glsl: Optimize min/max expression trees

Original patch by Petri Latvala <[email protected]>:

Add an optimization pass that drops min/max expression operands that
can be proven to not contribute to the final result. The algorithm is
similar to alpha-beta pruning on a minmax search, from the field of
AI.

This optimization pass can optimize min/max expressions where operands
are min/max expressions. Such code can appear in shaders by itself, or
as the result of clamp() or AMD_shader_trinary_minmax functions.

This optimization pass improves the generated code for piglit's
AMD_shader_trinary_minmax tests as follows:

total instructions in shared programs: 75 -> 67 (-10.67%)
instructions in affected programs:     60 -> 52 (-13.33%)
GAINED:                                0
LOST:                                  0

All tests (max3, min3, mid3) improved.

A full shader-db run:

total instructions in shared programs: 4293603 -> 4293575 (-0.00%)
instructions in affected programs:     1188 -> 1160 (-2.36%)
GAINED:                                0
LOST:                                  0

Improvements happen in Guacamelee and Serious Sam 3. One shader from
Dungeon Defenders is hurt by shader-db metrics (26 -> 28), because of
dropping of a (constant float (0.00000)) operand, which was
compiled to a saturate modifier.

Version 2 by Iago Toral Quiroga <[email protected]>:

Changes from review feedback:
- Squashed various cosmetic changes sent by Matt Turner.
- Make less_all_components return an enum rather than setting a class member.
  (Suggested by Mat Turner). Also, renamed it to compare_components.
- Make less_all_components, smaller_constant and larger_constant static.
  (Suggested by Mat Turner)
- Change mixmax_range to call its limits "low" and "high" instead of
  "range[0]" and "range[1]". (Suggested by Connor Abbot).
- Use ir_builder swizzle helpers in swizzle_if_required(). (Suggested by
  Connor Abbot).
- Make the logic more clearer by rearrenging the code and commenting.
  (Suggested by Connor Abbot).
- Added comment to explain why we need to recurse twice. (Suggested by
  Connor Abbot).
- If we cannot prune an expression, do not return early. Instead, attempt
  to prune its children. (Suggested by Connor Abbot).

Other changes:
- Instead of having a global "valid" visitor member, let the various functions
  that can determine this status return a boolean and check for its value
  to decide what to do in each case. This is more flexible and allows to
  recurse into children of parents that could not be prunned due to invalid
  ranges (so related to the last bullet in the review feedback).
- Make sure we always check if a range is valid before working with it. Since
  any use of get_range, combine_range or range_intersection can invalidate
  a range we should check for this situation every time we use any of these
  functions.

Version 3 by Iago Toral Quiroga <[email protected]>:

Changes from review feedback:
- Now we can make get_range, combine_range and range_intersection static too
  (suggested by Connor Abbot).
- Do not return NULL when looking for the larger or greater constant into
  mixed vector constants. Instead, produce a new constant by doing a
  component-wise minmax. With this we can also remove of the validations when
  we call into these functions (suggested by Connor Abbot).
- Add a comment explaining the meaning of the baserange argument in
  prune_expression (suggested by Connor Abbot).

Other changes:
- Eliminate minmax expressions operating on constant vectors with mixed values
  by resolving them.

No piglit regressions observed with Version 3.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=76861

Reviewed-by: Connor Abbott <[email protected]>

4 files changed, 478 insertions, 0 deletions

diff --git a/src/glsl/Makefile.sources b/src/glsl/Makefile.sources
index bfb6993536e..0c55327d608 100644
--- a/src/glsl/Makefile.sources
+++ b/src/glsl/Makefile.sources
@@ -96,6 +96,7 @@ LIBGLSL_FILES = \
 	$(GLSL_SRCDIR)/opt_flip_matrices.cpp \
 	$(GLSL_SRCDIR)/opt_function_inlining.cpp \
 	$(GLSL_SRCDIR)/opt_if_simplification.cpp \
+	$(GLSL_SRCDIR)/opt_minmax.cpp \
 	$(GLSL_SRCDIR)/opt_noop_swizzle.cpp \
 	$(GLSL_SRCDIR)/opt_rebalance_tree.cpp \
 	$(GLSL_SRCDIR)/opt_redundant_jumps.cpp \
diff --git a/src/glsl/glsl_parser_extras.cpp b/src/glsl/glsl_parser_extras.cpp
index cc7d2d746ff..79f849465c3 100644
--- a/src/glsl/glsl_parser_extras.cpp
+++ b/src/glsl/glsl_parser_extras.cpp
@@ -1613,6 +1613,7 @@ do_common_optimization(exec_list *ir, bool linked,
    else
       progress = do_constant_variable_unlinked(ir) || progress;
    progress = do_constant_folding(ir) || progress;
+   progress = do_minmax_prune(ir) || progress;
    progress = do_cse(ir) || progress;
    progress = do_rebalance_tree(ir) || progress;
    progress = do_algebraic(ir, native_integers, options) || progress;
diff --git a/src/glsl/ir_optimization.h b/src/glsl/ir_optimization.h
index 0c3a6383190..e25857ac579 100644
--- a/src/glsl/ir_optimization.h
+++ b/src/glsl/ir_optimization.h
@@ -99,6 +99,7 @@ bool opt_flatten_nested_if_blocks(exec_list *instructions);
 bool do_discard_simplification(exec_list *instructions);
 bool lower_if_to_cond_assign(exec_list *instructions, unsigned max_depth = 0);
 bool do_mat_op_to_vec(exec_list *instructions);
+bool do_minmax_prune(exec_list *instructions);
 bool do_noop_swizzle(exec_list *instructions);
 bool do_structure_splitting(exec_list *instructions);
 bool do_swizzle_swizzle(exec_list *instructions);
diff --git a/src/glsl/opt_minmax.cpp b/src/glsl/opt_minmax.cpp
new file mode 100644
index 00000000000..32fb2d7ea2a
--- /dev/null
+++ b/src/glsl/opt_minmax.cpp
@@ -0,0 +1,475 @@
+/*
+ * Copyright © 2014 Intel Corporation
+ *
+ * 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.
+ */
+
+/**
+ * \file opt_minmax.cpp
+ *
+ * Drop operands from an expression tree of only min/max operations if they
+ * can be proven to not contribute to the final result.
+ *
+ * The algorithm is similar to alpha-beta pruning on a minmax search.
+ */
+
+#include "ir.h"
+#include "ir_visitor.h"
+#include "ir_rvalue_visitor.h"
+#include "ir_optimization.h"
+#include "ir_builder.h"
+#include "program/prog_instruction.h"
+#include "glsl_types.h"
+#include "main/macros.h"
+
+using namespace ir_builder;
+
+namespace {
+
+enum compare_components_result {
+   LESS,
+   LESS_OR_EQUAL,
+   EQUAL,
+   GREATER_OR_EQUAL,
+   GREATER,
+   MIXED
+};
+
+class minmax_range {
+public:
+   minmax_range(ir_constant *low = NULL, ir_constant *high = NULL)
+   {
+      this->low = low;
+      this->high = high;
+   }
+
+   /* low is the lower limit of the range, high is the higher limit. NULL on
+    * low means negative infinity (unlimited) and on high positive infinity
+    * (unlimited). Because of the two interpretations of the value NULL,
+    * arbitrary comparison between ir_constants is impossible.
+    */
+   ir_constant *low;
+   ir_constant *high;
+};
+
+class ir_minmax_visitor : public ir_rvalue_enter_visitor {
+public:
+   ir_minmax_visitor()
+      : progress(false)
+   {
+   }
+
+   ir_rvalue *prune_expression(ir_expression *expr, minmax_range baserange);
+
+   void handle_rvalue(ir_rvalue **rvalue);
+
+   bool progress;
+};
+
+/*
+ * Returns LESS if all vector components of `a' are strictly lower than of `b',
+ * GREATER if all vector components of `a' are strictly greater than of `b',
+ * MIXED if some vector components of `a' are strictly lower than of `b' while
+ * others are strictly greater, or EQUAL otherwise.
+ */
+static enum compare_components_result
+compare_components(ir_constant *a, ir_constant *b)
+{
+   assert(a != NULL);
+   assert(b != NULL);
+
+   assert(a->type->base_type == b->type->base_type);
+
+   unsigned a_inc = a->type->is_scalar() ? 0 : 1;
+   unsigned b_inc = b->type->is_scalar() ? 0 : 1;
+   unsigned components = MAX2(a->type->components(), b->type->components());
+
+   bool foundless = false;
+   bool foundgreater = false;
+   bool foundequal = false;
+
+   for (unsigned i = 0, c0 = 0, c1 = 0;
+        i < components;
+        c0 += a_inc, c1 += b_inc, ++i) {
+      switch (a->type->base_type) {
+      case GLSL_TYPE_UINT:
+         if (a->value.u[c0] < b->value.u[c1])
+            foundless = true;
+         else if (a->value.u[c0] > b->value.u[c1])
+            foundgreater = true;
+         else
+            foundequal = true;
+         break;
+      case GLSL_TYPE_INT:
+         if (a->value.i[c0] < b->value.i[c1])
+            foundless = true;
+         else if (a->value.i[c0] > b->value.i[c1])
+            foundgreater = true;
+         else
+            foundequal = true;
+         break;
+      case GLSL_TYPE_FLOAT:
+         if (a->value.f[c0] < b->value.f[c1])
+            foundless = true;
+         else if (a->value.f[c0] > b->value.f[c1])
+            foundgreater = true;
+         else
+            foundequal = true;
+         break;
+      default:
+         unreachable("not reached");
+      }
+   }
+
+   if (foundless && foundgreater) {
+      /* Some components are strictly lower, others are strictly greater */
+      return MIXED;
+   }
+
+   if (foundequal) {
+       /* It is not mixed, but it is not strictly lower or greater */
+      if (foundless)
+         return LESS_OR_EQUAL;
+      if (foundgreater)
+         return GREATER_OR_EQUAL;
+      return EQUAL;
+   }
+
+   /* All components are strictly lower or strictly greater */
+   return foundless ? LESS : GREATER;
+}
+
+static ir_constant *
+combine_constant(bool ismin, ir_constant *a, ir_constant *b)
+{
+   void *mem_ctx = ralloc_parent(a);
+   ir_constant *c = a->clone(mem_ctx, NULL);
+   for (unsigned i = 0; i < c->type->components(); i++) {
+      switch (c->type->base_type) {
+      case GLSL_TYPE_UINT:
+         if ((ismin && b->value.u[i] < c->value.u[i]) ||
+             (!ismin && b->value.u[i] > c->value.u[i]))
+            c->value.u[i] = b->value.u[i];
+         break;
+      case GLSL_TYPE_INT:
+         if ((ismin && b->value.i[i] < c->value.i[i]) ||
+             (!ismin && b->value.i[i] > c->value.i[i]))
+            c->value.i[i] = b->value.i[i];
+         break;
+      case GLSL_TYPE_FLOAT:
+         if ((ismin && b->value.f[i] < c->value.f[i]) ||
+             (!ismin && b->value.f[i] > c->value.f[i]))
+            c->value.f[i] = b->value.f[i];
+         break;
+      default:
+         assert(!"not reached");
+      }
+   }
+   return c;
+}
+
+static ir_constant *
+smaller_constant(ir_constant *a, ir_constant *b)
+{
+   assert(a != NULL);
+   assert(b != NULL);
+
+   enum compare_components_result ret = compare_components(a, b);
+   if (ret == MIXED)
+      return combine_constant(true, a, b);
+   else if (ret < EQUAL)
+      return a;
+   else
+      return b;
+}
+
+static ir_constant *
+larger_constant(ir_constant *a, ir_constant *b)
+{
+   assert(a != NULL);
+   assert(b != NULL);
+
+   enum compare_components_result ret = compare_components(a, b);
+   if (ret == MIXED)
+      return combine_constant(false, a, b);
+   else if (ret < EQUAL)
+      return b;
+   else
+      return a;
+}
+
+/* Combines two ranges by doing an element-wise min() / max() depending on the
+ * operation.
+ */
+static minmax_range
+combine_range(minmax_range r0, minmax_range r1, bool ismin)
+{
+   minmax_range ret;
+
+   if (!r0.low) {
+      ret.low = ismin ? r0.low : r1.low;
+   } else if (!r1.low) {
+      ret.low = ismin ? r1.low : r0.low;
+   } else {
+      ret.low = ismin ? smaller_constant(r0.low, r1.low) :
+         larger_constant(r0.low, r1.low);
+   }
+
+   if (!r0.high) {
+      ret.high = ismin ? r1.high : r0.high;
+   } else if (!r1.high) {
+      ret.high = ismin ? r0.high : r1.high;
+   } else {
+      ret.high = ismin ? smaller_constant(r0.high, r1.high) :
+         larger_constant(r0.high, r1.high);
+   }
+
+   return ret;
+}
+
+/* Returns a range so that lower limit is the larger of the two lower limits,
+ * and higher limit is the smaller of the two higher limits.
+ */
+static minmax_range
+range_intersection(minmax_range r0, minmax_range r1)
+{
+   minmax_range ret;
+
+   if (!r0.low)
+      ret.low = r1.low;
+   else if (!r1.low)
+      ret.low = r0.low;
+   else
+      ret.low = larger_constant(r0.low, r1.low);
+
+   if (!r0.high)
+      ret.high = r1.high;
+   else if (!r1.high)
+      ret.high = r0.high;
+   else
+      ret.high = smaller_constant(r0.high, r1.high);
+
+   return ret;
+}
+
+static minmax_range
+get_range(ir_rvalue *rval)
+{
+   ir_expression *expr = rval->as_expression();
+   if (expr && (expr->operation == ir_binop_min ||
+                expr->operation == ir_binop_max)) {
+      minmax_range r0 = get_range(expr->operands[0]);
+      minmax_range r1 = get_range(expr->operands[1]);
+      return combine_range(r0, r1, expr->operation == ir_binop_min);
+   }
+
+   ir_constant *c = rval->as_constant();
+   if (c) {
+      return minmax_range(c, c);
+   }
+
+   return minmax_range();
+}
+
+/**
+ * Prunes a min/max expression considering the base range of the parent
+ * min/max expression.
+ *
+ * @param baserange the range that the parents of this min/max expression
+ * in the min/max tree will clamp its value to.
+ */
+ir_rvalue *
+ir_minmax_visitor::prune_expression(ir_expression *expr, minmax_range baserange)
+{
+   assert(expr->operation == ir_binop_min ||
+          expr->operation == ir_binop_max);
+
+   bool ismin = expr->operation == ir_binop_min;
+   minmax_range limits[2];
+
+   /* Recurse to get the ranges for each of the subtrees of this
+    * expression. We need to do this as a separate step because we need to
+    * know the ranges of each of the subtrees before we prune either one.
+    * Consider something like this:
+    *
+    *        max
+    *     /       \
+    *    max     max
+    *   /   \   /   \
+    *  3    a   b    2
+    *
+    * We would like to prune away the max on the bottom-right, but to do so
+    * we need to know the range of the expression on the left beforehand,
+    * and there's no guarantee that we will visit either subtree in a
+    * particular order.
+    */
+   for (unsigned i = 0; i < 2; ++i)
+      limits[i] = get_range(expr->operands[i]);
+
+   for (unsigned i = 0; i < 2; ++i) {
+      bool is_redundant = false;
+
+      enum compare_components_result cr = LESS;
+      if (ismin) {
+         /* If this operand will always be greater than the other one, it's
+          * redundant.
+          */
+         if (limits[i].low && limits[1 - i].high) {
+               cr = compare_components(limits[i].low, limits[1 - i].high);
+            if (cr >= EQUAL && cr != MIXED)
+               is_redundant = true;
+         }
+         /* If this operand is always greater than baserange, then even if
+          * it's smaller than the other one it'll get clamped, so it's
+          * redundant.
+          */
+         if (!is_redundant && limits[i].low && baserange.high) {
+            cr = compare_components(limits[i].low, baserange.high);
+            if (cr >= EQUAL && cr != MIXED)
+               is_redundant = true;
+         }
+      } else {
+         /* If this operand will always be lower than the other one, it's
+          * redundant.
+          */
+         if (limits[i].high && limits[1 - i].low) {
+            cr = compare_components(limits[i].high, limits[1 - i].low);
+            if (cr <= EQUAL)
+               is_redundant = true;
+         }
+         /* If this operand is always lower than baserange, then even if
+          * it's greater than the other one it'll get clamped, so it's
+          * redundant.
+          */
+         if (!is_redundant && limits[i].high && baserange.low) {
+            cr = compare_components(limits[i].high, baserange.low);
+            if (cr <= EQUAL)
+               is_redundant = true;
+         }
+      }
+
+      if (is_redundant) {
+         progress = true;
+
+         /* Recurse if necessary. */
+         ir_expression *op_expr = expr->operands[1 - i]->as_expression();
+         if (op_expr && (op_expr->operation == ir_binop_min ||
+                         op_expr->operation == ir_binop_max)) {
+            return prune_expression(op_expr, baserange);
+         }
+
+         return expr->operands[1 - i];
+      } else if (cr == MIXED) {
+         /* If we have mixed vector operands, we can try to resolve the minmax
+          * expression by doing a component-wise minmax:
+          *
+          *             min                          min
+          *           /    \                       /    \
+          *         min     a       ===>        [1,1]    a
+          *       /    \
+          *    [1,3]   [3,1]
+          *
+          */
+         ir_constant *a = expr->operands[0]->as_constant();
+         ir_constant *b = expr->operands[1]->as_constant();
+         if (a && b)
+            return combine_constant(ismin, a, b);
+      }
+   }
+
+   /* Now recurse to operands giving them the proper baserange. The baserange
+    * to pass is the intersection of our baserange and the other operand's
+    * limit with one of the ranges unlimited. If we can't compute a valid
+    * intersection, we use the current baserange.
+    */
+   for (unsigned i = 0; i < 2; ++i) {
+      ir_expression *op_expr = expr->operands[i]->as_expression();
+      if (op_expr && (op_expr->operation == ir_binop_min ||
+                      op_expr->operation == ir_binop_max)) {
+         /* We can only compute a new baserange for this operand if we managed
+          * to compute a valid range for the other operand.
+          */
+         if (ismin)
+            limits[1 - i].low = NULL;
+         else
+            limits[1 - i].high = NULL;
+         minmax_range base = range_intersection(limits[1 - i], baserange);
+         expr->operands[i] = prune_expression(op_expr, base);
+      }
+   }
+
+   /* If we got here we could not discard any of the operands of the minmax
+    * expression, but we can still try to resolve the expression if both
+    * operands are constant. We do this after the loop above, to make sure
+    * that if our operands are minmax expressions we have tried to prune them
+    * first (hopefully reducing them to constants).
+    */
+   ir_constant *a = expr->operands[0]->as_constant();
+   ir_constant *b = expr->operands[1]->as_constant();
+   if (a && b)
+      return combine_constant(ismin, a, b);
+
+   return expr;
+}
+
+static ir_rvalue *
+swizzle_if_required(ir_expression *expr, ir_rvalue *rval)
+{
+   if (expr->type->is_vector() && rval->type->is_scalar()) {
+      return swizzle(rval, SWIZZLE_XXXX, expr->type->vector_elements);
+   } else {
+      return rval;
+   }
+}
+
+void
+ir_minmax_visitor::handle_rvalue(ir_rvalue **rvalue)
+{
+   if (!*rvalue)
+      return;
+
+   ir_expression *expr = (*rvalue)->as_expression();
+   if (!expr || (expr->operation != ir_binop_min &&
+                 expr->operation != ir_binop_max))
+      return;
+
+   ir_rvalue *new_rvalue = prune_expression(expr, minmax_range());
+   if (new_rvalue == *rvalue)
+      return;
+
+   /* If the expression type is a vector and the optimization leaves a scalar
+    * as the result, we need to turn it into a vector.
+    */
+   *rvalue = swizzle_if_required(expr, new_rvalue);
+
+   progress = true;
+}
+
+}
+
+bool
+do_minmax_prune(exec_list *instructions)
+{
+   ir_minmax_visitor v;
+
+   visit_list_elements(&v, instructions);
+
+   return v.progress;
+}