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
|
/*
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 "brw_context.h"
#include "brw_state.h"
#include "intel_batchbuffer.h"
#include "intel_buffers.h"
/* This is used to initialize brw->state.atoms[]. We could use this
* list directly except for a single atom, brw_constant_buffer, which
* has a .dirty value which changes according to the parameters of the
* current fragment and vertex programs, and so cannot be a static
* value.
*/
static const struct brw_tracked_state *gen4_atoms[] =
{
&brw_check_fallback,
&brw_wm_input_sizes,
&brw_vs_prog, /* must do before GS prog, state base address. */
&brw_gs_prog, /* must do before state base address */
&brw_clip_prog, /* must do before state base address */
&brw_sf_prog, /* must do before state base address */
&brw_wm_prog, /* must do before state base address */
/* Once all the programs are done, we know how large urb entry
* sizes need to be and can decide if we need to change the urb
* layout.
*/
&brw_curbe_offsets,
&brw_recalculate_urb_fence,
&brw_cc_vp,
&brw_cc_unit,
&brw_vs_constants, /* Before vs_surfaces and constant_buffer */
&brw_wm_constants, /* Before wm_surfaces and constant_buffer */
&brw_vs_surfaces, /* must do before unit */
&brw_wm_constant_surface, /* must do before wm surfaces/bind bo */
&brw_wm_surfaces, /* must do before samplers and unit */
&brw_wm_binding_table,
&brw_wm_samplers,
/* These set up state for brw_psp_urb_cbs */
&brw_wm_unit,
&brw_sf_vp,
&brw_sf_unit,
&brw_vs_unit, /* always required, enabled or not */
&brw_clip_unit,
&brw_gs_unit,
/* Command packets:
*/
&brw_invarient_state,
&brw_state_base_address,
&brw_binding_table_pointers,
&brw_blend_constant_color,
&brw_depthbuffer,
&brw_polygon_stipple,
&brw_polygon_stipple_offset,
&brw_line_stipple,
&brw_aa_line_parameters,
&brw_psp_urb_cbs,
&brw_drawing_rect,
&brw_indices,
&brw_index_buffer,
&brw_vertices,
&brw_constant_buffer
};
static const struct brw_tracked_state *gen6_atoms[] =
{
&brw_check_fallback,
&brw_wm_input_sizes,
&brw_vs_prog, /* must do before state base address */
&brw_gs_prog, /* must do before state base address */
&brw_wm_prog, /* must do before state base address */
&gen6_clip_vp,
&gen6_sf_vp,
/* Command packets: */
&brw_invarient_state,
/* must do before binding table pointers, cc state ptrs */
&brw_state_base_address,
&brw_cc_vp,
&gen6_viewport_state, /* must do after *_vp stages */
&gen6_urb,
&gen6_blend_state, /* must do before cc unit */
&gen6_color_calc_state, /* must do before cc unit */
&gen6_depth_stencil_state, /* must do before cc unit */
&gen6_cc_state_pointers,
&brw_vs_constants, /* Before vs_surfaces and constant_buffer */
&brw_wm_constants, /* Before wm_surfaces and constant_buffer */
&gen6_vs_constants, /* Before vs_state */
&gen6_wm_constants, /* Before wm_state */
&brw_vs_surfaces, /* must do before unit */
&brw_wm_constant_surface, /* must do before wm surfaces/bind bo */
&brw_wm_surfaces, /* must do before samplers and unit */
&brw_wm_binding_table,
&brw_wm_samplers,
&gen6_sampler_state,
&gen6_vs_state,
&gen6_gs_state,
&gen6_clip_state,
&gen6_sf_state,
&gen6_wm_state,
&gen6_scissor_state,
&gen6_binding_table_pointers,
&brw_depthbuffer,
&brw_polygon_stipple,
&brw_polygon_stipple_offset,
&brw_line_stipple,
&brw_aa_line_parameters,
&brw_drawing_rect,
&brw_indices,
&brw_index_buffer,
&brw_vertices,
};
const struct brw_tracked_state *gen7_atoms[] =
{
&brw_check_fallback,
&brw_wm_input_sizes,
&brw_vs_prog,
&brw_gs_prog,
&brw_wm_prog,
/* Command packets: */
&brw_invarient_state,
/* must do before binding table pointers, cc state ptrs */
&brw_state_base_address,
&brw_cc_vp,
&gen7_cc_viewport_state_pointer, /* must do after brw_cc_vp */
&gen7_sf_clip_viewport,
&gen7_urb,
&gen6_blend_state, /* must do before cc unit */
&gen6_color_calc_state, /* must do before cc unit */
&gen6_depth_stencil_state, /* must do before cc unit */
&gen7_blend_state_pointer,
&gen7_cc_state_pointer,
&gen7_depth_stencil_state_pointer,
&brw_vs_constants, /* Before vs_surfaces and constant_buffer */
&brw_wm_constants, /* Before wm_surfaces and constant_buffer */
&gen6_vs_constants, /* Before vs_state */
&gen6_wm_constants, /* Before wm_surfaces and constant_buffer */
&brw_vs_surfaces, /* must do before unit */
&brw_wm_constant_surface, /* must do before wm surfaces/bind bo */
&brw_wm_surfaces, /* must do before samplers and unit */
&brw_wm_binding_table,
&gen7_samplers,
&gen7_disable_stages,
&gen7_vs_state,
&gen7_clip_state,
&gen7_sbe_state,
&gen7_sf_state,
&gen7_wm_state,
&gen7_ps_state,
&gen6_scissor_state,
&gen7_depthbuffer,
&brw_polygon_stipple,
&brw_polygon_stipple_offset,
&brw_line_stipple,
&brw_aa_line_parameters,
&brw_drawing_rect,
&brw_indices,
&brw_index_buffer,
&brw_vertices,
};
void brw_init_state( struct brw_context *brw )
{
const struct brw_tracked_state **atoms;
int num_atoms;
brw_init_caches(brw);
if (brw->intel.gen >= 7) {
atoms = gen7_atoms;
num_atoms = ARRAY_SIZE(gen7_atoms);
} else if (brw->intel.gen == 6) {
atoms = gen6_atoms;
num_atoms = ARRAY_SIZE(gen6_atoms);
} else {
atoms = gen4_atoms;
num_atoms = ARRAY_SIZE(gen4_atoms);
}
brw->atoms = atoms;
brw->num_atoms = num_atoms;
while (num_atoms--) {
assert((*atoms)->dirty.mesa |
(*atoms)->dirty.brw |
(*atoms)->dirty.cache);
assert((*atoms)->emit);
atoms++;
}
}
void brw_destroy_state( struct brw_context *brw )
{
brw_destroy_caches(brw);
}
/***********************************************************************
*/
static GLuint check_state( const struct brw_state_flags *a,
const struct brw_state_flags *b )
{
return ((a->mesa & b->mesa) |
(a->brw & b->brw) |
(a->cache & b->cache)) != 0;
}
static void accumulate_state( struct brw_state_flags *a,
const struct brw_state_flags *b )
{
a->mesa |= b->mesa;
a->brw |= b->brw;
a->cache |= b->cache;
}
static void xor_states( struct brw_state_flags *result,
const struct brw_state_flags *a,
const struct brw_state_flags *b )
{
result->mesa = a->mesa ^ b->mesa;
result->brw = a->brw ^ b->brw;
result->cache = a->cache ^ b->cache;
}
struct dirty_bit_map {
uint32_t bit;
char *name;
uint32_t count;
};
#define DEFINE_BIT(name) {name, #name, 0}
static struct dirty_bit_map mesa_bits[] = {
DEFINE_BIT(_NEW_MODELVIEW),
DEFINE_BIT(_NEW_PROJECTION),
DEFINE_BIT(_NEW_TEXTURE_MATRIX),
DEFINE_BIT(_NEW_COLOR),
DEFINE_BIT(_NEW_DEPTH),
DEFINE_BIT(_NEW_EVAL),
DEFINE_BIT(_NEW_FOG),
DEFINE_BIT(_NEW_HINT),
DEFINE_BIT(_NEW_LIGHT),
DEFINE_BIT(_NEW_LINE),
DEFINE_BIT(_NEW_PIXEL),
DEFINE_BIT(_NEW_POINT),
DEFINE_BIT(_NEW_POLYGON),
DEFINE_BIT(_NEW_POLYGONSTIPPLE),
DEFINE_BIT(_NEW_SCISSOR),
DEFINE_BIT(_NEW_STENCIL),
DEFINE_BIT(_NEW_TEXTURE),
DEFINE_BIT(_NEW_TRANSFORM),
DEFINE_BIT(_NEW_VIEWPORT),
DEFINE_BIT(_NEW_PACKUNPACK),
DEFINE_BIT(_NEW_ARRAY),
DEFINE_BIT(_NEW_RENDERMODE),
DEFINE_BIT(_NEW_BUFFERS),
DEFINE_BIT(_NEW_MULTISAMPLE),
DEFINE_BIT(_NEW_TRACK_MATRIX),
DEFINE_BIT(_NEW_PROGRAM),
DEFINE_BIT(_NEW_PROGRAM_CONSTANTS),
{0, 0, 0}
};
static struct dirty_bit_map brw_bits[] = {
DEFINE_BIT(BRW_NEW_URB_FENCE),
DEFINE_BIT(BRW_NEW_FRAGMENT_PROGRAM),
DEFINE_BIT(BRW_NEW_VERTEX_PROGRAM),
DEFINE_BIT(BRW_NEW_INPUT_DIMENSIONS),
DEFINE_BIT(BRW_NEW_CURBE_OFFSETS),
DEFINE_BIT(BRW_NEW_REDUCED_PRIMITIVE),
DEFINE_BIT(BRW_NEW_PRIMITIVE),
DEFINE_BIT(BRW_NEW_CONTEXT),
DEFINE_BIT(BRW_NEW_WM_INPUT_DIMENSIONS),
DEFINE_BIT(BRW_NEW_PROGRAM_CACHE),
DEFINE_BIT(BRW_NEW_PSP),
DEFINE_BIT(BRW_NEW_WM_SURFACES),
DEFINE_BIT(BRW_NEW_INDICES),
DEFINE_BIT(BRW_NEW_INDEX_BUFFER),
DEFINE_BIT(BRW_NEW_VERTICES),
DEFINE_BIT(BRW_NEW_BATCH),
DEFINE_BIT(BRW_NEW_VS_CONSTBUF),
DEFINE_BIT(BRW_NEW_WM_CONSTBUF),
DEFINE_BIT(BRW_NEW_VS_BINDING_TABLE),
DEFINE_BIT(BRW_NEW_GS_BINDING_TABLE),
DEFINE_BIT(BRW_NEW_PS_BINDING_TABLE),
DEFINE_BIT(BRW_NEW_STATE_BASE_ADDRESS),
{0, 0, 0}
};
static struct dirty_bit_map cache_bits[] = {
DEFINE_BIT(CACHE_NEW_BLEND_STATE),
DEFINE_BIT(CACHE_NEW_CC_VP),
DEFINE_BIT(CACHE_NEW_CC_UNIT),
DEFINE_BIT(CACHE_NEW_WM_PROG),
DEFINE_BIT(CACHE_NEW_SAMPLER),
DEFINE_BIT(CACHE_NEW_WM_UNIT),
DEFINE_BIT(CACHE_NEW_SF_PROG),
DEFINE_BIT(CACHE_NEW_SF_VP),
DEFINE_BIT(CACHE_NEW_SF_UNIT),
DEFINE_BIT(CACHE_NEW_VS_UNIT),
DEFINE_BIT(CACHE_NEW_VS_PROG),
DEFINE_BIT(CACHE_NEW_GS_UNIT),
DEFINE_BIT(CACHE_NEW_GS_PROG),
DEFINE_BIT(CACHE_NEW_CLIP_VP),
DEFINE_BIT(CACHE_NEW_CLIP_UNIT),
DEFINE_BIT(CACHE_NEW_CLIP_PROG),
{0, 0, 0}
};
static void
brw_update_dirty_count(struct dirty_bit_map *bit_map, int32_t bits)
{
int i;
for (i = 0; i < 32; i++) {
if (bit_map[i].bit == 0)
return;
if (bit_map[i].bit & bits)
bit_map[i].count++;
}
}
static void
brw_print_dirty_count(struct dirty_bit_map *bit_map, int32_t bits)
{
int i;
for (i = 0; i < 32; i++) {
if (bit_map[i].bit == 0)
return;
fprintf(stderr, "0x%08x: %12d (%s)\n",
bit_map[i].bit, bit_map[i].count, bit_map[i].name);
}
}
/***********************************************************************
* Emit all state:
*/
void brw_upload_state(struct brw_context *brw)
{
struct gl_context *ctx = &brw->intel.ctx;
struct intel_context *intel = &brw->intel;
struct brw_state_flags *state = &brw->state.dirty;
int i;
static int dirty_count = 0;
state->mesa |= brw->intel.NewGLState;
brw->intel.NewGLState = 0;
if (brw->emit_state_always) {
state->mesa |= ~0;
state->brw |= ~0;
state->cache |= ~0;
}
if (brw->fragment_program != ctx->FragmentProgram._Current) {
brw->fragment_program = ctx->FragmentProgram._Current;
brw->state.dirty.brw |= BRW_NEW_FRAGMENT_PROGRAM;
}
if (brw->vertex_program != ctx->VertexProgram._Current) {
brw->vertex_program = ctx->VertexProgram._Current;
brw->state.dirty.brw |= BRW_NEW_VERTEX_PROGRAM;
}
if ((state->mesa | state->cache | state->brw) == 0)
return;
brw->intel.Fallback = false; /* boolean, not bitfield */
intel_check_front_buffer_rendering(intel);
if (unlikely(INTEL_DEBUG)) {
/* Debug version which enforces various sanity checks on the
* state flags which are generated and checked to help ensure
* state atoms are ordered correctly in the list.
*/
struct brw_state_flags examined, prev;
memset(&examined, 0, sizeof(examined));
prev = *state;
for (i = 0; i < brw->num_atoms; i++) {
const struct brw_tracked_state *atom = brw->atoms[i];
struct brw_state_flags generated;
if (brw->intel.Fallback)
break;
if (check_state(state, &atom->dirty)) {
atom->emit(brw);
}
accumulate_state(&examined, &atom->dirty);
/* generated = (prev ^ state)
* if (examined & generated)
* fail;
*/
xor_states(&generated, &prev, state);
assert(!check_state(&examined, &generated));
prev = *state;
}
}
else {
for (i = 0; i < brw->num_atoms; i++) {
const struct brw_tracked_state *atom = brw->atoms[i];
if (brw->intel.Fallback)
break;
if (check_state(state, &atom->dirty)) {
atom->emit(brw);
}
}
}
if (unlikely(INTEL_DEBUG & DEBUG_STATE)) {
brw_update_dirty_count(mesa_bits, state->mesa);
brw_update_dirty_count(brw_bits, state->brw);
brw_update_dirty_count(cache_bits, state->cache);
if (dirty_count++ % 1000 == 0) {
brw_print_dirty_count(mesa_bits, state->mesa);
brw_print_dirty_count(brw_bits, state->brw);
brw_print_dirty_count(cache_bits, state->cache);
fprintf(stderr, "\n");
}
}
if (!brw->intel.Fallback)
memset(state, 0, sizeof(*state));
}
|