Implement remainder operator in aarch64 compiler

apps/interpreter/source/main.cpp

@@ -182,7 +182,7 @@ int main(int argc, char ** argv)
 		{
 			// TODO: remove, testing-only code; should execute entry point instead
 			auto offset = pcontext.symbols.at("test");
-			auto fptr = (std::uint32_t(*)(std::uint32_t))(executable.data.get() + offset);
+			auto fptr = (int(*)(int))(executable.data.get() + offset);
 			auto x = fptr(0);
 			std::cout << "Result: " << std::boolalpha << x << std::endl;
 		}

examples/jit_test.psl

@@ -10,5 +10,5 @@ func add_or_sub(add : bool) -> (u32 -> u32):
 	else:
 		return sub1
 
-func test(x: u32) -> u32:
+func test(x: i32) -> i32:
-	return (!0us) as u32
+	return 10 / 0

libs/jit/source/arch/aarch64/compiler.cpp

@@ -477,8 +477,19 @@ namespace pslang::jit::aarch64
 						}
 						break;
 					case ast::binary_operation_type::remainder:
-						// TODO: implement via div & mul & sub
+						if (types::is_signed_integer_type(*node.inferred_type))
-						throw std::runtime_error("remainder operator is not implemented");
+						{
+							builder.sdiv_reg(1, 0, 2);
+							builder.mul_reg(0, 2, 0);
+							builder.sub_reg(1, 0, 0);
+						}
+						else if (types::is_unsigned_integer_type(*node.inferred_type))
+						{
+							builder.udiv_reg(1, 0, 2);
+							builder.mul_reg(0, 2, 0);
+							builder.sub_reg(1, 0, 0);
+						}
+						break;
 					case ast::binary_operation_type::binary_and:
 						builder.and_reg(1, 0, 0);
 						break;

spec.txt

@@ -111,8 +111,8 @@ Arithmetic (only same integer/floating-point type):
     -x
     x + y
     x * y
-    x / y // in case of integers, rounds down when y>0 when x<0, consistently with %
+    x / y
-    x % y // integer only; mathematical, i.e. always in [0, y-1] when y>0 even when x<0; TODO: what if y<0?
+    x % y
 
 Pointer arithmetic (any pointer type + any integer type):
     p + x