summaryrefslogtreecommitdiff
path: root/src/mesa/drivers/dri/i965/brw_vec4.h
blob: 89b6f912272b967564388f404b00f5d83fe3efe6 (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
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
/*
 * Copyright © 2011 Intel Corporation
 *
 * 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 AUTHORS OR COPYRIGHT HOLDERS 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.
 */

#ifndef BRW_VEC4_H
#define BRW_VEC4_H

#include <stdint.h>
#include "brw_shader.h"
#include "main/compiler.h"
#include "program/hash_table.h"
#include "brw_program.h"

#ifdef __cplusplus
#include "brw_ir_vec4.h"

extern "C" {
#endif

#include "brw_context.h"
#include "brw_eu.h"
#include "intel_asm_annotation.h"

#ifdef __cplusplus
}; /* extern "C" */
#endif

#include "glsl/ir.h"
#include "glsl/nir/nir.h"


#ifdef __cplusplus
extern "C" {
#endif

#ifdef __cplusplus
} /* extern "C" */

namespace brw {

class vec4_live_variables;

/**
 * The vertex shader front-end.
 *
 * Translates either GLSL IR or Mesa IR (for ARB_vertex_program and
 * fixed-function) into VS IR.
 */
class vec4_visitor : public backend_shader
{
public:
   vec4_visitor(const struct brw_compiler *compiler,
                void *log_data,
                const struct brw_sampler_prog_key_data *key,
                struct brw_vue_prog_data *prog_data,
                const nir_shader *shader,
		void *mem_ctx,
                bool no_spills,
                int shader_time_index);
   virtual ~vec4_visitor();

   dst_reg dst_null_f()
   {
      return dst_reg(brw_null_reg());
   }

   dst_reg dst_null_d()
   {
      return dst_reg(retype(brw_null_reg(), BRW_REGISTER_TYPE_D));
   }

   dst_reg dst_null_ud()
   {
      return dst_reg(retype(brw_null_reg(), BRW_REGISTER_TYPE_UD));
   }

   const struct brw_sampler_prog_key_data * const key_tex;
   struct brw_vue_prog_data * const prog_data;
   char *fail_msg;
   bool failed;

   /**
    * GLSL IR currently being processed, which is associated with our
    * driver IR instructions for debugging purposes.
    */
   const void *base_ir;
   const char *current_annotation;

   int first_non_payload_grf;
   unsigned int max_grf;
   int *virtual_grf_start;
   int *virtual_grf_end;
   brw::vec4_live_variables *live_intervals;
   dst_reg userplane[MAX_CLIP_PLANES];

   bool need_all_constants_in_pull_buffer;

   /* Regs for vertex results.  Generated at ir_variable visiting time
    * for the ir->location's used.
    */
   dst_reg output_reg[BRW_VARYING_SLOT_COUNT];
   const char *output_reg_annotation[BRW_VARYING_SLOT_COUNT];
   int *uniform_size;
   int uniform_array_size; /*< Size of the uniform_size array */
   int uniforms;

   src_reg shader_start_time;

   bool run();
   void fail(const char *msg, ...);

   int setup_uniforms(int payload_reg);

   bool reg_allocate_trivial();
   bool reg_allocate();
   void evaluate_spill_costs(float *spill_costs, bool *no_spill);
   int choose_spill_reg(struct ra_graph *g);
   void spill_reg(int spill_reg);
   void move_grf_array_access_to_scratch();
   void move_uniform_array_access_to_pull_constants();
   void move_push_constants_to_pull_constants();
   void split_uniform_registers();
   void pack_uniform_registers();
   void calculate_live_intervals();
   void invalidate_live_intervals();
   void split_virtual_grfs();
   bool opt_vector_float();
   bool opt_reduce_swizzle();
   bool dead_code_eliminate();
   int var_range_start(unsigned v, unsigned n) const;
   int var_range_end(unsigned v, unsigned n) const;
   bool virtual_grf_interferes(int a, int b);
   bool opt_cmod_propagation();
   bool opt_copy_propagation(bool do_constant_prop = true);
   bool opt_cse_local(bblock_t *block);
   bool opt_cse();
   bool opt_algebraic();
   bool opt_register_coalesce();
   bool eliminate_find_live_channel();
   bool is_dep_ctrl_unsafe(const vec4_instruction *inst);
   void opt_set_dependency_control();
   void opt_schedule_instructions();

   vec4_instruction *emit(vec4_instruction *inst);

   vec4_instruction *emit(enum opcode opcode);
   vec4_instruction *emit(enum opcode opcode, const dst_reg &dst);
   vec4_instruction *emit(enum opcode opcode, const dst_reg &dst,
                          const src_reg &src0);
   vec4_instruction *emit(enum opcode opcode, const dst_reg &dst,
                          const src_reg &src0, const src_reg &src1);
   vec4_instruction *emit(enum opcode opcode, const dst_reg &dst,
                          const src_reg &src0, const src_reg &src1,
                          const src_reg &src2);

   vec4_instruction *emit_before(bblock_t *block,
                                 vec4_instruction *inst,
				 vec4_instruction *new_inst);

#define EMIT1(op) vec4_instruction *op(const dst_reg &, const src_reg &);
#define EMIT2(op) vec4_instruction *op(const dst_reg &, const src_reg &, const src_reg &);
#define EMIT3(op) vec4_instruction *op(const dst_reg &, const src_reg &, const src_reg &, const src_reg &);
   EMIT1(MOV)
   EMIT1(NOT)
   EMIT1(RNDD)
   EMIT1(RNDE)
   EMIT1(RNDZ)
   EMIT1(FRC)
   EMIT1(F32TO16)
   EMIT1(F16TO32)
   EMIT2(ADD)
   EMIT2(MUL)
   EMIT2(MACH)
   EMIT2(MAC)
   EMIT2(AND)
   EMIT2(OR)
   EMIT2(XOR)
   EMIT2(DP3)
   EMIT2(DP4)
   EMIT2(DPH)
   EMIT2(SHL)
   EMIT2(SHR)
   EMIT2(ASR)
   vec4_instruction *CMP(dst_reg dst, src_reg src0, src_reg src1,
			 enum brw_conditional_mod condition);
   vec4_instruction *IF(src_reg src0, src_reg src1,
                        enum brw_conditional_mod condition);
   vec4_instruction *IF(enum brw_predicate predicate);
   EMIT1(SCRATCH_READ)
   EMIT2(SCRATCH_WRITE)
   EMIT3(LRP)
   EMIT1(BFREV)
   EMIT3(BFE)
   EMIT2(BFI1)
   EMIT3(BFI2)
   EMIT1(FBH)
   EMIT1(FBL)
   EMIT1(CBIT)
   EMIT3(MAD)
   EMIT2(ADDC)
   EMIT2(SUBB)
#undef EMIT1
#undef EMIT2
#undef EMIT3

   int implied_mrf_writes(vec4_instruction *inst);

   vec4_instruction *emit_minmax(enum brw_conditional_mod conditionalmod, dst_reg dst,
                                 src_reg src0, src_reg src1);

   vec4_instruction *emit_lrp(const dst_reg &dst, const src_reg &x,
                              const src_reg &y, const src_reg &a);

   /**
    * Copy any live channel from \p src to the first channel of the
    * result.
    */
   src_reg emit_uniformize(const src_reg &src);

   src_reg fix_3src_operand(const src_reg &src);
   src_reg resolve_source_modifiers(const src_reg &src);

   vec4_instruction *emit_math(enum opcode opcode, const dst_reg &dst, const src_reg &src0,
                               const src_reg &src1 = src_reg());

   src_reg fix_math_operand(const src_reg &src);

   void emit_pack_half_2x16(dst_reg dst, src_reg src0);
   void emit_unpack_half_2x16(dst_reg dst, src_reg src0);
   void emit_unpack_unorm_4x8(const dst_reg &dst, src_reg src0);
   void emit_unpack_snorm_4x8(const dst_reg &dst, src_reg src0);
   void emit_pack_unorm_4x8(const dst_reg &dst, const src_reg &src0);
   void emit_pack_snorm_4x8(const dst_reg &dst, const src_reg &src0);

   void emit_texture(ir_texture_opcode op,
                     dst_reg dest,
                     const glsl_type *dest_type,
                     src_reg coordinate,
                     int coord_components,
                     src_reg shadow_comparitor,
                     src_reg lod, src_reg lod2,
                     src_reg sample_index,
                     uint32_t constant_offset,
                     src_reg offset_value,
                     src_reg mcs,
                     bool is_cube_array,
                     uint32_t sampler, src_reg sampler_reg);

   uint32_t gather_channel(unsigned gather_component, uint32_t sampler);
   src_reg emit_mcs_fetch(const glsl_type *coordinate_type, src_reg coordinate,
                          src_reg sampler);
   void emit_gen6_gather_wa(uint8_t wa, dst_reg dst);
   void swizzle_result(ir_texture_opcode op, dst_reg dest,
                       src_reg orig_val, uint32_t sampler,
                       const glsl_type *dest_type);

   void emit_ndc_computation();
   void emit_psiz_and_flags(dst_reg reg);
   vec4_instruction *emit_generic_urb_slot(dst_reg reg, int varying);
   virtual void emit_urb_slot(dst_reg reg, int varying);

   void emit_shader_time_begin();
   void emit_shader_time_end();
   void emit_shader_time_write(int shader_time_subindex, src_reg value);

   void emit_untyped_atomic(unsigned atomic_op, unsigned surf_index,
                            dst_reg dst, src_reg offset, src_reg src0,
                            src_reg src1);

   void emit_untyped_surface_read(unsigned surf_index, dst_reg dst,
                                  src_reg offset);

   src_reg get_scratch_offset(bblock_t *block, vec4_instruction *inst,
			      src_reg *reladdr, int reg_offset);
   src_reg get_pull_constant_offset(bblock_t *block, vec4_instruction *inst,
				    src_reg *reladdr, int reg_offset);
   void emit_scratch_read(bblock_t *block, vec4_instruction *inst,
			  dst_reg dst,
			  src_reg orig_src,
			  int base_offset);
   void emit_scratch_write(bblock_t *block, vec4_instruction *inst,
			   int base_offset);
   void emit_pull_constant_load(bblock_t *block, vec4_instruction *inst,
				dst_reg dst,
				src_reg orig_src,
				int base_offset);
   void emit_pull_constant_load_reg(dst_reg dst,
                                    src_reg surf_index,
                                    src_reg offset,
                                    bblock_t *before_block,
                                    vec4_instruction *before_inst);
   src_reg emit_resolve_reladdr(int scratch_loc[], bblock_t *block,
                                vec4_instruction *inst, src_reg src);

   void resolve_ud_negate(src_reg *reg);

   src_reg get_timestamp();

   void dump_instruction(backend_instruction *inst);
   void dump_instruction(backend_instruction *inst, FILE *file);

   bool is_high_sampler(src_reg sampler);

   virtual void emit_nir_code();
   virtual void nir_setup_inputs();
   virtual void nir_setup_uniforms();
   virtual void nir_setup_system_value_intrinsic(nir_intrinsic_instr *instr);
   virtual void nir_setup_system_values();
   virtual void nir_emit_impl(nir_function_impl *impl);
   virtual void nir_emit_cf_list(exec_list *list);
   virtual void nir_emit_if(nir_if *if_stmt);
   virtual void nir_emit_loop(nir_loop *loop);
   virtual void nir_emit_block(nir_block *block);
   virtual void nir_emit_instr(nir_instr *instr);
   virtual void nir_emit_load_const(nir_load_const_instr *instr);
   virtual void nir_emit_intrinsic(nir_intrinsic_instr *instr);
   virtual void nir_emit_alu(nir_alu_instr *instr);
   virtual void nir_emit_jump(nir_jump_instr *instr);
   virtual void nir_emit_texture(nir_tex_instr *instr);
   virtual void nir_emit_undef(nir_ssa_undef_instr *instr);
   virtual void nir_emit_ssbo_atomic(int op, nir_intrinsic_instr *instr);

   dst_reg get_nir_dest(nir_dest dest, enum brw_reg_type type);
   dst_reg get_nir_dest(nir_dest dest, nir_alu_type type);
   dst_reg get_nir_dest(nir_dest dest);
   src_reg get_nir_src(nir_src src, enum brw_reg_type type,
                       unsigned num_components = 4);
   src_reg get_nir_src(nir_src src, nir_alu_type type,
                       unsigned num_components = 4);
   src_reg get_nir_src(nir_src src,
                       unsigned num_components = 4);

   virtual dst_reg *make_reg_for_system_value(int location,
                                              const glsl_type *type) = 0;

   dst_reg *nir_locals;
   dst_reg *nir_ssa_values;
   src_reg *nir_inputs;
   dst_reg *nir_system_values;

protected:
   void emit_vertex();
   void lower_attributes_to_hw_regs(const int *attribute_map,
                                    bool interleaved);
   void setup_payload_interference(struct ra_graph *g, int first_payload_node,
                                   int reg_node_count);
   virtual void setup_payload() = 0;
   virtual void emit_prolog() = 0;
   virtual void emit_thread_end() = 0;
   virtual void emit_urb_write_header(int mrf) = 0;
   virtual vec4_instruction *emit_urb_write_opcode(bool complete) = 0;
   virtual void gs_emit_vertex(int stream_id);
   virtual void gs_end_primitive();

private:
   /**
    * If true, then register allocation should fail instead of spilling.
    */
   const bool no_spills;

   int shader_time_index;

   unsigned last_scratch; /**< measured in 32-byte (register size) units */
};


/**
 * The vertex shader code generator.
 *
 * Translates VS IR to actual i965 assembly code.
 */
class vec4_generator
{
public:
   vec4_generator(const struct brw_compiler *compiler, void *log_data,
                  struct brw_vue_prog_data *prog_data,
                  void *mem_ctx,
                  bool debug_flag,
                  const char *stage_name,
                  const char *stage_abbrev);
   ~vec4_generator();

   const unsigned *generate_assembly(const cfg_t *cfg, unsigned *asm_size,
                                     const nir_shader *nir);

private:
   void generate_code(const cfg_t *cfg, const nir_shader *nir);

   void generate_math1_gen4(vec4_instruction *inst,
			    struct brw_reg dst,
			    struct brw_reg src);
   void generate_math2_gen4(vec4_instruction *inst,
			    struct brw_reg dst,
			    struct brw_reg src0,
			    struct brw_reg src1);
   void generate_math_gen6(vec4_instruction *inst,
                           struct brw_reg dst,
                           struct brw_reg src0,
                           struct brw_reg src1);

   void generate_tex(vec4_instruction *inst,
                     struct brw_reg dst,
                     struct brw_reg src,
                     struct brw_reg sampler_index);

   void generate_vs_urb_write(vec4_instruction *inst);
   void generate_gs_urb_write(vec4_instruction *inst);
   void generate_gs_urb_write_allocate(vec4_instruction *inst);
   void generate_gs_thread_end(vec4_instruction *inst);
   void generate_gs_set_write_offset(struct brw_reg dst,
                                     struct brw_reg src0,
                                     struct brw_reg src1);
   void generate_gs_set_vertex_count(struct brw_reg dst,
                                     struct brw_reg src);
   void generate_gs_svb_write(vec4_instruction *inst,
                              struct brw_reg dst,
                              struct brw_reg src0,
                              struct brw_reg src1);
   void generate_gs_svb_set_destination_index(vec4_instruction *inst,
                                              struct brw_reg dst,
                                              struct brw_reg src);
   void generate_gs_set_dword_2(struct brw_reg dst, struct brw_reg src);
   void generate_gs_prepare_channel_masks(struct brw_reg dst);
   void generate_gs_set_channel_masks(struct brw_reg dst, struct brw_reg src);
   void generate_gs_get_instance_id(struct brw_reg dst);
   void generate_gs_ff_sync_set_primitives(struct brw_reg dst,
                                           struct brw_reg src0,
                                           struct brw_reg src1,
                                           struct brw_reg src2);
   void generate_gs_ff_sync(vec4_instruction *inst,
                            struct brw_reg dst,
                            struct brw_reg src0,
                            struct brw_reg src1);
   void generate_gs_set_primitive_id(struct brw_reg dst);
   void generate_oword_dual_block_offsets(struct brw_reg m1,
					  struct brw_reg index);
   void generate_scratch_write(vec4_instruction *inst,
			       struct brw_reg dst,
			       struct brw_reg src,
			       struct brw_reg index);
   void generate_scratch_read(vec4_instruction *inst,
			      struct brw_reg dst,
			      struct brw_reg index);
   void generate_pull_constant_load(vec4_instruction *inst,
				    struct brw_reg dst,
				    struct brw_reg index,
				    struct brw_reg offset);
   void generate_pull_constant_load_gen7(vec4_instruction *inst,
                                         struct brw_reg dst,
                                         struct brw_reg surf_index,
                                         struct brw_reg offset);
   void generate_set_simd4x2_header_gen9(vec4_instruction *inst,
                                         struct brw_reg dst);

   void generate_get_buffer_size(vec4_instruction *inst,
                                 struct brw_reg dst,
                                 struct brw_reg src,
                                 struct brw_reg index);

   void generate_unpack_flags(struct brw_reg dst);

   const struct brw_compiler *compiler;
   void *log_data; /* Passed to compiler->*_log functions */

   const struct brw_device_info *devinfo;

   struct brw_codegen *p;

   struct brw_vue_prog_data *prog_data;

   void *mem_ctx;
   const char *stage_name;
   const char *stage_abbrev;
   const bool debug_flag;
};

} /* namespace brw */
#endif /* __cplusplus */

#endif /* BRW_VEC4_H */