X-Git-Url: https://git.libre-soc.org/?a=blobdiff_plain;f=src%2Fmesa%2Fstate_tracker%2Fst_glsl_to_tgsi.cpp;h=89ad6cd8c287fe92cdd6beb0189a1cdd1ad82aea;hb=8498cb4a45e8ed53a2ee2b35d3c2cbb9963e1756;hp=a54ee17173af25ca34418c2732d3347dcc72a047;hpb=f4ec81032bb9c1460794d3d843d0ffe47a181291;p=mesa.git diff --git a/src/mesa/state_tracker/st_glsl_to_tgsi.cpp b/src/mesa/state_tracker/st_glsl_to_tgsi.cpp index a54ee17173a..89ad6cd8c28 100644 --- a/src/mesa/state_tracker/st_glsl_to_tgsi.cpp +++ b/src/mesa/state_tracker/st_glsl_to_tgsi.cpp @@ -1655,7 +1655,13 @@ glsl_to_tgsi_visitor::visit(ir_expression *ir) st_dst_reg temp_dst = st_dst_reg(temp); st_src_reg temp1 = st_src_reg(temp), temp2 = st_src_reg(temp); - emit_asm(ir, TGSI_OPCODE_SEQ, st_dst_reg(temp), op[0], op[1]); + if (ir->operands[0]->type->is_boolean() && + ir->operands[1]->as_constant() && + ir->operands[1]->as_constant()->is_one()) { + emit_asm(ir, TGSI_OPCODE_MOV, st_dst_reg(temp), op[0]); + } else { + emit_asm(ir, TGSI_OPCODE_SEQ, st_dst_reg(temp), op[0], op[1]); + } /* Emit 1-3 AND operations to combine the SEQ results. */ switch (ir->operands[0]->type->vector_elements) { @@ -1708,7 +1714,13 @@ glsl_to_tgsi_visitor::visit(ir_expression *ir) st_src_reg temp = get_temp(native_integers ? glsl_type::uvec4_type : glsl_type::vec4_type); - emit_asm(ir, TGSI_OPCODE_SNE, st_dst_reg(temp), op[0], op[1]); + if (ir->operands[0]->type->is_boolean() && + ir->operands[1]->as_constant() && + ir->operands[1]->as_constant()->is_zero()) { + emit_asm(ir, TGSI_OPCODE_MOV, st_dst_reg(temp), op[0]); + } else { + emit_asm(ir, TGSI_OPCODE_SNE, st_dst_reg(temp), op[0], op[1]); + } if (native_integers) { st_dst_reg temp_dst = st_dst_reg(temp); @@ -1764,89 +1776,6 @@ glsl_to_tgsi_visitor::visit(ir_expression *ir) } break; - case ir_unop_any: { - assert(ir->operands[0]->type->is_vector()); - - if (native_integers) { - int dst_swizzle = 0, op0_swizzle, i; - st_src_reg accum = op[0]; - - op0_swizzle = op[0].swizzle; - accum.swizzle = MAKE_SWIZZLE4(GET_SWZ(op0_swizzle, 0), - GET_SWZ(op0_swizzle, 0), - GET_SWZ(op0_swizzle, 0), - GET_SWZ(op0_swizzle, 0)); - for (i = 0; i < 4; i++) { - if (result_dst.writemask & (1 << i)) { - dst_swizzle = MAKE_SWIZZLE4(i, i, i, i); - break; - } - } - assert(i != 4); - assert(ir->operands[0]->type->is_boolean()); - - /* OR all the components together, since they should be either 0 or ~0 - */ - switch (ir->operands[0]->type->vector_elements) { - case 4: - op[0].swizzle = MAKE_SWIZZLE4(GET_SWZ(op0_swizzle, 3), - GET_SWZ(op0_swizzle, 3), - GET_SWZ(op0_swizzle, 3), - GET_SWZ(op0_swizzle, 3)); - emit_asm(ir, TGSI_OPCODE_OR, result_dst, accum, op[0]); - accum = st_src_reg(result_dst); - accum.swizzle = dst_swizzle; - /* fallthrough */ - case 3: - op[0].swizzle = MAKE_SWIZZLE4(GET_SWZ(op0_swizzle, 2), - GET_SWZ(op0_swizzle, 2), - GET_SWZ(op0_swizzle, 2), - GET_SWZ(op0_swizzle, 2)); - emit_asm(ir, TGSI_OPCODE_OR, result_dst, accum, op[0]); - accum = st_src_reg(result_dst); - accum.swizzle = dst_swizzle; - /* fallthrough */ - case 2: - op[0].swizzle = MAKE_SWIZZLE4(GET_SWZ(op0_swizzle, 1), - GET_SWZ(op0_swizzle, 1), - GET_SWZ(op0_swizzle, 1), - GET_SWZ(op0_swizzle, 1)); - emit_asm(ir, TGSI_OPCODE_OR, result_dst, accum, op[0]); - break; - default: - assert(!"Unexpected vector size"); - break; - } - } else { - /* After the dot-product, the value will be an integer on the - * range [0,4]. Zero stays zero, and positive values become 1.0. - */ - glsl_to_tgsi_instruction *const dp = - emit_dp(ir, result_dst, op[0], op[0], - ir->operands[0]->type->vector_elements); - if (this->prog->Target == GL_FRAGMENT_PROGRAM_ARB && - result_dst.type == GLSL_TYPE_FLOAT) { - /* The clamping to [0,1] can be done for free in the fragment - * shader with a saturate. - */ - dp->saturate = true; - } else if (result_dst.type == GLSL_TYPE_FLOAT) { - /* Negating the result of the dot-product gives values on the range - * [-4, 0]. Zero stays zero, and negative values become 1.0. This - * is achieved using SLT. - */ - st_src_reg slt_src = result_src; - slt_src.negate = ~slt_src.negate; - emit_asm(ir, TGSI_OPCODE_SLT, result_dst, slt_src, st_src_reg_for_float(0.0)); - } - else { - /* Use SNE 0 if integers are being used as boolean values. */ - emit_asm(ir, TGSI_OPCODE_SNE, result_dst, result_src, st_src_reg_for_int(0)); - } - } - break; - } - case ir_binop_logic_xor: if (native_integers) emit_asm(ir, TGSI_OPCODE_XOR, result_dst, op[0], op[1]); @@ -3236,6 +3165,8 @@ glsl_to_tgsi_visitor::visit(ir_texture *ir) case ir_texture_samples: opcode = TGSI_OPCODE_TXQS; break; + case ir_samples_identical: + unreachable("Unexpected ir_samples_identical opcode"); } if (ir->projector) { @@ -4334,138 +4265,6 @@ glsl_to_tgsi_visitor::renumber_registers(void) ralloc_free(first_reads); } -/** - * Returns a fragment program which implements the current pixel transfer ops. - * Based on get_pixel_transfer_program in st_atom_pixeltransfer.c. - */ -extern "C" void -get_pixel_transfer_visitor(struct st_fragment_program *fp, - glsl_to_tgsi_visitor *original, - int scale_and_bias, int pixel_maps) -{ - glsl_to_tgsi_visitor *v = new glsl_to_tgsi_visitor(); - struct st_context *st = st_context(original->ctx); - struct gl_program *prog = &fp->Base.Base; - struct gl_program_parameter_list *params = _mesa_new_parameter_list(); - st_src_reg coord, src0; - st_dst_reg dst0; - glsl_to_tgsi_instruction *inst; - - /* Copy attributes of the glsl_to_tgsi_visitor in the original shader. */ - v->ctx = original->ctx; - v->prog = prog; - v->shader_program = NULL; - v->shader = NULL; - v->glsl_version = original->glsl_version; - v->native_integers = original->native_integers; - v->options = original->options; - v->next_temp = original->next_temp; - v->num_address_regs = original->num_address_regs; - v->samplers_used = prog->SamplersUsed = original->samplers_used; - v->indirect_addr_consts = original->indirect_addr_consts; - memcpy(&v->immediates, &original->immediates, sizeof(v->immediates)); - v->num_immediates = original->num_immediates; - - /* - * Get initial pixel color from the texture. - * TEX colorTemp, fragment.texcoord[0], texture[0], 2D; - */ - coord = st_src_reg(PROGRAM_INPUT, VARYING_SLOT_TEX0, glsl_type::vec2_type); - src0 = v->get_temp(glsl_type::vec4_type); - dst0 = st_dst_reg(src0); - inst = v->emit_asm(NULL, TGSI_OPCODE_TEX, dst0, coord); - inst->sampler_array_size = 1; - inst->tex_target = TEXTURE_2D_INDEX; - - prog->InputsRead |= VARYING_BIT_TEX0; - prog->SamplersUsed |= (1 << 0); /* mark sampler 0 as used */ - v->samplers_used |= (1 << 0); - - if (scale_and_bias) { - static const gl_state_index scale_state[STATE_LENGTH] = - { STATE_INTERNAL, STATE_PT_SCALE, - (gl_state_index) 0, (gl_state_index) 0, (gl_state_index) 0 }; - static const gl_state_index bias_state[STATE_LENGTH] = - { STATE_INTERNAL, STATE_PT_BIAS, - (gl_state_index) 0, (gl_state_index) 0, (gl_state_index) 0 }; - GLint scale_p, bias_p; - st_src_reg scale, bias; - - scale_p = _mesa_add_state_reference(params, scale_state); - bias_p = _mesa_add_state_reference(params, bias_state); - - /* MAD colorTemp, colorTemp, scale, bias; */ - scale = st_src_reg(PROGRAM_STATE_VAR, scale_p, GLSL_TYPE_FLOAT); - bias = st_src_reg(PROGRAM_STATE_VAR, bias_p, GLSL_TYPE_FLOAT); - inst = v->emit_asm(NULL, TGSI_OPCODE_MAD, dst0, src0, scale, bias); - } - - if (pixel_maps) { - st_src_reg temp = v->get_temp(glsl_type::vec4_type); - st_dst_reg temp_dst = st_dst_reg(temp); - - assert(st->pixel_xfer.pixelmap_texture); - (void) st; - - /* With a little effort, we can do four pixel map look-ups with - * two TEX instructions: - */ - - /* TEX temp.rg, colorTemp.rgba, texture[1], 2D; */ - temp_dst.writemask = WRITEMASK_XY; /* write R,G */ - inst = v->emit_asm(NULL, TGSI_OPCODE_TEX, temp_dst, src0); - inst->sampler.index = 1; - inst->sampler_array_size = 1; - inst->tex_target = TEXTURE_2D_INDEX; - - /* TEX temp.ba, colorTemp.baba, texture[1], 2D; */ - src0.swizzle = MAKE_SWIZZLE4(SWIZZLE_Z, SWIZZLE_W, SWIZZLE_Z, SWIZZLE_W); - temp_dst.writemask = WRITEMASK_ZW; /* write B,A */ - inst = v->emit_asm(NULL, TGSI_OPCODE_TEX, temp_dst, src0); - inst->sampler.index = 1; - inst->sampler_array_size = 1; - inst->tex_target = TEXTURE_2D_INDEX; - - prog->SamplersUsed |= (1 << 1); /* mark sampler 1 as used */ - v->samplers_used |= (1 << 1); - - /* MOV colorTemp, temp; */ - inst = v->emit_asm(NULL, TGSI_OPCODE_MOV, dst0, temp); - } - - /* Now copy the instructions from the original glsl_to_tgsi_visitor into the - * new visitor. */ - foreach_in_list(glsl_to_tgsi_instruction, inst, &original->instructions) { - glsl_to_tgsi_instruction *newinst; - st_src_reg src_regs[4]; - - if (inst->dst[0].file == PROGRAM_OUTPUT) - prog->OutputsWritten |= BITFIELD64_BIT(inst->dst[0].index); - - for (int i = 0; i < 4; i++) { - src_regs[i] = inst->src[i]; - if (src_regs[i].file == PROGRAM_INPUT && - src_regs[i].index == VARYING_SLOT_COL0) { - src_regs[i].file = PROGRAM_TEMPORARY; - src_regs[i].index = src0.index; - } - else if (src_regs[i].file == PROGRAM_INPUT) - prog->InputsRead |= BITFIELD64_BIT(src_regs[i].index); - } - - newinst = v->emit_asm(NULL, inst->op, inst->dst[0], src_regs[0], src_regs[1], src_regs[2], src_regs[3]); - newinst->tex_target = inst->tex_target; - newinst->sampler_array_size = inst->sampler_array_size; - } - - /* Make modifications to fragment program info. */ - prog->Parameters = _mesa_combine_parameter_lists(params, - original->prog->Parameters); - _mesa_free_parameter_list(params); - count_resources(v, prog); - fp->glsl_to_tgsi = v; -} - /* ------------------------- TGSI conversion stuff -------------------------- */ struct label { unsigned branch_target; @@ -4540,6 +4339,7 @@ const unsigned _mesa_sysval_to_semantic[SYSTEM_VALUE_MAX] = { TGSI_SEMANTIC_SAMPLEID, TGSI_SEMANTIC_SAMPLEPOS, TGSI_SEMANTIC_SAMPLEMASK, + TGSI_SEMANTIC_HELPER_INVOCATION, /* Tessellation shaders */ @@ -5270,6 +5070,8 @@ st_translate_program( TGSI_SEMANTIC_BASEVERTEX); assert(_mesa_sysval_to_semantic[SYSTEM_VALUE_TESS_COORD] == TGSI_SEMANTIC_TESSCOORD); + assert(_mesa_sysval_to_semantic[SYSTEM_VALUE_HELPER_INVOCATION] == + TGSI_SEMANTIC_HELPER_INVOCATION); t = CALLOC_STRUCT(st_translate); if (!t) { @@ -5523,7 +5325,7 @@ st_translate_program( unsigned num_ubos = program->shader->NumUniformBlocks; for (i = 0; i < num_ubos; i++) { - unsigned size = program->shader->UniformBlocks[i].UniformBufferSize; + unsigned size = program->shader->UniformBlocks[i]->UniformBufferSize; unsigned num_const_vecs = (size + 15) / 16; unsigned first, last; assert(num_const_vecs > 0); @@ -5954,7 +5756,6 @@ st_link_shader(struct gl_context *ctx, struct gl_shader_program *prog) (!ctx->Const.NativeIntegers ? INT_DIV_TO_MUL_RCP : 0) | (options->EmitNoSat ? SAT_TO_CLAMP : 0)); - lower_ubo_reference(prog->_LinkedShaders[i], ir); do_vec_index_to_cond_assign(ir); lower_vector_insert(ir, true); lower_quadop_vector(ir, false);