Relax NeuralNetwork's VTS positive and negative base tests

There are some NN VTS tests that assume a service is able to generate a
model consisting only of a floating point add operation. However, some
drivers do not support floating point operations. This CL relaxes the
test requirements to allow a test to be skipped if the service does not
support floating point add.

Bug: 72764145
Test: mma
Test: VtsHalNeuralnetworksV1_0TargetTest
Change-Id: I6b0644432680fc2f8098b5187795dc2953df03f9
This commit is contained in:
Michael Butler 2018-02-26 15:24:46 -08:00
parent fe606d5aee
commit 4d5bb1097a
3 changed files with 128 additions and 91 deletions

View file

@ -186,35 +186,29 @@ void Execute(sp<V1_0::IDevice>& device, std::function<V1_0::Model(void)> create_
// see if service can handle model
bool fullySupportsModel = false;
ErrorStatus supportedStatus;
sp<PreparedModelCallback> preparedModelCallback = new PreparedModelCallback();
ASSERT_NE(nullptr, preparedModelCallback.get());
Return<void> supportedCall = device->getSupportedOperations(
model, [&](ErrorStatus status, const hidl_vec<bool>& supported) {
supportedStatus = status;
model, [&fullySupportsModel](ErrorStatus status, const hidl_vec<bool>& supported) {
ASSERT_EQ(ErrorStatus::NONE, status);
ASSERT_NE(0ul, supported.size());
fullySupportsModel =
std::all_of(supported.begin(), supported.end(), [](bool valid) { return valid; });
});
ASSERT_TRUE(supportedCall.isOk());
ASSERT_EQ(ErrorStatus::NONE, supportedStatus);
// launch prepare model
sp<PreparedModelCallback> preparedModelCallback = new PreparedModelCallback();
ASSERT_NE(nullptr, preparedModelCallback.get());
Return<ErrorStatus> prepareLaunchStatus = device->prepareModel(model, preparedModelCallback);
ASSERT_TRUE(prepareLaunchStatus.isOk());
ASSERT_EQ(ErrorStatus::NONE, static_cast<ErrorStatus>(prepareLaunchStatus));
// retrieve prepared model
preparedModelCallback->wait();
ErrorStatus prepareReturnStatus = preparedModelCallback->getStatus();
sp<IPreparedModel> preparedModel = preparedModelCallback->getPreparedModel();
if (fullySupportsModel) {
EXPECT_EQ(ErrorStatus::NONE, prepareReturnStatus);
} else {
EXPECT_TRUE(prepareReturnStatus == ErrorStatus::NONE ||
prepareReturnStatus == ErrorStatus::GENERAL_FAILURE);
}
// early termination if vendor service cannot fully prepare model
if (!fullySupportsModel && prepareReturnStatus == ErrorStatus::GENERAL_FAILURE) {
if (!fullySupportsModel && prepareReturnStatus != ErrorStatus::NONE) {
ASSERT_EQ(nullptr, preparedModel.get());
LOG(INFO) << "NN VTS: Early termination of test because vendor service cannot "
"prepare model that it does not support.";
@ -223,6 +217,7 @@ void Execute(sp<V1_0::IDevice>& device, std::function<V1_0::Model(void)> create_
<< std::endl;
return;
}
EXPECT_EQ(ErrorStatus::NONE, prepareReturnStatus);
ASSERT_NE(nullptr, preparedModel.get());
EvaluatePreparedModel(preparedModel, is_ignored, examples);
@ -235,36 +230,30 @@ void Execute(sp<V1_1::IDevice>& device, std::function<V1_1::Model(void)> create_
// see if service can handle model
bool fullySupportsModel = false;
ErrorStatus supportedStatus;
sp<PreparedModelCallback> preparedModelCallback = new PreparedModelCallback();
ASSERT_NE(nullptr, preparedModelCallback.get());
Return<void> supportedCall = device->getSupportedOperations_1_1(
model, [&](ErrorStatus status, const hidl_vec<bool>& supported) {
supportedStatus = status;
model, [&fullySupportsModel](ErrorStatus status, const hidl_vec<bool>& supported) {
ASSERT_EQ(ErrorStatus::NONE, status);
ASSERT_NE(0ul, supported.size());
fullySupportsModel =
std::all_of(supported.begin(), supported.end(), [](bool valid) { return valid; });
});
ASSERT_TRUE(supportedCall.isOk());
ASSERT_EQ(ErrorStatus::NONE, supportedStatus);
// launch prepare model
sp<PreparedModelCallback> preparedModelCallback = new PreparedModelCallback();
ASSERT_NE(nullptr, preparedModelCallback.get());
Return<ErrorStatus> prepareLaunchStatus =
device->prepareModel_1_1(model, preparedModelCallback);
ASSERT_TRUE(prepareLaunchStatus.isOk());
ASSERT_EQ(ErrorStatus::NONE, static_cast<ErrorStatus>(prepareLaunchStatus));
// retrieve prepared model
preparedModelCallback->wait();
ErrorStatus prepareReturnStatus = preparedModelCallback->getStatus();
sp<IPreparedModel> preparedModel = preparedModelCallback->getPreparedModel();
if (fullySupportsModel) {
EXPECT_EQ(ErrorStatus::NONE, prepareReturnStatus);
} else {
EXPECT_TRUE(prepareReturnStatus == ErrorStatus::NONE ||
prepareReturnStatus == ErrorStatus::GENERAL_FAILURE);
}
// early termination if vendor service cannot fully prepare model
if (!fullySupportsModel && prepareReturnStatus == ErrorStatus::GENERAL_FAILURE) {
if (!fullySupportsModel && prepareReturnStatus != ErrorStatus::NONE) {
ASSERT_EQ(nullptr, preparedModel.get());
LOG(INFO) << "NN VTS: Early termination of test because vendor service cannot "
"prepare model that it does not support.";
@ -273,6 +262,7 @@ void Execute(sp<V1_1::IDevice>& device, std::function<V1_1::Model(void)> create_
<< std::endl;
return;
}
EXPECT_EQ(ErrorStatus::NONE, prepareReturnStatus);
ASSERT_NE(nullptr, preparedModel.get());
// If in relaxed mode, set the error range to be 5ULP of FP16.

View file

@ -52,26 +52,51 @@ namespace functional {
using ::android::hardware::neuralnetworks::V1_0::implementation::ExecutionCallback;
using ::android::hardware::neuralnetworks::V1_0::implementation::PreparedModelCallback;
inline sp<IPreparedModel> doPrepareModelShortcut(sp<IDevice>& device) {
static void doPrepareModelShortcut(const sp<IDevice>& device, sp<IPreparedModel>* preparedModel) {
ASSERT_NE(nullptr, preparedModel);
Model model = createValidTestModel_1_0();
sp<PreparedModelCallback> preparedModelCallback = new PreparedModelCallback();
if (preparedModelCallback == nullptr) {
return nullptr;
}
Return<ErrorStatus> prepareLaunchStatus = device->prepareModel(model, preparedModelCallback);
if (!prepareLaunchStatus.isOk() || prepareLaunchStatus != ErrorStatus::NONE) {
return nullptr;
}
// see if service can handle model
bool fullySupportsModel = false;
Return<void> supportedOpsLaunchStatus = device->getSupportedOperations(
model, [&fullySupportsModel](ErrorStatus status, const hidl_vec<bool>& supported) {
ASSERT_EQ(ErrorStatus::NONE, status);
ASSERT_NE(0ul, supported.size());
fullySupportsModel =
std::all_of(supported.begin(), supported.end(), [](bool valid) { return valid; });
});
ASSERT_TRUE(supportedOpsLaunchStatus.isOk());
// launch prepare model
sp<PreparedModelCallback> preparedModelCallback = new PreparedModelCallback();
ASSERT_NE(nullptr, preparedModelCallback.get());
Return<ErrorStatus> prepareLaunchStatus = device->prepareModel(model, preparedModelCallback);
ASSERT_TRUE(prepareLaunchStatus.isOk());
ASSERT_EQ(ErrorStatus::NONE, static_cast<ErrorStatus>(prepareLaunchStatus));
// retrieve prepared model
preparedModelCallback->wait();
ErrorStatus prepareReturnStatus = preparedModelCallback->getStatus();
sp<IPreparedModel> preparedModel = preparedModelCallback->getPreparedModel();
if (prepareReturnStatus != ErrorStatus::NONE || preparedModel == nullptr) {
return nullptr;
}
*preparedModel = preparedModelCallback->getPreparedModel();
return preparedModel;
// The getSupportedOperations call returns a list of operations that are
// guaranteed not to fail if prepareModel is called, and
// 'fullySupportsModel' is true i.f.f. the entire model is guaranteed.
// If a driver has any doubt that it can prepare an operation, it must
// return false. So here, if a driver isn't sure if it can support an
// operation, but reports that it successfully prepared the model, the test
// can continue.
if (!fullySupportsModel && prepareReturnStatus != ErrorStatus::NONE) {
ASSERT_EQ(nullptr, preparedModel->get());
LOG(INFO) << "NN VTS: Early termination of test because vendor service cannot "
"prepare model that it does not support.";
std::cout << "[ ] Early termination of test because vendor service cannot "
"prepare model that it does not support."
<< std::endl;
return;
}
ASSERT_EQ(ErrorStatus::NONE, prepareReturnStatus);
ASSERT_NE(nullptr, preparedModel->get());
}
// create device test
@ -132,18 +157,8 @@ TEST_F(NeuralnetworksHidlTest, SupportedOperationsNegativeTest2) {
// prepare simple model positive test
TEST_F(NeuralnetworksHidlTest, SimplePrepareModelPositiveTest) {
Model model = createValidTestModel_1_0();
sp<PreparedModelCallback> preparedModelCallback = new PreparedModelCallback();
ASSERT_NE(nullptr, preparedModelCallback.get());
Return<ErrorStatus> prepareLaunchStatus = device->prepareModel(model, preparedModelCallback);
ASSERT_TRUE(prepareLaunchStatus.isOk());
EXPECT_EQ(ErrorStatus::NONE, static_cast<ErrorStatus>(prepareLaunchStatus));
preparedModelCallback->wait();
ErrorStatus prepareReturnStatus = preparedModelCallback->getStatus();
EXPECT_EQ(ErrorStatus::NONE, prepareReturnStatus);
sp<IPreparedModel> preparedModel = preparedModelCallback->getPreparedModel();
EXPECT_NE(nullptr, preparedModel.get());
sp<IPreparedModel> preparedModel;
doPrepareModelShortcut(device, &preparedModel);
}
// prepare simple model negative test 1
@ -184,8 +199,11 @@ TEST_F(NeuralnetworksHidlTest, SimpleExecuteGraphPositiveTest) {
std::vector<float> expectedData = {6.0f, 8.0f, 10.0f, 12.0f};
const uint32_t OUTPUT = 1;
sp<IPreparedModel> preparedModel = doPrepareModelShortcut(device);
ASSERT_NE(nullptr, preparedModel.get());
sp<IPreparedModel> preparedModel;
ASSERT_NO_FATAL_FAILURE(doPrepareModelShortcut(device, &preparedModel));
if (preparedModel == nullptr) {
return;
}
Request request = createValidTestRequest();
auto postWork = [&] {
@ -218,8 +236,11 @@ TEST_F(NeuralnetworksHidlTest, SimpleExecuteGraphPositiveTest) {
// execute simple graph negative test 1
TEST_F(NeuralnetworksHidlTest, SimpleExecuteGraphNegativeTest1) {
sp<IPreparedModel> preparedModel = doPrepareModelShortcut(device);
ASSERT_NE(nullptr, preparedModel.get());
sp<IPreparedModel> preparedModel;
ASSERT_NO_FATAL_FAILURE(doPrepareModelShortcut(device, &preparedModel));
if (preparedModel == nullptr) {
return;
}
Request request = createInvalidTestRequest1();
sp<ExecutionCallback> executionCallback = new ExecutionCallback();
@ -235,8 +256,11 @@ TEST_F(NeuralnetworksHidlTest, SimpleExecuteGraphNegativeTest1) {
// execute simple graph negative test 2
TEST_F(NeuralnetworksHidlTest, SimpleExecuteGraphNegativeTest2) {
sp<IPreparedModel> preparedModel = doPrepareModelShortcut(device);
ASSERT_NE(nullptr, preparedModel.get());
sp<IPreparedModel> preparedModel;
ASSERT_NO_FATAL_FAILURE(doPrepareModelShortcut(device, &preparedModel));
if (preparedModel == nullptr) {
return;
}
Request request = createInvalidTestRequest2();
sp<ExecutionCallback> executionCallback = new ExecutionCallback();

View file

@ -59,27 +59,52 @@ namespace functional {
using ::android::hardware::neuralnetworks::V1_0::implementation::ExecutionCallback;
using ::android::hardware::neuralnetworks::V1_0::implementation::PreparedModelCallback;
inline sp<IPreparedModel> doPrepareModelShortcut(sp<IDevice>& device) {
static void doPrepareModelShortcut(const sp<IDevice>& device, sp<IPreparedModel>* preparedModel) {
ASSERT_NE(nullptr, preparedModel);
Model model = createValidTestModel_1_1();
// see if service can handle model
bool fullySupportsModel = false;
Return<void> supportedOpsLaunchStatus = device->getSupportedOperations_1_1(
model, [&fullySupportsModel](ErrorStatus status, const hidl_vec<bool>& supported) {
ASSERT_EQ(ErrorStatus::NONE, status);
ASSERT_NE(0ul, supported.size());
fullySupportsModel =
std::all_of(supported.begin(), supported.end(), [](bool valid) { return valid; });
});
ASSERT_TRUE(supportedOpsLaunchStatus.isOk());
// launch prepare model
sp<PreparedModelCallback> preparedModelCallback = new PreparedModelCallback();
if (preparedModelCallback == nullptr) {
return nullptr;
}
ASSERT_NE(nullptr, preparedModelCallback.get());
Return<ErrorStatus> prepareLaunchStatus =
device->prepareModel_1_1(model, preparedModelCallback);
if (!prepareLaunchStatus.isOk() || prepareLaunchStatus != ErrorStatus::NONE) {
return nullptr;
}
ASSERT_TRUE(prepareLaunchStatus.isOk());
ASSERT_EQ(ErrorStatus::NONE, static_cast<ErrorStatus>(prepareLaunchStatus));
// retrieve prepared model
preparedModelCallback->wait();
ErrorStatus prepareReturnStatus = preparedModelCallback->getStatus();
sp<IPreparedModel> preparedModel = preparedModelCallback->getPreparedModel();
if (prepareReturnStatus != ErrorStatus::NONE || preparedModel == nullptr) {
return nullptr;
}
*preparedModel = preparedModelCallback->getPreparedModel();
return preparedModel;
// The getSupportedOperations call returns a list of operations that are
// guaranteed not to fail if prepareModel is called, and
// 'fullySupportsModel' is true i.f.f. the entire model is guaranteed.
// If a driver has any doubt that it can prepare an operation, it must
// return false. So here, if a driver isn't sure if it can support an
// operation, but reports that it successfully prepared the model, the test
// can continue.
if (!fullySupportsModel && prepareReturnStatus != ErrorStatus::NONE) {
ASSERT_EQ(nullptr, preparedModel->get());
LOG(INFO) << "NN VTS: Early termination of test because vendor service cannot "
"prepare model that it does not support.";
std::cout << "[ ] Early termination of test because vendor service cannot "
"prepare model that it does not support."
<< std::endl;
return;
}
ASSERT_EQ(ErrorStatus::NONE, prepareReturnStatus);
ASSERT_NE(nullptr, preparedModel->get());
}
// create device test
@ -142,19 +167,8 @@ TEST_F(NeuralnetworksHidlTest, SupportedOperationsNegativeTest2) {
// prepare simple model positive test
TEST_F(NeuralnetworksHidlTest, SimplePrepareModelPositiveTest) {
Model model = createValidTestModel_1_1();
sp<PreparedModelCallback> preparedModelCallback = new PreparedModelCallback();
ASSERT_NE(nullptr, preparedModelCallback.get());
Return<ErrorStatus> prepareLaunchStatus =
device->prepareModel_1_1(model, preparedModelCallback);
ASSERT_TRUE(prepareLaunchStatus.isOk());
EXPECT_EQ(ErrorStatus::NONE, static_cast<ErrorStatus>(prepareLaunchStatus));
preparedModelCallback->wait();
ErrorStatus prepareReturnStatus = preparedModelCallback->getStatus();
EXPECT_EQ(ErrorStatus::NONE, prepareReturnStatus);
sp<IPreparedModel> preparedModel = preparedModelCallback->getPreparedModel();
EXPECT_NE(nullptr, preparedModel.get());
sp<IPreparedModel> preparedModel;
doPrepareModelShortcut(device, &preparedModel);
}
// prepare simple model negative test 1
@ -197,8 +211,11 @@ TEST_F(NeuralnetworksHidlTest, SimpleExecuteGraphPositiveTest) {
std::vector<float> expectedData = {6.0f, 8.0f, 10.0f, 12.0f};
const uint32_t OUTPUT = 1;
sp<IPreparedModel> preparedModel = doPrepareModelShortcut(device);
ASSERT_NE(nullptr, preparedModel.get());
sp<IPreparedModel> preparedModel;
ASSERT_NO_FATAL_FAILURE(doPrepareModelShortcut(device, &preparedModel));
if (preparedModel == nullptr) {
return;
}
Request request = createValidTestRequest();
auto postWork = [&] {
@ -231,8 +248,11 @@ TEST_F(NeuralnetworksHidlTest, SimpleExecuteGraphPositiveTest) {
// execute simple graph negative test 1
TEST_F(NeuralnetworksHidlTest, SimpleExecuteGraphNegativeTest1) {
sp<IPreparedModel> preparedModel = doPrepareModelShortcut(device);
ASSERT_NE(nullptr, preparedModel.get());
sp<IPreparedModel> preparedModel;
ASSERT_NO_FATAL_FAILURE(doPrepareModelShortcut(device, &preparedModel));
if (preparedModel == nullptr) {
return;
}
Request request = createInvalidTestRequest1();
sp<ExecutionCallback> executionCallback = new ExecutionCallback();
@ -248,8 +268,11 @@ TEST_F(NeuralnetworksHidlTest, SimpleExecuteGraphNegativeTest1) {
// execute simple graph negative test 2
TEST_F(NeuralnetworksHidlTest, SimpleExecuteGraphNegativeTest2) {
sp<IPreparedModel> preparedModel = doPrepareModelShortcut(device);
ASSERT_NE(nullptr, preparedModel.get());
sp<IPreparedModel> preparedModel;
ASSERT_NO_FATAL_FAILURE(doPrepareModelShortcut(device, &preparedModel));
if (preparedModel == nullptr) {
return;
}
Request request = createInvalidTestRequest2();
sp<ExecutionCallback> executionCallback = new ExecutionCallback();