mirror of
https://github.com/RGBCube/serenity
synced 2025-10-31 17:12:43 +00:00
LibWasm: Implement 15 more SIMD instructions
This commit is contained in:
Ali Mohammad Pur
Jelle Raaijmakers
parent
e8bdb7e5f8
commit
82b7368de1
4 changed files with 327 additions and 10 deletions
|
|
@ -296,8 +296,8 @@ void BytecodeInterpreter::call_address(Configuration& configuration, FunctionAdd
|
|||
configuration.stack().entries().unchecked_append(move(entry));
|
||||
}
|
||||
|
||||
template<typename PopTypeLHS, typename PushType, typename Operator, typename PopTypeRHS>
|
||||
void BytecodeInterpreter::binary_numeric_operation(Configuration& configuration)
|
||||
template<typename PopTypeLHS, typename PushType, typename Operator, typename PopTypeRHS, typename... Args>
|
||||
void BytecodeInterpreter::binary_numeric_operation(Configuration& configuration, Args&&... args)
|
||||
{
|
||||
auto rhs_entry = configuration.stack().pop();
|
||||
auto& lhs_entry = configuration.stack().peek();
|
||||
|
|
@ -306,7 +306,7 @@ void BytecodeInterpreter::binary_numeric_operation(Configuration& configuration)
|
|||
auto rhs = rhs_ptr->to<PopTypeRHS>();
|
||||
auto lhs = lhs_ptr->to<PopTypeLHS>();
|
||||
PushType result;
|
||||
auto call_result = Operator {}(lhs.value(), rhs.value());
|
||||
auto call_result = Operator { forward<Args>(args)... }(lhs.value(), rhs.value());
|
||||
if constexpr (IsSpecializationOf<decltype(call_result), AK::Result>) {
|
||||
if (call_result.is_error()) {
|
||||
trap_if_not(false, call_result.error());
|
||||
|
|
@ -320,13 +320,13 @@ void BytecodeInterpreter::binary_numeric_operation(Configuration& configuration)
|
|||
lhs_entry = Value(result);
|
||||
}
|
||||
|
||||
template<typename PopType, typename PushType, typename Operator>
|
||||
void BytecodeInterpreter::unary_operation(Configuration& configuration)
|
||||
template<typename PopType, typename PushType, typename Operator, typename... Args>
|
||||
void BytecodeInterpreter::unary_operation(Configuration& configuration, Args&&... args)
|
||||
{
|
||||
auto& entry = configuration.stack().peek();
|
||||
auto entry_ptr = entry.get_pointer<Value>();
|
||||
auto value = entry_ptr->to<PopType>();
|
||||
auto call_result = Operator {}(*value);
|
||||
auto call_result = Operator { forward<Args>(args)... }(*value);
|
||||
PushType result;
|
||||
if constexpr (IsSpecializationOf<decltype(call_result), AK::Result>) {
|
||||
if (call_result.is_error()) {
|
||||
|
|
@ -1241,6 +1241,225 @@ void BytecodeInterpreter::interpret(Configuration& configuration, InstructionPoi
|
|||
return binary_numeric_operation<u128, u128, Operators::VectorShiftRight<2, MakeUnsigned>, i32>(configuration);
|
||||
case Instructions::i64x2_shr_s.value():
|
||||
return binary_numeric_operation<u128, u128, Operators::VectorShiftRight<2, MakeSigned>, i32>(configuration);
|
||||
case Instructions::i8x16_swizzle.value():
|
||||
return binary_numeric_operation<u128, u128, Operators::VectorSwizzle>(configuration);
|
||||
case Instructions::i8x16_extract_lane_s.value():
|
||||
return unary_operation<u128, i8, Operators::VectorExtractLane<16, MakeSigned>>(configuration, instruction.arguments().get<Instruction::LaneIndex>().lane);
|
||||
case Instructions::i8x16_extract_lane_u.value():
|
||||
return unary_operation<u128, u8, Operators::VectorExtractLane<16, MakeUnsigned>>(configuration, instruction.arguments().get<Instruction::LaneIndex>().lane);
|
||||
case Instructions::i16x8_extract_lane_s.value():
|
||||
return unary_operation<u128, i16, Operators::VectorExtractLane<8, MakeSigned>>(configuration, instruction.arguments().get<Instruction::LaneIndex>().lane);
|
||||
case Instructions::i16x8_extract_lane_u.value():
|
||||
return unary_operation<u128, u16, Operators::VectorExtractLane<8, MakeUnsigned>>(configuration, instruction.arguments().get<Instruction::LaneIndex>().lane);
|
||||
case Instructions::i32x4_extract_lane.value():
|
||||
return unary_operation<u128, i32, Operators::VectorExtractLane<4, MakeSigned>>(configuration, instruction.arguments().get<Instruction::LaneIndex>().lane);
|
||||
case Instructions::i64x2_extract_lane.value():
|
||||
return unary_operation<u128, i64, Operators::VectorExtractLane<2, MakeSigned>>(configuration, instruction.arguments().get<Instruction::LaneIndex>().lane);
|
||||
case Instructions::f32x4_extract_lane.value():
|
||||
return unary_operation<u128, float, Operators::VectorExtractLaneFloat<4>>(configuration, instruction.arguments().get<Instruction::LaneIndex>().lane);
|
||||
case Instructions::f64x2_extract_lane.value():
|
||||
return unary_operation<u128, double, Operators::VectorExtractLaneFloat<2>>(configuration, instruction.arguments().get<Instruction::LaneIndex>().lane);
|
||||
case Instructions::i8x16_replace_lane.value():
|
||||
return binary_numeric_operation<u128, u128, Operators::VectorReplaceLane<16, i32>, i32>(configuration, instruction.arguments().get<Instruction::LaneIndex>().lane);
|
||||
case Instructions::i16x8_replace_lane.value():
|
||||
return binary_numeric_operation<u128, u128, Operators::VectorReplaceLane<8, i32>, i32>(configuration, instruction.arguments().get<Instruction::LaneIndex>().lane);
|
||||
case Instructions::i32x4_replace_lane.value():
|
||||
return binary_numeric_operation<u128, u128, Operators::VectorReplaceLane<4>, i32>(configuration, instruction.arguments().get<Instruction::LaneIndex>().lane);
|
||||
case Instructions::i64x2_replace_lane.value():
|
||||
return binary_numeric_operation<u128, u128, Operators::VectorReplaceLane<2>, i64>(configuration, instruction.arguments().get<Instruction::LaneIndex>().lane);
|
||||
case Instructions::f32x4_replace_lane.value():
|
||||
return binary_numeric_operation<u128, u128, Operators::VectorReplaceLane<4, float>, float>(configuration, instruction.arguments().get<Instruction::LaneIndex>().lane);
|
||||
case Instructions::f64x2_replace_lane.value():
|
||||
return binary_numeric_operation<u128, u128, Operators::VectorReplaceLane<2, double>, double>(configuration, instruction.arguments().get<Instruction::LaneIndex>().lane);
|
||||
case Instructions::i8x16_eq.value():
|
||||
case Instructions::i8x16_ne.value():
|
||||
case Instructions::i8x16_lt_s.value():
|
||||
case Instructions::i8x16_lt_u.value():
|
||||
case Instructions::i8x16_gt_s.value():
|
||||
case Instructions::i8x16_gt_u.value():
|
||||
case Instructions::i8x16_le_s.value():
|
||||
case Instructions::i8x16_le_u.value():
|
||||
case Instructions::i8x16_ge_s.value():
|
||||
case Instructions::i8x16_ge_u.value():
|
||||
case Instructions::i16x8_eq.value():
|
||||
case Instructions::i16x8_ne.value():
|
||||
case Instructions::i16x8_lt_s.value():
|
||||
case Instructions::i16x8_lt_u.value():
|
||||
case Instructions::i16x8_gt_s.value():
|
||||
case Instructions::i16x8_gt_u.value():
|
||||
case Instructions::i16x8_le_s.value():
|
||||
case Instructions::i16x8_le_u.value():
|
||||
case Instructions::i16x8_ge_s.value():
|
||||
case Instructions::i16x8_ge_u.value():
|
||||
case Instructions::i32x4_eq.value():
|
||||
case Instructions::i32x4_ne.value():
|
||||
case Instructions::i32x4_lt_s.value():
|
||||
case Instructions::i32x4_lt_u.value():
|
||||
case Instructions::i32x4_gt_s.value():
|
||||
case Instructions::i32x4_gt_u.value():
|
||||
case Instructions::i32x4_le_s.value():
|
||||
case Instructions::i32x4_le_u.value():
|
||||
case Instructions::i32x4_ge_s.value():
|
||||
case Instructions::i32x4_ge_u.value():
|
||||
case Instructions::f32x4_eq.value():
|
||||
case Instructions::f32x4_ne.value():
|
||||
case Instructions::f32x4_lt.value():
|
||||
case Instructions::f32x4_gt.value():
|
||||
case Instructions::f32x4_le.value():
|
||||
case Instructions::f32x4_ge.value():
|
||||
case Instructions::f64x2_eq.value():
|
||||
case Instructions::f64x2_ne.value():
|
||||
case Instructions::f64x2_lt.value():
|
||||
case Instructions::f64x2_gt.value():
|
||||
case Instructions::f64x2_le.value():
|
||||
case Instructions::f64x2_ge.value():
|
||||
case Instructions::v128_not.value():
|
||||
case Instructions::v128_and.value():
|
||||
case Instructions::v128_andnot.value():
|
||||
case Instructions::v128_or.value():
|
||||
case Instructions::v128_xor.value():
|
||||
case Instructions::v128_bitselect.value():
|
||||
case Instructions::v128_any_true.value():
|
||||
case Instructions::v128_load8_lane.value():
|
||||
case Instructions::v128_load16_lane.value():
|
||||
case Instructions::v128_load32_lane.value():
|
||||
case Instructions::v128_load64_lane.value():
|
||||
case Instructions::v128_store8_lane.value():
|
||||
case Instructions::v128_store16_lane.value():
|
||||
case Instructions::v128_store32_lane.value():
|
||||
case Instructions::v128_store64_lane.value():
|
||||
case Instructions::v128_load32_zero.value():
|
||||
case Instructions::v128_load64_zero.value():
|
||||
case Instructions::f32x4_demote_f64x2_zero.value():
|
||||
case Instructions::f64x2_promote_low_f32x4.value():
|
||||
case Instructions::i8x16_abs.value():
|
||||
case Instructions::i8x16_neg.value():
|
||||
case Instructions::i8x16_popcnt.value():
|
||||
case Instructions::i8x16_all_true.value():
|
||||
case Instructions::i8x16_bitmask.value():
|
||||
case Instructions::i8x16_narrow_i16x8_s.value():
|
||||
case Instructions::i8x16_narrow_i16x8_u.value():
|
||||
case Instructions::f32x4_ceil.value():
|
||||
case Instructions::f32x4_floor.value():
|
||||
case Instructions::f32x4_trunc.value():
|
||||
case Instructions::f32x4_nearest.value():
|
||||
case Instructions::i8x16_add.value():
|
||||
case Instructions::i8x16_add_sat_s.value():
|
||||
case Instructions::i8x16_add_sat_u.value():
|
||||
case Instructions::i8x16_sub.value():
|
||||
case Instructions::i8x16_sub_sat_s.value():
|
||||
case Instructions::i8x16_sub_sat_u.value():
|
||||
case Instructions::f64x2_ceil.value():
|
||||
case Instructions::f64x2_floor.value():
|
||||
case Instructions::i8x16_min_s.value():
|
||||
case Instructions::i8x16_min_u.value():
|
||||
case Instructions::i8x16_max_s.value():
|
||||
case Instructions::i8x16_max_u.value():
|
||||
case Instructions::f64x2_trunc.value():
|
||||
case Instructions::i8x16_avgr_u.value():
|
||||
case Instructions::i16x8_extadd_pairwise_i8x16_s.value():
|
||||
case Instructions::i16x8_extadd_pairwise_i8x16_u.value():
|
||||
case Instructions::i32x4_extadd_pairwise_i16x8_s.value():
|
||||
case Instructions::i32x4_extadd_pairwise_i16x8_u.value():
|
||||
case Instructions::i16x8_abs.value():
|
||||
case Instructions::i16x8_neg.value():
|
||||
case Instructions::i16x8_q15mulr_sat_s.value():
|
||||
case Instructions::i16x8_all_true.value():
|
||||
case Instructions::i16x8_bitmask.value():
|
||||
case Instructions::i16x8_narrow_i32x4_s.value():
|
||||
case Instructions::i16x8_narrow_i32x4_u.value():
|
||||
case Instructions::i16x8_extend_low_i8x16_s.value():
|
||||
case Instructions::i16x8_extend_high_i8x16_s.value():
|
||||
case Instructions::i16x8_extend_low_i8x16_u.value():
|
||||
case Instructions::i16x8_extend_high_i8x16_u.value():
|
||||
case Instructions::i16x8_add.value():
|
||||
case Instructions::i16x8_add_sat_s.value():
|
||||
case Instructions::i16x8_add_sat_u.value():
|
||||
case Instructions::i16x8_sub.value():
|
||||
case Instructions::i16x8_sub_sat_s.value():
|
||||
case Instructions::i16x8_sub_sat_u.value():
|
||||
case Instructions::f64x2_nearest.value():
|
||||
case Instructions::i16x8_mul.value():
|
||||
case Instructions::i16x8_min_s.value():
|
||||
case Instructions::i16x8_min_u.value():
|
||||
case Instructions::i16x8_max_s.value():
|
||||
case Instructions::i16x8_max_u.value():
|
||||
case Instructions::i16x8_avgr_u.value():
|
||||
case Instructions::i16x8_extmul_low_i8x16_s.value():
|
||||
case Instructions::i16x8_extmul_high_i8x16_s.value():
|
||||
case Instructions::i16x8_extmul_low_i8x16_u.value():
|
||||
case Instructions::i16x8_extmul_high_i8x16_u.value():
|
||||
case Instructions::i32x4_abs.value():
|
||||
case Instructions::i32x4_neg.value():
|
||||
case Instructions::i32x4_all_true.value():
|
||||
case Instructions::i32x4_bitmask.value():
|
||||
case Instructions::i32x4_extend_low_i16x8_s.value():
|
||||
case Instructions::i32x4_extend_high_i16x8_s.value():
|
||||
case Instructions::i32x4_extend_low_i16x8_u.value():
|
||||
case Instructions::i32x4_extend_high_i16x8_u.value():
|
||||
case Instructions::i32x4_add.value():
|
||||
case Instructions::i32x4_sub.value():
|
||||
case Instructions::i32x4_mul.value():
|
||||
case Instructions::i32x4_min_s.value():
|
||||
case Instructions::i32x4_min_u.value():
|
||||
case Instructions::i32x4_max_s.value():
|
||||
case Instructions::i32x4_max_u.value():
|
||||
case Instructions::i32x4_dot_i16x8_s.value():
|
||||
case Instructions::i32x4_extmul_low_i16x8_s.value():
|
||||
case Instructions::i32x4_extmul_high_i16x8_s.value():
|
||||
case Instructions::i32x4_extmul_low_i16x8_u.value():
|
||||
case Instructions::i32x4_extmul_high_i16x8_u.value():
|
||||
case Instructions::i64x2_abs.value():
|
||||
case Instructions::i64x2_neg.value():
|
||||
case Instructions::i64x2_all_true.value():
|
||||
case Instructions::i64x2_bitmask.value():
|
||||
case Instructions::i64x2_extend_low_i32x4_s.value():
|
||||
case Instructions::i64x2_extend_high_i32x4_s.value():
|
||||
case Instructions::i64x2_extend_low_i32x4_u.value():
|
||||
case Instructions::i64x2_extend_high_i32x4_u.value():
|
||||
case Instructions::i64x2_add.value():
|
||||
case Instructions::i64x2_sub.value():
|
||||
case Instructions::i64x2_mul.value():
|
||||
case Instructions::i64x2_eq.value():
|
||||
case Instructions::i64x2_ne.value():
|
||||
case Instructions::i64x2_lt_s.value():
|
||||
case Instructions::i64x2_gt_s.value():
|
||||
case Instructions::i64x2_le_s.value():
|
||||
case Instructions::i64x2_ge_s.value():
|
||||
case Instructions::i64x2_extmul_low_i32x4_s.value():
|
||||
case Instructions::i64x2_extmul_high_i32x4_s.value():
|
||||
case Instructions::i64x2_extmul_low_i32x4_u.value():
|
||||
case Instructions::i64x2_extmul_high_i32x4_u.value():
|
||||
case Instructions::f32x4_abs.value():
|
||||
case Instructions::f32x4_neg.value():
|
||||
case Instructions::f32x4_sqrt.value():
|
||||
case Instructions::f32x4_add.value():
|
||||
case Instructions::f32x4_sub.value():
|
||||
case Instructions::f32x4_mul.value():
|
||||
case Instructions::f32x4_div.value():
|
||||
case Instructions::f32x4_min.value():
|
||||
case Instructions::f32x4_max.value():
|
||||
case Instructions::f32x4_pmin.value():
|
||||
case Instructions::f32x4_pmax.value():
|
||||
case Instructions::f64x2_abs.value():
|
||||
case Instructions::f64x2_neg.value():
|
||||
case Instructions::f64x2_sqrt.value():
|
||||
case Instructions::f64x2_add.value():
|
||||
case Instructions::f64x2_sub.value():
|
||||
case Instructions::f64x2_mul.value():
|
||||
case Instructions::f64x2_div.value():
|
||||
case Instructions::f64x2_min.value():
|
||||
case Instructions::f64x2_max.value():
|
||||
case Instructions::f64x2_pmin.value():
|
||||
case Instructions::f64x2_pmax.value():
|
||||
case Instructions::i32x4_trunc_sat_f32x4_s.value():
|
||||
case Instructions::i32x4_trunc_sat_f32x4_u.value():
|
||||
case Instructions::f32x4_convert_i32x4_s.value():
|
||||
case Instructions::f32x4_convert_i32x4_u.value():
|
||||
case Instructions::i32x4_trunc_sat_f64x2_s_zero.value():
|
||||
case Instructions::i32x4_trunc_sat_f64x2_u_zero.value():
|
||||
case Instructions::f64x2_convert_low_i32x4_s.value():
|
||||
case Instructions::f64x2_convert_low_i32x4_u.value():
|
||||
case Instructions::table_init.value():
|
||||
case Instructions::elem_drop.value():
|
||||
case Instructions::table_copy.value():
|
||||
|
|
|
|||
|
|
@ -65,11 +65,11 @@ protected:
|
|||
void store_to_memory(Configuration&, Instruction const&, ReadonlyBytes data, i32 base);
|
||||
void call_address(Configuration&, FunctionAddress);
|
||||
|
||||
template<typename PopTypeLHS, typename PushType, typename Operator, typename PopTypeRHS = PopTypeLHS>
|
||||
void binary_numeric_operation(Configuration&);
|
||||
template<typename PopTypeLHS, typename PushType, typename Operator, typename PopTypeRHS = PopTypeLHS, typename... Args>
|
||||
void binary_numeric_operation(Configuration&, Args&&...);
|
||||
|
||||
template<typename PopType, typename PushType, typename Operator>
|
||||
void unary_operation(Configuration&);
|
||||
template<typename PopType, typename PushType, typename Operator, typename... Args>
|
||||
void unary_operation(Configuration&, Args&&...);
|
||||
|
||||
template<typename V, typename T>
|
||||
MakeUnsigned<T> checked_unsigned_truncate(V);
|
||||
|
|
|
|||
|
|
@ -172,6 +172,101 @@ struct VectorShiftRight {
|
|||
}
|
||||
};
|
||||
|
||||
struct VectorSwizzle {
|
||||
auto operator()(u128 c1, u128 c2) const
|
||||
{
|
||||
// https://webassembly.github.io/spec/core/bikeshed/#-mathsfi8x16hrefsyntax-instr-vecmathsfswizzle%E2%91%A0
|
||||
auto i = bit_cast<Native128ByteVectorOf<i8, MakeSigned>>(c2);
|
||||
auto j = bit_cast<Native128ByteVectorOf<i8, MakeSigned>>(c1);
|
||||
auto result = AK::SIMD::shuffle(i, j);
|
||||
return bit_cast<u128>(result);
|
||||
}
|
||||
static StringView name() { return "vec(8x16).swizzle"sv; }
|
||||
};
|
||||
|
||||
template<size_t VectorSize, template<typename> typename SetSign>
|
||||
struct VectorExtractLane {
|
||||
size_t lane;
|
||||
|
||||
auto operator()(u128 c) const
|
||||
{
|
||||
auto result = bit_cast<Native128ByteVectorOf<NativeIntegralType<128 / VectorSize>, SetSign>>(c);
|
||||
return result[lane];
|
||||
}
|
||||
|
||||
static StringView name()
|
||||
{
|
||||
switch (VectorSize) {
|
||||
case 16:
|
||||
return "vec(8x16).extract_lane"sv;
|
||||
case 8:
|
||||
return "vec(16x8).extract_lane"sv;
|
||||
case 4:
|
||||
return "vec(32x4).extract_lane"sv;
|
||||
case 2:
|
||||
return "vec(64x2).extract_lane"sv;
|
||||
default:
|
||||
VERIFY_NOT_REACHED();
|
||||
}
|
||||
}
|
||||
};
|
||||
|
||||
template<size_t VectorSize>
|
||||
struct VectorExtractLaneFloat {
|
||||
size_t lane;
|
||||
|
||||
auto operator()(u128 c) const
|
||||
{
|
||||
auto result = bit_cast<NativeFloatingVectorType<128 / VectorSize, VectorSize>>(c);
|
||||
return result[lane];
|
||||
}
|
||||
|
||||
static StringView name()
|
||||
{
|
||||
switch (VectorSize) {
|
||||
case 16:
|
||||
return "vec(8x16).extract_lane"sv;
|
||||
case 8:
|
||||
return "vec(16x8).extract_lane"sv;
|
||||
case 4:
|
||||
return "vec(32x4).extract_lane"sv;
|
||||
case 2:
|
||||
return "vec(64x2).extract_lane"sv;
|
||||
default:
|
||||
VERIFY_NOT_REACHED();
|
||||
}
|
||||
}
|
||||
};
|
||||
|
||||
template<size_t VectorSize, typename TrueValueType = NativeIntegralType<128 / VectorSize>>
|
||||
struct VectorReplaceLane {
|
||||
size_t lane;
|
||||
using ValueType = Conditional<IsFloatingPoint<TrueValueType>, NativeFloatingType<128 / VectorSize>, NativeIntegralType<128 / VectorSize>>;
|
||||
|
||||
auto operator()(u128 c, TrueValueType value) const
|
||||
{
|
||||
auto result = bit_cast<Native128ByteVectorOf<ValueType, MakeUnsigned>>(c);
|
||||
result[lane] = static_cast<ValueType>(value);
|
||||
return bit_cast<u128>(result);
|
||||
}
|
||||
|
||||
static StringView name()
|
||||
{
|
||||
switch (VectorSize) {
|
||||
case 16:
|
||||
return "vec(8x16).replace_lane"sv;
|
||||
case 8:
|
||||
return "vec(16x8).replace_lane"sv;
|
||||
case 4:
|
||||
return "vec(32x4).replace_lane"sv;
|
||||
case 2:
|
||||
return "vec(64x2).replace_lane"sv;
|
||||
default:
|
||||
VERIFY_NOT_REACHED();
|
||||
}
|
||||
}
|
||||
};
|
||||
|
||||
struct Minimum {
|
||||
template<typename Lhs, typename Rhs>
|
||||
auto operator()(Lhs lhs, Rhs rhs) const
|
||||
|
|
|
|||
|
|
@ -29,6 +29,9 @@ using NativeFloatingType = Conditional<M == 32, f32, Conditional<M == 64, f64, v
|
|||
template<size_t M, size_t N, template<typename> typename SetSign, typename ElementType = SetSign<NativeIntegralType<M>>>
|
||||
using NativeVectorType __attribute__((vector_size(N * sizeof(ElementType)))) = ElementType;
|
||||
|
||||
template<size_t M, size_t N, typename ElementType = NativeFloatingType<M>>
|
||||
using NativeFloatingVectorType __attribute__((vector_size(N * sizeof(ElementType)))) = ElementType;
|
||||
|
||||
template<typename T, template<typename> typename SetSign>
|
||||
using Native128ByteVectorOf = NativeVectorType<sizeof(T) * 8, 16 / sizeof(T), SetSign, T>;
|
||||
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue