util: rename list_empty() to list_is_empty()

[mesa.git] / src / compiler / nir / nir_lower_vec_to_movs.c

diff --git a/src/compiler/nir/nir_lower_vec_to_movs.c b/src/compiler/nir/nir_lower_vec_to_movs.c
index b383fb8ccf2b3c2fad2a5704523f65ccfdebd600..eec994f15e53881580659ecfcbce91713b0f5db2 100644 (file)
--- a/src/compiler/nir/nir_lower_vec_to_movs.c
+++ b/src/compiler/nir/nir_lower_vec_to_movs.c
@@ -57,7 +57,7 @@ insert_mov(nir_alu_instr *vec, unsigned start_idx, nir_shader *shader)
 {
    assert(start_idx < nir_op_infos[vec->op].num_inputs);
 
-   nir_alu_instr *mov = nir_alu_instr_create(shader, nir_op_imov);
+   nir_alu_instr *mov = nir_alu_instr_create(shader, nir_op_mov);
    nir_alu_src_copy(&mov->src[0], &vec->src[start_idx], mov);
    nir_alu_dest_copy(&mov->dest, &vec->dest, mov);
 
@@ -118,7 +118,7 @@ has_replicated_dest(nir_alu_instr *alu)
  * can then call insert_mov as normal.
  */
 static unsigned
-try_coalesce(nir_alu_instr *vec, unsigned start_idx, nir_shader *shader)
+try_coalesce(nir_alu_instr *vec, unsigned start_idx)
 {
    assert(start_idx < nir_op_infos[vec->op].num_inputs);
 
@@ -140,7 +140,7 @@ try_coalesce(nir_alu_instr *vec, unsigned start_idx, nir_shader *shader)
          return 0;
    }
 
-   if (!list_empty(&vec->src[start_idx].src.ssa->if_uses))
+   if (!list_is_empty(&vec->src[start_idx].src.ssa->if_uses))
       return 0;
 
    if (vec->src[start_idx].src.ssa->parent_instr->type != nir_instr_type_alu)
@@ -230,6 +230,7 @@ lower_vec_to_movs_block(nir_block *block, nir_function_impl *impl)
          continue; /* The loop */
       }
 
+      bool vec_had_ssa_dest = vec->dest.dest.is_ssa;
       if (vec->dest.dest.is_ssa) {
          /* Since we insert multiple MOVs, we have a register destination. */
          nir_register *reg = nir_local_reg_create(impl);
@@ -263,8 +264,12 @@ lower_vec_to_movs_block(nir_block *block, nir_function_impl *impl)
          if (!(vec->dest.write_mask & (1 << i)))
             continue;
 
-         if (!(finished_write_mask & (1 << i)))
-            finished_write_mask |= try_coalesce(vec, i, shader);
+         /* Coalescing moves the register writes from the vec up to the ALU
+          * instruction in the source.  We can only do this if the original
+          * vecN had an SSA destination.
+          */
+         if (vec_had_ssa_dest && !(finished_write_mask & (1 << i)))
+            finished_write_mask |= try_coalesce(vec, i);
 
          if (!(finished_write_mask & (1 << i)))
             finished_write_mask |= insert_mov(vec, i, shader);