src/glsl/ir_optimization.h

   1 /*
   2  * Copyright © 2010 Intel Corporation
   3  *
   4  * Permission is hereby granted, free of charge, to any person obtaining a
   5  * copy of this software and associated documentation files (the "Software"),
   6  * to deal in the Software without restriction, including without limitation
   7  * the rights to use, copy, modify, merge, publish, distribute, sublicense,
   8  * and/or sell copies of the Software, and to permit persons to whom the
   9  * Software is furnished to do so, subject to the following conditions:
  10  *
  11  * The above copyright notice and this permission notice (including the next
  12  * paragraph) shall be included in all copies or substantial portions of the
  13  * Software.
  14  *
  15  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  16  * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  17  * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
  18  * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
  19  * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
  20  * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
  21  * DEALINGS IN THE SOFTWARE.
  22  */
  23
  24
  25 /**
  26  * \file ir_optimization.h
  27  *
  28  * Prototypes for optimization passes to be called by the compiler and drivers.
  29  */
  30
  31 /* Operations for lower_instructions() */
  32 #define SUB_TO_ADD_NEG     0x01
  33 #define DIV_TO_MUL_RCP     0x02
  34 #define EXP_TO_EXP2        0x04
  35 #define POW_TO_EXP2        0x08
  36 #define LOG_TO_LOG2        0x10
  37 #define MOD_TO_FRACT       0x20
  38 #define INT_DIV_TO_MUL_RCP 0x40
  39 #define LRP_TO_ARITH       0x80
  40
  41 /**
  42  * \see class lower_packing_builtins_visitor
  43  */
  44 enum lower_packing_builtins_op {
  45    LOWER_PACK_UNPACK_NONE               = 0x0000,
  46
  47    LOWER_PACK_SNORM_2x16                = 0x0001,
  48    LOWER_UNPACK_SNORM_2x16              = 0x0002,
  49
  50    LOWER_PACK_UNORM_2x16                = 0x0004,
  51    LOWER_UNPACK_UNORM_2x16              = 0x0008,
  52
  53    LOWER_PACK_HALF_2x16                 = 0x0010,
  54    LOWER_UNPACK_HALF_2x16               = 0x0020,
  55
  56    LOWER_PACK_HALF_2x16_TO_SPLIT        = 0x0040,
  57    LOWER_UNPACK_HALF_2x16_TO_SPLIT      = 0x0080,
  58
  59    LOWER_PACK_SNORM_4x8                 = 0x0100,
  60    LOWER_UNPACK_SNORM_4x8               = 0x0200,
  61
  62    LOWER_PACK_UNORM_4x8                 = 0x0400,
  63    LOWER_UNPACK_UNORM_4x8               = 0x0800,
  64 };
  65
  66 bool do_common_optimization(exec_list *ir, bool linked,
  67                             bool uniform_locations_assigned,
  68                             unsigned max_unroll_iterations);
  69
  70 bool do_algebraic(exec_list *instructions);
  71 bool do_constant_folding(exec_list *instructions);
  72 bool do_constant_variable(exec_list *instructions);
  73 bool do_constant_variable_unlinked(exec_list *instructions);
  74 bool do_copy_propagation(exec_list *instructions);
  75 bool do_copy_propagation_elements(exec_list *instructions);
  76 bool do_constant_propagation(exec_list *instructions);
  77 bool do_dead_code(exec_list *instructions, bool uniform_locations_assigned);
  78 bool do_dead_code_local(exec_list *instructions);
  79 bool do_dead_code_unlinked(exec_list *instructions);
  80 bool do_dead_functions(exec_list *instructions);
  81 bool do_function_inlining(exec_list *instructions);
  82 bool do_lower_jumps(exec_list *instructions, bool pull_out_jumps = true, bool lower_sub_return = true, bool lower_main_return = false, bool lower_continue = false, bool lower_break = false);
  83 bool do_lower_texture_projection(exec_list *instructions);
  84 bool do_if_simplification(exec_list *instructions);
  85 bool opt_flatten_nested_if_blocks(exec_list *instructions);
  86 bool do_discard_simplification(exec_list *instructions);
  87 bool lower_if_to_cond_assign(exec_list *instructions, unsigned max_depth = 0);
  88 bool do_mat_op_to_vec(exec_list *instructions);
  89 bool do_noop_swizzle(exec_list *instructions);
  90 bool do_structure_splitting(exec_list *instructions);
  91 bool do_swizzle_swizzle(exec_list *instructions);
  92 bool do_tree_grafting(exec_list *instructions);
  93 bool do_vec_index_to_cond_assign(exec_list *instructions);
  94 bool do_vec_index_to_swizzle(exec_list *instructions);
  95 bool lower_discard(exec_list *instructions);
  96 void lower_discard_flow(exec_list *instructions);
  97 bool lower_instructions(exec_list *instructions, unsigned what_to_lower);
  98 bool lower_noise(exec_list *instructions);
  99 bool lower_variable_index_to_cond_assign(exec_list *instructions,
 100     bool lower_input, bool lower_output, bool lower_temp, bool lower_uniform);
 101 bool lower_quadop_vector(exec_list *instructions, bool dont_lower_swz);
 102 bool lower_clip_distance(gl_shader *shader);
 103 void lower_output_reads(exec_list *instructions);
 104 bool lower_packing_builtins(exec_list *instructions, int op_mask);
 105 void lower_ubo_reference(struct gl_shader *shader, exec_list *instructions);
 106 void lower_packed_varyings(void *mem_ctx, unsigned location_base,
 107                            unsigned locations_used, ir_variable_mode mode,
 108                            gl_shader *shader);
 109 bool optimize_redundant_jumps(exec_list *instructions);
 110 bool optimize_split_arrays(exec_list *instructions, bool linked);
 111
 112 ir_rvalue *
 113 compare_index_block(exec_list *instructions, ir_variable *index,
 114                     unsigned base, unsigned components, void *mem_ctx);