X-Git-Url: https://git.libre-soc.org/?a=blobdiff_plain;f=src%2Fglsl%2Fir_optimization.h;h=be86f547f7752a599d91ace7bfeeba55b23a8714;hb=a39a8fbbaa129f4e52f2a3ad2747182e9a74d910;hp=59a040751d96a20998fae5aad3980bf9b0763e88;hpb=f919547f3785b1d8839b9fc5c00ac397e30896a1;p=mesa.git diff --git a/src/glsl/ir_optimization.h b/src/glsl/ir_optimization.h index 59a040751d9..be86f547f77 100644 --- a/src/glsl/ir_optimization.h +++ b/src/glsl/ir_optimization.h @@ -29,43 +29,119 @@ */ /* Operations for lower_instructions() */ -#define SUB_TO_ADD_NEG 0x01 -#define DIV_TO_MUL_RCP 0x02 -#define EXP_TO_EXP2 0x04 -#define POW_TO_EXP2 0x08 -#define LOG_TO_LOG2 0x10 -#define MOD_TO_FRACT 0x20 +#define SUB_TO_ADD_NEG 0x01 +#define DIV_TO_MUL_RCP 0x02 +#define EXP_TO_EXP2 0x04 +#define POW_TO_EXP2 0x08 +#define LOG_TO_LOG2 0x10 +#define MOD_TO_FLOOR 0x20 +#define INT_DIV_TO_MUL_RCP 0x40 +#define LDEXP_TO_ARITH 0x80 +#define CARRY_TO_ARITH 0x100 +#define BORROW_TO_ARITH 0x200 +#define SAT_TO_CLAMP 0x400 +#define DOPS_TO_DFRAC 0x800 +#define DFREXP_DLDEXP_TO_ARITH 0x1000 -bool do_common_optimization(exec_list *ir, bool linked, unsigned max_unroll_iterations); +/** + * \see class lower_packing_builtins_visitor + */ +enum lower_packing_builtins_op { + LOWER_PACK_UNPACK_NONE = 0x0000, + + LOWER_PACK_SNORM_2x16 = 0x0001, + LOWER_UNPACK_SNORM_2x16 = 0x0002, + + LOWER_PACK_UNORM_2x16 = 0x0004, + LOWER_UNPACK_UNORM_2x16 = 0x0008, + + LOWER_PACK_HALF_2x16 = 0x0010, + LOWER_UNPACK_HALF_2x16 = 0x0020, + + LOWER_PACK_HALF_2x16_TO_SPLIT = 0x0040, + LOWER_UNPACK_HALF_2x16_TO_SPLIT = 0x0080, + + LOWER_PACK_SNORM_4x8 = 0x0100, + LOWER_UNPACK_SNORM_4x8 = 0x0200, -bool do_algebraic(exec_list *instructions); + LOWER_PACK_UNORM_4x8 = 0x0400, + LOWER_UNPACK_UNORM_4x8 = 0x0800, + + LOWER_PACK_USE_BFI = 0x1000, + LOWER_PACK_USE_BFE = 0x2000, +}; + +bool do_common_optimization(exec_list *ir, bool linked, + bool uniform_locations_assigned, + const struct gl_shader_compiler_options *options, + bool native_integers); + +bool do_rebalance_tree(exec_list *instructions); +bool do_algebraic(exec_list *instructions, bool native_integers, + const struct gl_shader_compiler_options *options); +bool opt_conditional_discard(exec_list *instructions); bool do_constant_folding(exec_list *instructions); bool do_constant_variable(exec_list *instructions); bool do_constant_variable_unlinked(exec_list *instructions); bool do_copy_propagation(exec_list *instructions); bool do_copy_propagation_elements(exec_list *instructions); bool do_constant_propagation(exec_list *instructions); -bool do_dead_code(exec_list *instructions); +void do_dead_builtin_varyings(struct gl_context *ctx, + gl_shader *producer, gl_shader *consumer, + unsigned num_tfeedback_decls, + class tfeedback_decl *tfeedback_decls); +bool do_dead_code(exec_list *instructions, bool uniform_locations_assigned); bool do_dead_code_local(exec_list *instructions); bool do_dead_code_unlinked(exec_list *instructions); bool do_dead_functions(exec_list *instructions); +bool opt_flip_matrices(exec_list *instructions); bool do_function_inlining(exec_list *instructions); 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); bool do_lower_texture_projection(exec_list *instructions); bool do_if_simplification(exec_list *instructions); +bool opt_flatten_nested_if_blocks(exec_list *instructions); bool do_discard_simplification(exec_list *instructions); bool lower_if_to_cond_assign(exec_list *instructions, unsigned max_depth = 0); bool do_mat_op_to_vec(exec_list *instructions); +bool do_minmax_prune(exec_list *instructions); bool do_noop_swizzle(exec_list *instructions); bool do_structure_splitting(exec_list *instructions); bool do_swizzle_swizzle(exec_list *instructions); +bool do_vectorize(exec_list *instructions); bool do_tree_grafting(exec_list *instructions); bool do_vec_index_to_cond_assign(exec_list *instructions); bool do_vec_index_to_swizzle(exec_list *instructions); bool lower_discard(exec_list *instructions); +void lower_discard_flow(exec_list *instructions); bool lower_instructions(exec_list *instructions, unsigned what_to_lower); bool lower_noise(exec_list *instructions); -bool lower_variable_index_to_cond_assign(exec_list *instructions, - bool lower_input, bool lower_output, bool lower_temp, bool lower_uniform); +bool lower_variable_index_to_cond_assign(gl_shader_stage stage, + exec_list *instructions, bool lower_input, bool lower_output, + bool lower_temp, bool lower_uniform); bool lower_quadop_vector(exec_list *instructions, bool dont_lower_swz); +bool lower_const_arrays_to_uniforms(exec_list *instructions); +bool lower_clip_distance(gl_shader *shader); +void lower_output_reads(unsigned stage, exec_list *instructions); +bool lower_packing_builtins(exec_list *instructions, int op_mask); +void lower_shared_reference(struct gl_shader *shader, unsigned *shared_size); +void lower_ubo_reference(struct gl_shader *shader); +void lower_packed_varyings(void *mem_ctx, + unsigned locations_used, ir_variable_mode mode, + unsigned gs_input_vertices, gl_shader *shader); +bool lower_vector_insert(exec_list *instructions, bool lower_nonconstant_index); +bool lower_vector_derefs(gl_shader *shader); +void lower_named_interface_blocks(void *mem_ctx, gl_shader *shader); bool optimize_redundant_jumps(exec_list *instructions); +bool optimize_split_arrays(exec_list *instructions, bool linked); +bool lower_offset_arrays(exec_list *instructions); +void optimize_dead_builtin_variables(exec_list *instructions, + enum ir_variable_mode other); +bool lower_tess_level(gl_shader *shader); + +bool lower_vertex_id(gl_shader *shader); + +bool lower_subroutine(exec_list *instructions, struct _mesa_glsl_parse_state *state); + +ir_rvalue * +compare_index_block(exec_list *instructions, ir_variable *index, + unsigned base, unsigned components, void *mem_ctx);