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
|
/*
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 "intel_batchbuffer.h"
#include "intel_regions.h"
#include "brw_context.h"
#include "brw_state.h"
#include "brw_defines.h"
/***********************************************************************
* Blend color
*/
static void upload_blend_constant_color(struct brw_context *brw)
{
GLcontext *ctx = &brw->intel.ctx;
struct brw_blend_constant_color bcc;
memset(&bcc, 0, sizeof(bcc));
bcc.header.opcode = CMD_BLEND_CONSTANT_COLOR;
bcc.header.length = sizeof(bcc)/4-2;
bcc.blend_constant_color[0] = ctx->Color.BlendColor[0];
bcc.blend_constant_color[1] = ctx->Color.BlendColor[1];
bcc.blend_constant_color[2] = ctx->Color.BlendColor[2];
bcc.blend_constant_color[3] = ctx->Color.BlendColor[3];
BRW_CACHED_BATCH_STRUCT(brw, &bcc);
}
const struct brw_tracked_state brw_blend_constant_color = {
.dirty = {
.mesa = _NEW_COLOR,
.brw = BRW_NEW_CONTEXT,
.cache = 0
},
.emit = upload_blend_constant_color
};
/* Constant single cliprect for framebuffer object or DRI2 drawing */
static void upload_drawing_rect(struct brw_context *brw)
{
struct intel_context *intel = &brw->intel;
GLcontext *ctx = &intel->ctx;
BEGIN_BATCH(4);
OUT_BATCH(_3DSTATE_DRAWRECT_INFO_I965);
OUT_BATCH(0); /* xmin, ymin */
OUT_BATCH(((ctx->DrawBuffer->Width - 1) & 0xffff) |
((ctx->DrawBuffer->Height - 1) << 16));
OUT_BATCH(0);
ADVANCE_BATCH();
}
const struct brw_tracked_state brw_drawing_rect = {
.dirty = {
.mesa = _NEW_BUFFERS,
.brw = BRW_NEW_CONTEXT,
.cache = 0
},
.emit = upload_drawing_rect
};
static void prepare_binding_table_pointers(struct brw_context *brw)
{
brw_add_validated_bo(brw, brw->vs.bind_bo);
brw_add_validated_bo(brw, brw->wm.bind_bo);
}
/**
* Upload the binding table pointers, which point each stage's array of surface
* state pointers.
*
* The binding table pointers are relative to the surface state base address,
* which is 0.
*/
static void upload_binding_table_pointers(struct brw_context *brw)
{
struct intel_context *intel = &brw->intel;
BEGIN_BATCH(6);
OUT_BATCH(CMD_BINDING_TABLE_PTRS << 16 | (6 - 2));
if (brw->vs.bind_bo != NULL)
OUT_RELOC(brw->vs.bind_bo, I915_GEM_DOMAIN_SAMPLER, 0, 0); /* vs */
else
OUT_BATCH(0);
OUT_BATCH(0); /* gs */
OUT_BATCH(0); /* clip */
OUT_BATCH(0); /* sf */
OUT_RELOC(brw->wm.bind_bo, I915_GEM_DOMAIN_SAMPLER, 0, 0); /* wm/ps */
ADVANCE_BATCH();
}
const struct brw_tracked_state brw_binding_table_pointers = {
.dirty = {
.mesa = 0,
.brw = BRW_NEW_BATCH,
.cache = CACHE_NEW_SURF_BIND,
},
.prepare = prepare_binding_table_pointers,
.emit = upload_binding_table_pointers,
};
/**
* Upload the binding table pointers, which point each stage's array of surface
* state pointers.
*
* The binding table pointers are relative to the surface state base address,
* which is 0.
*/
static void upload_gen6_binding_table_pointers(struct brw_context *brw)
{
struct intel_context *intel = &brw->intel;
BEGIN_BATCH(4);
OUT_BATCH(CMD_BINDING_TABLE_PTRS << 16 |
GEN6_BINDING_TABLE_MODIFY_VS |
GEN6_BINDING_TABLE_MODIFY_GS |
GEN6_BINDING_TABLE_MODIFY_PS |
(4 - 2));
if (brw->vs.bind_bo != NULL)
OUT_RELOC(brw->vs.bind_bo, I915_GEM_DOMAIN_SAMPLER, 0, 0); /* vs */
else
OUT_BATCH(0);
OUT_BATCH(0); /* gs */
OUT_RELOC(brw->wm.bind_bo, I915_GEM_DOMAIN_SAMPLER, 0, 0); /* wm/ps */
ADVANCE_BATCH();
}
const struct brw_tracked_state gen6_binding_table_pointers = {
.dirty = {
.mesa = 0,
.brw = BRW_NEW_BATCH,
.cache = CACHE_NEW_SURF_BIND,
},
.prepare = prepare_binding_table_pointers,
.emit = upload_gen6_binding_table_pointers,
};
/**
* Upload pointers to the per-stage state.
*
* The state pointers in this packet are all relative to the general state
* base address set by CMD_STATE_BASE_ADDRESS, which is 0.
*/
static void upload_pipelined_state_pointers(struct brw_context *brw )
{
struct intel_context *intel = &brw->intel;
BEGIN_BATCH(7);
OUT_BATCH(CMD_PIPELINED_STATE_POINTERS << 16 | (7 - 2));
OUT_RELOC(brw->vs.state_bo, I915_GEM_DOMAIN_INSTRUCTION, 0, 0);
if (brw->gs.prog_active)
OUT_RELOC(brw->gs.state_bo, I915_GEM_DOMAIN_INSTRUCTION, 0, 1);
else
OUT_BATCH(0);
OUT_RELOC(brw->clip.state_bo, I915_GEM_DOMAIN_INSTRUCTION, 0, 1);
OUT_RELOC(brw->sf.state_bo, I915_GEM_DOMAIN_INSTRUCTION, 0, 0);
OUT_RELOC(brw->wm.state_bo, I915_GEM_DOMAIN_INSTRUCTION, 0, 0);
OUT_RELOC(brw->cc.state_bo, I915_GEM_DOMAIN_INSTRUCTION, 0, 0);
ADVANCE_BATCH();
brw->state.dirty.brw |= BRW_NEW_PSP;
}
static void prepare_psp_urb_cbs(struct brw_context *brw)
{
brw_add_validated_bo(brw, brw->vs.state_bo);
brw_add_validated_bo(brw, brw->gs.state_bo);
brw_add_validated_bo(brw, brw->clip.state_bo);
brw_add_validated_bo(brw, brw->sf.state_bo);
brw_add_validated_bo(brw, brw->wm.state_bo);
brw_add_validated_bo(brw, brw->cc.state_bo);
}
static void upload_psp_urb_cbs(struct brw_context *brw )
{
upload_pipelined_state_pointers(brw);
brw_upload_urb_fence(brw);
brw_upload_cs_urb_state(brw);
}
const struct brw_tracked_state brw_psp_urb_cbs = {
.dirty = {
.mesa = 0,
.brw = BRW_NEW_URB_FENCE | BRW_NEW_BATCH,
.cache = (CACHE_NEW_VS_UNIT |
CACHE_NEW_GS_UNIT |
CACHE_NEW_GS_PROG |
CACHE_NEW_CLIP_UNIT |
CACHE_NEW_SF_UNIT |
CACHE_NEW_WM_UNIT |
CACHE_NEW_CC_UNIT)
},
.prepare = prepare_psp_urb_cbs,
.emit = upload_psp_urb_cbs,
};
static void prepare_depthbuffer(struct brw_context *brw)
{
struct intel_region *region = brw->state.depth_region;
if (region != NULL)
brw_add_validated_bo(brw, region->buffer);
}
static void emit_depthbuffer(struct brw_context *brw)
{
struct intel_context *intel = &brw->intel;
struct intel_region *region = brw->state.depth_region;
unsigned int len;
if (intel->gen >= 6)
len = 7;
else if (intel->is_g4x || intel->is_ironlake)
len = 6;
else
len = 5;
if (region == NULL) {
BEGIN_BATCH(len);
OUT_BATCH(CMD_DEPTH_BUFFER << 16 | (len - 2));
OUT_BATCH((BRW_DEPTHFORMAT_D32_FLOAT << 18) |
(BRW_SURFACE_NULL << 29));
OUT_BATCH(0);
OUT_BATCH(0);
OUT_BATCH(0);
if (intel->is_g4x || intel->is_ironlake || intel->gen >= 6)
OUT_BATCH(0);
if (intel->gen >= 6)
OUT_BATCH(0);
ADVANCE_BATCH();
} else {
unsigned int format;
switch (region->cpp) {
case 2:
format = BRW_DEPTHFORMAT_D16_UNORM;
break;
case 4:
if (intel->depth_buffer_is_float)
format = BRW_DEPTHFORMAT_D32_FLOAT;
else
format = BRW_DEPTHFORMAT_D24_UNORM_S8_UINT;
break;
default:
assert(0);
return;
}
assert(region->tiling != I915_TILING_X);
if (IS_GEN6(intel->intelScreen->deviceID))
assert(region->tiling != I915_TILING_NONE);
BEGIN_BATCH(len);
OUT_BATCH(CMD_DEPTH_BUFFER << 16 | (len - 2));
OUT_BATCH(((region->pitch * region->cpp) - 1) |
(format << 18) |
(BRW_TILEWALK_YMAJOR << 26) |
((region->tiling != I915_TILING_NONE) << 27) |
(BRW_SURFACE_2D << 29));
OUT_RELOC(region->buffer,
I915_GEM_DOMAIN_RENDER, I915_GEM_DOMAIN_RENDER,
0);
OUT_BATCH((BRW_SURFACE_MIPMAPLAYOUT_BELOW << 1) |
((region->pitch - 1) << 6) |
((region->height - 1) << 19));
OUT_BATCH(0);
if (intel->is_g4x || intel->is_ironlake || intel->gen >= 6)
OUT_BATCH(0);
if (intel->gen >= 6)
OUT_BATCH(0);
ADVANCE_BATCH();
}
}
const struct brw_tracked_state brw_depthbuffer = {
.dirty = {
.mesa = 0,
.brw = BRW_NEW_DEPTH_BUFFER | BRW_NEW_BATCH,
.cache = 0,
},
.prepare = prepare_depthbuffer,
.emit = emit_depthbuffer,
};
/***********************************************************************
* Polygon stipple packet
*/
static void upload_polygon_stipple(struct brw_context *brw)
{
GLcontext *ctx = &brw->intel.ctx;
struct brw_polygon_stipple bps;
GLuint i;
memset(&bps, 0, sizeof(bps));
bps.header.opcode = CMD_POLY_STIPPLE_PATTERN;
bps.header.length = sizeof(bps)/4-2;
/* Polygon stipple is provided in OpenGL order, i.e. bottom
* row first. If we're rendering to a window (i.e. the
* default frame buffer object, 0), then we need to invert
* it to match our pixel layout. But if we're rendering
* to a FBO (i.e. any named frame buffer object), we *don't*
* need to invert - we already match the layout.
*/
if (ctx->DrawBuffer->Name == 0) {
for (i = 0; i < 32; i++)
bps.stipple[i] = ctx->PolygonStipple[31 - i]; /* invert */
}
else {
for (i = 0; i < 32; i++)
bps.stipple[i] = ctx->PolygonStipple[i]; /* don't invert */
}
BRW_CACHED_BATCH_STRUCT(brw, &bps);
}
const struct brw_tracked_state brw_polygon_stipple = {
.dirty = {
.mesa = _NEW_POLYGONSTIPPLE,
.brw = BRW_NEW_CONTEXT,
.cache = 0
},
.emit = upload_polygon_stipple
};
/***********************************************************************
* Polygon stipple offset packet
*/
static void upload_polygon_stipple_offset(struct brw_context *brw)
{
GLcontext *ctx = &brw->intel.ctx;
struct brw_polygon_stipple_offset bpso;
memset(&bpso, 0, sizeof(bpso));
bpso.header.opcode = CMD_POLY_STIPPLE_OFFSET;
bpso.header.length = sizeof(bpso)/4-2;
/* If we're drawing to a system window (ctx->DrawBuffer->Name == 0),
* we have to invert the Y axis in order to match the OpenGL
* pixel coordinate system, and our offset must be matched
* to the window position. If we're drawing to a FBO
* (ctx->DrawBuffer->Name != 0), then our native pixel coordinate
* system works just fine, and there's no window system to
* worry about.
*/
if (brw->intel.ctx.DrawBuffer->Name == 0) {
bpso.bits0.x_offset = 0;
bpso.bits0.y_offset = (32 - (ctx->DrawBuffer->Height & 31)) & 31;
}
else {
bpso.bits0.y_offset = 0;
bpso.bits0.x_offset = 0;
}
BRW_CACHED_BATCH_STRUCT(brw, &bpso);
}
#define _NEW_WINDOW_POS 0x40000000
const struct brw_tracked_state brw_polygon_stipple_offset = {
.dirty = {
.mesa = _NEW_WINDOW_POS,
.brw = BRW_NEW_CONTEXT,
.cache = 0
},
.emit = upload_polygon_stipple_offset
};
/**********************************************************************
* AA Line parameters
*/
static void upload_aa_line_parameters(struct brw_context *brw)
{
struct brw_aa_line_parameters balp;
if (!brw->has_aa_line_parameters)
return;
/* use legacy aa line coverage computation */
memset(&balp, 0, sizeof(balp));
balp.header.opcode = CMD_AA_LINE_PARAMETERS;
balp.header.length = sizeof(balp) / 4 - 2;
BRW_CACHED_BATCH_STRUCT(brw, &balp);
}
const struct brw_tracked_state brw_aa_line_parameters = {
.dirty = {
.mesa = 0,
.brw = BRW_NEW_CONTEXT,
.cache = 0
},
.emit = upload_aa_line_parameters
};
/***********************************************************************
* Line stipple packet
*/
static void upload_line_stipple(struct brw_context *brw)
{
GLcontext *ctx = &brw->intel.ctx;
struct brw_line_stipple bls;
GLfloat tmp;
GLint tmpi;
memset(&bls, 0, sizeof(bls));
bls.header.opcode = CMD_LINE_STIPPLE_PATTERN;
bls.header.length = sizeof(bls)/4 - 2;
bls.bits0.pattern = ctx->Line.StipplePattern;
bls.bits1.repeat_count = ctx->Line.StippleFactor;
tmp = 1.0 / (GLfloat) ctx->Line.StippleFactor;
tmpi = tmp * (1<<13);
bls.bits1.inverse_repeat_count = tmpi;
BRW_CACHED_BATCH_STRUCT(brw, &bls);
}
const struct brw_tracked_state brw_line_stipple = {
.dirty = {
.mesa = _NEW_LINE,
.brw = BRW_NEW_CONTEXT,
.cache = 0
},
.emit = upload_line_stipple
};
/***********************************************************************
* Misc invarient state packets
*/
static void upload_invarient_state( struct brw_context *brw )
{
struct intel_context *intel = &brw->intel;
{
/* 0x61040000 Pipeline Select */
/* PipelineSelect : 0 */
struct brw_pipeline_select ps;
memset(&ps, 0, sizeof(ps));
ps.header.opcode = brw->CMD_PIPELINE_SELECT;
ps.header.pipeline_select = 0;
BRW_BATCH_STRUCT(brw, &ps);
}
if (intel->gen < 6) {
struct brw_global_depth_offset_clamp gdo;
memset(&gdo, 0, sizeof(gdo));
/* Disable depth offset clamping.
*/
gdo.header.opcode = CMD_GLOBAL_DEPTH_OFFSET_CLAMP;
gdo.header.length = sizeof(gdo)/4 - 2;
gdo.depth_offset_clamp = 0.0;
BRW_BATCH_STRUCT(brw, &gdo);
}
/* 0x61020000 State Instruction Pointer */
{
struct brw_system_instruction_pointer sip;
memset(&sip, 0, sizeof(sip));
sip.header.opcode = CMD_STATE_INSN_POINTER;
sip.header.length = 0;
sip.bits0.pad = 0;
sip.bits0.system_instruction_pointer = 0;
BRW_BATCH_STRUCT(brw, &sip);
}
{
struct brw_vf_statistics vfs;
memset(&vfs, 0, sizeof(vfs));
vfs.opcode = brw->CMD_VF_STATISTICS;
if (INTEL_DEBUG & DEBUG_STATS)
vfs.statistics_enable = 1;
BRW_BATCH_STRUCT(brw, &vfs);
}
}
const struct brw_tracked_state brw_invarient_state = {
.dirty = {
.mesa = 0,
.brw = BRW_NEW_CONTEXT,
.cache = 0
},
.emit = upload_invarient_state
};
/**
* Define the base addresses which some state is referenced from.
*
* This allows us to avoid having to emit relocations in many places for
* cached state, and instead emit pointers inside of large, mostly-static
* state pools. This comes at the expense of memory, and more expensive cache
* misses.
*/
static void upload_state_base_address( struct brw_context *brw )
{
struct intel_context *intel = &brw->intel;
/* Output the structure (brw_state_base_address) directly to the
* batchbuffer, so we can emit relocations inline.
*/
if (intel->gen >= 6) {
BEGIN_BATCH(10);
OUT_BATCH(CMD_STATE_BASE_ADDRESS << 16 | (10 - 2));
OUT_BATCH(1); /* General state base address */
OUT_BATCH(1); /* Surface state base address */
OUT_BATCH(1); /* Dynamic state base address */
OUT_BATCH(1); /* Indirect object base address */
OUT_BATCH(1); /* Instruction base address */
OUT_BATCH(1); /* General state upper bound */
OUT_BATCH(1); /* Dynamic state upper bound */
OUT_BATCH(1); /* Indirect object upper bound */
OUT_BATCH(1); /* Instruction access upper bound */
ADVANCE_BATCH();
} else if (intel->is_ironlake) {
BEGIN_BATCH(8);
OUT_BATCH(CMD_STATE_BASE_ADDRESS << 16 | (8 - 2));
OUT_BATCH(1); /* General state base address */
OUT_BATCH(1); /* Surface state base address */
OUT_BATCH(1); /* Indirect object base address */
OUT_BATCH(1); /* Instruction base address */
OUT_BATCH(1); /* General state upper bound */
OUT_BATCH(1); /* Indirect object upper bound */
OUT_BATCH(1); /* Instruction access upper bound */
ADVANCE_BATCH();
} else {
BEGIN_BATCH(6);
OUT_BATCH(CMD_STATE_BASE_ADDRESS << 16 | (6 - 2));
OUT_BATCH(1); /* General state base address */
OUT_BATCH(1); /* Surface state base address */
OUT_BATCH(1); /* Indirect object base address */
OUT_BATCH(1); /* General state upper bound */
OUT_BATCH(1); /* Indirect object upper bound */
ADVANCE_BATCH();
}
}
const struct brw_tracked_state brw_state_base_address = {
.dirty = {
.mesa = 0,
.brw = BRW_NEW_CONTEXT,
.cache = 0,
},
.emit = upload_state_base_address
};
|