summaryrefslogtreecommitdiff
path: root/src/mesa/drivers/dri/savage/savagerender.c
blob: c369bb124c20606ad9e5f3b84273a140b7a50517 (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
/*
 * Copyright 2005  Felix Kuehling
 * 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 FELIX KUEHLING 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.
 */

/*
 * Render unclipped vertex buffers by emitting vertices directly to
 * dma buffers.  Use strip/fan hardware primitives where possible.
 * Simulate missing primitives with indexed vertices.
 */
#include "main/glheader.h"
#include "main/context.h"
#include "main/macros.h"
#include "main/imports.h"
#include "main/mtypes.h"

#include "tnl/t_context.h"

#include "savagecontext.h"
#include "savagestate.h"
#include "savageioctl.h"

/*
 * Standard render tab for Savage4 and smooth shading on Savage3D
 */
#define HAVE_POINTS      0
#define HAVE_LINES       0
#define HAVE_LINE_STRIPS 0
#define HAVE_TRIANGLES   1
#define HAVE_TRI_STRIPS  1
#define HAVE_TRI_STRIP_1 0
#define HAVE_TRI_FANS    1
#define HAVE_POLYGONS    0
#define HAVE_QUADS       0
#define HAVE_QUAD_STRIPS 0

#define HAVE_ELTS        1

#define LOCAL_VARS savageContextPtr imesa = SAVAGE_CONTEXT(ctx) 
#define INIT( prim ) do {						\
   if (0) fprintf(stderr, "%s\n", __FUNCTION__);			\
   savageFlushVertices(imesa);						\
   switch (prim) {							\
   case GL_TRIANGLES:	   imesa->HwPrim = SAVAGE_PRIM_TRILIST; break;	\
   case GL_TRIANGLE_STRIP: imesa->HwPrim = SAVAGE_PRIM_TRISTRIP; break;	\
   case GL_TRIANGLE_FAN:   imesa->HwPrim = SAVAGE_PRIM_TRIFAN; break;	\
   }									\
} while (0)
#define FLUSH()		savageFlushElts(imesa), savageFlushVertices(imesa)

#define GET_CURRENT_VB_MAX_VERTS() \
   ((imesa->bufferSize/4 - imesa->vtxBuf->used) / imesa->HwVertexSize)
#define GET_SUBSEQUENT_VB_MAX_VERTS() \
   (imesa->bufferSize/4 / imesa->HwVertexSize)

#define ALLOC_VERTS( nr ) \
	savageAllocVtxBuf( imesa, (nr) * imesa->HwVertexSize )
#define EMIT_VERTS( ctx, j, nr, buf ) \
	_tnl_emit_vertices_to_buffer(ctx, j, (j)+(nr), buf )

#define ELTS_VARS( buf ) GLushort *dest = buf, firstElt = imesa->firstElt
#define ELT_INIT( prim ) INIT(prim)

/* (size - used - 1 qword for drawing command) * 4 elts per qword */
#define GET_CURRENT_VB_MAX_ELTS() \
   ((imesa->cmdBuf.size - (imesa->cmdBuf.write - imesa->cmdBuf.base) - 1)*4)
/* (size - space for initial state - 1 qword for drawing command) * 4 elts
 * imesa is not defined in validate_render :( */
#define GET_SUBSEQUENT_VB_MAX_ELTS()					\
   ((SAVAGE_CONTEXT(ctx)->cmdBuf.size - 				\
     (SAVAGE_CONTEXT(ctx)->cmdBuf.start - 				\
      SAVAGE_CONTEXT(ctx)->cmdBuf.base) - 1)*4)

#define ALLOC_ELTS(nr) savageAllocElts(imesa, nr)
#define EMIT_ELT(offset, x) do {					\
   (dest)[offset] = (GLushort) ((x)+firstElt);				\
} while (0)
#define EMIT_TWO_ELTS(offset, x, y) do {				\
   *(GLuint *)(dest + offset) = (((y)+firstElt) << 16) |		\
				((x)+firstElt);				\
} while (0)

#define INCR_ELTS( nr ) dest += nr
#define ELTPTR dest
#define RELEASE_ELT_VERTS() \
   savageReleaseIndexedVerts(imesa)

#define EMIT_INDEXED_VERTS( ctx, start, count ) do {			\
   GLuint *buf = savageAllocIndexedVerts(imesa, count-start);		\
   EMIT_VERTS(ctx, start, count-start, buf);				\
} while (0)

#define TAG(x) savage_##x
#include "tnl_dd/t_dd_dmatmp.h"

/*
 * On Savage3D triangle fans and strips are broken with flat
 * shading. With triangles it wants the color for flat shading in the
 * first vertex! So we make another template instance which uses
 * triangles only (with reordered vertices: SAVAGE_PRIM_TRILIST_201).
 * The reordering is done by the DRM.
 */
#undef  HAVE_TRI_STRIPS
#undef  HAVE_TRI_FANS
#define HAVE_TRI_STRIPS	0
#define HAVE_TRI_FANS	0

#undef  INIT
#define INIT( prim ) do {						\
   if (0) fprintf(stderr, "%s\n", __FUNCTION__);			\
   savageFlushVertices(imesa);						\
   imesa->HwPrim = SAVAGE_PRIM_TRILIST_201;				\
} while(0)

#undef  TAG
#define TAG(x) savage_flat_##x##_s3d
#include "tnl_dd/t_dd_dmatmp.h"


/**********************************************************************/
/*                          Render pipeline stage                     */
/**********************************************************************/

static GLboolean savage_run_render( GLcontext *ctx,
				    struct tnl_pipeline_stage *stage )
{
   savageContextPtr imesa = SAVAGE_CONTEXT(ctx);
   TNLcontext *tnl = TNL_CONTEXT(ctx);
   struct vertex_buffer *VB = &tnl->vb; 
   tnl_render_func *tab, *tab_elts;
   GLboolean valid;
   GLuint i;

   if (savageHaveIndexedVerts(imesa))
      savageReleaseIndexedVerts(imesa);

   if (imesa->savageScreen->chipset < S3_SAVAGE4 &&
       (ctx->_TriangleCaps & DD_FLATSHADE)) {
      tab = savage_flat_render_tab_verts_s3d;
      tab_elts = savage_flat_render_tab_elts_s3d;
      valid = savage_flat_validate_render_s3d( ctx, VB );
   } else {
      tab = savage_render_tab_verts;
      tab_elts = savage_render_tab_elts;
      valid = savage_validate_render( ctx, VB );
   }

   /* Don't handle clipping or vertex manipulations.
    */
   if (imesa->RenderIndex != 0 || !valid) {
      return GL_TRUE;
   }
   
   tnl->Driver.Render.Start( ctx );
   /* Check RenderIndex again. The ptexHack is detected late in RenderStart.
    * Also check for ptex fallbacks detected late.
    */
   if (imesa->RenderIndex != 0 || imesa->Fallback != 0) {
      return GL_TRUE;
   }

   /* setup for hardware culling */
   imesa->raster_primitive = GL_TRIANGLES;
   imesa->new_state |= SAVAGE_NEW_CULL;

   /* update and emit state */
   savageDDUpdateHwState(ctx);
   savageEmitChangedState(imesa);

   if (VB->Elts) {
      tab = tab_elts;
      if (!savageHaveIndexedVerts(imesa)) {
	 if (VB->Count > GET_SUBSEQUENT_VB_MAX_VERTS())
	    return GL_TRUE;
	 EMIT_INDEXED_VERTS(ctx, 0, VB->Count);
      }
   }

   for (i = 0 ; i < VB->PrimitiveCount ; i++)
   {
      GLuint prim = _tnl_translate_prim(&VB->Primitive[i]);
      GLuint start = VB->Primitive[i].start;
      GLuint length = VB->Primitive[i].count;

      if (length)
	 tab[prim & PRIM_MODE_MASK]( ctx, start, start+length, prim);
   }

   tnl->Driver.Render.Finish( ctx );

   return GL_FALSE;		/* finished the pipe */
}

struct tnl_pipeline_stage _savage_render_stage = 
{ 
   "savage render",
   NULL,
   NULL,
   NULL,
   NULL,
   savage_run_render		/* run */
};


/**********************************************************************/
/*         Pipeline stage for texture coordinate normalization        */
/**********************************************************************/
struct texnorm_stage_data {
   GLboolean active;
   GLvector4f texcoord[MAX_TEXTURE_UNITS];
};

#define TEXNORM_STAGE_DATA(stage) ((struct texnorm_stage_data *)stage->privatePtr)


static GLboolean run_texnorm_stage( GLcontext *ctx,
				    struct tnl_pipeline_stage *stage )
{
   struct texnorm_stage_data *store = TEXNORM_STAGE_DATA(stage);
   savageContextPtr imesa = SAVAGE_CONTEXT(ctx);
   TNLcontext *tnl = TNL_CONTEXT(ctx);
   struct vertex_buffer *VB = &tnl->vb;
   GLuint i;

   if (imesa->Fallback || !store->active)
      return GL_TRUE;

   for (i = 0 ; i < ctx->Const.MaxTextureUnits ; i++) {
      const GLbitfield reallyEnabled = ctx->Texture.Unit[i]._ReallyEnabled;
      if (reallyEnabled) {
         const struct gl_texture_object *texObj = ctx->Texture.Unit[i]._Current;
         const GLboolean normalizeS = (texObj->WrapS == GL_REPEAT);
         const GLboolean normalizeT = (reallyEnabled & TEXTURE_2D_BIT) &&
            (texObj->WrapT == GL_REPEAT);
         const GLfloat *in = (GLfloat *)VB->AttribPtr[_TNL_ATTRIB_TEX0 + i]->data;
         const GLint instride = VB->AttribPtr[_TNL_ATTRIB_TEX0 + i]->stride;
         GLfloat (*out)[4] = store->texcoord[i].data;
         GLint j;

         if (!ctx->Texture.Unit[i]._ReallyEnabled ||
             VB->AttribPtr[_TNL_ATTRIB_TEX0 + i]->size == 4)
            /* Never try to normalize homogenous tex coords! */
            continue;

         if (normalizeS && normalizeT) {
            /* take first texcoords as rough estimate of mean value */
            GLfloat correctionS = -floor(in[0]+0.5);
            GLfloat correctionT = -floor(in[1]+0.5);
            for (j = 0; j < VB->Count; ++j) {
               out[j][0] = in[0] + correctionS;
               out[j][1] = in[1] + correctionT;
               in = (GLfloat *)((GLubyte *)in + instride);
            }
         } else if (normalizeS) {
            /* take first texcoords as rough estimate of mean value */
            GLfloat correctionS = -floor(in[0]+0.5);
            if (reallyEnabled & TEXTURE_2D_BIT) {
               for (j = 0; j < VB->Count; ++j) {
                  out[j][0] = in[0] + correctionS;
                  out[j][1] = in[1];
                  in = (GLfloat *)((GLubyte *)in + instride);
               }
            } else {
               for (j = 0; j < VB->Count; ++j) {
                  out[j][0] = in[0] + correctionS;
                  in = (GLfloat *)((GLubyte *)in + instride);
               }
            }
         } else if (normalizeT) {
            /* take first texcoords as rough estimate of mean value */
            GLfloat correctionT = -floor(in[1]+0.5);
            for (j = 0; j < VB->Count; ++j) {
               out[j][0] = in[0];
               out[j][1] = in[1] + correctionT;
               in = (GLfloat *)((GLubyte *)in + instride);
            }
         }

         if (normalizeS || normalizeT)
            VB->AttribPtr[_TNL_ATTRIB_TEX0 + i] = &store->texcoord[i];
      }
   }

   return GL_TRUE;
}

/* Called the first time stage->run() is invoked.
 */
static GLboolean alloc_texnorm_data( GLcontext *ctx,
				     struct tnl_pipeline_stage *stage )
{
   struct vertex_buffer *VB = &TNL_CONTEXT(ctx)->vb;
   struct texnorm_stage_data *store;
   GLuint i;

   stage->privatePtr = CALLOC(sizeof(*store));
   store = TEXNORM_STAGE_DATA(stage);
   if (!store)
      return GL_FALSE;

   for (i = 0 ; i < ctx->Const.MaxTextureUnits ; i++)
      _mesa_vector4f_alloc( &store->texcoord[i], 0, VB->Size, 32 );
   
   return GL_TRUE;
}

static void validate_texnorm( GLcontext *ctx,
			      struct tnl_pipeline_stage *stage )
{
   struct texnorm_stage_data *store = TEXNORM_STAGE_DATA(stage);
   GLuint flags = 0;

   if (((ctx->Texture.Unit[0]._ReallyEnabled & (TEXTURE_1D_BIT|TEXTURE_2D_BIT)) &&
	(ctx->Texture.Unit[0]._Current->WrapS == GL_REPEAT)) ||
       ((ctx->Texture.Unit[0]._ReallyEnabled & TEXTURE_2D_BIT) &&
	(ctx->Texture.Unit[0]._Current->WrapT == GL_REPEAT)))
      flags |= VERT_BIT_TEX0;

   if (((ctx->Texture.Unit[1]._ReallyEnabled & (TEXTURE_1D_BIT|TEXTURE_2D_BIT)) &&
	(ctx->Texture.Unit[1]._Current->WrapS == GL_REPEAT)) ||
       ((ctx->Texture.Unit[1]._ReallyEnabled & TEXTURE_2D_BIT) &&
	(ctx->Texture.Unit[1]._Current->WrapT == GL_REPEAT)))
      flags |= VERT_BIT_TEX1;

   store->active = (flags != 0);
}

static void free_texnorm_data( struct tnl_pipeline_stage *stage )
{
   struct texnorm_stage_data *store = TEXNORM_STAGE_DATA(stage);
   GLuint i;

   if (store) {
      for (i = 0 ; i < MAX_TEXTURE_UNITS ; i++)
	 if (store->texcoord[i].data)
	    _mesa_vector4f_free( &store->texcoord[i] );
      FREE( store );
      stage->privatePtr = 0;
   }
}

struct tnl_pipeline_stage _savage_texnorm_stage =
{
   "savage texture coordinate normalization stage", /* name */
   NULL,				/* private data */
   alloc_texnorm_data,			/* run -- initially set to init */
   free_texnorm_data,			/* destructor */
   validate_texnorm,
   run_texnorm_stage
};