1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
|
/* $XFree86$ */
/**************************************************************************
Copyright 1999, 2000 ATI Technologies Inc. and Precision Insight, Inc.,
Cedar Park, Texas.
All Rights Reserved.
Permission is hereby granted, free of charge, to any person obtaining a
copy of this software and associated documentation files (the "Software"),
to deal in the Software without restriction, including without limitation
on the rights to use, copy, modify, merge, publish, distribute, sub
license, and/or sell copies of the Software, and to permit persons to whom
the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice (including the next
paragraph) shall be included in all copies or substantial portions of the
Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
ATI, PRECISION INSIGHT AND/OR THEIR SUPPLIERS BE LIABLE FOR ANY CLAIM,
DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
USE OR OTHER DEALINGS IN THE SOFTWARE.
**************************************************************************/
/*
* Authors:
* Kevin E. Martin <kevin@precisioninsight.com>
*
*/
#include "r128_init.h"
#include "r128_mesa.h"
#include "r128_state.h"
#include "r128_vb.h"
#include "r128_tris.h"
#include "r128_fastpath.h"
#include "mmath.h"
#include "cva.h"
#include "vertices.h"
/* FIXME: These routines were copied from the i810 driver, and were only
slightly modified for the Rage 128. They still need to be optmizied
and cleaned up. Also, support for USE_RHW2 needs to be added. */
typedef struct r128_fast_table {
r128BuildVerticesFunc build_vertices;
r128InterpFunc interp;
} r128FastPathTable;
#define POINT(x) r128DrawPoint(r128ctx, &vert[x].v, psize)
#define LINE(x,y) r128DrawLine(r128ctx, &vert[x].v, &vert[y].v, lwidth)
#define TRI(x,y,z) r128DrawTriangle(r128ctx, &vert[x].v, &vert[y].v, &vert[z].v)
/* Direct, and no clipping required. The clip funcs have not been
written yet, so this is only useful for the fast path. */
#define RENDER_POINTS(start, count) \
do { \
GLuint e; \
for (e = start; e <= count; e++) \
POINT(elt[e]); \
} while (0)
#define RENDER_LINE(i1, i) \
do { \
GLuint e1 = elt[i1], e = elt[i]; \
LINE(e1, e); \
} while (0)
#define RENDER_TRI(i2, i1, i, pv, parity) \
do { \
GLuint e2 = elt[i2], e1 = elt[i1], e = elt[i]; \
if (parity) { \
GLuint tmp = e2; \
e2 = e1; \
e1 = tmp; \
} \
TRI(e2, e1, e); \
} while (0)
#define RENDER_QUAD(i3, i2, i1, i, pv) \
do { \
GLuint e3 = elt[i3], e2 = elt[i2], e1 = elt[i1], e = elt[i]; \
TRI(e3, e2, e); \
TRI(e2, e1, e); \
} while (0)
#define LOCAL_VARS \
r128VertexPtr vert = R128_DRIVER_DATA(VB)->verts; \
const GLuint *elt = VB->EltPtr->data; \
GLcontext *ctx = VB->ctx; \
r128ContextPtr r128ctx = R128_CONTEXT(ctx); \
const GLfloat lwidth = ctx->Line.Width; \
const GLfloat psize = ctx->Point.Size; \
(void) lwidth; (void)psize; (void) vert;
#define TAG(x) x##_r128_smooth_indirect
#include "render_tmp.h"
#define NEGATIVE(f) (f < 0)
#define DIFFERENT_SIGNS(a,b) ((a*b) < 0)
#define LINTERP(T, A, B) ((A) + (T) * ((B) - (A)))
#define INTERP_RGBA(t, out, a, b) \
do { \
int i; \
for (i = 0; i < 4; i++) { \
GLfloat fa = UBYTE_COLOR_TO_FLOAT_COLOR(a[i]); \
GLfloat fb = UBYTE_COLOR_TO_FLOAT_COLOR(b[i]); \
GLfloat fo = LINTERP(t, fa, fb); \
FLOAT_COLOR_TO_UBYTE_COLOR(out[i], fo); \
} \
} while (0)
#define CLIP(SGN, V, PLANE) \
do { \
if (mask & PLANE) { \
GLuint *indata = inlist[in]; \
GLuint *outdata = inlist[in ^= 1]; \
GLuint nr = n; \
GLfloat *J = verts[indata[nr-1]].f; \
GLfloat dpJ = (SGN J[V]) + J[3]; \
\
inlist[0] = vlist1; \
for (i = n = 0 ; i < nr ; i++) { \
GLuint elt_i = indata[i]; \
GLfloat *I = verts[elt_i].f; \
GLfloat dpI = (SGN I[V]) + I[3]; \
\
if (DIFFERENT_SIGNS(dpI, dpJ)) { \
GLfloat *O = verts[next_vert].f; \
GLfloat t, *in, *out; \
\
if (NEGATIVE(dpI)) { \
t = dpI / (dpI - dpJ); \
in = I; \
out = J; \
} else { \
t = dpJ / (dpJ - dpI); \
in = J; \
out = I; \
} \
\
interp(t, O, in, out); \
\
clipmask[next_vert] = 0; \
outdata[n++] = next_vert++; \
} \
\
clipmask[elt_i] |= PLANE; /* don't set up */ \
\
if (!NEGATIVE(dpI)) { \
outdata[n++] = elt_i; \
clipmask[elt_i] &= ~PLANE; /* set up after all */ \
} \
\
J = I; \
dpJ = dpI; \
} \
\
if (n < 3) return; \
} \
} while (0)
#define LINE_CLIP(x,y,z,w,PLANE) \
do { \
if (mask & PLANE) { \
GLfloat dpI = DOT4V(I,x,y,z,w); \
GLfloat dpJ = DOT4V(J,x,y,z,w); \
\
if (DIFFERENT_SIGNS(dpI, dpJ)) { \
GLfloat *O = verts[next_vert].f; \
GLfloat t = dpI / (dpI - dpJ); \
\
interp(t, O, I, J); \
\
clipmask[next_vert] = 0; \
\
if (NEGATIVE(dpI)) { \
clipmask[elts[0]] |= PLANE; \
I = O; \
elts[0] = next_vert++; \
} else { \
clipmask[elts[1]] |= PLANE; \
J = O; \
elts[1] = next_vert++; \
} \
} else if (NEGATIVE(dpI)) return; \
} \
} while (0)
static void r128TriClip(GLuint **p_elts,
r128Vertex *verts,
GLubyte *clipmask,
GLuint *p_next_vert,
GLubyte mask,
r128InterpFunc interp)
{
GLuint *elts = *p_elts;
GLuint next_vert = *p_next_vert;
GLuint in = 0;
GLuint n = 3;
GLuint vlist1[VB_MAX_CLIPPED_VERTS];
GLuint vlist2[VB_MAX_CLIPPED_VERTS];
GLuint *inlist[2];
GLuint *out;
GLuint i;
inlist[0] = elts;
inlist[1] = vlist2;
CLIP(-,0,CLIP_RIGHT_BIT);
CLIP(+,0,CLIP_LEFT_BIT);
CLIP(-,1,CLIP_TOP_BIT);
CLIP(+,1,CLIP_BOTTOM_BIT);
CLIP(-,2,CLIP_FAR_BIT);
CLIP(+,2,CLIP_NEAR_BIT);
/* Convert the planar polygon to a list of triangles */
out = inlist[in];
for (i = 2 ; i < n ; i++) {
elts[0] = out[0];
elts[1] = out[i-1];
elts[2] = out[i];
elts += 3;
}
*p_next_vert = next_vert;
*p_elts = elts;
}
static void r128LineClip(GLuint **p_elts,
r128Vertex *verts,
GLubyte *clipmask,
GLuint *p_next_vert,
GLubyte mask,
r128InterpFunc interp)
{
GLuint *elts = *p_elts;
GLfloat *I = verts[elts[0]].f;
GLfloat *J = verts[elts[1]].f;
GLuint next_vert = *p_next_vert;
LINE_CLIP(1,0,0,-1,CLIP_LEFT_BIT);
LINE_CLIP(-1,0,0,1,CLIP_RIGHT_BIT);
LINE_CLIP(0,1,0,-1,CLIP_TOP_BIT);
LINE_CLIP(0,-1,0,1,CLIP_BOTTOM_BIT);
LINE_CLIP(0,0,1,-1,CLIP_FAR_BIT);
LINE_CLIP(0,0,-1,1,CLIP_NEAR_BIT);
*p_next_vert = next_vert;
*p_elts += 2;
}
#define CLIP_POINT(e) \
do { \
if (mask[e]) *out++ = e; \
} while (0)
#define CLIP_LINE(e1, e0) \
do { \
GLubyte ormask = mask[e0] | mask[e1]; \
out[0] = e1; \
out[1] = e0; \
out += 2; \
if (ormask) { \
out-=2; \
if (!(mask[e0] & mask[e1])) { \
r128LineClip(&out, verts, mask, &next_vert, ormask, interp); \
} \
} \
} while (0)
#define CLIP_TRIANGLE(e2, e1, e0) \
do { \
GLubyte ormask; \
out[0] = e2; \
out[1] = e1; \
out[2] = e0; \
out += 3; \
ormask = mask[e2] | mask[e1] | mask[e0]; \
if (ormask) { \
out -= 3; \
if (!(mask[e2] & mask[e1] & mask[e0])) { \
r128TriClip(&out, verts, mask, &next_vert, ormask, interp); \
} \
} \
} while (0)
/* Build a table of functions to clip each primitive type. These
* produce a list of elements in the appropriate 'reduced' primitive,
* ie (points, lines, triangles) containing all the clipped and
* unclipped primitives from the original list.
*/
#define LOCAL_VARS \
r128ContextPtr r128ctx = R128_CONTEXT(VB->ctx); \
r128VertexBufferPtr r128VB = R128_DRIVER_DATA(VB); \
GLuint *elt = VB->EltPtr->data; \
r128Vertex *verts = r128VB->verts; \
GLuint next_vert = r128VB->last_vert; \
GLuint *out = r128VB->clipped_elements.data; \
GLubyte *mask = VB->ClipMask; \
r128InterpFunc interp = r128ctx->interp; \
(void) interp; (void) verts;
#define POSTFIX \
r128VB->clipped_elements.count = out - r128VB->clipped_elements.data; \
r128VB->last_vert = next_vert;
#define INIT(x)
#define RENDER_POINTS(start, count) \
do { \
GLuint i; \
for (i = start; i < count; i++) \
CLIP_POINT(elt[i]); \
} while (0)
#define RENDER_LINE(i1, i0) \
do { \
CLIP_LINE(elt[i1], elt[i0]); \
} while (0)
#define RENDER_TRI(i2, i1, i0, pv, parity) \
do { \
GLuint e2 = elt[i2], e1 = elt[i1], e0 = elt[i0]; \
if (parity) e2 = elt[i1], e1 = elt[i2]; \
CLIP_TRIANGLE(e2, e1, e0); \
} while (0)
#define RENDER_QUAD(i3, i2, i1, i0, pv) \
do { \
CLIP_TRIANGLE(elt[i3], elt[i2], elt[i0]); \
CLIP_TRIANGLE(elt[i2], elt[i1], elt[i0]); \
} while (0)
#define TAG(x) r128_clip_##x##_elt
#include "render_tmp.h"
/* Pack rgba and/or texture into the remaining half of a 32 byte vertex.
*/
#define CLIP_UBYTE_COLOR 4
#define CLIP_UBYTE_B 0
#define CLIP_UBYTE_G 1
#define CLIP_UBYTE_R 2
#define CLIP_UBYTE_A 3
#define CLIP_S0 6
#define CLIP_T0 7
#define CLIP_S1 8
#define CLIP_T1 9
#define TYPE (0)
#define TAG(x) x
#include "r128_fasttmp.h"
#define TYPE (R128_RGBA_BIT)
#define TAG(x) x##_RGBA
#include "r128_fasttmp.h"
#define TYPE (R128_TEX0_BIT)
#define TAG(x) x##_TEX0
#include "r128_fasttmp.h"
#define TYPE (R128_RGBA_BIT | R128_TEX0_BIT)
#define TAG(x) x##_RGBA_TEX0
#include "r128_fasttmp.h"
#define TYPE (R128_RGBA_BIT | R128_TEX0_BIT | R128_TEX1_BIT)
#define TAG(x) x##_RGBA_TEX0_TEX1
#include "r128_fasttmp.h"
/* This one *could* get away with sneaking TEX1 into the color and
* specular slots, thus fitting inside a cache line. Would be even
* better to switch to a smaller vertex.
*/
#define TYPE (R128_TEX0_BIT | R128_TEX1_BIT)
#define TAG(x) x##_TEX0_TEX1
#include "r128_fasttmp.h"
/* Render elements directly from original list of vertices. */
static void r128RenderElementsDirect(struct vertex_buffer *VB)
{
GLcontext *ctx = VB->ctx;
r128ContextPtr r128ctx = R128_CONTEXT(ctx);
GLenum prim = ctx->CVA.elt_mode;
GLuint nr = VB->EltPtr->count;
render_func func = render_tab_r128_smooth_indirect[prim];
GLuint p = 0;
if (r128ctx->dirty) r128UpdateHWState(r128ctx);
do {
func(VB, 0, nr, 0);
} while (ctx->Driver.MultipassFunc &&
ctx->Driver.MultipassFunc(VB, ++p));
}
/* Project vertices from clip to device space */
static void r128ProjectVertices(struct vertex_buffer *VB)
{
GLcontext *ctx = VB->ctx;
GLmatrix *mat = &ctx->Viewport.WindowMap;
r128ContextPtr r128ctx = R128_CONTEXT(ctx);
r128VertexBufferPtr r128VB = R128_DRIVER_DATA(VB);
GLfloat m[16];
m[MAT_SX] = mat->m[MAT_SX];
m[MAT_TX] = mat->m[MAT_TX];
m[MAT_SY] = -mat->m[MAT_SY];
m[MAT_TY] = -mat->m[MAT_TY] + r128ctx->driDrawable->h;
m[MAT_SZ] = mat->m[MAT_SZ] * (1.0 / 0x10000);
m[MAT_TZ] = mat->m[MAT_TZ] * (1.0 / 0x10000);
#if USE_RHW2
/* FIXME: Handle RHW2?? */
gl_project_v16(r128VB->verts[VB->CopyStart].f,
r128VB->verts[r128VB->last_vert].f,
m,
16 * 4);
#else
gl_project_v16(r128VB->verts[VB->CopyStart].f,
r128VB->verts[r128VB->last_vert].f,
m,
16 * 4);
#endif
}
/* Project clipped vertices from clip to device space */
static void r128ProjectClippedVertices(struct vertex_buffer *VB)
{
GLcontext *ctx = VB->ctx;
GLmatrix *mat = &ctx->Viewport.WindowMap;
r128ContextPtr r128ctx = R128_CONTEXT(ctx);
r128VertexBufferPtr r128VB = R128_DRIVER_DATA(VB);
GLfloat m[16];
m[MAT_SX] = mat->m[MAT_SX];
m[MAT_TX] = mat->m[MAT_TX];
m[MAT_SY] = -mat->m[MAT_SY];
m[MAT_TY] = -mat->m[MAT_TY] + r128ctx->driDrawable->h;
m[MAT_SZ] = mat->m[MAT_SZ] * (1.0 / 0x10000);
m[MAT_TZ] = mat->m[MAT_TZ] * (1.0 / 0x10000);
#if USE_RHW2
/* FIXME: Handle RHW2?? */
gl_project_clipped_v16(r128VB->verts[VB->CopyStart].f,
r128VB->verts[r128VB->last_vert].f,
m,
16 * 4,
VB->ClipMask + VB->CopyStart);
#else
gl_project_clipped_v16(r128VB->verts[VB->CopyStart].f,
r128VB->verts[r128VB->last_vert].f,
m,
16 * 4,
VB->ClipMask + VB->CopyStart);
#endif
}
static r128FastPathTable r128FastTab[0x80];
/* Initialize the table of fast path support functions */
void r128FastPathInit(void)
{
r128_clip_render_init_elt();
render_init_r128_smooth_indirect();
r128_init_fastpath(&r128FastTab[0]);
r128_init_fastpath_RGBA(&r128FastTab[R128_RGBA_BIT]);
r128_init_fastpath_TEX0(&r128FastTab[R128_TEX0_BIT]);
r128_init_fastpath_RGBA_TEX0(&r128FastTab[R128_RGBA_BIT|R128_TEX0_BIT]);
r128_init_fastpath_TEX0_TEX1(&r128FastTab[R128_TEX0_BIT|R128_TEX1_BIT]);
r128_init_fastpath_RGBA_TEX0_TEX1(&r128FastTab[R128_RGBA_BIT|R128_TEX0_BIT|
R128_TEX1_BIT]);
}
#define VALID_SETUP (R128_RGBA_BIT | R128_TEX0_BIT | R128_TEX1_BIT)
void r128FastPath(struct vertex_buffer *VB)
{
GLcontext *ctx = VB->ctx;
GLenum prim = ctx->CVA.elt_mode;
r128ContextPtr r128ctx = R128_CONTEXT(ctx);
r128FastPathTable *tab = &r128FastTab[r128ctx->SetupIndex & VALID_SETUP];
GLuint do_cliptest = 1;
gl_prepare_arrays_cva(VB); /* still need this */
/* Reserve enough space for the pathological case */
if (VB->EltPtr->count * 12 > R128_DRIVER_DATA(VB)->size) {
r128ResizeVB(VB, VB->EltPtr->count * 12);
do_cliptest = 1;
}
tab->build_vertices(VB, do_cliptest); /* object->clip space */
if (r128ctx->dirty) r128UpdateHWState(r128ctx);
if (VB->ClipOrMask) {
if (!VB->ClipAndMask) {
render_func *clip = r128_clip_render_tab_elt;
r128ctx->interp = tab->interp;
clip[prim](VB, 0, VB->EltPtr->count, 0); /* build new elts */
ctx->CVA.elt_mode = gl_reduce_prim[prim];
VB->EltPtr = &(R128_DRIVER_DATA(VB)->clipped_elements);
r128ProjectClippedVertices(VB); /* clip->device space */
r128RenderElementsDirect(VB); /* render using new list */
}
} else {
r128ProjectVertices(VB); /* clip->device space */
r128RenderElementsDirect(VB); /* render using orig list */
}
/* This indicates that there is no cached data to reuse */
VB->pipeline->data_valid = 0;
VB->pipeline->new_state = 0;
}
|
