aboutsummaryrefslogtreecommitdiffstats
path: root/src/gallium/auxiliary/draw/draw_pipe_aaline.c
blob: 66a943aff425d80a84870c612f78818971a8f430 (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
/**************************************************************************
 *
 * Copyright 2007-2018 VMware, Inc.
 * 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 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 VMWARE 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.
 *
 **************************************************************************/

/**
 * AA line stage:  AA lines are converted triangles (with extra generic)
 *
 * Authors:  Brian Paul
 */


#include "pipe/p_context.h"
#include "pipe/p_defines.h"
#include "pipe/p_shader_tokens.h"
#include "util/u_inlines.h"

#include "util/u_format.h"
#include "util/u_math.h"
#include "util/u_memory.h"

#include "tgsi/tgsi_transform.h"
#include "tgsi/tgsi_dump.h"

#include "draw_context.h"
#include "draw_private.h"
#include "draw_pipe.h"


/** Approx number of new tokens for instructions in aa_transform_inst() */
#define NUM_NEW_TOKENS 53


/**
 * Subclass of pipe_shader_state to carry extra fragment shader info.
 */
struct aaline_fragment_shader
{
   struct pipe_shader_state state;
   void *driver_fs;
   void *aaline_fs;
   int generic_attrib;  /**< generic used for distance */
};


/**
 * Subclass of draw_stage
 */
struct aaline_stage
{
   struct draw_stage stage;

   float half_line_width;

   /** For AA lines, this is the vertex attrib slot for new generic */
   uint coord_slot;
   /** position, not necessarily output zero */
   uint pos_slot;


   /*
    * Currently bound state
    */
   struct aaline_fragment_shader *fs;

   /*
    * Driver interface/override functions
    */
   void * (*driver_create_fs_state)(struct pipe_context *,
                                    const struct pipe_shader_state *);
   void (*driver_bind_fs_state)(struct pipe_context *, void *);
   void (*driver_delete_fs_state)(struct pipe_context *, void *);
};



/**
 * Subclass of tgsi_transform_context, used for transforming the
 * user's fragment shader to add the special AA instructions.
 */
struct aa_transform_context {
   struct tgsi_transform_context base;
   uint64_t tempsUsed;  /**< bitmask */
   int colorOutput; /**< which output is the primary color */
   int maxInput, maxGeneric;  /**< max input index found */
   int colorTemp, aaTemp;  /**< temp registers */
};

/**
 * TGSI declaration transform callback.
 * Look for a free input attrib, and two free temp regs.
 */
static void
aa_transform_decl(struct tgsi_transform_context *ctx,
                  struct tgsi_full_declaration *decl)
{
   struct aa_transform_context *aactx = (struct aa_transform_context *)ctx;

   if (decl->Declaration.File == TGSI_FILE_OUTPUT &&
       decl->Semantic.Name == TGSI_SEMANTIC_COLOR &&
       decl->Semantic.Index == 0) {
      aactx->colorOutput = decl->Range.First;
   }
   else if (decl->Declaration.File == TGSI_FILE_INPUT) {
      if ((int) decl->Range.Last > aactx->maxInput)
         aactx->maxInput = decl->Range.Last;
      if (decl->Semantic.Name == TGSI_SEMANTIC_GENERIC &&
           (int) decl->Semantic.Index > aactx->maxGeneric) {
         aactx->maxGeneric = decl->Semantic.Index;
      }
   }
   else if (decl->Declaration.File == TGSI_FILE_TEMPORARY) {
      uint i;
      for (i = decl->Range.First;
           i <= decl->Range.Last; i++) {
         /*
          * XXX this bitfield doesn't really cut it...
          */
         aactx->tempsUsed |= UINT64_C(1) << i;
      }
   }

   ctx->emit_declaration(ctx, decl);
}


/**
 * Find the lowest zero bit, or -1 if bitfield is all ones.
 */
static int
free_bit(uint64_t bitfield)
{
   return ffsll(~bitfield) - 1;
}


/**
 * TGSI transform prolog callback.
 */
static void
aa_transform_prolog(struct tgsi_transform_context *ctx)
{
   struct aa_transform_context *aactx = (struct aa_transform_context *) ctx;
   uint64_t usedTemps = aactx->tempsUsed;

   /* find two free temp regs */
   aactx->colorTemp = free_bit(usedTemps);
   usedTemps |= UINT64_C(1) << aactx->colorTemp;
   aactx->aaTemp = free_bit(usedTemps);
   assert(aactx->colorTemp >= 0);
   assert(aactx->aaTemp >= 0);

   /* declare new generic input/texcoord */
   tgsi_transform_input_decl(ctx, aactx->maxInput + 1,
                             TGSI_SEMANTIC_GENERIC, aactx->maxGeneric + 1,
                             TGSI_INTERPOLATE_LINEAR);

   /* declare new temp regs */
   tgsi_transform_temp_decl(ctx, aactx->aaTemp);
   tgsi_transform_temp_decl(ctx, aactx->colorTemp);
}


/**
 * TGSI transform epilog callback.
 */
static void
aa_transform_epilog(struct tgsi_transform_context *ctx)
{
   struct aa_transform_context *aactx = (struct aa_transform_context *) ctx;

   if (aactx->colorOutput != -1) {
      struct tgsi_full_instruction inst;
      /* insert distance-based coverage code for antialiasing. */

      /* saturate(linewidth - fabs(interpx), linelength - fabs(interpz) */
      inst = tgsi_default_full_instruction();
      inst.Instruction.Saturate = true;
      inst.Instruction.Opcode = TGSI_OPCODE_ADD;
      inst.Instruction.NumDstRegs = 1;
      tgsi_transform_dst_reg(&inst.Dst[0], TGSI_FILE_TEMPORARY,
                             aactx->aaTemp, TGSI_WRITEMASK_XZ);
      inst.Instruction.NumSrcRegs = 2;
      tgsi_transform_src_reg(&inst.Src[1], TGSI_FILE_INPUT, aactx->maxInput + 1,
                             TGSI_SWIZZLE_X, TGSI_SWIZZLE_X,
                             TGSI_SWIZZLE_Z, TGSI_SWIZZLE_Z);
      tgsi_transform_src_reg(&inst.Src[0], TGSI_FILE_INPUT, aactx->maxInput + 1,
                             TGSI_SWIZZLE_Y, TGSI_SWIZZLE_Y,
                             TGSI_SWIZZLE_W, TGSI_SWIZZLE_W);
      inst.Src[1].Register.Absolute = true;
      inst.Src[1].Register.Negate = true;
      ctx->emit_instruction(ctx, &inst);

      /* MUL width / height alpha */
      tgsi_transform_op2_swz_inst(ctx, TGSI_OPCODE_MUL,
                                  TGSI_FILE_TEMPORARY, aactx->aaTemp,
                                  TGSI_WRITEMASK_W,
                                  TGSI_FILE_TEMPORARY, aactx->aaTemp,
                                  TGSI_SWIZZLE_X,
                                  TGSI_FILE_TEMPORARY, aactx->aaTemp,
                                  TGSI_SWIZZLE_Z, false);

      /* MOV rgb */
      tgsi_transform_op1_inst(ctx, TGSI_OPCODE_MOV,
                              TGSI_FILE_OUTPUT, aactx->colorOutput,
                              TGSI_WRITEMASK_XYZ,
                              TGSI_FILE_TEMPORARY, aactx->colorTemp);

      /* MUL alpha */
      tgsi_transform_op2_inst(ctx, TGSI_OPCODE_MUL,
                              TGSI_FILE_OUTPUT, aactx->colorOutput,
                              TGSI_WRITEMASK_W,
                              TGSI_FILE_TEMPORARY, aactx->colorTemp,
                              TGSI_FILE_TEMPORARY, aactx->aaTemp, false);
   }
}


/**
 * TGSI instruction transform callback.
 * Replace writes to result.color w/ a temp reg.
 */
static void
aa_transform_inst(struct tgsi_transform_context *ctx,
                  struct tgsi_full_instruction *inst)
{
   struct aa_transform_context *aactx = (struct aa_transform_context *) ctx;
   uint i;

   /*
    * Look for writes to result.color and replace with colorTemp reg.
    */
   for (i = 0; i < inst->Instruction.NumDstRegs; i++) {
      struct tgsi_full_dst_register *dst = &inst->Dst[i];
      if (dst->Register.File == TGSI_FILE_OUTPUT &&
          dst->Register.Index == aactx->colorOutput) {
         dst->Register.File = TGSI_FILE_TEMPORARY;
         dst->Register.Index = aactx->colorTemp;
      }
   }

   ctx->emit_instruction(ctx, inst);
}


/**
 * Generate the frag shader we'll use for drawing AA lines.
 * This will be the user's shader plus some arithmetic instructions.
 */
static boolean
generate_aaline_fs(struct aaline_stage *aaline)
{
   struct pipe_context *pipe = aaline->stage.draw->pipe;
   const struct pipe_shader_state *orig_fs = &aaline->fs->state;
   struct pipe_shader_state aaline_fs;
   struct aa_transform_context transform;
   const uint newLen = tgsi_num_tokens(orig_fs->tokens) + NUM_NEW_TOKENS;

   aaline_fs = *orig_fs; /* copy to init */
   aaline_fs.tokens = tgsi_alloc_tokens(newLen);
   if (aaline_fs.tokens == NULL)
      return FALSE;

   memset(&transform, 0, sizeof(transform));
   transform.colorOutput = -1;
   transform.maxInput = -1;
   transform.maxGeneric = -1;
   transform.colorTemp = -1;
   transform.aaTemp = -1;
   transform.base.prolog = aa_transform_prolog;
   transform.base.epilog = aa_transform_epilog;
   transform.base.transform_instruction = aa_transform_inst;
   transform.base.transform_declaration = aa_transform_decl;

   tgsi_transform_shader(orig_fs->tokens,
                         (struct tgsi_token *) aaline_fs.tokens,
                         newLen, &transform.base);

#if 0 /* DEBUG */
   debug_printf("draw_aaline, orig shader:\n");
   tgsi_dump(orig_fs->tokens, 0);
   debug_printf("draw_aaline, new shader:\n");
   tgsi_dump(aaline_fs.tokens, 0);
#endif

   aaline->fs->aaline_fs = aaline->driver_create_fs_state(pipe, &aaline_fs);
   if (aaline->fs->aaline_fs == NULL)
      goto fail;

   aaline->fs->generic_attrib = transform.maxGeneric + 1;
   FREE((void *)aaline_fs.tokens);
   return TRUE;

fail:
   FREE((void *)aaline_fs.tokens);
   return FALSE;
}


/**
 * When we're about to draw our first AA line in a batch, this function is
 * called to tell the driver to bind our modified fragment shader.
 */
static boolean
bind_aaline_fragment_shader(struct aaline_stage *aaline)
{
   struct draw_context *draw = aaline->stage.draw;
   struct pipe_context *pipe = draw->pipe;

   if (!aaline->fs->aaline_fs && !generate_aaline_fs(aaline))
      return FALSE;

   draw->suspend_flushing = TRUE;
   aaline->driver_bind_fs_state(pipe, aaline->fs->aaline_fs);
   draw->suspend_flushing = FALSE;

   return TRUE;
}



static inline struct aaline_stage *
aaline_stage(struct draw_stage *stage)
{
   return (struct aaline_stage *) stage;
}


/**
 * Draw a wide line by drawing a quad, using geometry which will
 * fullfill GL's antialiased line requirements.
 */
static void
aaline_line(struct draw_stage *stage, struct prim_header *header)
{
   const struct aaline_stage *aaline = aaline_stage(stage);
   const float half_width = aaline->half_line_width;
   struct prim_header tri;
   struct vertex_header *v[8];
   uint coordPos = aaline->coord_slot;
   uint posPos = aaline->pos_slot;
   float *pos, *tex;
   float dx = header->v[1]->data[posPos][0] - header->v[0]->data[posPos][0];
   float dy = header->v[1]->data[posPos][1] - header->v[0]->data[posPos][1];
   float a = atan2f(dy, dx);
   float c_a = cosf(a), s_a = sinf(a);
   float half_length;
   float t_l, t_w;
   uint i;

   half_length = 0.5f * sqrtf(dx * dx + dy * dy);

   if (half_length < 0.5f) {
      /*
       * The logic we use for "normal" sized segments is incorrect
       * for very short segments (basically because we only have
       * one value to interpolate, not a distance to each endpoint).
       * Therefore, we calculate half_length differently, so that for
       * original line length (near) 0, we get alpha 0 - otherwise
       * max alpha would still be 0.5. This also prevents us from
       * artifacts due to degenerated lines (the endpoints being
       * identical, which would still receive anywhere from alpha
       * 0-0.5 otherwise) (at least the pstipple stage may generate
       * such lines due to float inaccuracies if line length is very
       * close to a integer).
       * Might not be fully accurate neither (because the "strength" of
       * the line is going to be determined by how close to the pixel
       * center those 1 or 2 fragments are) but it's probably the best
       * we can do.
       */
      half_length = 2.0f * half_length;
   } else {
      half_length = half_length + 0.5f;
   }

   t_w = half_width;
   t_l = 0.5f;

   /* allocate/dup new verts */
   for (i = 0; i < 4; i++) {
      v[i] = dup_vert(stage, header->v[i/2], i);
   }

   /*
    * Quad strip for line from v0 to v1 (*=endpoints):
    *
    *  1                             3
    *  +-----------------------------+
    *  |                             |
    *  | *v0                     v1* |
    *  |                             |
    *  +-----------------------------+
    *  0                             2
    */

   /*
    * We increase line length by 0.5 pixels (at each endpoint),
    * and calculate the tri endpoints by moving them half-width
    * distance away perpendicular to the line.
    * XXX: since we change line endpoints (by 0.5 pixel), should
    * actually re-interpolate all other values?
    */

   /* new verts */
   pos = v[0]->data[posPos];
   pos[0] += (-t_l * c_a -  t_w * s_a);
   pos[1] += (-t_l * s_a +  t_w * c_a);

   pos = v[1]->data[posPos];
   pos[0] += (-t_l * c_a - -t_w * s_a);
   pos[1] += (-t_l * s_a + -t_w * c_a);

   pos = v[2]->data[posPos];
   pos[0] += (t_l * c_a -  t_w * s_a);
   pos[1] += (t_l * s_a +  t_w * c_a);

   pos = v[3]->data[posPos];
   pos[0] += (t_l * c_a - -t_w * s_a);
   pos[1] += (t_l * s_a + -t_w * c_a);

   /* new texcoords */
   tex = v[0]->data[coordPos];
   ASSIGN_4V(tex, -half_width, half_width, -half_length, half_length);

   tex = v[1]->data[coordPos];
   ASSIGN_4V(tex, half_width, half_width, -half_length, half_length);

   tex = v[2]->data[coordPos];
   ASSIGN_4V(tex, -half_width, half_width, half_length, half_length);

   tex = v[3]->data[coordPos];
   ASSIGN_4V(tex, half_width, half_width, half_length, half_length);

   tri.v[0] = v[2];  tri.v[1] = v[1];  tri.v[2] = v[0];
   stage->next->tri(stage->next, &tri);

   tri.v[0] = v[3];  tri.v[1] = v[1];  tri.v[2] = v[2];
   stage->next->tri(stage->next, &tri);
}


static void
aaline_first_line(struct draw_stage *stage, struct prim_header *header)
{
   auto struct aaline_stage *aaline = aaline_stage(stage);
   struct draw_context *draw = stage->draw;
   struct pipe_context *pipe = draw->pipe;
   const struct pipe_rasterizer_state *rast = draw->rasterizer;
   void *r;

   assert(draw->rasterizer->line_smooth);

   if (draw->rasterizer->line_width <= 1.0)
      aaline->half_line_width = 1.0;
   else
      aaline->half_line_width = 0.5f * draw->rasterizer->line_width + 0.5f;

   if (!draw->rasterizer->half_pixel_center)
      /*
       * The tex coords probably would need adjustments?
       */
      debug_printf("aa lines without half pixel center may be wrong\n");

   /*
    * Bind (generate) our fragprog
    */
   if (!bind_aaline_fragment_shader(aaline)) {
      stage->line = draw_pipe_passthrough_line;
      stage->line(stage, header);
      return;
   }

   draw_aaline_prepare_outputs(draw, draw->pipeline.aaline);

   draw->suspend_flushing = TRUE;

   /* Disable triangle culling, stippling, unfilled mode etc. */
   r = draw_get_rasterizer_no_cull(draw, rast->scissor, rast->flatshade);
   pipe->bind_rasterizer_state(pipe, r);

   draw->suspend_flushing = FALSE;

   /* now really draw first line */
   stage->line = aaline_line;
   stage->line(stage, header);
}


static void
aaline_flush(struct draw_stage *stage, unsigned flags)
{
   struct draw_context *draw = stage->draw;
   struct aaline_stage *aaline = aaline_stage(stage);
   struct pipe_context *pipe = draw->pipe;

   stage->line = aaline_first_line;
   stage->next->flush(stage->next, flags);

   /* restore original frag shader */
   draw->suspend_flushing = TRUE;
   aaline->driver_bind_fs_state(pipe, aaline->fs ? aaline->fs->driver_fs : NULL);

   /* restore original rasterizer state */
   if (draw->rast_handle) {
      pipe->bind_rasterizer_state(pipe, draw->rast_handle);
   }

   draw->suspend_flushing = FALSE;

   draw_remove_extra_vertex_attribs(draw);
}


static void
aaline_reset_stipple_counter(struct draw_stage *stage)
{
   stage->next->reset_stipple_counter(stage->next);
}


static void
aaline_destroy(struct draw_stage *stage)
{
   struct aaline_stage *aaline = aaline_stage(stage);
   struct pipe_context *pipe = stage->draw->pipe;

   draw_free_temp_verts(stage);

   /* restore the old entry points */
   pipe->create_fs_state = aaline->driver_create_fs_state;
   pipe->bind_fs_state = aaline->driver_bind_fs_state;
   pipe->delete_fs_state = aaline->driver_delete_fs_state;

   FREE(stage);
}


static struct aaline_stage *
draw_aaline_stage(struct draw_context *draw)
{
   struct aaline_stage *aaline = CALLOC_STRUCT(aaline_stage);
   if (!aaline)
      return NULL;

   aaline->stage.draw = draw;
   aaline->stage.name = "aaline";
   aaline->stage.next = NULL;
   aaline->stage.point = draw_pipe_passthrough_point;
   aaline->stage.line = aaline_first_line;
   aaline->stage.tri = draw_pipe_passthrough_tri;
   aaline->stage.flush = aaline_flush;
   aaline->stage.reset_stipple_counter = aaline_reset_stipple_counter;
   aaline->stage.destroy = aaline_destroy;

   if (!draw_alloc_temp_verts(&aaline->stage, 8))
      goto fail;

   return aaline;

 fail:
   aaline->stage.destroy(&aaline->stage);

   return NULL;
}


static struct aaline_stage *
aaline_stage_from_pipe(struct pipe_context *pipe)
{
   struct draw_context *draw = (struct draw_context *) pipe->draw;

   if (draw) {
      return aaline_stage(draw->pipeline.aaline);
   } else {
      return NULL;
   }
}


/**
 * This function overrides the driver's create_fs_state() function and
 * will typically be called by the state tracker.
 */
static void *
aaline_create_fs_state(struct pipe_context *pipe,
                       const struct pipe_shader_state *fs)
{
   struct aaline_stage *aaline = aaline_stage_from_pipe(pipe);
   struct aaline_fragment_shader *aafs = NULL;

   if (!aaline)
      return NULL;

   aafs = CALLOC_STRUCT(aaline_fragment_shader);

   if (!aafs)
      return NULL;

   aafs->state.tokens = tgsi_dup_tokens(fs->tokens);

   /* pass-through */
   aafs->driver_fs = aaline->driver_create_fs_state(pipe, fs);

   return aafs;
}


static void
aaline_bind_fs_state(struct pipe_context *pipe, void *fs)
{
   struct aaline_stage *aaline = aaline_stage_from_pipe(pipe);
   struct aaline_fragment_shader *aafs = (struct aaline_fragment_shader *) fs;

   if (!aaline) {
      return;
   }

   /* save current */
   aaline->fs = aafs;
   /* pass-through */
   aaline->driver_bind_fs_state(pipe, (aafs ? aafs->driver_fs : NULL));
}


static void
aaline_delete_fs_state(struct pipe_context *pipe, void *fs)
{
   struct aaline_stage *aaline = aaline_stage_from_pipe(pipe);
   struct aaline_fragment_shader *aafs = (struct aaline_fragment_shader *) fs;

   if (!aafs) {
      return;
   }

   if (aaline) {
      /* pass-through */
      aaline->driver_delete_fs_state(pipe, aafs->driver_fs);

      if (aafs->aaline_fs)
         aaline->driver_delete_fs_state(pipe, aafs->aaline_fs);
   }

   FREE((void*)aafs->state.tokens);
   FREE(aafs);
}


void
draw_aaline_prepare_outputs(struct draw_context *draw,
                            struct draw_stage *stage)
{
   struct aaline_stage *aaline = aaline_stage(stage);
   const struct pipe_rasterizer_state *rast = draw->rasterizer;

   /* update vertex attrib info */
   aaline->pos_slot = draw_current_shader_position_output(draw);

   if (!rast->line_smooth)
      return;

   /* allocate the extra post-transformed vertex attribute */
   aaline->coord_slot = draw_alloc_extra_vertex_attrib(draw,
                                                     TGSI_SEMANTIC_GENERIC,
                                                     aaline->fs->generic_attrib);
}

/**
 * Called by drivers that want to install this AA line prim stage
 * into the draw module's pipeline.  This will not be used if the
 * hardware has native support for AA lines.
 */
boolean
draw_install_aaline_stage(struct draw_context *draw, struct pipe_context *pipe)
{
   struct aaline_stage *aaline;

   pipe->draw = (void *) draw;

   /*
    * Create / install AA line drawing / prim stage
    */
   aaline = draw_aaline_stage(draw);
   if (!aaline)
      goto fail;

   /* save original driver functions */
   aaline->driver_create_fs_state = pipe->create_fs_state;
   aaline->driver_bind_fs_state = pipe->bind_fs_state;
   aaline->driver_delete_fs_state = pipe->delete_fs_state;

   /* override the driver's functions */
   pipe->create_fs_state = aaline_create_fs_state;
   pipe->bind_fs_state = aaline_bind_fs_state;
   pipe->delete_fs_state = aaline_delete_fs_state;

   /* Install once everything is known to be OK:
    */
   draw->pipeline.aaline = &aaline->stage;

   return TRUE;

fail:
   if (aaline)
      aaline->stage.destroy(&aaline->stage);

   return FALSE;
}