summaryrefslogtreecommitdiffstats
path: root/src/mesa/tnl/t_vb_program.c
blob: 0a959a04af16bccdf46acf637a8f42ecb4d0fb3c (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
/*
 * Mesa 3-D graphics library
 * Version:  6.5.3
 *
 * Copyright (C) 1999-2007  Brian Paul   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, 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 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
 * BRIAN PAUL 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.
 */


/**
 * \file tnl/t_vb_program.c
 * \brief Pipeline stage for executing NVIDIA vertex programs.
 * \author Brian Paul,  Keith Whitwell
 */


#include "glheader.h"
#include "context.h"
#include "macros.h"
#include "imports.h"
#include "prog_instruction.h"
#include "prog_statevars.h"
#include "prog_execute.h"

#include "tnl.h"
#include "t_context.h"
#include "t_pipeline.h"



/*!
 * Private storage for the vertex program pipeline stage.
 */
struct vp_stage_data {
   /** The results of running the vertex program go into these arrays. */
   GLvector4f results[VERT_RESULT_MAX];

   GLvector4f ndcCoords;              /**< normalized device coords */
   GLubyte *clipmask;                 /**< clip flags */
   GLubyte ormask, andmask;           /**< for clipping */
};


#define VP_STAGE_DATA(stage) ((struct vp_stage_data *)(stage->privatePtr))


/**
 * Initialize virtual machine state prior to executing vertex program.
 */
static void
init_machine(GLcontext *ctx, struct gl_program_machine *machine)
{
   /* Input registers get initialized from the current vertex attribs */
   MEMCPY(machine->VertAttribs, ctx->Current.Attrib,
          MAX_VERTEX_PROGRAM_ATTRIBS * 4 * sizeof(GLfloat));

   if (ctx->VertexProgram.Current->IsNVProgram) {
      GLuint i;
      /* Output/result regs are initialized to [0,0,0,1] */
      for (i = 0; i < MAX_NV_VERTEX_PROGRAM_OUTPUTS; i++) {
         ASSIGN_4V(machine->Outputs[i], 0.0F, 0.0F, 0.0F, 1.0F);
      }
      /* Temp regs are initialized to [0,0,0,0] */
      for (i = 0; i < MAX_NV_VERTEX_PROGRAM_TEMPS; i++) {
         ASSIGN_4V(machine->Temporaries[i], 0.0F, 0.0F, 0.0F, 0.0F);
      }
      for (i = 0; i < MAX_VERTEX_PROGRAM_ADDRESS_REGS; i++) {
         ASSIGN_4V(machine->AddressReg[i], 0, 0, 0, 0);
      }
   }

   /* init condition codes */
   machine->CondCodes[0] = COND_EQ;
   machine->CondCodes[1] = COND_EQ;
   machine->CondCodes[2] = COND_EQ;
   machine->CondCodes[3] = COND_EQ;

   /* init call stack */
   machine->StackDepth = 0;
}


/**
 * Copy the 16 elements of a matrix into four consecutive program
 * registers starting at 'pos'.
 */
static void
load_matrix(GLfloat registers[][4], GLuint pos, const GLfloat mat[16])
{
   GLuint i;
   for (i = 0; i < 4; i++) {
      registers[pos + i][0] = mat[0 + i];
      registers[pos + i][1] = mat[4 + i];
      registers[pos + i][2] = mat[8 + i];
      registers[pos + i][3] = mat[12 + i];
   }
}


/**
 * As above, but transpose the matrix.
 */
static void
load_transpose_matrix(GLfloat registers[][4], GLuint pos,
                      const GLfloat mat[16])
{
   MEMCPY(registers[pos], mat, 16 * sizeof(GLfloat));
}


/**
 * Load current vertex program's parameter registers with tracked
 * matrices (if NV program).  This only needs to be done per
 * glBegin/glEnd, not per-vertex.
 */
void
_mesa_load_tracked_matrices(GLcontext *ctx)
{
   GLuint i;

   for (i = 0; i < MAX_NV_VERTEX_PROGRAM_PARAMS / 4; i++) {
      /* point 'mat' at source matrix */
      GLmatrix *mat;
      if (ctx->VertexProgram.TrackMatrix[i] == GL_MODELVIEW) {
         mat = ctx->ModelviewMatrixStack.Top;
      }
      else if (ctx->VertexProgram.TrackMatrix[i] == GL_PROJECTION) {
         mat = ctx->ProjectionMatrixStack.Top;
      }
      else if (ctx->VertexProgram.TrackMatrix[i] == GL_TEXTURE) {
         mat = ctx->TextureMatrixStack[ctx->Texture.CurrentUnit].Top;
      }
      else if (ctx->VertexProgram.TrackMatrix[i] == GL_COLOR) {
         mat = ctx->ColorMatrixStack.Top;
      }
      else if (ctx->VertexProgram.TrackMatrix[i]==GL_MODELVIEW_PROJECTION_NV) {
         /* XXX verify the combined matrix is up to date */
         mat = &ctx->_ModelProjectMatrix;
      }
      else if (ctx->VertexProgram.TrackMatrix[i] >= GL_MATRIX0_NV &&
               ctx->VertexProgram.TrackMatrix[i] <= GL_MATRIX7_NV) {
         GLuint n = ctx->VertexProgram.TrackMatrix[i] - GL_MATRIX0_NV;
         ASSERT(n < MAX_PROGRAM_MATRICES);
         mat = ctx->ProgramMatrixStack[n].Top;
      }
      else {
         /* no matrix is tracked, but we leave the register values as-is */
         assert(ctx->VertexProgram.TrackMatrix[i] == GL_NONE);
         continue;
      }

      /* load the matrix values into sequential registers */
      if (ctx->VertexProgram.TrackMatrixTransform[i] == GL_IDENTITY_NV) {
         load_matrix(ctx->VertexProgram.Parameters, i*4, mat->m);
      }
      else if (ctx->VertexProgram.TrackMatrixTransform[i] == GL_INVERSE_NV) {
         _math_matrix_analyse(mat); /* update the inverse */
         ASSERT(!_math_matrix_is_dirty(mat));
         load_matrix(ctx->VertexProgram.Parameters, i*4, mat->inv);
      }
      else if (ctx->VertexProgram.TrackMatrixTransform[i] == GL_TRANSPOSE_NV) {
         load_transpose_matrix(ctx->VertexProgram.Parameters, i*4, mat->m);
      }
      else {
         assert(ctx->VertexProgram.TrackMatrixTransform[i]
                == GL_INVERSE_TRANSPOSE_NV);
         _math_matrix_analyse(mat); /* update the inverse */
         ASSERT(!_math_matrix_is_dirty(mat));
         load_transpose_matrix(ctx->VertexProgram.Parameters, i*4, mat->inv);
      }
   }
}


/**
 * This function executes vertex programs
 */
static GLboolean
run_vp( GLcontext *ctx, struct tnl_pipeline_stage *stage )
{
   TNLcontext *tnl = TNL_CONTEXT(ctx);
   struct vp_stage_data *store = VP_STAGE_DATA(stage);
   struct vertex_buffer *VB = &tnl->vb;
   struct gl_vertex_program *program = ctx->VertexProgram._Current;
   struct gl_program_machine machine;
   GLuint outputs[VERT_RESULT_MAX], numOutputs;
   GLuint i, j;

#define FORCE_PROG_EXECUTE_C 1
#if FORCE_PROG_EXECUTE_C
   if (!program)
      return GL_TRUE;
#else
   if (!program || !program->IsNVProgram)
      return GL_TRUE;
#endif

   if (ctx->VertexProgram.Current->IsNVProgram) {
      _mesa_load_tracked_matrices(ctx);
   }
   else {
      _mesa_load_state_parameters(ctx, program->Base.Parameters);
   }

   numOutputs = 0;
   for (i = 0; i < VERT_RESULT_MAX; i++) {
      if (program->Base.OutputsWritten & (1 << i)) {
         outputs[numOutputs++] = i;
      }
   }

   for (i = 0; i < VB->Count; i++) {
      GLuint attr;

      init_machine(ctx, &machine);

#if 0
      printf("Input  %d: %f, %f, %f, %f\n", i,
             VB->AttribPtr[0]->data[i][0],
             VB->AttribPtr[0]->data[i][1],
             VB->AttribPtr[0]->data[i][2],
             VB->AttribPtr[0]->data[i][3]);
      printf("   color: %f, %f, %f, %f\n",
             VB->AttribPtr[3]->data[i][0],
             VB->AttribPtr[3]->data[i][1],
             VB->AttribPtr[3]->data[i][2],
             VB->AttribPtr[3]->data[i][3]);
      printf("  normal: %f, %f, %f, %f\n",
             VB->AttribPtr[2]->data[i][0],
             VB->AttribPtr[2]->data[i][1],
             VB->AttribPtr[2]->data[i][2],
             VB->AttribPtr[2]->data[i][3]);
#endif

      /* the vertex array case */
      for (attr = 0; attr < VERT_ATTRIB_MAX; attr++) {
	 if (program->Base.InputsRead & (1 << attr)) {
	    const GLubyte *ptr = (const GLubyte*) VB->AttribPtr[attr]->data;
	    const GLuint size = VB->AttribPtr[attr]->size;
	    const GLuint stride = VB->AttribPtr[attr]->stride;
	    const GLfloat *data = (GLfloat *) (ptr + stride * i);
	    COPY_CLEAN_4V(machine.VertAttribs[attr], size, data);
	 }
      }

      /* execute the program */
      _mesa_execute_program(ctx, &program->Base, &machine);

      /* Fixup fog an point size results if needed */
      if (ctx->Fog.Enabled &&
          (program->Base.OutputsWritten & (1 << VERT_RESULT_FOGC)) == 0) {
         machine.Outputs[VERT_RESULT_FOGC][0] = 1.0;
      }

      if (ctx->VertexProgram.PointSizeEnabled &&
          (program->Base.OutputsWritten & (1 << VERT_RESULT_PSIZ)) == 0) {
         machine.Outputs[VERT_RESULT_PSIZ][0] = ctx->Point.Size;
      }

      /* copy the output registers into the VB->attribs arrays */
      for (j = 0; j < numOutputs; j++) {
         const GLuint attr = outputs[j];
         COPY_4V(store->results[attr].data[i], machine.Outputs[attr]);
      }
#if 0
      printf("HPOS: %f %f %f %f\n",
             machine.Outputs[0][0], 
             machine.Outputs[0][1], 
             machine.Outputs[0][2], 
             machine.Outputs[0][3]);
#endif
   }

   /* Setup the VB pointers so that the next pipeline stages get
    * their data from the right place (the program output arrays).
    */
   VB->ClipPtr = &store->results[VERT_RESULT_HPOS];
   VB->ClipPtr->size = 4;
   VB->ClipPtr->count = VB->Count;
   VB->ColorPtr[0] = &store->results[VERT_RESULT_COL0];
   VB->ColorPtr[1] = &store->results[VERT_RESULT_BFC0];
   VB->SecondaryColorPtr[0] = &store->results[VERT_RESULT_COL1];
   VB->SecondaryColorPtr[1] = &store->results[VERT_RESULT_BFC1];
   VB->FogCoordPtr = &store->results[VERT_RESULT_FOGC];

   VB->AttribPtr[VERT_ATTRIB_COLOR0] = &store->results[VERT_RESULT_COL0];
   VB->AttribPtr[VERT_ATTRIB_COLOR1] = &store->results[VERT_RESULT_COL1];
   VB->AttribPtr[VERT_ATTRIB_FOG] = &store->results[VERT_RESULT_FOGC];
   VB->AttribPtr[_TNL_ATTRIB_POINTSIZE] = &store->results[VERT_RESULT_PSIZ];

   for (i = 0; i < ctx->Const.MaxTextureCoordUnits; i++) {
      VB->TexCoordPtr[i] = 
      VB->AttribPtr[_TNL_ATTRIB_TEX0 + i]
         = &store->results[VERT_RESULT_TEX0 + i];
   }

   for (i = 0; i < ctx->Const.MaxVarying; i++) {
      if (program->Base.OutputsWritten & (1 << (VERT_RESULT_VAR0 + i))) {
         /* Note: varying results get put into the generic attributes */
	 VB->AttribPtr[VERT_ATTRIB_GENERIC0+i]
            = &store->results[VERT_RESULT_VAR0 + i];
      }
   }

   /* Cliptest and perspective divide.  Clip functions must clear
    * the clipmask.
    */
   store->ormask = 0;
   store->andmask = CLIP_FRUSTUM_BITS;

   if (tnl->NeedNdcCoords) {
      VB->NdcPtr =
         _mesa_clip_tab[VB->ClipPtr->size]( VB->ClipPtr,
                                            &store->ndcCoords,
                                            store->clipmask,
                                            &store->ormask,
                                            &store->andmask );
   }
   else {
      VB->NdcPtr = NULL;
      _mesa_clip_np_tab[VB->ClipPtr->size]( VB->ClipPtr,
                                            NULL,
                                            store->clipmask,
                                            &store->ormask,
                                            &store->andmask );
   }

   if (store->andmask)  /* All vertices are outside the frustum */
      return GL_FALSE;


   /* This is where we'd do clip testing against the user-defined
    * clipping planes, but they're not supported by vertex programs.
    */

   VB->ClipOrMask = store->ormask;
   VB->ClipMask = store->clipmask;

   return GL_TRUE;
}


/**
 * Called the first time stage->run is called.  In effect, don't
 * allocate data until the first time the stage is run.
 */
static GLboolean init_vp( GLcontext *ctx,
			  struct tnl_pipeline_stage *stage )
{
   TNLcontext *tnl = TNL_CONTEXT(ctx);
   struct vertex_buffer *VB = &(tnl->vb);
   struct vp_stage_data *store;
   const GLuint size = VB->Size;
   GLuint i;

   stage->privatePtr = MALLOC(sizeof(*store));
   store = VP_STAGE_DATA(stage);
   if (!store)
      return GL_FALSE;

   /* Allocate arrays of vertex output values */
   for (i = 0; i < VERT_RESULT_MAX; i++) {
      _mesa_vector4f_alloc( &store->results[i], 0, size, 32 );
      store->results[i].size = 4;
   }

   /* a few other misc allocations */
   _mesa_vector4f_alloc( &store->ndcCoords, 0, size, 32 );
   store->clipmask = (GLubyte *) ALIGN_MALLOC(sizeof(GLubyte)*size, 32 );

   return GL_TRUE;
}


/**
 * Destructor for this pipeline stage.
 */
static void dtr( struct tnl_pipeline_stage *stage )
{
   struct vp_stage_data *store = VP_STAGE_DATA(stage);

   if (store) {
      GLuint i;

      /* free the vertex program result arrays */
      for (i = 0; i < VERT_RESULT_MAX; i++)
         _mesa_vector4f_free( &store->results[i] );

      /* free misc arrays */
      _mesa_vector4f_free( &store->ndcCoords );
      ALIGN_FREE( store->clipmask );

      FREE( store );
      stage->privatePtr = NULL;
   }
}


/**
 * Public description of this pipeline stage.
 */
const struct tnl_pipeline_stage _tnl_vertex_program_stage =
{
   "vertex-program",
   NULL,			/* private_data */
   init_vp,			/* create */
   dtr,				/* destroy */
   NULL, 			/* validate */
   run_vp			/* run -- initially set to ctr */
};