r0[2] = MAT(m,0,2), r0[3] = MAT(m,0,3), r0[4] = 1.0, r0[5] = r0[6] = r0[7] = 0.0, r1[0] = MAT(m,1,0), r1[1] = MAT(m,1,1), r1[2] = MAT(m,1,2), r1[3] = MAT(m,1,3), r1[5] = 1.0, r1[4] = r1[6] = r1[7] = 0.0, r2[0] = MAT(m,2,0), r2[1] = MAT(m,2,1), r2[2] = MAT(m,2,2), r2[3] = MAT(m,2,3), r2[6] = 1.0, r2[4] = r2[5] = r2[7] = 0.0, r3[0] = MAT(m,3,0), r3[1] = MAT(m,3,1), r3[2] = MAT(m,3,2), r3[3] = MAT(m,3,3), r3[7] = 1.0, r3[4] = r3[5] = r3[6] = 0.0; /* choose pivot - or die */ if (FABSF(r3[0])>FABSF(r2[0])) SWAP_ROWS(r3, r2); if (FABSF(r2[0])>FABSF(r1[0])) SWAP_ROWS(r2, r1); if (FABSF(r1[0])>FABSF(r0[0])) SWAP_ROWS(r1, r0); if (0.0 == r0[0]) return GL_FALSE; /* eliminate first variable */ m1 = r1[0]/r0[0]; m2 = r2[0]/r0[0]; m3 = r3[0]/r0[0]; s = r0[1]; r1[1] -= m1 * s; r2[1] -= m2 * s; r3[1] -= m3 * s; s = r0[2]; r1[2] -= m1 * s; r2[2] -= m2 * s; r3[2] -= m3 * s; s = r0[3]; r1[3] -= m1 * s; r2[3] -= m2 * s; r3[3] -= m3 * s; s = r0[4]; if (s != 0.0) { r1[4] -= m1 * s; r2[4] -= m2 * s; r3[4] -= m3 * s; } s = r0[5]; if (s != 0.0) { r1[5] -= m1 * s; r2[5] -= m2 * s; r3[5] -= m3 * s; } s = r0[6]; if (s != 0.0) { r1[6] -= m1 * s; r2[6] -= m2 * s; r3[6] -= m3 * s; } s = r0[7]; if (s != 0.0) { r1[7] -= m1 * s; r2[7] -= m2 * s; r3[7] -= m3 * s; } /* choose pivot - or die */ if (FABSF(r3[1])>FABSF(r2[1])) SWAP_ROWS(r3, r2); if (FABSF(r2[1])>FABSF(r1[1])) SWAP_ROWS(r2, r1); if (0.0 == r1[1]) return GL_FALSE; /* eliminate second variable */ m2 = r2[1]/r1[1]; m3 = r3[1]/r1[1]; r2[2] -= m2 * r1[2]; r3[2] -= m3 * r1[2]; r2[3] -= m2 * r1[3]; r3[3] -= m3 * r1[3]; s = r1[4]; if (0.0 != s) { r2[4] -= m2 * s; r3[4] -= m3 * s; } s = r1[5]; if (0.0 != s) { r2[5] -= m2 * s; r3[5] -= m3 * s; } s = r1[6]; if (0.0 != s) { r2[6] -= m2 * s; r3[6] -= m3 * s; } s = r1[7]; if (0.0 != s) { r2[7] -= m2 * s; r3[7] -= m3 * s; } /* choose pivot - or die */ if (FABSF(r3[2])>FABSF(r2[2])) SWAP_ROWS(r3, r2); if (0.0 == r2[2]) return GL_FALSE; /* eliminate third variable */ m3 = r3[2]/r2[2]; r3[3] -= m3 * r2[3], r3[4] -= m3 * r2[4], r3[5] -= m3 * r2[5], r3[6] -= m3 * r2[6], r3[7] -= m3 * r2[7]; /* last check */ if (0.0 == r3[3]) return GL_FALSE; s = 1.0F/r3[3]; /* now back substitute row 3 */ r3[4] *= s; r3[5] *= s; r3[6] *= s; r3[7] *= s; m2 = r2[3]; /* now back substitute row 2 */ s = 1.0F/r2[2]; r2[4] = s * (r2[4] - r3[4] * m2), r2[5] = s * (r2[5] - r3[5] * m2), r2[6] = s * (r2[6] - r3[6] * m2), r2[7] = s * (r2[7] - r3[7] * m2); m1 = r1[3]; r1[4] -= r3[4] * m1, r1[5] -= r3[5] * m1, r1[6] -= r3[6] * m1, r1[7] -= r3[7] * m1; m0 = r0[3]; r0[4] -= r3[4] * m0, r0[5] -= r3[5] * m0, r0[6] -= r3[6] * m0, r0[7] -= r3[7] * m0; m1 = r1[2]; /* now back substitute row 1 */ s = 1.0F/r1[1]; r1[4] = s * (r1[4] - r2[4] * m1), r1[5] = s * (r1[5] - r2[5] * m1), r1[6] = s * (r1[6] - r2[6] * m1), r1[7] = s * (r1[7] - r2[7] * m1); m0 = r0[2]; r0[4] -= r2[4] * m0, r0[5] -= r2[5] * m0, r0[6] -= r2[6] * m0, r0[7] -= r2[7] * m0; m0 = r0[1]; /* now back substitute row 0 */ s = 1.0F/r0[0]; r0[4] = s * (r0[4] - r1[4] * m0), r0[5] = s * (r0[5] - r1[5] * m0), r0[6] = s * (r0[6] - r1[6] * m0), r0[7] = s * (r0[7] - r1[7] * m0); MAT(out,0,0) = r0[4]; MAT(out,0,1) = r0[5], MAT(out,0,2) = r0[6]; MAT(out,0,3) = r0[7], MAT(out,1,0) = r1[4]; MAT(out,1,1) = r1[5], MAT(out,1,2) = r1[6]; MAT(out,1,3) = r1[7], MAT(out,2,0) = r2[4]; MAT(out,2,1) = r2[5], MAT(out,2,2) = r2[6]; MAT(out,2,3) = r2[7], MAT(out,3,0) = r3[4]; MAT(out,3,1) = r3[5], MAT(out,3,2) = r3[6]; MAT(out,3,3) = r3[7]; return GL_TRUE; }
/*
* As above, but write stencil values.
*/
void
_swrast_write_zoomed_stencil_span( GLcontext *ctx,
GLuint n, GLint x, GLint y,
const GLstencil stencil[], GLint y0,
GLint skipPixels )
{
GLint m;
GLint r0, r1, row, r;
GLint i, j, skipcol;
GLstencil zstencil[MAX_WIDTH]; /* zoomed stencil values */
GLint maxwidth = MIN2( ctx->DrawBuffer->Width, MAX_WIDTH );
(void) skipPixels; /* XXX this shouldn't be ignored */
/* compute width of output row */
m = (GLint) FABSF( n * ctx->Pixel.ZoomX );
if (m==0) {
return;
}
if (ctx->Pixel.ZoomX<0.0) {
/* adjust x coordinate for left/right mirroring */
x = x - m;
}
/* compute which rows to draw */
row = y - y0;
r0 = y0 + (GLint) (row * ctx->Pixel.ZoomY);
r1 = y0 + (GLint) ((row+1) * ctx->Pixel.ZoomY);
if (r0==r1) {
return;
}
else if (r1<r0) {
GLint rtmp = r1;
r1 = r0;
r0 = rtmp;
}
/* return early if r0...r1 is above or below window */
if (r0<0 && r1<0) {
/* below window */
return;
}
if (r0 >= (GLint) ctx->DrawBuffer->Height &&
r1 >= (GLint) ctx->DrawBuffer->Height) {
/* above window */
return;
}
/* check if left edge is outside window */
skipcol = 0;
if (x<0) {
skipcol = -x;
m += x;
}
/* make sure span isn't too long or short */
if (m>maxwidth) {
m = maxwidth;
}
else if (m<=0) {
return;
}
ASSERT( m <= MAX_WIDTH );
/* zoom the span horizontally */
if (ctx->Pixel.ZoomX==-1.0F) {
/* n==m */
for (j=0;j<m;j++) {
i = n - (j+skipcol) - 1;
zstencil[j] = stencil[i];
}
}
else {
GLfloat xscale = 1.0F / ctx->Pixel.ZoomX;
for (j=0;j<m;j++) {
i = (GLint) ((j+skipcol) * xscale);
if (i<0) i = n + i - 1;
zstencil[j] = stencil[i];
}
}
/* write the span */
for (r=r0; r<r1; r++) {
_swrast_write_stencil_span( ctx, m, x+skipcol, r, zstencil );
}
}
static GLboolean
run_fog_stage(struct gl_context *ctx, struct tnl_pipeline_stage *stage)
{
TNLcontext *tnl = TNL_CONTEXT(ctx);
struct vertex_buffer *VB = &tnl->vb;
struct fog_stage_data *store = FOG_STAGE_DATA(stage);
GLvector4f *input;
if (!ctx->Fog.Enabled)
return GL_TRUE;
if (ctx->Fog.FogCoordinateSource == GL_FRAGMENT_DEPTH_EXT && !ctx->VertexProgram._Current) {
GLuint i;
GLfloat *coord;
/* Fog is computed from vertex or fragment Z values */
/* source = VB->AttribPtr[_TNL_ATTRIB_POS] or VB->EyePtr coords */
/* dest = VB->AttribPtr[_TNL_ATTRIB_FOG] = fog stage private storage */
VB->AttribPtr[_TNL_ATTRIB_FOG] = &store->fogcoord;
if (!ctx->_NeedEyeCoords) {
/* compute fog coords from object coords */
const GLfloat *m = ctx->ModelviewMatrixStack.Top->m;
GLfloat plane[4];
/* Use this to store calculated eye z values:
*/
input = &store->fogcoord;
plane[0] = m[2];
plane[1] = m[6];
plane[2] = m[10];
plane[3] = m[14];
/* Full eye coords weren't required, just calculate the
* eye Z values.
*/
_mesa_dotprod_tab[VB->AttribPtr[_TNL_ATTRIB_POS]->size]
( (GLfloat *) input->data,
4 * sizeof(GLfloat),
VB->AttribPtr[_TNL_ATTRIB_POS], plane );
input->count = VB->AttribPtr[_TNL_ATTRIB_POS]->count;
/* make sure coords are really positive
NOTE should avoid going through array twice */
coord = input->start;
for (i = 0; i < input->count; i++) {
*coord = FABSF(*coord);
STRIDE_F(coord, input->stride);
}
}
else {
/* fog coordinates = eye Z coordinates - need to copy for ABS */
input = &store->fogcoord;
if (VB->EyePtr->size < 2)
_mesa_vector4f_clean_elem( VB->EyePtr, VB->Count, 2 );
input->stride = 4 * sizeof(GLfloat);
input->count = VB->EyePtr->count;
coord = VB->EyePtr->start;
for (i = 0 ; i < VB->EyePtr->count; i++) {
input->data[i][0] = FABSF(coord[2]);
STRIDE_F(coord, VB->EyePtr->stride);
}
}
}
else {
/* use glFogCoord() coordinates */
input = VB->AttribPtr[_TNL_ATTRIB_FOG]; /* source data */
/* input->count may be one if glFogCoord was only called once
* before glBegin. But we need to compute fog for all vertices.
*/
input->count = VB->AttribPtr[_TNL_ATTRIB_POS]->count;
VB->AttribPtr[_TNL_ATTRIB_FOG] = &store->fogcoord; /* dest data */
}
if (tnl->_DoVertexFog) {
/* compute blend factors from fog coordinates */
compute_fog_blend_factors( ctx, VB->AttribPtr[_TNL_ATTRIB_FOG], input );
}
else {
/* results = incoming fog coords (compute fog per-fragment later) */
VB->AttribPtr[_TNL_ATTRIB_FOG] = input;
}
return GL_TRUE;
}
/**
* Execute the given vertex program
*/
void
_mesa_exec_vertex_program(GLcontext *ctx, const struct vertex_program *program)
{
struct vertex_program_state *state = &ctx->VertexProgram;
const struct vp_instruction *inst;
ctx->_CurrentProgram = GL_VERTEX_PROGRAM_ARB; /* or NV, doesn't matter */
/* If the program is position invariant, multiply the input
* position and the MVP matrix and stick it into the output pos slot
*/
if (ctx->VertexProgram.Current->IsPositionInvariant) {
TRANSFORM_POINT( ctx->VertexProgram.Outputs[0],
ctx->_ModelProjectMatrix.m,
ctx->VertexProgram.Inputs[0]);
/* XXX: This could go elsewhere */
ctx->VertexProgram.Current->OutputsWritten |= 0x1;
}
for (inst = program->Instructions; /*inst->Opcode != VP_OPCODE_END*/; inst++) {
if (ctx->VertexProgram.CallbackEnabled &&
ctx->VertexProgram.Callback) {
ctx->VertexProgram.CurrentPosition = inst->StringPos;
ctx->VertexProgram.Callback(program->Base.Target,
ctx->VertexProgram.CallbackData);
}
switch (inst->Opcode) {
case VP_OPCODE_MOV:
{
GLfloat t[4];
fetch_vector4( &inst->SrcReg[0], state, t );
store_vector4( &inst->DstReg, state, t );
}
break;
case VP_OPCODE_LIT:
{
const GLfloat epsilon = 1.0e-5F; /* XXX fix? */
GLfloat t[4], lit[4];
fetch_vector4( &inst->SrcReg[0], state, t );
if (t[3] < -(128.0F - epsilon))
t[3] = - (128.0F - epsilon);
else if (t[3] > 128.0F - epsilon)
t[3] = 128.0F - epsilon;
if (t[0] < 0.0)
t[0] = 0.0;
if (t[1] < 0.0)
t[1] = 0.0;
lit[0] = 1.0;
lit[1] = t[0];
lit[2] = (t[0] > 0.0) ? (GLfloat) exp(t[3] * log(t[1])) : 0.0F;
lit[3] = 1.0;
store_vector4( &inst->DstReg, state, lit );
}
break;
case VP_OPCODE_RCP:
{
GLfloat t[4];
fetch_vector1( &inst->SrcReg[0], state, t );
if (t[0] != 1.0F)
t[0] = 1.0F / t[0]; /* div by zero is infinity! */
t[1] = t[2] = t[3] = t[0];
store_vector4( &inst->DstReg, state, t );
}
break;
case VP_OPCODE_RSQ:
{
GLfloat t[4];
fetch_vector1( &inst->SrcReg[0], state, t );
t[0] = INV_SQRTF(FABSF(t[0]));
t[1] = t[2] = t[3] = t[0];
store_vector4( &inst->DstReg, state, t );
}
break;
case VP_OPCODE_EXP:
{
GLfloat t[4], q[4], floor_t0;
fetch_vector1( &inst->SrcReg[0], state, t );
floor_t0 = (float) floor(t[0]);
if (floor_t0 > FLT_MAX_EXP) {
SET_POS_INFINITY(q[0]);
SET_POS_INFINITY(q[2]);
}
else if (floor_t0 < FLT_MIN_EXP) {
q[0] = 0.0F;
q[2] = 0.0F;
}
else {
#ifdef USE_IEEE
GLint ii = (GLint) floor_t0;
ii = (ii < 23) + 0x3f800000;
SET_FLOAT_BITS(q[0], ii);
q[0] = *((GLfloat *) &ii);
#else
q[0] = (GLfloat) pow(2.0, floor_t0);
#endif
q[2] = (GLfloat) (q[0] * LOG2(q[1]));
}
q[1] = t[0] - floor_t0;
q[3] = 1.0F;
store_vector4( &inst->DstReg, state, q );
}
break;
case VP_OPCODE_LOG:
{
GLfloat t[4], q[4], abs_t0;
fetch_vector1( &inst->SrcReg[0], state, t );
abs_t0 = (GLfloat) fabs(t[0]);
if (abs_t0 != 0.0F) {
/* Since we really can't handle infinite values on VMS
* like other OSes we'll use __MAXFLOAT to represent
* infinity. This may need some tweaking.
*/
#ifdef VMS
if (abs_t0 == __MAXFLOAT)
#else
if (IS_INF_OR_NAN(abs_t0))
#endif
{
SET_POS_INFINITY(q[0]);
q[1] = 1.0F;
SET_POS_INFINITY(q[2]);
}
else {
int exponent;
double mantissa = frexp(t[0], &exponent);
q[0] = (GLfloat) (exponent - 1);
q[1] = (GLfloat) (2.0 * mantissa); /* map [.5, 1) -> [1, 2) */
q[2] = (GLfloat) (q[0] + LOG2(q[1]));
}
}
else {
SET_NEG_INFINITY(q[0]);
q[1] = 1.0F;
SET_NEG_INFINITY(q[2]);
}
q[3] = 1.0;
store_vector4( &inst->DstReg, state, q );
}
break;
case VP_OPCODE_MUL:
{
GLfloat t[4], u[4], prod[4];
fetch_vector4( &inst->SrcReg[0], state, t );
fetch_vector4( &inst->SrcReg[1], state, u );
prod[0] = t[0] * u[0];
prod[1] = t[1] * u[1];
prod[2] = t[2] * u[2];
prod[3] = t[3] * u[3];
store_vector4( &inst->DstReg, state, prod );
}
break;
case VP_OPCODE_ADD:
{
GLfloat t[4], u[4], sum[4];
fetch_vector4( &inst->SrcReg[0], state, t );
fetch_vector4( &inst->SrcReg[1], state, u );
sum[0] = t[0] + u[0];
sum[1] = t[1] + u[1];
sum[2] = t[2] + u[2];
sum[3] = t[3] + u[3];
store_vector4( &inst->DstReg, state, sum );
}
break;
case VP_OPCODE_DP3:
{
GLfloat t[4], u[4], dot[4];
fetch_vector4( &inst->SrcReg[0], state, t );
fetch_vector4( &inst->SrcReg[1], state, u );
dot[0] = t[0] * u[0] + t[1] * u[1] + t[2] * u[2];
dot[1] = dot[2] = dot[3] = dot[0];
store_vector4( &inst->DstReg, state, dot );
}
break;
case VP_OPCODE_DP4:
{
GLfloat t[4], u[4], dot[4];
fetch_vector4( &inst->SrcReg[0], state, t );
fetch_vector4( &inst->SrcReg[1], state, u );
dot[0] = t[0] * u[0] + t[1] * u[1] + t[2] * u[2] + t[3] * u[3];
dot[1] = dot[2] = dot[3] = dot[0];
store_vector4( &inst->DstReg, state, dot );
}
break;
case VP_OPCODE_DST:
{
GLfloat t[4], u[4], dst[4];
fetch_vector4( &inst->SrcReg[0], state, t );
fetch_vector4( &inst->SrcReg[1], state, u );
dst[0] = 1.0F;
dst[1] = t[1] * u[1];
dst[2] = t[2];
dst[3] = u[3];
store_vector4( &inst->DstReg, state, dst );
}
break;
case VP_OPCODE_MIN:
{
GLfloat t[4], u[4], min[4];
fetch_vector4( &inst->SrcReg[0], state, t );
fetch_vector4( &inst->SrcReg[1], state, u );
min[0] = (t[0] < u[0]) ? t[0] : u[0];
min[1] = (t[1] < u[1]) ? t[1] : u[1];
min[2] = (t[2] < u[2]) ? t[2] : u[2];
min[3] = (t[3] < u[3]) ? t[3] : u[3];
store_vector4( &inst->DstReg, state, min );
}
break;
case VP_OPCODE_MAX:
{
GLfloat t[4], u[4], max[4];
fetch_vector4( &inst->SrcReg[0], state, t );
fetch_vector4( &inst->SrcReg[1], state, u );
max[0] = (t[0] > u[0]) ? t[0] : u[0];
max[1] = (t[1] > u[1]) ? t[1] : u[1];
max[2] = (t[2] > u[2]) ? t[2] : u[2];
max[3] = (t[3] > u[3]) ? t[3] : u[3];
store_vector4( &inst->DstReg, state, max );
}
break;
case VP_OPCODE_SLT:
{
GLfloat t[4], u[4], slt[4];
fetch_vector4( &inst->SrcReg[0], state, t );
fetch_vector4( &inst->SrcReg[1], state, u );
slt[0] = (t[0] < u[0]) ? 1.0F : 0.0F;
slt[1] = (t[1] < u[1]) ? 1.0F : 0.0F;
slt[2] = (t[2] < u[2]) ? 1.0F : 0.0F;
slt[3] = (t[3] < u[3]) ? 1.0F : 0.0F;
store_vector4( &inst->DstReg, state, slt );
}
break;
case VP_OPCODE_SGE:
{
GLfloat t[4], u[4], sge[4];
fetch_vector4( &inst->SrcReg[0], state, t );
fetch_vector4( &inst->SrcReg[1], state, u );
sge[0] = (t[0] >= u[0]) ? 1.0F : 0.0F;
sge[1] = (t[1] >= u[1]) ? 1.0F : 0.0F;
sge[2] = (t[2] >= u[2]) ? 1.0F : 0.0F;
sge[3] = (t[3] >= u[3]) ? 1.0F : 0.0F;
store_vector4( &inst->DstReg, state, sge );
}
break;
case VP_OPCODE_MAD:
{
GLfloat t[4], u[4], v[4], sum[4];
fetch_vector4( &inst->SrcReg[0], state, t );
fetch_vector4( &inst->SrcReg[1], state, u );
fetch_vector4( &inst->SrcReg[2], state, v );
sum[0] = t[0] * u[0] + v[0];
sum[1] = t[1] * u[1] + v[1];
sum[2] = t[2] * u[2] + v[2];
sum[3] = t[3] * u[3] + v[3];
store_vector4( &inst->DstReg, state, sum );
}
break;
case VP_OPCODE_ARL:
{
GLfloat t[4];
fetch_vector4( &inst->SrcReg[0], state, t );
state->AddressReg[0] = (GLint) floor(t[0]);
}
break;
case VP_OPCODE_DPH:
{
GLfloat t[4], u[4], dot[4];
fetch_vector4( &inst->SrcReg[0], state, t );
fetch_vector4( &inst->SrcReg[1], state, u );
dot[0] = t[0] * u[0] + t[1] * u[1] + t[2] * u[2] + u[3];
dot[1] = dot[2] = dot[3] = dot[0];
store_vector4( &inst->DstReg, state, dot );
}
break;
case VP_OPCODE_RCC:
{
GLfloat t[4], u;
fetch_vector1( &inst->SrcReg[0], state, t );
if (t[0] == 1.0F)
u = 1.0F;
else
u = 1.0F / t[0];
if (u > 0.0F) {
if (u > 1.884467e+019F) {
u = 1.884467e+019F; /* IEEE 32-bit binary value 0x5F800000 */
}
else if (u < 5.42101e-020F) {
u = 5.42101e-020F; /* IEEE 32-bit binary value 0x1F800000 */
}
}
else {
if (u < -1.884467e+019F) {
u = -1.884467e+019F; /* IEEE 32-bit binary value 0xDF800000 */
}
else if (u > -5.42101e-020F) {
u = -5.42101e-020F; /* IEEE 32-bit binary value 0x9F800000 */
}
}
t[0] = t[1] = t[2] = t[3] = u;
store_vector4( &inst->DstReg, state, t );
}
break;
case VP_OPCODE_SUB: /* GL_NV_vertex_program1_1 */
{
GLfloat t[4], u[4], sum[4];
fetch_vector4( &inst->SrcReg[0], state, t );
fetch_vector4( &inst->SrcReg[1], state, u );
sum[0] = t[0] - u[0];
sum[1] = t[1] - u[1];
sum[2] = t[2] - u[2];
sum[3] = t[3] - u[3];
store_vector4( &inst->DstReg, state, sum );
}
break;
case VP_OPCODE_ABS: /* GL_NV_vertex_program1_1 */
{
GLfloat t[4];
fetch_vector4( &inst->SrcReg[0], state, t );
if (t[0] < 0.0) t[0] = -t[0];
if (t[1] < 0.0) t[1] = -t[1];
if (t[2] < 0.0) t[2] = -t[2];
if (t[3] < 0.0) t[3] = -t[3];
store_vector4( &inst->DstReg, state, t );
}
break;
case VP_OPCODE_FLR: /* GL_ARB_vertex_program */
{
GLfloat t[4];
fetch_vector4( &inst->SrcReg[0], state, t );
t[0] = FLOORF(t[0]);
t[1] = FLOORF(t[1]);
t[2] = FLOORF(t[2]);
t[3] = FLOORF(t[3]);
store_vector4( &inst->DstReg, state, t );
}
break;
case VP_OPCODE_FRC: /* GL_ARB_vertex_program */
{
GLfloat t[4];
fetch_vector4( &inst->SrcReg[0], state, t );
t[0] = t[0] - FLOORF(t[0]);
t[1] = t[1] - FLOORF(t[1]);
t[2] = t[2] - FLOORF(t[2]);
t[3] = t[3] - FLOORF(t[3]);
store_vector4( &inst->DstReg, state, t );
}
break;
case VP_OPCODE_EX2: /* GL_ARB_vertex_program */
{
GLfloat t[4];
fetch_vector1( &inst->SrcReg[0], state, t );
t[0] = t[1] = t[2] = t[3] = (GLfloat)_mesa_pow(2.0, t[0]);
store_vector4( &inst->DstReg, state, t );
}
break;
case VP_OPCODE_LG2: /* GL_ARB_vertex_program */
{
GLfloat t[4];
fetch_vector1( &inst->SrcReg[0], state, t );
t[0] = t[1] = t[2] = t[3] = LOG2(t[0]);
store_vector4( &inst->DstReg, state, t );
}
break;
case VP_OPCODE_POW: /* GL_ARB_vertex_program */
{
GLfloat t[4], u[4];
fetch_vector1( &inst->SrcReg[0], state, t );
fetch_vector1( &inst->SrcReg[1], state, u );
t[0] = t[1] = t[2] = t[3] = (GLfloat)_mesa_pow(t[0], u[0]);
store_vector4( &inst->DstReg, state, t );
}
break;
case VP_OPCODE_XPD: /* GL_ARB_vertex_program */
{
GLfloat t[4], u[4], cross[4];
fetch_vector4( &inst->SrcReg[0], state, t );
fetch_vector4( &inst->SrcReg[1], state, u );
cross[0] = t[1] * u[2] - t[2] * u[1];
cross[1] = t[2] * u[0] - t[0] * u[2];
cross[2] = t[0] * u[1] - t[1] * u[0];
store_vector4( &inst->DstReg, state, cross );
}
break;
case VP_OPCODE_SWZ: /* GL_ARB_vertex_program */
{
const struct vp_src_register *source = &inst->SrcReg[0];
const GLfloat *src = get_register_pointer(source, state);
GLfloat result[4];
GLuint i;
/* do extended swizzling here */
for (i = 0; i < 3; i++) {
if (source->Swizzle[i] == SWIZZLE_ZERO)
result[i] = 0.0;
else if (source->Swizzle[i] == SWIZZLE_ONE)
result[i] = -1.0;
else
result[i] = -src[source->Swizzle[i]];
if (source->Negate)
result[i] = -result[i];
}
store_vector4( &inst->DstReg, state, result );
}
break;
case VP_OPCODE_END:
ctx->_CurrentProgram = 0;
return;
default:
/* bad instruction opcode */
_mesa_problem(ctx, "Bad VP Opcode in _mesa_exec_vertex_program");
ctx->_CurrentProgram = 0;
return;
} /* switch */
} /* for */
ctx->_CurrentProgram = 0;
}