Mask generator optimizations

Author: Mohammad Rezaei

Changelog.md

@@ -6,6 +6,7 @@ Language Features:
 Compiler Features:
 * NatSpec: Capture Natspec documentation of `enum` values in the AST.
 
+* Constant Optimizer: Compute masks using shifts when optimizing for size; use an ``--optimizer-runs`` value less than 200 for maximum size reduction.
 
 Bugfixes:
 * SMTChecker: Do not consider loop conditions as constant-condition verification target as this could cause incorrect reports and internal compiler errors.

libevmasm/ConstantOptimiser.cpp

@@ -253,7 +253,47 @@ AssemblyItems ComputeMethod::findRepresentation(u256 const& _value)
 	if (_value < 0x10000)
 		// Very small value, not worth computing
 		return AssemblyItems{_value};
-	else if (numberEncodingSize(~_value) < numberEncodingSize(_value))
+
+	// check for masks first
+	unsigned lowZeros = 0;
+	unsigned highOnes = 0;
+	for (; ((_value >> lowZeros) & 1) == 0 && lowZeros < 256; lowZeros++) {}
+	for (; ((_value >> (lowZeros + highOnes)) & 1) == 1 && highOnes < 256; highOnes++) {}
+	if (m_params.evmVersion.hasBitwiseShifting() && highOnes > 32 &&
+		((_value >> (lowZeros + highOnes)) == 0) &&
+		((lowZeros + highOnes < 256) || lowZeros > 16))
+	{
+		// this is a big enough mask to use zero negation
+		AssemblyItems newRoutine = AssemblyItems{u256(0), Instruction::NOT};
+		if ((highOnes + lowZeros) != 256)
+			newRoutine += AssemblyItems{u256(256 - highOnes), Instruction::SHR};
+		if (lowZeros > 0)
+			newRoutine += AssemblyItems{u256(lowZeros), Instruction::SHL};
+		return newRoutine;
+	}
+	// check powers of 10
+	u256 divBy10 = _value;
+	unsigned pow10 = 0;
+	while (divBy10 > 0 && divBy10 % 10 == 0)
+	{
+		divBy10 = divBy10 / 10;
+		pow10++;
+	}
+	// 10^9 = 0x3b9aca00; requires 5 bytes with PUSH4; also 5 bytes with PUSH 5 PUSH 10 EXP
+	if (pow10 > 9 && divBy10 == 1)
+	{
+		// pure power of 10
+		return AssemblyItems{u256(pow10), u256(10), Instruction::EXP};
+	}
+	// 10^x * y can be encoded as: push x push 10 exp push y mul. That's about 7 bytes more than plain push
+	if (pow10 > 12)
+	{
+		AssemblyItems newRoutine = findRepresentation(divBy10);
+		newRoutine += AssemblyItems{u256(pow10), u256(10), Instruction::EXP, Instruction::MUL};
+		return newRoutine;
+	}
+	if (numberEncodingSize(~_value) < numberEncodingSize(_value) &&
+		(lowZeros+highOnes < 256 || highOnes > 16))
 		// Negated is shorter to represent
 		return findRepresentation(~_value) + AssemblyItems{Instruction::NOT};
 	else

test/cmdlineTests/optimizer_BlockDeDuplicator/output

@@ -7,11 +7,11 @@ EVM assembly:
   0x00
   dup1
   sload
-  not(sub(shl(0x40, 0x01), 0x01))
+  shl(0x40, not(0x00))
   and
     /* "input.sol":201:206  fun_x */
   or(tag_0_7, shl(0x20, tag_2))
-  sub(shl(0x40, 0x01), 0x01)
+  shr(0xc0, not(0x00))
     /* "input.sol":179:210  function() r = true ? fun_x : f */
   and
   or

test/cmdlineTests/optimizer_inliner_dynamic_reference/output

@@ -84,7 +84,7 @@ sub_0: assembly {
         /* "input.sol":147:152  x = f */
       dup1
       sload
-      not(0xffffffffffffffff)
+      shl(0x40, not(0x00))
       and
         /* "input.sol":151:152  f */
       tag_17

test/cmdlineTests/optimizer_inliner_dynamic_reference_constructor/output

@@ -17,11 +17,11 @@ tag_1:
     /* "input.sol":93:98  x = f */
   dup1
   sload
-  not(sub(shl(0x40, 0x01), 0x01))
+  shl(0x40, not(0x00))
   and
     /* "input.sol":97:98  f */
   or(tag_0_12, shl(0x20, tag_4))
-  sub(shl(0x40, 0x01), 0x01)
+  shr(0xc0, not(0x00))
     /* "input.sol":93:98  x = f */
   and
   or

test/cmdlineTests/standard_debug_info_in_evm_asm_via_ir_location/output.json

@@ -20,7 +20,7 @@
   dup4
   add
   swap2
-  sub(shl(0x40, 0x01), 0x01)
+  shr(0xc0, not(0x00))
   dup4
   gt
   dup5
@@ -104,7 +104,7 @@ sub_0: assembly {
       jumpi(tag_26, callvalue)
       jumpi(tag_26, slt(add(not(0x03), calldatasize), 0x00))
       sload(0x00)
-      sub(shl(0xff, 0x01), 0x01)
+      shr(0x01, not(0x00))
       dup2
       eq
       tag_14
@@ -350,7 +350,7 @@ sub_0: assembly {
   dup4
   add
   swap2
-  sub(shl(0x40, 0x01), 0x01)
+  shr(0xc0, not(0x00))
   dup4
   gt
   dup5
@@ -450,7 +450,7 @@ sub_0: assembly {
       jumpi(tag_26, callvalue)
       jumpi(tag_26, slt(add(not(0x03), calldatasize), 0x00))
       sload(0x00)
-      sub(shl(0xff, 0x01), 0x01)
+      shr(0x01, not(0x00))
       dup2
       eq
       tag_14

test/cmdlineTests/viair_subobject_optimization/output

@@ -18,7 +18,7 @@ EVM assembly:
   dup4
   add
   swap2
-  sub(shl(0x40, 0x01), 0x01)
+  shr(0xc0, not(0x00))
   dup4
   gt
   dup5
@@ -223,7 +223,7 @@ sub_0: assembly {
           dup4
           add
           swap2
-          sub(shl(0x40, 0x01), 0x01)
+          shr(0xc0, not(0x00))
           dup4
           gt
           dup5