summaryrefslogtreecommitdiff
path: root/src/mesa/drivers/dri/r600/r700_render.c
blob: 58b4491d00dd6cb3e5f86c9564984eab5060292f (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
/*
 * Copyright (C) 2008-2009  Advanced Micro Devices, Inc.
 *
 * 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
 * THE COPYRIGHT HOLDER(S) 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:
 *   Richard Li <RichardZ.Li@amd.com>, <richardradeon@gmail.com>
 *   CooperYuan <cooper.yuan@amd.com>, <cooperyuan@gmail.com>
 */

#include "main/glheader.h"
#include "main/state.h"
#include "main/imports.h"
#include "main/enums.h"
#include "main/macros.h"
#include "main/context.h"
#include "main/dd.h"
#include "main/simple_list.h"
#include "main/api_arrayelt.h"
#include "swrast/swrast.h"
#include "swrast_setup/swrast_setup.h"
#include "vbo/vbo.h"

#include "tnl/tnl.h"
#include "tnl/t_vp_build.h"
#include "tnl/t_context.h"
#include "tnl/t_vertex.h"
#include "tnl/t_pipeline.h"

#include "r600_context.h"
#include "r600_cmdbuf.h"

#include "r600_tex.h"

#include "r700_vertprog.h"
#include "r700_fragprog.h"
#include "r700_state.h"

void r700WaitForIdle(context_t *context);
void r700WaitForIdleClean(context_t *context);
void r700Start3D(context_t *context);
GLboolean r700SendTextureState(context_t *context);
static unsigned int r700PrimitiveType(int prim);
void r600UpdateTextureState(GLcontext * ctx);
GLboolean r700SyncSurf(context_t *context,
		       struct radeon_bo *pbo,
		       uint32_t read_domain,
		       uint32_t write_domain,
		       uint32_t sync_type);

void r700WaitForIdle(context_t *context)
{
    BATCH_LOCALS(&context->radeon);
    BEGIN_BATCH_NO_AUTOSTATE(3);

    R600_OUT_BATCH(CP_PACKET3(R600_IT_SET_CONFIG_REG, 1));
    R600_OUT_BATCH(mmWAIT_UNTIL - ASIC_CONFIG_BASE_INDEX);
    R600_OUT_BATCH(WAIT_3D_IDLE_bit);

    END_BATCH();
    COMMIT_BATCH();
}

void r700WaitForIdleClean(context_t *context)
{
    BATCH_LOCALS(&context->radeon);
    BEGIN_BATCH_NO_AUTOSTATE(5);

    R600_OUT_BATCH(CP_PACKET3(R600_IT_EVENT_WRITE, 0));
    R600_OUT_BATCH(CACHE_FLUSH_AND_INV_EVENT);

    R600_OUT_BATCH(CP_PACKET3(R600_IT_SET_CONFIG_REG, 1));
    R600_OUT_BATCH(mmWAIT_UNTIL - ASIC_CONFIG_BASE_INDEX);
    R600_OUT_BATCH(WAIT_3D_IDLE_bit | WAIT_3D_IDLECLEAN_bit);

    END_BATCH();
    COMMIT_BATCH();
}

void r700Start3D(context_t *context)
{
    BATCH_LOCALS(&context->radeon);
    if (context->radeon.radeonScreen->chip_family < CHIP_FAMILY_RV770)
    {
        BEGIN_BATCH_NO_AUTOSTATE(2);
        R600_OUT_BATCH(CP_PACKET3(R600_IT_START_3D_CMDBUF, 0));
        R600_OUT_BATCH(0);
        END_BATCH();
    }

    BEGIN_BATCH_NO_AUTOSTATE(3);
    R600_OUT_BATCH(CP_PACKET3(R600_IT_CONTEXT_CONTROL, 1));
    R600_OUT_BATCH(0x80000000);
    R600_OUT_BATCH(0x80000000);
    END_BATCH();

    COMMIT_BATCH();

    r700WaitForIdleClean(context);
}

static GLboolean r700SetupShaders(GLcontext * ctx)
{
    context_t *context = R700_CONTEXT(ctx);

    R700_CHIP_CONTEXT *r700 = (R700_CHIP_CONTEXT*)(&context->hw);

    GLuint exportCount;

    r700->ps.SQ_PGM_RESOURCES_PS.u32All = 0;
    r700->vs.SQ_PGM_RESOURCES_VS.u32All = 0;

    SETbit(r700->ps.SQ_PGM_RESOURCES_PS.u32All, PGM_RESOURCES__PRIME_CACHE_ON_DRAW_bit);
    SETbit(r700->vs.SQ_PGM_RESOURCES_VS.u32All, PGM_RESOURCES__PRIME_CACHE_ON_DRAW_bit);

    r700SetupVertexProgram(ctx);

    r700SetupFragmentProgram(ctx);

    exportCount = (r700->ps.SQ_PGM_EXPORTS_PS.u32All & EXPORT_MODE_mask) / (1 << EXPORT_MODE_shift);
    r700->CB_SHADER_CONTROL.u32All = (1 << exportCount) - 1;

    r600UpdateTextureState(ctx);

    r700SendFSState(context); // FIXME just a place holder for now
    r700SendPSState(context);
    r700SendVSState(context);

    r700SendTextureState(context);
    r700SetupStreams(ctx);

    return GL_TRUE;
}

GLboolean r700SyncSurf(context_t *context,
		       struct radeon_bo *pbo,
		       uint32_t read_domain,
		       uint32_t write_domain,
		       uint32_t sync_type)
{
    BATCH_LOCALS(&context->radeon);
    uint32_t cp_coher_size;

    if (!pbo)
	    return GL_FALSE;

    if (pbo->size == 0xffffffff)
	    cp_coher_size = 0xffffffff;
    else
	    cp_coher_size = ((pbo->size + 255) >> 8);

    BEGIN_BATCH_NO_AUTOSTATE(5 + 2);
    R600_OUT_BATCH(CP_PACKET3(R600_IT_SURFACE_SYNC, 3));
    R600_OUT_BATCH(sync_type);
    R600_OUT_BATCH(cp_coher_size);
    R600_OUT_BATCH(0);
    R600_OUT_BATCH(10);
    R600_OUT_BATCH_RELOC(0,
			 pbo,
			 0,
			 read_domain, write_domain, 0); // ???

    END_BATCH();
    COMMIT_BATCH();

    return GL_TRUE;
}

static unsigned int r700PrimitiveType(int prim)
{
    switch (prim & PRIM_MODE_MASK)
    {
    case GL_POINTS:
        return DI_PT_POINTLIST;
        break;
    case GL_LINES:
        return DI_PT_LINELIST;
        break;
    case GL_LINE_STRIP:
        return DI_PT_LINESTRIP;
        break;
    case GL_LINE_LOOP:
        return DI_PT_LINELOOP;
        break;
    case GL_TRIANGLES:
        return DI_PT_TRILIST;
        break;
    case GL_TRIANGLE_STRIP:
        return DI_PT_TRISTRIP;
        break;
    case GL_TRIANGLE_FAN:
        return DI_PT_TRIFAN;
        break;
    case GL_QUADS:
        return DI_PT_QUADLIST;
        break;
    case GL_QUAD_STRIP:
        return DI_PT_QUADSTRIP;
        break;
    case GL_POLYGON:
        return DI_PT_POLYGON;
        break;
    default:
        assert(0);
        return -1;
        break;
    }
}

static int r700NumVerts(int num_verts, int prim)
{
	int verts_off = 0;

	switch (prim & PRIM_MODE_MASK) {
	case GL_POINTS:
		verts_off = 0;
		break;
	case GL_LINES:
		verts_off = num_verts % 2;
		break;
	case GL_LINE_STRIP:
		if (num_verts < 2)
			verts_off = num_verts;
		break;
	case GL_LINE_LOOP:
		if (num_verts < 2)
			verts_off = num_verts;
		break;
	case GL_TRIANGLES:
		verts_off = num_verts % 3;
		break;
	case GL_TRIANGLE_STRIP:
		if (num_verts < 3)
			verts_off = num_verts;
		break;
	case GL_TRIANGLE_FAN:
		if (num_verts < 3)
			verts_off = num_verts;
		break;
	case GL_QUADS:
		verts_off = num_verts % 4;
		break;
	case GL_QUAD_STRIP:
		if (num_verts < 4)
			verts_off = num_verts;
		else
			verts_off = num_verts % 2;
		break;
	case GL_POLYGON:
		if (num_verts < 3)
			verts_off = num_verts;
		break;
	default:
		assert(0);
		return -1;
		break;
	}

	return num_verts - verts_off;
}

static void r700RunRenderPrimitive(GLcontext * ctx, int start, int end, int prim)
{
	context_t *context = R700_CONTEXT(ctx);
	BATCH_LOCALS(&context->radeon);
	int type, i, total_emit;
	int num_indices;
	uint32_t vgt_draw_initiator = 0;
	uint32_t vgt_index_type     = 0;
	uint32_t vgt_primitive_type = 0;
	uint32_t vgt_num_indices    = 0;

	type = r700PrimitiveType(prim);
	num_indices = r700NumVerts(end - start, prim);

	if (type < 0 || num_indices <= 0)
		return;

        total_emit =   3 /* VGT_PRIMITIVE_TYPE */
		     + 2 /* VGT_INDEX_TYPE */
		     + 2 /* NUM_INSTANCES */
                     + num_indices + 3; /* DRAW_INDEX_IMMD */

        BEGIN_BATCH_NO_AUTOSTATE(total_emit);
	// prim
        SETfield(vgt_primitive_type, type,
		 VGT_PRIMITIVE_TYPE__PRIM_TYPE_shift, VGT_PRIMITIVE_TYPE__PRIM_TYPE_mask);
        R600_OUT_BATCH(CP_PACKET3(R600_IT_SET_CONFIG_REG, 1));
        R600_OUT_BATCH(mmVGT_PRIMITIVE_TYPE - ASIC_CONFIG_BASE_INDEX);
        R600_OUT_BATCH(vgt_primitive_type);

	// index type
        SETfield(vgt_index_type, DI_INDEX_SIZE_32_BIT, INDEX_TYPE_shift, INDEX_TYPE_mask);
        R600_OUT_BATCH(CP_PACKET3(R600_IT_INDEX_TYPE, 0));
        R600_OUT_BATCH(vgt_index_type);

	// num instances
	R600_OUT_BATCH(CP_PACKET3(R600_IT_NUM_INSTANCES, 0));
        R600_OUT_BATCH(1);

	// draw packet
        vgt_num_indices = num_indices;
        SETfield(vgt_draw_initiator, DI_SRC_SEL_IMMEDIATE, SOURCE_SELECT_shift, SOURCE_SELECT_mask);
	SETfield(vgt_draw_initiator, DI_MAJOR_MODE_0, MAJOR_MODE_shift, MAJOR_MODE_mask);

        R600_OUT_BATCH(CP_PACKET3(R600_IT_DRAW_INDEX_IMMD, (num_indices + 1)));
        R600_OUT_BATCH(vgt_num_indices);
        R600_OUT_BATCH(vgt_draw_initiator);

        for (i = start; i < (start + num_indices); i++) {
            R600_OUT_BATCH(i);
        }
        END_BATCH();
        COMMIT_BATCH();

}

void r700EmitState(GLcontext * ctx)
{
	context_t *context = R700_CONTEXT(ctx);
	radeonContextPtr radeon = &context->radeon;

	if (radeon->cmdbuf.cs->cdw && !radeon->hw.is_dirty && !radeon->hw.all_dirty)
		return;

	rcommonEnsureCmdBufSpace(&context->radeon,
				 context->radeon.hw.max_state_size, __FUNCTION__);

	r700SendSQConfig(context);

	r700SendUCPState(context);
	r700SendContextStates(context);
	r700SendViewportState(context, 0);
	r700SendRenderTargetState(context, 0);
	r700SendDepthTargetState(context);

}

static GLboolean r700RunRender(GLcontext * ctx,
			                   struct tnl_pipeline_stage *stage)
{
    context_t *context = R700_CONTEXT(ctx);
    unsigned int i;
    TNLcontext *tnl = TNL_CONTEXT(ctx);
    struct vertex_buffer *vb = &tnl->vb;

    r700Start3D(context);

    r700UpdateShaders(ctx);
    r700SetScissor(context);
    r700SetupShaders(ctx);

    r700EmitState(ctx);

    /* richard test code */
    for (i = 0; i < vb->PrimitiveCount; i++) {
        GLuint prim = _tnl_translate_prim(&vb->Primitive[i]);
        GLuint start = vb->Primitive[i].start;
        GLuint end = vb->Primitive[i].start + vb->Primitive[i].count;
	r700RunRenderPrimitive(ctx, start, end, prim);
    }

    /* Flush render op cached for last several quads. */
    r700WaitForIdleClean(context);

    radeonReleaseArrays(ctx, ~0);

    return GL_FALSE;
}

static GLboolean r700RunNonTCLRender(GLcontext * ctx,
				     struct tnl_pipeline_stage *stage) /* -------------------- */
{
	GLboolean bRet = GL_TRUE;
	
	return bRet;
}

static GLboolean r700RunTCLRender(GLcontext * ctx,  /*----------------------*/
				  struct tnl_pipeline_stage *stage)
{
	GLboolean bRet = GL_FALSE;

    /* TODO : sw fallback */

    /**
    * Ensure all enabled and complete textures are uploaded along with any buffers being used.
    */
    if(!r600ValidateBuffers(ctx))
    {
        return GL_TRUE;
    }

    bRet = r700RunRender(ctx, stage);

    return bRet;
	//GL_FALSE will stop to do other pipe stage in _tnl_run_pipeline
    //The render here DOES finish the whole pipe, so GL_FALSE should be returned for success.
}

const struct tnl_pipeline_stage _r700_render_stage = {
	"r700 Hardware Rasterization",
	NULL,
	NULL,
	NULL,
	NULL,
	r700RunNonTCLRender
};

const struct tnl_pipeline_stage _r700_tcl_stage = {
	"r700 Hardware Transform, Clipping and Lighting",
	NULL,
	NULL,
	NULL,
	NULL,
	r700RunTCLRender
};

const struct tnl_pipeline_stage *r700_pipeline[] = 
{
    &_r700_tcl_stage,
    &_tnl_vertex_transform_stage,
	&_tnl_normal_transform_stage,
	&_tnl_lighting_stage,
	&_tnl_fog_coordinate_stage,
	&_tnl_texgen_stage,
	&_tnl_texture_transform_stage,
	&_tnl_vertex_program_stage,

    &_r700_render_stage,
    &_tnl_render_stage,
    0,
};