mirror of
https://github.com/dart-lang/sdk
synced 2024-11-05 18:22:09 +00:00
f21b7cafbc
Before: `NoSuchMethodError: The method 'FfiTrampoline' was called on null.` After: `Invalid argument(s): argument value for ':ffi_param2' is null`. Makes the ArgumentNullError RTE lookup the name of the argument in the code source map when reporting a null argument. Makes the FFI call arguments and FFI callbacks use kArgumentError instead of the default kNoSuchMethod so that we target this RTE instead. This changes the Error type from `NoSuchMethodError` to `ArgumentError`. Because `Error`s should not be caught [1], this is fine. Since FFI trampolines are created from type arguments, the arguments do not have names. The arguments are assigned names programmatically. See the related bug. Also, this CL cleans up the SourcePosition of the `CheckNullOptimized`, it was never passed. [1] https://dart.dev/guides/language/effective-dart/usage#dont-explicitly-catch-error-or-types-that-implement-it TEST=tests/ffi/function_test.dart Closes: https://github.com/dart-lang/sdk/issues/47094 Bug: https://github.com/dart-lang/sdk/issues/36780 Change-Id: I15e7de4d026e034bde0eda3ba7fe3785f0da5057 Cq-Include-Trybots: luci.dart.try:vm-precomp-ffi-qemu-linux-release-arm-try,vm-ffi-android-debug-arm-try,vm-kernel-precomp-dwarf-linux-product-x64-try,vm-kernel-precomp-linux-debug-x64-try,app-kernel-linux-debug-x64-try,vm-kernel-reload-rollback-linux-debug-x64-try,vm-kernel-reload-linux-debug-x64-try,vm-ffi-android-debug-arm64-try,vm-kernel-nnbd-mac-debug-x64-try,vm-kernel-precomp-nnbd-linux-debug-x64-try Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/212462 Commit-Queue: Daco Harkes <dacoharkes@google.com> Reviewed-by: Clement Skau <cskau@google.com> Reviewed-by: Tess Strickland <sstrickl@google.com>
473 lines
15 KiB
Dart
473 lines
15 KiB
Dart
// Copyright (c) 2019, the Dart project authors. Please see the AUTHORS file
|
|
// for details. All rights reserved. Use of this source code is governed by a
|
|
// BSD-style license that can be found in the LICENSE file.
|
|
//
|
|
// Dart test program for testing dart:ffi function pointers.
|
|
//
|
|
// VMOptions=
|
|
// VMOptions=--deterministic --optimization-counter-threshold=10
|
|
// VMOptions=--use-slow-path
|
|
// VMOptions=--use-slow-path --stacktrace-every=100
|
|
// VMOptions=--write-protect-code --no-dual-map-code
|
|
// VMOptions=--write-protect-code --no-dual-map-code --use-slow-path
|
|
// VMOptions=--write-protect-code --no-dual-map-code --stacktrace-every=100
|
|
// SharedObjects=ffi_test_functions
|
|
|
|
import 'dart:ffi';
|
|
|
|
import "package:ffi/ffi.dart";
|
|
import "package:expect/expect.dart";
|
|
|
|
import 'dylib_utils.dart';
|
|
|
|
void main() {
|
|
for (int i = 0; i < 100; ++i) {
|
|
testNativeFunctionFromCast();
|
|
testNativeFunctionFromLookup();
|
|
test64bitInterpretations();
|
|
testExtension();
|
|
testTruncation();
|
|
testNativeFunctionDoubles();
|
|
testNativeFunctionFloats();
|
|
testNativeFunctionManyArguments1();
|
|
testNativeFunctionManyArguments2();
|
|
testNativeFunctionManyArguments3();
|
|
testNativeFunctionManyArguments4();
|
|
testNativeFunctionManyArguments5();
|
|
testNativeFunctionPointer();
|
|
testNullPointers();
|
|
testFloatRounding();
|
|
testVoidReturn();
|
|
testNoArgs();
|
|
testNativeFunctionNullableInt();
|
|
}
|
|
}
|
|
|
|
final ffiTestFunctions = dlopenPlatformSpecific("ffi_test_functions");
|
|
|
|
typedef NativeBinaryOp = Int32 Function(Int32, Int32);
|
|
typedef UnaryOp = int Function(int);
|
|
typedef BinaryOp = int Function(int, int);
|
|
typedef GenericBinaryOp<T> = int Function(int, T);
|
|
|
|
void testNativeFunctionFromCast() {
|
|
Pointer<IntPtr> p1 = calloc();
|
|
Pointer<NativeFunction<NativeBinaryOp>> p2 = p1.cast();
|
|
p2.asFunction<BinaryOp>();
|
|
p2.asFunction<GenericBinaryOp<int>>();
|
|
calloc.free(p1);
|
|
}
|
|
|
|
typedef NativeQuadOpSigned = Int64 Function(Int8, Int16, Int32, Int64);
|
|
typedef QuadOp = int Function(int, int, int, int);
|
|
typedef NativeQuadOpUnsigned = Uint64 Function(Uint8, Uint16, Uint32, Uint64);
|
|
|
|
BinaryOp sumPlus42 =
|
|
ffiTestFunctions.lookupFunction<NativeBinaryOp, BinaryOp>("SumPlus42");
|
|
|
|
QuadOp intComputation = ffiTestFunctions
|
|
.lookupFunction<NativeQuadOpSigned, QuadOp>("IntComputation");
|
|
|
|
void testNativeFunctionFromLookup() {
|
|
Expect.equals(49, sumPlus42(3, 4));
|
|
Expect.equals(49, (sumPlus42 as dynamic)(3, 4));
|
|
Expect.throwsNoSuchMethodError(() => (sumPlus42 as dynamic)());
|
|
Expect.throwsTypeError(() => (sumPlus42 as dynamic)(3, 4.0));
|
|
|
|
Expect.equals(625, intComputation(125, 250, 500, 1000));
|
|
Expect.equals(625, (intComputation as dynamic)(125, 250, 500, 1000));
|
|
|
|
Expect.equals(
|
|
0x7FFFFFFFFFFFFFFF, intComputation(0, 0, 0, 0x7FFFFFFFFFFFFFFF));
|
|
Expect.equals(
|
|
-0x8000000000000000, intComputation(0, 0, 0, -0x8000000000000000));
|
|
|
|
Expect.equals(0x7FFFFFFFFFFFFFFF,
|
|
(intComputation as dynamic)(0, 0, 0, 0x7FFFFFFFFFFFFFFF));
|
|
Expect.equals(-0x8000000000000000,
|
|
(intComputation as dynamic)(0, 0, 0, -0x8000000000000000));
|
|
}
|
|
|
|
typedef NativeReturnMaxUint8 = Uint8 Function();
|
|
int Function() returnMaxUint8 = ffiTestFunctions
|
|
.lookup("ReturnMaxUint8")
|
|
.cast<NativeFunction<NativeReturnMaxUint8>>()
|
|
.asFunction();
|
|
int Function() returnMaxUint8v2 = ffiTestFunctions
|
|
.lookup("ReturnMaxUint8v2")
|
|
.cast<NativeFunction<NativeReturnMaxUint8>>()
|
|
.asFunction();
|
|
|
|
typedef NativeReturnMaxUint16 = Uint16 Function();
|
|
int Function() returnMaxUint16 = ffiTestFunctions
|
|
.lookup("ReturnMaxUint16")
|
|
.cast<NativeFunction<NativeReturnMaxUint16>>()
|
|
.asFunction();
|
|
int Function() returnMaxUint16v2 = ffiTestFunctions
|
|
.lookup("ReturnMaxUint16v2")
|
|
.cast<NativeFunction<NativeReturnMaxUint16>>()
|
|
.asFunction();
|
|
|
|
typedef NativeReturnMaxUint32 = Uint32 Function();
|
|
int Function() returnMaxUint32 = ffiTestFunctions
|
|
.lookup("ReturnMaxUint32")
|
|
.cast<NativeFunction<NativeReturnMaxUint32>>()
|
|
.asFunction();
|
|
int Function() returnMaxUint32v2 = ffiTestFunctions
|
|
.lookup("ReturnMaxUint32v2")
|
|
.cast<NativeFunction<NativeReturnMaxUint32>>()
|
|
.asFunction();
|
|
|
|
typedef NativeReturnMinInt8 = Int8 Function();
|
|
int Function() returnMinInt8 = ffiTestFunctions
|
|
.lookup("ReturnMinInt8")
|
|
.cast<NativeFunction<NativeReturnMinInt8>>()
|
|
.asFunction();
|
|
int Function() returnMinInt8v2 = ffiTestFunctions
|
|
.lookup("ReturnMinInt8v2")
|
|
.cast<NativeFunction<NativeReturnMinInt8>>()
|
|
.asFunction();
|
|
|
|
typedef NativeReturnMinInt16 = Int16 Function();
|
|
int Function() returnMinInt16 = ffiTestFunctions
|
|
.lookup("ReturnMinInt16")
|
|
.cast<NativeFunction<NativeReturnMinInt16>>()
|
|
.asFunction();
|
|
int Function() returnMinInt16v2 = ffiTestFunctions
|
|
.lookup("ReturnMinInt16v2")
|
|
.cast<NativeFunction<NativeReturnMinInt16>>()
|
|
.asFunction();
|
|
|
|
typedef NativeReturnMinInt32 = Int32 Function();
|
|
int Function() returnMinInt32 = ffiTestFunctions
|
|
.lookup("ReturnMinInt32")
|
|
.cast<NativeFunction<NativeReturnMinInt32>>()
|
|
.asFunction();
|
|
int Function() returnMinInt32v2 = ffiTestFunctions
|
|
.lookup("ReturnMinInt32v2")
|
|
.cast<NativeFunction<NativeReturnMinInt32>>()
|
|
.asFunction();
|
|
|
|
typedef NativeTakeMaxUint8 = IntPtr Function(Uint8);
|
|
int Function(int) takeMaxUint8 = ffiTestFunctions
|
|
.lookup("TakeMaxUint8")
|
|
.cast<NativeFunction<NativeTakeMaxUint8>>()
|
|
.asFunction();
|
|
|
|
typedef NativeTakeMaxUint16 = IntPtr Function(Uint16);
|
|
int Function(int) takeMaxUint16 = ffiTestFunctions
|
|
.lookup("TakeMaxUint16")
|
|
.cast<NativeFunction<NativeTakeMaxUint16>>()
|
|
.asFunction();
|
|
|
|
typedef NativeTakeMaxUint32 = IntPtr Function(Uint32);
|
|
int Function(int) takeMaxUint32 = ffiTestFunctions
|
|
.lookup("TakeMaxUint32")
|
|
.cast<NativeFunction<NativeTakeMaxUint32>>()
|
|
.asFunction();
|
|
|
|
typedef NativeTakeMinInt8 = IntPtr Function(Int8);
|
|
int Function(int) takeMinInt8 = ffiTestFunctions
|
|
.lookup("TakeMinInt8")
|
|
.cast<NativeFunction<NativeTakeMinInt8>>()
|
|
.asFunction();
|
|
|
|
typedef NativeTakeMinInt16 = IntPtr Function(Int16);
|
|
int Function(int) takeMinInt16 = ffiTestFunctions
|
|
.lookup("TakeMinInt16")
|
|
.cast<NativeFunction<NativeTakeMinInt16>>()
|
|
.asFunction();
|
|
|
|
typedef NativeTakeMinInt32 = IntPtr Function(Int32);
|
|
int Function(int) takeMinInt32 = ffiTestFunctions
|
|
.lookup("TakeMinInt32")
|
|
.cast<NativeFunction<NativeTakeMinInt32>>()
|
|
.asFunction();
|
|
|
|
typedef NativeTakeMaxUint8x10 = IntPtr Function(
|
|
Uint8, Uint8, Uint8, Uint8, Uint8, Uint8, Uint8, Uint8, Uint8, Uint8);
|
|
int Function(int, int, int, int, int, int, int, int, int, int) takeMaxUint8x10 =
|
|
ffiTestFunctions
|
|
.lookup("TakeMaxUint8x10")
|
|
.cast<NativeFunction<NativeTakeMaxUint8x10>>()
|
|
.asFunction();
|
|
|
|
void testExtension() {
|
|
// Sign extension on the way back to Dart.
|
|
Expect.equals(0xff, returnMaxUint8());
|
|
Expect.equals(0xffff, returnMaxUint16());
|
|
Expect.equals(0xffffffff, returnMaxUint32());
|
|
Expect.equals(-0x80, returnMinInt8());
|
|
Expect.equals(-0x8000, returnMinInt16());
|
|
Expect.equals(-0x80000000, returnMinInt32());
|
|
// Truncation in C, and sign extension back to Dart.
|
|
Expect.equals(0xff, returnMaxUint8v2());
|
|
Expect.equals(0xffff, returnMaxUint16v2());
|
|
Expect.equals(0xffffffff, returnMaxUint32v2());
|
|
Expect.equals(-0x80, returnMinInt8v2());
|
|
Expect.equals(-0x8000, returnMinInt16v2());
|
|
Expect.equals(-0x80000000, returnMinInt32v2());
|
|
|
|
// Upper bits propper, should work without truncation.
|
|
Expect.equals(1, takeMaxUint8(0xff));
|
|
Expect.equals(1, takeMaxUint16(0xffff));
|
|
Expect.equals(1, takeMaxUint32(0xffffffff));
|
|
Expect.equals(1, takeMinInt8(-0x80));
|
|
Expect.equals(1, takeMinInt16(-0x8000));
|
|
Expect.equals(1, takeMinInt32(-0x80000000));
|
|
Expect.equals(
|
|
1,
|
|
takeMaxUint8x10(
|
|
0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff));
|
|
// Upper bits garbage, needs to truncate.
|
|
Expect.equals(1, takeMaxUint8(0xabcff));
|
|
Expect.equals(1, takeMaxUint16(0xabcffff));
|
|
Expect.equals(1, takeMaxUint32(0xabcffffffff));
|
|
Expect.equals(1, takeMinInt8(0x8abc80));
|
|
Expect.equals(1, takeMinInt16(0x8abc8000));
|
|
Expect.equals(1, takeMinInt32(0x8abc80000000));
|
|
Expect.equals(
|
|
1,
|
|
takeMaxUint8x10(0xabcff, 0xabcff, 0xabcff, 0xabcff, 0xabcff, 0xabcff,
|
|
0xabcff, 0xabcff, 0xabcff, 0xabcff));
|
|
}
|
|
|
|
QuadOp uintComputation = ffiTestFunctions
|
|
.lookupFunction<NativeQuadOpUnsigned, QuadOp>("UintComputation");
|
|
|
|
void test64bitInterpretations() {
|
|
// 2 ^ 63 - 1
|
|
Expect.equals(
|
|
0x7FFFFFFFFFFFFFFF, uintComputation(0, 0, 0, 0x7FFFFFFFFFFFFFFF));
|
|
// -2 ^ 63 interpreted as 2 ^ 63
|
|
Expect.equals(
|
|
-0x8000000000000000, uintComputation(0, 0, 0, -0x8000000000000000));
|
|
// -1 interpreted as 2 ^ 64 - 1
|
|
Expect.equals(-1, uintComputation(0, 0, 0, -1));
|
|
}
|
|
|
|
typedef NativeSenaryOp = Int64 Function(
|
|
Int8, Int16, Int32, Uint8, Uint16, Uint32);
|
|
typedef SenaryOp = int Function(int, int, int, int, int, int);
|
|
|
|
SenaryOp sumSmallNumbers = ffiTestFunctions
|
|
.lookupFunction<NativeSenaryOp, SenaryOp>("SumSmallNumbers");
|
|
|
|
void testTruncation() {
|
|
sumSmallNumbers(128, 0, 0, 0, 0, 0);
|
|
sumSmallNumbers(-129, 0, 0, 0, 0, 0);
|
|
sumSmallNumbers(0, 0, 0, 256, 0, 0);
|
|
sumSmallNumbers(0, 0, 0, -1, 0, 0);
|
|
|
|
sumSmallNumbers(0, 0x8000, 0, 0, 0, 0);
|
|
sumSmallNumbers(0, 0xFFFFFFFFFFFF7FFF, 0, 0, 0, 0);
|
|
sumSmallNumbers(0, 0, 0, 0, 0x10000, 0);
|
|
sumSmallNumbers(0, 0, 0, 0, -1, 0);
|
|
|
|
Expect.equals(0xFFFFFFFF80000000, sumSmallNumbers(0, 0, 0x80000000, 0, 0, 0));
|
|
Expect.equals(
|
|
0x000000007FFFFFFF, sumSmallNumbers(0, 0, 0xFFFFFFFF7FFFFFFF, 0, 0, 0));
|
|
Expect.equals(0, sumSmallNumbers(0, 0, 0, 0, 0, 0x100000000));
|
|
Expect.equals(0xFFFFFFFF, sumSmallNumbers(0, 0, 0, 0, 0, -1));
|
|
}
|
|
|
|
typedef NativeDoubleUnaryOp = Double Function(Double);
|
|
typedef DoubleUnaryOp = double Function(double);
|
|
|
|
DoubleUnaryOp times1_337Double = ffiTestFunctions
|
|
.lookupFunction<NativeDoubleUnaryOp, DoubleUnaryOp>("Times1_337Double");
|
|
|
|
void testNativeFunctionDoubles() {
|
|
Expect.approxEquals(2.0 * 1.337, times1_337Double(2.0));
|
|
}
|
|
|
|
typedef NativeFloatUnaryOp = Float Function(Float);
|
|
|
|
DoubleUnaryOp times1_337Float = ffiTestFunctions
|
|
.lookupFunction<NativeFloatUnaryOp, DoubleUnaryOp>("Times1_337Float");
|
|
|
|
void testNativeFunctionFloats() {
|
|
Expect.approxEquals(1337.0, times1_337Float(1000.0));
|
|
}
|
|
|
|
typedef NativeDecenaryOp = IntPtr Function(IntPtr, IntPtr, IntPtr, IntPtr,
|
|
IntPtr, IntPtr, IntPtr, IntPtr, IntPtr, IntPtr);
|
|
typedef NativeDecenaryOp2 = Int16 Function(
|
|
Int8, Int16, Int8, Int16, Int8, Int16, Int8, Int16, Int8, Int16);
|
|
typedef DecenaryOp = int Function(
|
|
int, int, int, int, int, int, int, int, int, int);
|
|
|
|
DecenaryOp sumManyInts = ffiTestFunctions
|
|
.lookupFunction<NativeDecenaryOp, DecenaryOp>("SumManyInts");
|
|
|
|
void testNativeFunctionManyArguments1() {
|
|
Expect.equals(55, sumManyInts(1, 2, 3, 4, 5, 6, 7, 8, 9, 10));
|
|
}
|
|
|
|
DecenaryOp sumManySmallInts = ffiTestFunctions
|
|
.lookupFunction<NativeDecenaryOp2, DecenaryOp>("SumManySmallInts");
|
|
|
|
void testNativeFunctionManyArguments5() {
|
|
Expect.equals(55, sumManySmallInts(1, 2, 3, 4, 5, 6, 7, 8, 9, 10));
|
|
}
|
|
|
|
typedef NativeUndenaryOp = IntPtr Function(IntPtr, IntPtr, IntPtr, IntPtr,
|
|
IntPtr, IntPtr, IntPtr, IntPtr, IntPtr, IntPtr, IntPtr);
|
|
typedef UndenaryOp = int Function(
|
|
int, int, int, int, int, int, int, int, int, int, int);
|
|
|
|
UndenaryOp sumManyIntsOdd = ffiTestFunctions
|
|
.lookupFunction<NativeUndenaryOp, UndenaryOp>("SumManyIntsOdd");
|
|
|
|
void testNativeFunctionManyArguments4() {
|
|
Expect.equals(66, sumManyIntsOdd(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11));
|
|
}
|
|
|
|
typedef NativeDoubleDecenaryOp = Double Function(Double, Double, Double, Double,
|
|
Double, Double, Double, Double, Double, Double);
|
|
typedef DoubleDecenaryOp = double Function(double, double, double, double,
|
|
double, double, double, double, double, double);
|
|
|
|
DoubleDecenaryOp sumManyDoubles = ffiTestFunctions
|
|
.lookupFunction<NativeDoubleDecenaryOp, DoubleDecenaryOp>("SumManyDoubles");
|
|
|
|
void testNativeFunctionManyArguments2() {
|
|
Expect.approxEquals(
|
|
55.0, sumManyDoubles(1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0));
|
|
}
|
|
|
|
typedef NativeVigesimalOp = Double Function(
|
|
IntPtr,
|
|
Float,
|
|
IntPtr,
|
|
Double,
|
|
IntPtr,
|
|
Float,
|
|
IntPtr,
|
|
Double,
|
|
IntPtr,
|
|
Float,
|
|
IntPtr,
|
|
Double,
|
|
IntPtr,
|
|
Float,
|
|
IntPtr,
|
|
Double,
|
|
IntPtr,
|
|
Float,
|
|
IntPtr,
|
|
Double);
|
|
typedef VigesimalOp = double Function(
|
|
int,
|
|
double,
|
|
int,
|
|
double,
|
|
int,
|
|
double,
|
|
int,
|
|
double,
|
|
int,
|
|
double,
|
|
int,
|
|
double,
|
|
int,
|
|
double,
|
|
int,
|
|
double,
|
|
int,
|
|
double,
|
|
int,
|
|
double);
|
|
|
|
VigesimalOp sumManyNumbers = ffiTestFunctions
|
|
.lookupFunction<NativeVigesimalOp, VigesimalOp>("SumManyNumbers");
|
|
|
|
void testNativeFunctionManyArguments3() {
|
|
Expect.approxEquals(
|
|
210.0,
|
|
sumManyNumbers(1, 2.0, 3, 4.0, 5, 6.0, 7, 8.0, 9, 10.0, 11, 12.0, 13,
|
|
14.0, 15, 16.0, 17, 18.0, 19, 20.0));
|
|
}
|
|
|
|
typedef Int64PointerUnOp = Pointer<Int64> Function(Pointer<Int64>);
|
|
|
|
Int64PointerUnOp assign1337Index1 = ffiTestFunctions
|
|
.lookupFunction<Int64PointerUnOp, Int64PointerUnOp>("Assign1337Index1");
|
|
|
|
void testNativeFunctionPointer() {
|
|
Pointer<Int64> p2 = calloc(2);
|
|
p2.value = 42;
|
|
p2[1] = 1000;
|
|
Pointer<Int64> result = assign1337Index1(p2);
|
|
Expect.equals(1337, result.value);
|
|
Expect.equals(1337, p2[1]);
|
|
Expect.equals(p2.elementAt(1).address, result.address);
|
|
calloc.free(p2);
|
|
}
|
|
|
|
Int64PointerUnOp nullableInt64ElemAt1 = ffiTestFunctions
|
|
.lookupFunction<Int64PointerUnOp, Int64PointerUnOp>("NullableInt64ElemAt1");
|
|
|
|
void testNullPointers() {
|
|
Pointer<Int64> result = nullableInt64ElemAt1(nullptr);
|
|
Expect.equals(result, nullptr);
|
|
|
|
Pointer<Int64> p2 = calloc(2);
|
|
result = nullableInt64ElemAt1(p2);
|
|
Expect.notEquals(result, nullptr);
|
|
calloc.free(p2);
|
|
}
|
|
|
|
typedef NativeFloatPointerToBool = Uint8 Function(Pointer<Float>);
|
|
typedef FloatPointerToBool = int Function(Pointer<Float>);
|
|
|
|
FloatPointerToBool isRoughly1337 = ffiTestFunctions.lookupFunction<
|
|
NativeFloatPointerToBool, FloatPointerToBool>("IsRoughly1337");
|
|
|
|
void testFloatRounding() {
|
|
Pointer<Float> p2 = calloc();
|
|
p2.value = 1337.0;
|
|
|
|
int result = isRoughly1337(p2);
|
|
Expect.equals(1, result);
|
|
|
|
calloc.free(p2);
|
|
}
|
|
|
|
typedef NativeFloatToVoid = Void Function(Float);
|
|
typedef DoubleToVoid = void Function(double);
|
|
|
|
DoubleToVoid devNullFloat = ffiTestFunctions
|
|
.lookupFunction<NativeFloatToVoid, DoubleToVoid>("DevNullFloat");
|
|
|
|
void testVoidReturn() {
|
|
devNullFloat(1337.0);
|
|
|
|
dynamic loseSignature = devNullFloat;
|
|
dynamic result = loseSignature(1337.0);
|
|
Expect.isNull(result);
|
|
}
|
|
|
|
typedef NativeVoidToFloat = Float Function();
|
|
typedef VoidToDouble = double Function();
|
|
|
|
VoidToDouble inventFloatValue = ffiTestFunctions
|
|
.lookupFunction<NativeVoidToFloat, VoidToDouble>("InventFloatValue");
|
|
|
|
void testNoArgs() {
|
|
double result = inventFloatValue();
|
|
Expect.approxEquals(1337.0, result);
|
|
}
|
|
|
|
void testNativeFunctionNullableInt() {
|
|
final sumPlus42 = ffiTestFunctions.lookupFunction<
|
|
Int32 Function(Int32, Int32), int Function(int, int?)>("SumPlus42");
|
|
|
|
try {
|
|
sumPlus42(3, null);
|
|
} catch (e) {
|
|
// TODO(http://dartbug.com/47098): Save param names to dwarf.
|
|
Expect.isTrue(e.toString().contains('ffi_param2') ||
|
|
e.toString().contains('<optimized out>'));
|
|
}
|
|
}
|