You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

114 lines
4.2 KiB

/*
* Copyright (C) 2019 The Android Open Source Project
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#define LOG_TAG "TestControlFlow"
#include <ControlFlow.h>
#include <android-base/logging.h>
#include <gtest/gtest.h>
#include "TestNeuralNetworksWrapper.h"
namespace android::nn {
namespace {
using test_wrapper::Compilation;
using test_wrapper::Execution;
using test_wrapper::Model;
using test_wrapper::OperandType;
using test_wrapper::Result;
using test_wrapper::Type;
constexpr uint64_t kMillisecondsInNanosecond = 1'000'000;
constexpr int32_t kNoActivation = ANEURALNETWORKS_FUSED_NONE;
class ControlFlowTest : public ::testing::Test {};
TEST_F(ControlFlowTest, InfiniteLoop) {
// Expected result: execution aborted after the specified timeout.
// Model: given n <= 1.0, never returns.
//
// i = 1.0
// while i >= n:
// i = i + 1.0
OperandType boolType(Type::TENSOR_BOOL8, {1});
OperandType activationType(Type::INT32, {});
OperandType counterType(Type::TENSOR_FLOAT32, {1});
Model conditionModel;
{
uint32_t i = conditionModel.addOperand(&counterType);
uint32_t n = conditionModel.addOperand(&counterType);
uint32_t out = conditionModel.addOperand(&boolType);
conditionModel.addOperation(ANEURALNETWORKS_GREATER_EQUAL, {i, n}, {out});
conditionModel.identifyInputsAndOutputs({i, n}, {out});
ASSERT_EQ(conditionModel.finish(), Result::NO_ERROR);
ASSERT_TRUE(conditionModel.isValid());
}
Model bodyModel;
{
uint32_t i = bodyModel.addOperand(&counterType);
uint32_t n = bodyModel.addOperand(&counterType);
uint32_t one = bodyModel.addConstantOperand(&counterType, 1.0f);
uint32_t noActivation = bodyModel.addConstantOperand(&activationType, kNoActivation);
uint32_t iOut = bodyModel.addOperand(&counterType);
bodyModel.addOperation(ANEURALNETWORKS_ADD, {i, one, noActivation}, {iOut});
bodyModel.identifyInputsAndOutputs({i, n}, {iOut});
ASSERT_EQ(bodyModel.finish(), Result::NO_ERROR);
ASSERT_TRUE(bodyModel.isValid());
}
Model model;
{
uint32_t iInit = model.addConstantOperand(&counterType, 1.0f);
uint32_t n = model.addOperand(&counterType);
uint32_t conditionOperand = model.addModelOperand(&conditionModel);
uint32_t bodyOperand = model.addModelOperand(&bodyModel);
uint32_t iOut = model.addOperand(&counterType);
model.addOperation(ANEURALNETWORKS_WHILE, {conditionOperand, bodyOperand, iInit, n},
{iOut});
model.identifyInputsAndOutputs({n}, {iOut});
ASSERT_EQ(model.finish(), Result::NO_ERROR);
ASSERT_TRUE(model.isValid());
}
Compilation compilation(&model);
ASSERT_EQ(compilation.finish(), Result::NO_ERROR);
float input = 0;
float output;
Execution execution(&compilation);
ASSERT_EQ(execution.setInput(0, &input), Result::NO_ERROR);
ASSERT_EQ(execution.setOutput(0, &output), Result::NO_ERROR);
ASSERT_EQ(execution.setLoopTimeout(1 * kMillisecondsInNanosecond), Result::NO_ERROR);
Result result = execution.compute();
ASSERT_TRUE(result == Result::MISSED_DEADLINE_TRANSIENT ||
result == Result::MISSED_DEADLINE_PERSISTENT)
<< "result = " << static_cast<int>(result);
}
TEST_F(ControlFlowTest, GetLoopTimeouts) {
uint64_t defaultTimeout = ANeuralNetworks_getDefaultLoopTimeout();
uint64_t maximumTimeout = ANeuralNetworks_getMaximumLoopTimeout();
ASSERT_EQ(defaultTimeout, operation_while::kTimeoutNsDefault);
ASSERT_EQ(maximumTimeout, operation_while::kTimeoutNsMaximum);
}
} // end namespace
} // namespace android::nn