From d5c9028be4d5e85a424cc1e62509775045150a80 Mon Sep 17 00:00:00 2001
From: Rerumu <rerumu@protonmail.com>
Date: Thu, 19 May 2022 00:10:31 -0400
Subject: [PATCH] Start adding integer support for Luau

---
 codegen-luau/runtime/runtime.lua | 381 ++++++++++++++++++++-----------
 codegen-luau/src/lib.rs          |   7 +-
 2 files changed, 248 insertions(+), 140 deletions(-)

diff --git a/codegen-luau/runtime/runtime.lua b/codegen-luau/runtime/runtime.lua
index db4b9ee..e822957 100644
--- a/codegen-luau/runtime/runtime.lua
+++ b/codegen-luau/runtime/runtime.lua
@@ -1,37 +1,23 @@
 local module = {}
 
-local math_floor = math.floor
-local math_ceil = math.ceil
+local MAX_SIGNED = 0x7fffffff
+local BIT_SET_32 = 0x100000000
 
-local bit32 = bit32
-local bit_band = bit32.band
+local to_u32 = bit32.band
 
-local function no_op(x)
-	return x
-end
+local num_from_u32 = I64.from_u32
+local num_into_u32 = I64.into_u32
 
-local function to_u32(x)
-	return bit_band(x, 0xFFFFFFFF)
-end
-
-local function to_i32(x)
-	if x > 0x7FFFFFFF then
-		x = x - 0x100000000
+local function to_i32(num)
+	if num > MAX_SIGNED then
+		num = num - BIT_SET_32
 	end
 
-	return x
+	return num
 end
 
-local function wrap_i32(x)
-	return to_i32(to_u32(x))
-end
-
-local function truncate(num)
-	if num >= 0 then
-		return math_floor(num)
-	else
-		return math_ceil(num)
-	end
+local function no_op(num)
+	return num
 end
 
 do
@@ -40,33 +26,45 @@ do
 	local mul = {}
 	local div = {}
 
+	local assert = assert
+
 	function add.i32(a, b)
-		return wrap_i32(a + b)
+		return to_u32(a + b)
 	end
 
+	add.i64 = I64.add
+
 	function sub.i32(a, b)
-		return wrap_i32(a - b)
+		return to_u32(a - b)
 	end
 
+	sub.i64 = I64.subtract
+
 	function mul.i32(a, b)
-		return wrap_i32(a * b)
+		return to_u32(a * b)
 	end
 
+	mul.i64 = I64.multiply
+
 	function div.i32(lhs, rhs)
 		assert(rhs ~= 0, "division by zero")
 
-		return truncate(lhs / rhs)
+		lhs = to_i32(lhs)
+		rhs = to_i32(rhs)
+
+		return to_u32(lhs / rhs)
 	end
 
+	div.i64 = I64.divide_signed
+
 	function div.u32(lhs, rhs)
 		assert(rhs ~= 0, "division by zero")
 
-		lhs = to_u32(lhs)
-		rhs = to_u32(rhs)
-
-		return to_i32(math.floor(lhs / rhs))
+		return to_u32(lhs / rhs)
 	end
 
+	div.u64 = I64.divide_unsigned
+
 	module.add = add
 	module.sub = sub
 	module.mul = mul
@@ -78,6 +76,8 @@ do
 	local ctz = {}
 	local popcnt = {}
 
+	local bit_and = bit32.band
+
 	clz.i32 = bit32.countlz
 	ctz.i32 = bit32.countrz
 
@@ -85,7 +85,7 @@ do
 		local count = 0
 
 		while num ~= 0 do
-			num = bit32.band(num, num - 1)
+			num = bit_and(num, num - 1)
 			count = count + 1
 		end
 
@@ -98,71 +98,55 @@ do
 end
 
 do
-	local eqz = {}
-	local eq = {}
-	local ne = {}
 	local le = {}
 	local lt = {}
 	local ge = {}
 	local gt = {}
 
-	local function to_boolean(cond)
-		if cond then
-			return 1
-		else
-			return 0
-		end
-	end
-
-	function eq.i32(lhs, rhs)
-		return to_boolean(lhs == rhs)
-	end
-	function eq.num(lhs, rhs)
-		return to_boolean(lhs == rhs)
-	end
-
-	function eqz.i32(lhs)
-		return to_boolean(lhs == 0)
-	end
-
-	function ne.i32(lhs, rhs)
-		return to_boolean(lhs ~= rhs)
-	end
-	function ne.num(lhs, rhs)
-		return to_boolean(lhs ~= rhs)
-	end
+	local num_is_equal = I64.is_equal
+	local num_is_greater_signed = I64.is_greater_signed
+	local num_is_greater_unsigned = I64.is_greater_unsigned
+	local num_is_less_signed = I64.is_less_signed
+	local num_is_less_unsigned = I64.is_less_unsigned
 
 	function ge.i32(lhs, rhs)
-		return to_boolean(lhs >= rhs)
+		return to_i32(lhs) >= to_i32(rhs)
 	end
-	function ge.u32(lhs, rhs)
-		return to_boolean(to_u32(lhs) >= to_u32(rhs))
+
+	function ge.i64(lhs, rhs)
+		return num_is_greater_signed(lhs, rhs) or num_is_equal(lhs, rhs)
+	end
+
+	function ge.u64(lhs, rhs)
+		return num_is_greater_unsigned(lhs, rhs) or num_is_equal(lhs, rhs)
 	end
 
 	function gt.i32(lhs, rhs)
-		return to_boolean(lhs > rhs)
-	end
-	function gt.u32(lhs, rhs)
-		return to_boolean(to_u32(lhs) > to_u32(rhs))
+		return to_i32(lhs) > to_i32(rhs)
 	end
 
+	gt.i64 = num_is_greater_signed
+	gt.u64 = num_is_greater_unsigned
+
 	function le.i32(lhs, rhs)
-		return to_boolean(lhs <= rhs)
+		return to_i32(lhs) <= to_i32(rhs)
 	end
-	function le.u32(lhs, rhs)
-		return to_boolean(to_u32(lhs) <= to_u32(rhs))
+
+	function le.i64(lhs, rhs)
+		return num_is_less_signed(lhs, rhs) or num_is_equal(lhs, rhs)
+	end
+
+	function le.u64(lhs, rhs)
+		return num_is_less_unsigned(lhs, rhs) or num_is_equal(lhs, rhs)
 	end
 
 	function lt.i32(lhs, rhs)
-		return to_boolean(lhs < rhs)
-	end
-	function lt.u32(lhs, rhs)
-		return to_boolean(to_u32(lhs) < to_u32(rhs))
+		return to_i32(lhs) < to_i32(rhs)
 	end
 
-	module.eqz = eqz
-	module.eq = eq
-	module.ne = ne
+	lt.i64 = num_is_less_signed
+	lt.u64 = num_is_less_unsigned
+
 	module.le = le
 	module.lt = lt
 	module.ge = ge
@@ -176,12 +160,16 @@ do
 	local bnot = {}
 
 	band.i32 = bit32.band
+	band.i64 = I64.bit_and
 
 	bnot.i32 = bit32.bnot
+	bnot.i64 = I64.bit_not
 
 	bor.i32 = bit32.bor
+	bor.i64 = I64.bit_or
 
 	bxor.i32 = bit32.bxor
+	bxor.i64 = I64.bit_xor
 
 	module.band = band
 	module.bor = bor
@@ -196,14 +184,20 @@ do
 	local rotr = {}
 
 	rotl.i32 = bit32.lrotate
+	rotl.i64 = bit32.lrotate
 
 	rotr.i32 = bit32.rrotate
+	rotr.i64 = bit32.rrotate
 
 	shl.i32 = bit32.lshift
+	shl.i64 = bit32.lshift
 	shl.u32 = bit32.lshift
+	shl.u64 = bit32.lshift
 
 	shr.i32 = bit32.arshift
+	shr.i64 = bit32.arshift
 	shr.u32 = bit32.rshift
+	shr.u64 = bit32.rshift
 
 	module.shl = shl
 	module.shr = shr
@@ -218,19 +212,123 @@ do
 	local convert = {}
 	local reinterpret = {}
 
-	trunc.i32_f32 = truncate
-	trunc.i32_f64 = truncate
-	trunc.u32_f32 = truncate
-	trunc.u32_f64 = truncate
+	local math_ceil = math.ceil
+	local math_floor = math.floor
 
-	extend.i64_i32 = no_op
+	local string_pack = string.pack
+	local string_unpack = string.unpack
 
-	function convert.f32_i32(num)
-		return num
+	local num_from_u64 = I64.from_u64
+	local num_into_u64 = I64.into_u64
+
+	local num_negate = I64.negate
+	local num_is_negative = I64.is_negative
+
+	function wrap.i32_i64(num)
+		local data_1, _ = num_into_u32(num)
+
+		return data_1
 	end
 
-	function convert.f64_i32(num)
-		return num
+	trunc.i32_f32 = to_u32
+	trunc.i32_f64 = to_u32
+	trunc.u32_f32 = no_op
+	trunc.u32_f64 = no_op
+
+	function trunc.i64_f32(num)
+		if num < 0 then
+			local temp = num_from_u64(-math_ceil(num))
+
+			return num_negate(temp)
+		else
+			local temp = math_floor(num)
+
+			return num_from_u64(temp)
+		end
+	end
+
+	function trunc.i64_f64(num)
+		if num < 0 then
+			local temp = num_from_u64(-math_ceil(num))
+
+			return num_negate(temp)
+		else
+			local temp = math_floor(num)
+
+			return num_from_u64(temp)
+		end
+	end
+
+	trunc.u64_f32 = num_from_u64
+	trunc.u64_f64 = num_from_u64
+
+	function extend.i64_i32(num)
+		if num > MAX_SIGNED then
+			local temp = num_from_u32(-num + BIT_SET_32, 0)
+
+			return num_negate(temp)
+		else
+			return num_from_u32(num, 0)
+		end
+	end
+
+	function extend.u64_i32(num)
+		return num_from_u32(num, 0)
+	end
+
+	convert.f32_i32 = no_op
+	convert.f32_u32 = no_op
+
+	function convert.f32_i64(num)
+		if num_is_negative(num) then
+			local temp = num_negate(num)
+
+			return -num_into_u64(temp)
+		else
+			return num_into_u64(num)
+		end
+	end
+
+	convert.f32_u64 = num_into_u64
+	convert.f64_i32 = to_i32
+	convert.f64_u32 = no_op
+
+	function convert.f64_i64(num)
+		if num_is_negative(num) then
+			local temp = num_negate(num)
+
+			return -num_into_u64(temp)
+		else
+			return num_into_u64(num)
+		end
+	end
+
+	convert.f64_u64 = num_into_u64
+
+	function reinterpret.i32_f32(num)
+		local packed = string_pack("f", num)
+
+		return string_unpack("<I4", packed)
+	end
+
+	function reinterpret.i64_f64(num)
+		local packed = string_pack("d", num)
+		local data_1, data_2 = string_unpack("<I4I4", packed)
+
+		return num_from_u32(data_1, data_2)
+	end
+
+	function reinterpret.f32_i32(num)
+		local packed = string_pack("<I4", num)
+
+		return string_unpack("f", packed)
+	end
+
+	function reinterpret.f64_i64(num)
+		local data_1, data_2 = num_into_u32(num)
+		local packed = string_pack("<I4I4", data_1, data_2)
+
+		return string_unpack("d", packed)
 	end
 
 	module.wrap = wrap
@@ -245,38 +343,32 @@ do
 	local store = {}
 	local allocator = {}
 
-	local function rip_u64(x)
-		return math.floor(x / 0x100000000), x % 0x100000000
+	local bit_extract = bit32.extract
+	local bit_replace = bit32.replace
+
+	local bit_bor = bit32.bor
+	local bit_lshift = bit32.lshift
+	local bit_rshift = bit32.rshift
+
+	local math_floor = math.floor
+
+	local string_byte = string.byte
+	local string_unpack = string.unpack
+
+	local function load_byte(data, addr)
+		local value = data[math_floor(addr / 4)] or 0
+
+		return bit_extract(value, addr % 4 * 8, 8)
 	end
 
-	local function merge_u64(hi, lo)
-		return hi * 0x100000000 + lo
-	end
+	local function store_byte(data, addr, value)
+		local adjust = math_floor(addr / 4)
 
-	local function black_mask_byte(value, offset)
-		local mask = bit32.lshift(0xFF, offset * 8)
-
-		return bit32.band(value, bit32.bnot(mask))
-	end
-
-	local function load_byte(memory, addr)
-		local offset = addr % 4
-		local value = memory.data[(addr - offset) / 4] or 0
-
-		return bit32.band(bit32.rshift(value, offset * 8), 0xFF)
-	end
-
-	local function store_byte(memory, addr, value)
-		local offset = addr % 4
-		local adjust = (addr - offset) / 4
-		local lhs = bit32.lshift(bit32.band(value, 0xFF), offset * 8)
-		local rhs = black_mask_byte(memory.data[adjust] or 0, offset)
-
-		memory.data[adjust] = bit32.bor(lhs, rhs)
+		data[adjust] = bit_replace(data[adjust] or 0, value, addr % 4 * 8, 8)
 	end
 
 	function load.i32_i8(memory, addr)
-		local b = load_byte(memory, addr)
+		local b = load_byte(memory.data, addr)
 
 		if b > 0x7F then
 			b = b - 0x100
@@ -285,71 +377,82 @@ do
 		return b
 	end
 
-	load.i32_u8 = load_byte
+	function load.i32_u8(memory, addr)
+		return load_byte(memory.data, addr)
+	end
 
 	function load.i32(memory, addr)
+		local data = memory.data
+
 		if addr % 4 == 0 then
 			-- aligned read
-			return memory.data[addr / 4] or 0
+			return data[addr / 4] or 0
 		else
 			-- unaligned read
-			local b1 = load_byte(memory, addr)
-			local b2 = bit32.lshift(load_byte(memory, addr + 1), 8)
-			local b3 = bit32.lshift(load_byte(memory, addr + 2), 16)
-			local b4 = bit32.lshift(load_byte(memory, addr + 3), 24)
+			local b1 = load_byte(data, addr)
+			local b2 = bit_lshift(load_byte(data, addr + 1), 8)
+			local b3 = bit_lshift(load_byte(data, addr + 2), 16)
+			local b4 = bit_lshift(load_byte(data, addr + 3), 24)
 
-			return bit32.bor(b1, b2, b3, b4)
+			return bit_bor(b1, b2, b3, b4)
 		end
 	end
 
+	local load_i32 = load.i32
+
 	function load.i64(memory, addr)
-		local hi = load.i32(memory, addr + 4)
-		local lo = load.i32(memory, addr)
+		local data_1 = load_i32(memory, addr)
+		local data_2 = load_i32(memory, addr + 4)
 
-		return merge_u64(hi, lo)
+		return num_from_u32(data_1, data_2)
 	end
 
-	store.i32_n8 = store_byte
+	function store.i32_i8(memory, addr, value)
+		store_byte(memory.data, addr, value)
+	end
 
 	function store.i32(memory, addr, value)
+		local data = memory.data
+
 		if addr % 4 == 0 then
 			-- aligned write
-			memory.data[addr / 4] = value
+			data[addr / 4] = value
 		else
 			-- unaligned write
-			store_byte(memory, addr, value)
-			store_byte(memory, addr + 1, bit32.rshift(value, 8))
-			store_byte(memory, addr + 2, bit32.rshift(value, 16))
-			store_byte(memory, addr + 3, bit32.rshift(value, 24))
+			store_byte(data, addr, value)
+			store_byte(data, addr + 1, bit_rshift(value, 8))
+			store_byte(data, addr + 2, bit_rshift(value, 16))
+			store_byte(data, addr + 3, bit_rshift(value, 24))
 		end
 	end
 
-	function store.i64(memory, addr, value)
-		local hi, lo = rip_u64(value)
+	local store_i32 = store.i32
 
-		store.i32(memory, addr, lo)
-		store.i32(memory, addr + 4, hi)
+	function store.i64(memory, addr, value)
+		local data_1, data_2 = num_into_u32(value)
+
+		store_i32(memory, addr, data_1)
+		store_i32(memory, addr + 4, data_2)
 	end
 
 	function allocator.new(min, max)
 		return { min = min, max = max, data = {} }
 	end
 
-	function allocator.init(memory, offset, data)
-		local store_i8 = module.store.i32_n8
-		local store_i32 = module.store.i32
+	local store_i8 = store.i32_n8
 
+	function allocator.init(memory, offset, data)
 		local len = #data
 		local rem = len % 4
 
 		for i = 1, len - rem, 4 do
-			local v = string.unpack("<I4", data, i)
+			local v = string_unpack("<I4", data, i)
 
 			store_i32(memory, offset + i - 1, v)
 		end
 
 		for i = len - rem + 1, len do
-			local v = string.byte(data, i)
+			local v = string_byte(data, i)
 
 			store_i8(memory, offset + i - 1, v)
 		end
diff --git a/codegen-luau/src/lib.rs b/codegen-luau/src/lib.rs
index b28d6c0..ee3b5d9 100644
--- a/codegen-luau/src/lib.rs
+++ b/codegen-luau/src/lib.rs
@@ -1,4 +1,9 @@
-pub static RUNTIME: &str = include_str!("../runtime/runtime.lua");
+pub static RUNTIME: &str = concat!(
+	"local I64 = (function() ",
+	include_str!("../runtime/numeric.lua"),
+	" end)()\n",
+	include_str!("../runtime/runtime.lua")
+);
 
 pub use translator::translate;