/*
 * Copyright © 2014-2015 Broadcom
 * Copyright © 2021 Google
 *
 * 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.
 */

#include "nir_builder.h"

nir_ssa_def *
nir_builder_alu_instr_finish_and_insert(nir_builder *build, nir_alu_instr *instr)
{
   const nir_op_info *op_info = &nir_op_infos[instr->op];

   instr->exact = build->exact;

   /* Guess the number of components the destination temporary should have
    * based on our input sizes, if it's not fixed for the op.
    */
   unsigned num_components = op_info->output_size;
   if (num_components == 0) {
      for (unsigned i = 0; i < op_info->num_inputs; i++) {
         if (op_info->input_sizes[i] == 0)
            num_components = MAX2(num_components,
                                  instr->src[i].src.ssa->num_components);
      }
   }
   assert(num_components != 0);

   /* Figure out the bitwidth based on the source bitwidth if the instruction
    * is variable-width.
    */
   unsigned bit_size = nir_alu_type_get_type_size(op_info->output_type);
   if (bit_size == 0) {
      for (unsigned i = 0; i < op_info->num_inputs; i++) {
         unsigned src_bit_size = instr->src[i].src.ssa->bit_size;
         if (nir_alu_type_get_type_size(op_info->input_types[i]) == 0) {
            if (bit_size)
               assert(src_bit_size == bit_size);
            else
               bit_size = src_bit_size;
         } else {
            assert(src_bit_size ==
               nir_alu_type_get_type_size(op_info->input_types[i]));
         }
      }
   }

   /* When in doubt, assume 32. */
   if (bit_size == 0)
      bit_size = 32;

   /* Make sure we don't swizzle from outside of our source vector (like if a
    * scalar value was passed into a multiply with a vector).
    */
   for (unsigned i = 0; i < op_info->num_inputs; i++) {
      for (unsigned j = instr->src[i].src.ssa->num_components;
           j < NIR_MAX_VEC_COMPONENTS; j++) {
         instr->src[i].swizzle[j] = instr->src[i].src.ssa->num_components - 1;
      }
   }

   nir_ssa_dest_init(&instr->instr, &instr->dest.dest, num_components,
                     bit_size, NULL);
   instr->dest.write_mask = (1 << num_components) - 1;

   nir_builder_instr_insert(build, &instr->instr);

   return &instr->dest.dest.ssa;
}