summaryrefslogtreecommitdiff
path: root/src/gallium/drivers/i965/brw_sf_emit.c
blob: 497634ec9ed8111951502c57b4ba937f2f97ba34 (plain)
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
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
/*
 Copyright (C) Intel Corp.  2006.  All Rights Reserved.
 Intel funded Tungsten Graphics (http://www.tungstengraphics.com) to
 develop this 3D driver.
 
 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 the rights to use, copy, modify, merge, publish,
 distribute, sublicense, 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 NONINFRINGEMENT.
 IN NO EVENT SHALL THE COPYRIGHT OWNER(S) AND/OR ITS 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:
  *   Keith Whitwell <keith@tungstengraphics.com>
  */
   

#include "brw_batchbuffer.h"

#include "brw_defines.h"
#include "brw_context.h"
#include "brw_eu.h"
#include "brw_sf.h"


static struct brw_reg get_vert_attr(struct brw_sf_compile *c,
				    struct brw_reg vert,
				    GLuint attr)
{
   GLuint off = attr / 2;
   GLuint sub = attr % 2;

   return brw_vec4_grf(vert.nr + off, sub * 4);
}


/*********************************************************************** 
 * Twoside lighting
 */
static void copy_bfc( struct brw_sf_compile *c,
		      struct brw_reg vert )
{
   struct brw_compile *p = &c->func;

   if (c->key.attr_col0 && c->key.attr_bfc0)
      brw_MOV(p, 
	      get_vert_attr(c, vert, c->key.attr_col0), 
	      get_vert_attr(c, vert, c->key.attr_bfc0));

   if (c->key.attr_col1 && c->key.attr_bfc1)
      brw_MOV(p, 
	      get_vert_attr(c, vert, c->key.attr_col1), 
	      get_vert_attr(c, vert, c->key.attr_bfc1));
}


static void do_twoside_color( struct brw_sf_compile *c )
{
   struct brw_compile *p = &c->func;
   struct brw_instruction *if_insn;
   GLuint backface_conditional = c->key.frontface_ccw ? BRW_CONDITIONAL_G : BRW_CONDITIONAL_L;

   /* Already done in clip program:
    */
   if (c->key.primitive == SF_UNFILLED_TRIS)
      return;

   /* XXX: What happens if BFC isn't present?  This could only happen
    * for user-supplied vertex programs, as t_vp_build.c always does
    * the right thing.
    */
   if (!(c->key.attr_col0 && c->key.attr_bfc0) &&
       !(c->key.attr_col1 && c->key.attr_bfc1))
      return;
   
   /* Need to use BRW_EXECUTE_4 and also do an 4-wide compare in order
    * to get all channels active inside the IF.  In the clipping code
    * we run with NoMask, so it's not an option and we can use
    * BRW_EXECUTE_1 for all comparisions.
    */
   brw_push_insn_state(p);
   brw_CMP(p, vec4(brw_null_reg()), backface_conditional, c->det, brw_imm_f(0));
   if_insn = brw_IF(p, BRW_EXECUTE_4); 
   {
      switch (c->nr_verts) {
      case 3: copy_bfc(c, c->vert[2]);
      case 2: copy_bfc(c, c->vert[1]);
      case 1: copy_bfc(c, c->vert[0]);
      }
   }
   brw_ENDIF(p, if_insn);
   brw_pop_insn_state(p);
}



/***********************************************************************
 * Flat shading
 */

#define VERT_RESULT_COLOR_BITS ((1<<VERT_RESULT_COL0) | \
                                 (1<<VERT_RESULT_COL1))

static void copy_colors( struct brw_sf_compile *c,
		     struct brw_reg dst,
		     struct brw_reg src)
{
   struct brw_compile *p = &c->func;

   if (c->key.attr_col0)
      brw_MOV(p, 
	      get_vert_attr(c, dst, c->key.attr_col0), 
	      get_vert_attr(c, src, c->key.attr_col0));

   if (c->key.attr_col1)
      brw_MOV(p, 
	      get_vert_attr(c, dst, c->key.attr_col1), 
	      get_vert_attr(c, src, c->key.attr_col1));

}



/* Need to use a computed jump to copy flatshaded attributes as the
 * vertices are ordered according to y-coordinate before reaching this
 * point, so the PV could be anywhere.
 */
static void do_flatshade_triangle( struct brw_sf_compile *c )
{
   struct brw_compile *p = &c->func;
   struct brw_reg ip = brw_ip_reg();
   GLuint jmpi = 1;
   GLuint nr = 0;

   if (c->key.attr_col0)
      nr++;

   if (c->key.attr_col1)
      nr++;

   if (nr == 0)
      return;

   /* Already done in clip program:
    */
   if (c->key.primitive == SF_UNFILLED_TRIS)
      return;

   if (BRW_IS_IGDNG(p->brw))
       jmpi = 2;

   brw_push_insn_state(p);
   
   brw_MUL(p, c->pv, c->pv, brw_imm_d(jmpi*(nr*2+1)));
   brw_JMPI(p, ip, ip, c->pv);

   copy_colors(c, c->vert[1], c->vert[0]);
   copy_colors(c, c->vert[2], c->vert[0]);
   brw_JMPI(p, ip, ip, brw_imm_d(jmpi*(nr*4+1)));

   copy_colors(c, c->vert[0], c->vert[1]);
   copy_colors(c, c->vert[2], c->vert[1]);
   brw_JMPI(p, ip, ip, brw_imm_d(jmpi*nr*2));

   copy_colors(c, c->vert[0], c->vert[2]);
   copy_colors(c, c->vert[1], c->vert[2]);

   brw_pop_insn_state(p);
}
	

static void do_flatshade_line( struct brw_sf_compile *c )
{
   struct brw_compile *p = &c->func;
   struct brw_reg ip = brw_ip_reg();
   GLuint jmpi = 1;
   GLuint nr = 0;

   if (c->key.attr_col0)
      nr++;

   if (c->key.attr_col1)
      nr++;

   if (nr == 0)
      return;

   /* Already done in clip program: 
    */
   if (c->key.primitive == SF_UNFILLED_TRIS)
      return;

   if (BRW_IS_IGDNG(p->brw))
       jmpi = 2;

   brw_push_insn_state(p);
   
   brw_MUL(p, c->pv, c->pv, brw_imm_d(jmpi*(nr+1)));
   brw_JMPI(p, ip, ip, c->pv);
   copy_colors(c, c->vert[1], c->vert[0]);

   brw_JMPI(p, ip, ip, brw_imm_ud(jmpi*nr));
   copy_colors(c, c->vert[0], c->vert[1]);

   brw_pop_insn_state(p);
}

	

/***********************************************************************
 * Triangle setup.
 */


static void alloc_regs( struct brw_sf_compile *c )
{
   GLuint reg, i;

   /* Values computed by fixed function unit:
    */
   c->pv  = retype(brw_vec1_grf(1, 1), BRW_REGISTER_TYPE_D);
   c->det = brw_vec1_grf(1, 2);
   c->dx0 = brw_vec1_grf(1, 3);
   c->dx2 = brw_vec1_grf(1, 4);
   c->dy0 = brw_vec1_grf(1, 5);
   c->dy2 = brw_vec1_grf(1, 6);

   /* z and 1/w passed in seperately:
    */
   c->z[0]     = brw_vec1_grf(2, 0);
   c->inv_w[0] = brw_vec1_grf(2, 1);
   c->z[1]     = brw_vec1_grf(2, 2);
   c->inv_w[1] = brw_vec1_grf(2, 3);
   c->z[2]     = brw_vec1_grf(2, 4);
   c->inv_w[2] = brw_vec1_grf(2, 5);
   
   /* The vertices:
    */
   reg = 3;
   for (i = 0; i < c->nr_verts; i++) {
      c->vert[i] = brw_vec8_grf(reg, 0);
      reg += c->nr_attr_regs;
   }

   /* Temporaries, allocated after last vertex reg.
    */
   c->inv_det = brw_vec1_grf(reg, 0);  reg++;
   c->a1_sub_a0 = brw_vec8_grf(reg, 0);  reg++;
   c->a2_sub_a0 = brw_vec8_grf(reg, 0);  reg++;
   c->tmp = brw_vec8_grf(reg, 0);  reg++;

   /* Note grf allocation:
    */
   c->prog_data.total_grf = reg;
   

   /* Outputs of this program - interpolation coefficients for
    * rasterization:
    */
   c->m1Cx = brw_vec8_reg(BRW_MESSAGE_REGISTER_FILE, 1, 0);
   c->m2Cy = brw_vec8_reg(BRW_MESSAGE_REGISTER_FILE, 2, 0);
   c->m3C0 = brw_vec8_reg(BRW_MESSAGE_REGISTER_FILE, 3, 0);
}


static void copy_z_inv_w( struct brw_sf_compile *c )
{
   struct brw_compile *p = &c->func;
   GLuint i;

   brw_push_insn_state(p);
	
   /* Copy both scalars with a single MOV:
    */
   for (i = 0; i < c->nr_verts; i++)
      brw_MOV(p, vec2(suboffset(c->vert[i], 2)), vec2(c->z[i]));
	 
   brw_pop_insn_state(p);
}


static void invert_det( struct brw_sf_compile *c)
{
   /* Looks like we invert all 8 elements just to get 1/det in
    * position 2 !?!
    */
   brw_math(&c->func, 
	    c->inv_det, 
	    BRW_MATH_FUNCTION_INV,
	    BRW_MATH_SATURATE_NONE,
	    0, 
	    c->det,
	    BRW_MATH_DATA_SCALAR,
	    BRW_MATH_PRECISION_FULL);

}


/* Two attributes packed into a wide register.  Figure out if either
 * or both of them need linear/perspective interpolation.  Constant
 * regs are left as-is.
 */
static GLboolean calculate_masks( struct brw_sf_compile *c,
				  GLuint reg,
				  GLushort *pc,
				  GLushort *pc_persp,
				  GLushort *pc_linear)
{
   GLboolean is_last_attr = (reg == c->nr_setup_regs - 1);
   GLuint persp_mask = c->key.persp_attrs;
   GLuint linear_mask = (c->key.persp_attrs | c->key.linear_attrs);

   *pc_persp = 0;
   *pc_linear = 0;
   *pc = 0xf;
      
   if (persp_mask & (1 << (reg*2))) 
      *pc_persp = 0xf;

   if (linear_mask & (1 << (reg*2))) 
      *pc_linear = 0xf;

   /* Maybe only processs one attribute on the final round:
    */
   if (reg*2+1 < c->nr_setup_attrs) {
      *pc |= 0xf0;

      if (persp_mask & (1 << (reg*2+1))) 
	 *pc_persp |= 0xf0;

      if (linear_mask & (1 << (reg*2+1))) 
	 *pc_linear |= 0xf0;
   }

   return is_last_attr;
}


void brw_emit_null_setup( struct brw_sf_compile *c )
{
   struct brw_compile *p = &c->func;

   /* m0 is implicitly copied from r0 in the send instruction:
    */	 
   brw_urb_WRITE(p, 
                 brw_null_reg(),
                 0,
                 brw_vec8_grf(0, 0), /* r0, will be copied to m0 */
                 0, 	/* allocate */
                 1,	/* used */
                 1, 	/* msg len */
                 0,	/* response len */
                 1,	/* eot */
                 1, 	/* writes complete */
                 0,	/* offset */
                 BRW_URB_SWIZZLE_TRANSPOSE); 
}

void brw_emit_tri_setup( struct brw_sf_compile *c, GLboolean allocate)
{
   struct brw_compile *p = &c->func;
   GLuint i;

   c->nr_verts = 3;

   if (allocate)
      alloc_regs(c);

   invert_det(c);
   copy_z_inv_w(c);

   if (c->key.do_twoside_color) 
      do_twoside_color(c);

   if (c->key.do_flat_shading)
      do_flatshade_triangle(c);
      
   
   for (i = 0; i < c->nr_setup_regs; i++)
   {
      /* Pair of incoming attributes:
       */
      struct brw_reg a0 = offset(c->vert[0], i);
      struct brw_reg a1 = offset(c->vert[1], i);
      struct brw_reg a2 = offset(c->vert[2], i);
      GLushort pc, pc_persp, pc_linear;
      GLboolean last = calculate_masks(c, i, &pc, &pc_persp, &pc_linear);

      if (pc_persp)
      {
	 brw_set_predicate_control_flag_value(p, pc_persp);
	 brw_MUL(p, a0, a0, c->inv_w[0]);
	 brw_MUL(p, a1, a1, c->inv_w[1]);
	 brw_MUL(p, a2, a2, c->inv_w[2]);
      }
      
      
      /* Calculate coefficients for interpolated values:
       */      
      if (pc_linear)
      {
	 brw_set_predicate_control_flag_value(p, pc_linear);

	 brw_ADD(p, c->a1_sub_a0, a1, negate(a0));
	 brw_ADD(p, c->a2_sub_a0, a2, negate(a0));

	 /* calculate dA/dx
	  */
	 brw_MUL(p, brw_null_reg(), c->a1_sub_a0, c->dy2);
	 brw_MAC(p, c->tmp, c->a2_sub_a0, negate(c->dy0));
	 brw_MUL(p, c->m1Cx, c->tmp, c->inv_det);
		
	 /* calculate dA/dy
	  */
	 brw_MUL(p, brw_null_reg(), c->a2_sub_a0, c->dx0);
	 brw_MAC(p, c->tmp, c->a1_sub_a0, negate(c->dx2));
	 brw_MUL(p, c->m2Cy, c->tmp, c->inv_det);
      }

      {
	 brw_set_predicate_control_flag_value(p, pc); 
	 /* start point for interpolation
	  */
	 brw_MOV(p, c->m3C0, a0);
      
	 /* Copy m0..m3 to URB.  m0 is implicitly copied from r0 in
	  * the send instruction:
	  */	 
	 brw_urb_WRITE(p, 
		       brw_null_reg(),
		       0,
		       brw_vec8_grf(0, 0), /* r0, will be copied to m0 */
		       0, 	/* allocate */
		       1,	/* used */
		       4, 	/* msg len */
		       0,	/* response len */
		       last,	/* eot */
		       last, 	/* writes complete */
		       i*4,	/* offset */
		       BRW_URB_SWIZZLE_TRANSPOSE); /* XXX: Swizzle control "SF to windower" */
      }
   }
}



void brw_emit_line_setup( struct brw_sf_compile *c, GLboolean allocate)
{
   struct brw_compile *p = &c->func;
   GLuint i;


   c->nr_verts = 2;

   if (allocate)
      alloc_regs(c);

   invert_det(c);
   copy_z_inv_w(c);

   if (c->key.do_flat_shading)
      do_flatshade_line(c);

   for (i = 0; i < c->nr_setup_regs; i++)
   {
      /* Pair of incoming attributes:
       */
      struct brw_reg a0 = offset(c->vert[0], i);
      struct brw_reg a1 = offset(c->vert[1], i);
      GLushort pc, pc_persp, pc_linear;
      GLboolean last = calculate_masks(c, i, &pc, &pc_persp, &pc_linear);

      if (pc_persp)
      {
	 brw_set_predicate_control_flag_value(p, pc_persp);
	 brw_MUL(p, a0, a0, c->inv_w[0]);
	 brw_MUL(p, a1, a1, c->inv_w[1]);
      }

      /* Calculate coefficients for position, color:
       */
      if (pc_linear) {
	 brw_set_predicate_control_flag_value(p, pc_linear); 

	 brw_ADD(p, c->a1_sub_a0, a1, negate(a0));

 	 brw_MUL(p, c->tmp, c->a1_sub_a0, c->dx0); 
	 brw_MUL(p, c->m1Cx, c->tmp, c->inv_det);
		
	 brw_MUL(p, c->tmp, c->a1_sub_a0, c->dy0);
	 brw_MUL(p, c->m2Cy, c->tmp, c->inv_det);
      }

      {
	 brw_set_predicate_control_flag_value(p, pc); 

	 /* start point for interpolation
	  */
	 brw_MOV(p, c->m3C0, a0);

	 /* Copy m0..m3 to URB. 
	  */
	 brw_urb_WRITE(p, 
		       brw_null_reg(),
		       0,
		       brw_vec8_grf(0, 0),
		       0, 	/* allocate */
		       1, 	/* used */
		       4, 	/* msg len */
		       0,	/* response len */
		       last, 	/* eot */
		       last, 	/* writes complete */
		       i*4,	/* urb destination offset */
		       BRW_URB_SWIZZLE_TRANSPOSE); 
      }
   } 
}

void brw_emit_point_sprite_setup( struct brw_sf_compile *c, GLboolean allocate)
{
   struct brw_compile *p = &c->func;
   GLuint i;

   c->nr_verts = 1;

   if (allocate)
      alloc_regs(c);

   copy_z_inv_w(c);

   for (i = 0; i < c->nr_setup_regs; i++)
   {
      /* XXX: only seems to check point_coord_replace_attrs for every
       * second attribute?!?
       */
      boolean coord_replace = !!(c->key.point_coord_replace_attrs & (1<<(2*i)));
      struct brw_reg a0 = offset(c->vert[0], i);
      GLushort pc, pc_persp, pc_linear;
      GLboolean last = calculate_masks(c, i, &pc, &pc_persp, &pc_linear);
            
      if (pc_persp)
      {				
	 if (coord_replace) {
	    brw_set_predicate_control_flag_value(p, pc_persp);
	    brw_MUL(p, a0, a0, c->inv_w[0]);
	 }
      }

      if (coord_replace) {
	 /* Caculate 1.0/PointWidth */
	 brw_math(&c->func,
		  c->tmp,
		  BRW_MATH_FUNCTION_INV,
		  BRW_MATH_SATURATE_NONE,
		  0,
		  c->dx0,
		  BRW_MATH_DATA_SCALAR,
		  BRW_MATH_PRECISION_FULL);

	 if (c->key.sprite_origin_lower_left) {
	    brw_MUL(p, c->m1Cx, c->tmp, c->inv_w[0]);
	    brw_MOV(p, vec1(suboffset(c->m1Cx, 1)), brw_imm_f(0.0));
	    brw_MUL(p, c->m2Cy, c->tmp, negate(c->inv_w[0]));
	    brw_MOV(p, vec1(suboffset(c->m2Cy, 0)), brw_imm_f(0.0));
	 } 
	 else {
	    brw_MUL(p, c->m1Cx, c->tmp, c->inv_w[0]);
	    brw_MOV(p, vec1(suboffset(c->m1Cx, 1)), brw_imm_f(0.0));
	    brw_MUL(p, c->m2Cy, c->tmp, c->inv_w[0]);
	    brw_MOV(p, vec1(suboffset(c->m2Cy, 0)), brw_imm_f(0.0));
	 }
      } 
      else {
	 brw_MOV(p, c->m1Cx, brw_imm_ud(0));
	 brw_MOV(p, c->m2Cy, brw_imm_ud(0));
      }

      {
	 brw_set_predicate_control_flag_value(p, pc); 
	 if (coord_replace) {
	    if (c->key.sprite_origin_lower_left) {
	       brw_MUL(p, c->m3C0, c->inv_w[0], brw_imm_f(1.0));
	       brw_MOV(p, vec1(suboffset(c->m3C0, 0)), brw_imm_f(0.0));
	    }
	    else {
	       brw_MOV(p, c->m3C0, brw_imm_f(0.0));
	    }
	 } 
	 else {
	    brw_MOV(p, c->m3C0, a0); /* constant value */
	 }

	 /* Copy m0..m3 to URB. 
	  */
	 brw_urb_WRITE(p, 
		       brw_null_reg(),
		       0,
		       brw_vec8_grf(0, 0),
		       0, 	/* allocate */
		       1,	/* used */
		       4, 	/* msg len */
		       0,	/* response len */
		       last, 	/* eot */
		       last, 	/* writes complete */
		       i*4,	/* urb destination offset */
		       BRW_URB_SWIZZLE_TRANSPOSE);
      }
   }
}

/* Points setup - several simplifications as all attributes are
 * constant across the face of the point (point sprites excluded!)
 */
void brw_emit_point_setup( struct brw_sf_compile *c, GLboolean allocate)
{
   struct brw_compile *p = &c->func;
   GLuint i;

   c->nr_verts = 1;
   
   if (allocate)
      alloc_regs(c);

   copy_z_inv_w(c);

   brw_MOV(p, c->m1Cx, brw_imm_ud(0)); /* zero - move out of loop */
   brw_MOV(p, c->m2Cy, brw_imm_ud(0)); /* zero - move out of loop */

   for (i = 0; i < c->nr_setup_regs; i++)
   {
      struct brw_reg a0 = offset(c->vert[0], i);
      GLushort pc, pc_persp, pc_linear;
      GLboolean last = calculate_masks(c, i, &pc, &pc_persp, &pc_linear);
            
      if (pc_persp)
      {				
	 /* This seems odd as the values are all constant, but the
	  * fragment shader will be expecting it:
	  */
	 brw_set_predicate_control_flag_value(p, pc_persp);
	 brw_MUL(p, a0, a0, c->inv_w[0]);
      }


      /* The delta values are always zero, just send the starting
       * coordinate.  Again, this is to fit in with the interpolation
       * code in the fragment shader.
       */
      {
	 brw_set_predicate_control_flag_value(p, pc); 

	 brw_MOV(p, c->m3C0, a0); /* constant value */

	 /* Copy m0..m3 to URB. 
	  */
	 brw_urb_WRITE(p, 
		       brw_null_reg(),
		       0,
		       brw_vec8_grf(0, 0),
		       0, 	/* allocate */
		       1,	/* used */
		       4, 	/* msg len */
		       0,	/* response len */
		       last, 	/* eot */
		       last, 	/* writes complete */
		       i*4,	/* urb destination offset */
		       BRW_URB_SWIZZLE_TRANSPOSE);
      }
   }
}

void brw_emit_anyprim_setup( struct brw_sf_compile *c )
{
   struct brw_compile *p = &c->func;
   struct brw_reg ip = brw_ip_reg();
   struct brw_reg payload_prim = brw_uw1_reg(BRW_GENERAL_REGISTER_FILE, 1, 0);
   struct brw_reg payload_attr = get_element_ud(brw_vec1_reg(BRW_GENERAL_REGISTER_FILE, 1, 0), 0); 
   struct brw_reg primmask;
   struct brw_instruction *jmp;
   struct brw_reg v1_null_ud = vec1(retype(brw_null_reg(), BRW_REGISTER_TYPE_UD));
   
   GLuint saveflag;

   c->nr_verts = 3;
   alloc_regs(c);

   primmask = retype(get_element(c->tmp, 0), BRW_REGISTER_TYPE_UD);

   brw_MOV(p, primmask, brw_imm_ud(1));
   brw_SHL(p, primmask, primmask, payload_prim);

   brw_set_conditionalmod(p, BRW_CONDITIONAL_Z);
   brw_AND(p, v1_null_ud, primmask, brw_imm_ud((1<<_3DPRIM_TRILIST) |
					       (1<<_3DPRIM_TRISTRIP) |
					       (1<<_3DPRIM_TRIFAN) |
					       (1<<_3DPRIM_TRISTRIP_REVERSE) |
					       (1<<_3DPRIM_POLYGON) |
					       (1<<_3DPRIM_RECTLIST) |
					       (1<<_3DPRIM_TRIFAN_NOSTIPPLE)));
   jmp = brw_JMPI(p, ip, ip, brw_imm_d(0));
   {
      saveflag = p->flag_value;
      brw_push_insn_state(p); 
      brw_emit_tri_setup( c, GL_FALSE );
      brw_pop_insn_state(p);
      p->flag_value = saveflag;
      /* note - thread killed in subroutine, so must
       * restore the flag which is changed when building
       * the subroutine. fix #13240
       */
   }
   brw_land_fwd_jump(p, jmp);

   brw_set_conditionalmod(p, BRW_CONDITIONAL_Z);
   brw_AND(p, v1_null_ud, primmask, brw_imm_ud((1<<_3DPRIM_LINELIST) |
					       (1<<_3DPRIM_LINESTRIP) |
					       (1<<_3DPRIM_LINELOOP) |
					       (1<<_3DPRIM_LINESTRIP_CONT) |
					       (1<<_3DPRIM_LINESTRIP_BF) |
					       (1<<_3DPRIM_LINESTRIP_CONT_BF)));
   jmp = brw_JMPI(p, ip, ip, brw_imm_d(0));
   {
      saveflag = p->flag_value;
      brw_push_insn_state(p); 
      brw_emit_line_setup( c, GL_FALSE );
      brw_pop_insn_state(p);
      p->flag_value = saveflag;
      /* note - thread killed in subroutine */
   }
   brw_land_fwd_jump(p, jmp); 

   brw_set_conditionalmod(p, BRW_CONDITIONAL_Z);
   brw_AND(p, v1_null_ud, payload_attr, brw_imm_ud(1<<BRW_SPRITE_POINT_ENABLE));
   jmp = brw_JMPI(p, ip, ip, brw_imm_d(0));
   {
      saveflag = p->flag_value;
      brw_push_insn_state(p); 
      brw_emit_point_sprite_setup( c, GL_FALSE );
      brw_pop_insn_state(p);
      p->flag_value = saveflag;
   }
   brw_land_fwd_jump(p, jmp); 

   brw_emit_point_setup( c, GL_FALSE );
}