nir/algebraic: Eliminate useless fsat() on operand of comparison w/value in (0, 1)

v2: Fix copy-and-paste bug in a cmp b vs b cmp a cases.

All Gen7+ platforms had similar results. (Ice Lake shown)
total instructions in shared programs: 17224337 -> 17224269 (<.01%)
instructions in affected programs: 13578 -> 13510 (-0.50%)
helped: 68
HURT: 0
helped stats (abs) min: 1 max: 1 x̄: 1.00 x̃: 1
helped stats (rel) min: 0.31% max: 3.12% x̄: 0.84% x̃: 0.42%
95% mean confidence interval for instructions value: -1.00 -1.00
95% mean confidence interval for instructions %-change: -1.05% -0.63%
Instructions are helped.

total cycles in shared programs: 360826090 -> 360825137 (<.01%)
cycles in affected programs: 94867 -> 93914 (-1.00%)
helped: 58
HURT: 1
helped stats (abs) min: 2 max: 28 x̄: 17.74 x̃: 18
helped stats (rel) min: 0.08% max: 3.17% x̄: 1.39% x̃: 1.22%
HURT stats (abs)   min: 76 max: 76 x̄: 76.00 x̃: 76
HURT stats (rel)   min: 2.86% max: 2.86% x̄: 2.86% x̃: 2.86%
95% mean confidence interval for cycles value: -19.53 -12.78
95% mean confidence interval for cycles %-change: -1.56% -1.08%
Cycles are helped.

No changes on any other Intel platform.

Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Thomas Helland <thomashelland90@gmail.com>

diff --git a/src/compiler/nir/nir_opt_algebraic.py b/src/compiler/nir/nir_opt_algebraic.py
index 5fc6fd8..79058c0 100644 (file)
--- a/src/compiler/nir/nir_opt_algebraic.py
+++ b/src/compiler/nir/nir_opt_algebraic.py
@@ -213,6 +213,18 @@ optimizations = [
    (('fne', ('fneg', a), -1.0), ('fne', 1.0, a)),
    (('feq', -1.0, ('fneg', a)), ('feq', a, 1.0)),
 
+   (('flt', ('fsat(is_used_once)', a), '#b(is_gt_0_and_lt_1)'), ('flt', a, b)),
+   (('flt', '#b(is_gt_0_and_lt_1)', ('fsat(is_used_once)', a)), ('flt', b, a)),
+   (('fge', ('fsat(is_used_once)', a), '#b(is_gt_0_and_lt_1)'), ('fge', a, b)),
+   (('fge', '#b(is_gt_0_and_lt_1)', ('fsat(is_used_once)', a)), ('fge', b, a)),
+   (('feq', ('fsat(is_used_once)', a), '#b(is_gt_0_and_lt_1)'), ('feq', a, b)),
+   (('fne', ('fsat(is_used_once)', a), '#b(is_gt_0_and_lt_1)'), ('fne', a, b)),
+
+   (('fge', ('fsat(is_used_once)', a), 1.0), ('fge', a, 1.0)),
+   (('flt', ('fsat(is_used_once)', a), 1.0), ('flt', a, 1.0)),
+   (('fge', 0.0, ('fsat(is_used_once)', a)), ('fge', 0.0, a)),
+   (('flt', 0.0, ('fsat(is_used_once)', a)), ('flt', 0.0, a)),
+
    # 0.0 >= b2f(a)
    # b2f(a) <= 0.0
    # b2f(a) == 0.0 because b2f(a) can only be 0 or 1
@@ -1136,6 +1148,21 @@ late_optimizations = [
    # optimization loop can prevent other optimizations.
    (('fneg', ('fneg', a)), a),
 
+   # These are duplicated from the main optimizations table.  The late
+   # patterns that rearrange expressions like x - .5 < 0 to x < .5 can create
+   # new patterns like these.  The patterns that compare with zero are removed
+   # because they are unlikely to be created in by anything in
+   # late_optimizations.
+   (('flt', ('fsat(is_used_once)', a), '#b(is_gt_0_and_lt_1)'), ('flt', a, b)),
+   (('flt', '#b(is_gt_0_and_lt_1)', ('fsat(is_used_once)', a)), ('flt', b, a)),
+   (('fge', ('fsat(is_used_once)', a), '#b(is_gt_0_and_lt_1)'), ('fge', a, b)),
+   (('fge', '#b(is_gt_0_and_lt_1)', ('fsat(is_used_once)', a)), ('fge', b, a)),
+   (('feq', ('fsat(is_used_once)', a), '#b(is_gt_0_and_lt_1)'), ('feq', a, b)),
+   (('fne', ('fsat(is_used_once)', a), '#b(is_gt_0_and_lt_1)'), ('fne', a, b)),
+
+   (('fge', ('fsat(is_used_once)', a), 1.0), ('fge', a, 1.0)),
+   (('flt', ('fsat(is_used_once)', a), 1.0), ('flt', a, 1.0)),
+
    (('~fge', ('fmin(is_used_once)', ('fadd(is_used_once)', a, b), ('fadd', c, d)), 0.0), ('iand', ('fge', a, ('fneg', b)), ('fge', c, ('fneg', d)))),
 
    (('flt', ('fneg', a), ('fneg', b)), ('flt', b, a)),

diff --git a/src/compiler/nir/nir_search_helpers.h b/src/compiler/nir/nir_search_helpers.h
index dd220df..9e03c61 100644 (file)
--- a/src/compiler/nir/nir_search_helpers.h
+++ b/src/compiler/nir/nir_search_helpers.h
@@ -109,6 +109,36 @@ is_zero_to_one(nir_alu_instr *instr, unsigned src, unsigned num_components,
    return true;
 }
 
+/**
+ * Exclusive compare with (0, 1).
+ *
+ * This differs from \c is_zero_to_one because that function tests 0 <= src <=
+ * 1 while this function tests 0 < src < 1.
+ */
+static inline bool
+is_gt_0_and_lt_1(nir_alu_instr *instr, unsigned src, unsigned num_components,
+                 const uint8_t *swizzle)
+{
+   /* only constant srcs: */
+   if (!nir_src_is_const(instr->src[src].src))
+      return false;
+
+   for (unsigned i = 0; i < num_components; i++) {
+      switch (nir_op_infos[instr->op].input_types[src]) {
+      case nir_type_float: {
+         double val = nir_src_comp_as_float(instr->src[src].src, swizzle[i]);
+         if (isnan(val) || val <= 0.0f || val >= 1.0f)
+            return false;
+         break;
+      }
+      default:
+         return false;
+      }
+   }
+
+   return true;
+}
+
 static inline bool
 is_not_const_zero(nir_alu_instr *instr, unsigned src, unsigned num_components,
                   const uint8_t *swizzle)