[Mesa-dev] [PATCH v2 5/5] nir: link time opt duplicate varyings

[Timothy Arceri]

If we are outputting the same value to more than one output
component rewrite the inputs to read from a single component.

This will allow the duplicate varying components to be optimised
away by the existing opts.

shader-db results i965 (SKL):

total instructions in shared programs: 12869230 -> 12860886 (-0.06%)
instructions in affected programs: 322601 -> 314257 (-2.59%)
helped: 3080
HURT: 8

total cycles in shared programs: 317792574 -> 317730593 (-0.02%)
cycles in affected programs: 2584925 -> 2522944 (-2.40%)
helped: 2975
HURT: 477

shader-db results radeonsi (VEGA):

SGPRS: 31576 -> 31664 (0.28 %)
VGPRS: 17484 -> 17064 (-2.40 %)
Spilled SGPRs: 184 -> 167 (-9.24 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Private memory VGPRs: 0 -> 0 (0.00 %)
Scratch size: 0 -> 0 (0.00 %) dwords per thread
Code Size: 583340 -> 569368 (-2.40 %) bytes
LDS: 0 -> 0 (0.00 %) blocks
Max Waves: 6162 -> 6270 (1.75 %)
Wait states: 0 -> 0 (0.00 %)

vkpipeline-db results RADV (VEGA):

Totals from affected shaders:
SGPRS: 14880 -> 15080 (1.34 %)
VGPRS: 10872 -> 10888 (0.15 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Private memory VGPRs: 0 -> 0 (0.00 %)
Scratch size: 0 -> 0 (0.00 %) dwords per thread
Code Size: 674016 -> 668396 (-0.83 %) bytes
LDS: 0 -> 0 (0.00 %) blocks
Max Waves: 2708 -> 2704 (-0.15 %)
Wait states: 0 -> 0 (0.00 %

V2: bunch of tidy ups suggested by Jason
---
 src/compiler/nir/nir_linking_helpers.c | 83 ++++++++++++++++++++++++++
 1 file changed, 83 insertions(+)

diff --git a/src/compiler/nir/nir_linking_helpers.c b/src/compiler/nir/nir_linking_helpers.c
index 2a3e6f407c..b74f57e6d8 100644
--- a/src/compiler/nir/nir_linking_helpers.c
+++ b/src/compiler/nir/nir_linking_helpers.c
@@ -699,6 +699,21 @@ nir_link_xfb_varyings(nir_shader *producer, nir_shader *consumer)
    }
 }
 
+static nir_variable *
+get_matching_input_var(nir_shader *consumer, nir_variable *out_var)
+{
+   nir_foreach_variable(var, &consumer->inputs) {
+      if (var->data.location == out_var->data.location &&
+          var->data.location_frac == out_var->data.location_frac &&
+          var->data.interpolation == out_var->data.interpolation &&
+          get_interp_loc(var) == get_interp_loc(out_var)) {
+         return var;
+      }
+   }
+
+   return NULL;
+}
+
 static bool
 can_replace_varying(nir_variable *out_var)
 {
@@ -774,6 +789,54 @@ replace_constant_input(nir_shader *shader, nir_intrinsic_instr *store_intr)
    return progress;
 }
 
+static bool
+replace_duplicate_input(nir_shader *shader, nir_variable *input_var,
+                         nir_intrinsic_instr *dup_store_intr)
+{
+   assert(input_var);
+
+   nir_function_impl *impl = nir_shader_get_entrypoint(shader);
+
+   nir_builder b;
+   nir_builder_init(&b, impl);
+
+   nir_variable *dup_out_var =
+      nir_deref_instr_get_variable(nir_src_as_deref(dup_store_intr->src[0]));
+
+   bool progress = false;
+   nir_foreach_block(block, impl) {
+      nir_foreach_instr(instr, block) {
+         if (instr->type != nir_instr_type_intrinsic)
+            continue;
+
+         nir_intrinsic_instr *intr = nir_instr_as_intrinsic(instr);
+         if (intr->intrinsic != nir_intrinsic_load_deref)
+            continue;
+
+         nir_deref_instr *in_deref = nir_src_as_deref(intr->src[0]);
+         if (in_deref->mode != nir_var_shader_in)
+            continue;
+
+         nir_variable *in_var = nir_deref_instr_get_variable(in_deref);
+
+         if (in_var->data.location != dup_out_var->data.location ||
+             in_var->data.location_frac != dup_out_var->data.location_frac ||
+             in_var->data.interpolation != input_var->data.interpolation ||
+             get_interp_loc(in_var) != get_interp_loc(input_var))
+            continue;
+
+         b.cursor = nir_before_instr(instr);
+
+         nir_ssa_def *load = nir_load_var(&b, input_var);
+         nir_ssa_def_rewrite_uses(&intr->dest.ssa, nir_src_for_ssa(load));
+
+         progress = true;
+      }
+   }
+
+   return progress;
+}
+
 bool
 nir_link_opt_varyings(nir_shader *producer, nir_shader *consumer)
 {
@@ -787,6 +850,10 @@ nir_link_opt_varyings(nir_shader *producer, nir_shader *consumer)
 
    nir_function_impl *impl = nir_shader_get_entrypoint(producer);
 
+   struct hash_table *varying_values =
+      _mesa_hash_table_create(NULL,  _mesa_hash_pointer,
+                              _mesa_key_pointer_equal);
+
    /* If we find a store in the last block of the producer we can be sure this
     * is the only possible value for this output.
     */
@@ -810,8 +877,24 @@ nir_link_opt_varyings(nir_shader *producer, nir_shader *consumer)
 
       if (intr->src[1].ssa->parent_instr->type == nir_instr_type_load_const) {
          progress |= replace_constant_input(consumer, intr);
+      } else {
+         struct hash_entry *entry =
+               _mesa_hash_table_search(varying_values, intr->src[1].ssa);
+         if (entry) {
+            progress |= replace_duplicate_input(consumer,
+                                                (nir_variable *) entry->data,
+                                                intr);
+         } else {
+            nir_variable *in_var = get_matching_input_var(consumer, out_var);
+            if (in_var) {
+               _mesa_hash_table_insert(varying_values, intr->src[1].ssa,
+                                       in_var);
+            }
+         }
       }
    }
 
+   _mesa_hash_table_destroy(varying_values, NULL);
+
    return progress;
 }
-- 
2.19.2