Use softer error reporting instead of CHECK*
- CHECK(x) checks whether condition x holds and LOG(FATAL) if not, which
will result in calling abort().
- This change uses nnOpsCheck which would log the failing condition and
return false to the runtime, allowing graceful failures.
Bug: 63905942
Test: NeuralNetworkTests pass
Change-Id: I8b1217f777638f974c91fa429449e39d37218af6
diff --git a/common/OperationsUtils.cpp b/common/OperationsUtils.cpp
index ae7c827..f38fcc1 100644
--- a/common/OperationsUtils.cpp
+++ b/common/OperationsUtils.cpp
@@ -66,47 +66,56 @@
}
-void QuantizeMultiplierSmallerThanOne(double double_multiplier,
+// Macro to check if the input parameters for operation are valid or not.
+#define NN_OPS_CHECK(v) \
+ if (!(v)) { \
+ LOG(ERROR) << "NN_OPS_CHECK failed: " << #v << "'\n"; \
+ return false; \
+ }
+
+bool QuantizeMultiplierSmallerThanOne(double double_multiplier,
int32_t* quantized_multiplier,
int32_t* right_shift) {
- CHECK(double_multiplier >= 0.);
- CHECK(double_multiplier < 1.);
+ NN_OPS_CHECK(double_multiplier >= 0.);
+ NN_OPS_CHECK(double_multiplier < 1.);
if (double_multiplier == 0.) {
*quantized_multiplier = 0;
*right_shift = 0;
- return;
+ return true;
}
- CHECK(double_multiplier > 0.);
+ NN_OPS_CHECK(double_multiplier > 0.);
const double q = std::frexp(double_multiplier, right_shift);
*right_shift *= -1;
int64_t q_fixed = static_cast<int64_t>(std::round(q * (1ll << 31)));
- CHECK(q_fixed <= (1ll << 31));
+ NN_OPS_CHECK(q_fixed <= (1ll << 31));
if (q_fixed == (1ll << 31)) {
q_fixed /= 2;
--*right_shift;
}
- CHECK_GE(*right_shift, 0);
- CHECK_LE(q_fixed, std::numeric_limits<int32_t>::max());
+ NN_OPS_CHECK(*right_shift >= 0);
+ NN_OPS_CHECK(q_fixed <= std::numeric_limits<int32_t>::max());
*quantized_multiplier = static_cast<int32_t>(q_fixed);
+ return true;
}
-void QuantizeMultiplierGreaterThanOne(double double_multiplier,
+bool QuantizeMultiplierGreaterThanOne(double double_multiplier,
int32_t* quantized_multiplier,
int* left_shift) {
- CHECK(double_multiplier > 1.);
+ NN_OPS_CHECK(double_multiplier > 1.);
const double q = std::frexp(double_multiplier, left_shift);
int64_t q_fixed = static_cast<int64_t>(std::round(q * (1ll << 31)));
- CHECK(q_fixed <= (1ll << 31));
+ NN_OPS_CHECK(q_fixed <= (1ll << 31));
if (q_fixed == (1ll << 31)) {
q_fixed /= 2;
++*left_shift;
}
- CHECK_GE(*left_shift, 0);
- CHECK_LE(q_fixed, std::numeric_limits<int32_t>::max());
+ NN_OPS_CHECK(*left_shift >= 0);
+ NN_OPS_CHECK(q_fixed <= std::numeric_limits<int32_t>::max());
*quantized_multiplier = static_cast<int32_t>(q_fixed);
+ return true;
}
-void GetQuantizedConvolutionMultipler(const Shape& inputShape,
+bool GetQuantizedConvolutionMultipler(const Shape& inputShape,
const Shape& filterShape,
const Shape& biasShape,
const Shape& outputShape,
@@ -116,11 +125,12 @@
const float output_scale = outputShape.scale;
// The following conditions must be guaranteed by the training pipeline.
- CHECK(std::abs(input_product_scale - bias_scale) <=
+ NN_OPS_CHECK(std::abs(input_product_scale - bias_scale) <=
1e-6 * std::min(input_product_scale, bias_scale));
- CHECK(input_product_scale >= 0);
- CHECK(input_product_scale < output_scale);
+ NN_OPS_CHECK(input_product_scale >= 0);
+ NN_OPS_CHECK(input_product_scale < output_scale);
*multiplier = input_product_scale / output_scale;
+ return true;
}
void CalculateActivationRangeUint8(int32_t activation,
@@ -162,16 +172,8 @@
return static_cast<int32_t>(std::floor(max_input_rescaled));
}
-
-// Macro to check if the input parameters for operation are valid or not.
-#define nnOpsCheck(v) \
- if (!(v)) { \
- LOG(ERROR) << "nnOpsCheck failed: " << #v << "'\n"; \
- return false; \
- }
-
bool addMulPrepare(const Shape& in1, const Shape& in2, Shape* out) {
- nnOpsCheck(getNumberOfDimensions(in1) <= 4 && getNumberOfDimensions(in2) <= 4);
+ NN_OPS_CHECK(getNumberOfDimensions(in1) <= 4 && getNumberOfDimensions(in2) <= 4);
if (SameShape(in1, in2)) {
return SetShape(in1, out);
} else {
@@ -224,14 +226,14 @@
int32_t padding_top, int32_t padding_bottom,
int32_t stride_width, int32_t stride_height,
Shape* output) {
- nnOpsCheck(getNumberOfDimensions(input) == 4);
- nnOpsCheck(getNumberOfDimensions(filter) == 4);
- nnOpsCheck(getNumberOfDimensions(bias) == 1);
+ NN_OPS_CHECK(getNumberOfDimensions(input) == 4);
+ NN_OPS_CHECK(getNumberOfDimensions(filter) == 4);
+ NN_OPS_CHECK(getNumberOfDimensions(bias) == 1);
- nnOpsCheck(getSizeOfDimension(filter, 0) == getSizeOfDimension(bias, 0));
- nnOpsCheck(getSizeOfDimension(filter, 3) == getSizeOfDimension(input, 3));
+ NN_OPS_CHECK(getSizeOfDimension(filter, 0) == getSizeOfDimension(bias, 0));
+ NN_OPS_CHECK(getSizeOfDimension(filter, 3) == getSizeOfDimension(input, 3));
- nnOpsCheck(stride_width == stride_height);
+ NN_OPS_CHECK(stride_width == stride_height);
uint32_t channels_out = getSizeOfDimension(filter, 0);
uint32_t width = getSizeOfDimension(input, 2);
@@ -257,13 +259,13 @@
int32_t padding_top, int32_t padding_bottom,
int32_t stride_width, int32_t stride_height,
Shape* output) {
- nnOpsCheck(getNumberOfDimensions(input) == 4);
- nnOpsCheck(getNumberOfDimensions(filter) == 4);
- nnOpsCheck(getNumberOfDimensions(bias) == 1);
+ NN_OPS_CHECK(getNumberOfDimensions(input) == 4);
+ NN_OPS_CHECK(getNumberOfDimensions(filter) == 4);
+ NN_OPS_CHECK(getNumberOfDimensions(bias) == 1);
- nnOpsCheck(getSizeOfDimension(filter, 3) == getSizeOfDimension(bias, 0));
+ NN_OPS_CHECK(getSizeOfDimension(filter, 3) == getSizeOfDimension(bias, 0));
- nnOpsCheck(stride_width == stride_height);
+ NN_OPS_CHECK(stride_width == stride_height);
uint32_t channels_out = getSizeOfDimension(filter, 3);
uint32_t width = getSizeOfDimension(input, 2);
@@ -289,8 +291,8 @@
int32_t stride_width, int32_t stride_height,
int32_t filter_width, int32_t filter_height,
Shape* output) {
- nnOpsCheck(getNumberOfDimensions(input) == 4);
- nnOpsCheck(stride_width == stride_height);
+ NN_OPS_CHECK(getNumberOfDimensions(input) == 4);
+ NN_OPS_CHECK(stride_width == stride_height);
uint32_t batches = getSizeOfDimension(input, 0);
uint32_t width = getSizeOfDimension(input, 2);
@@ -310,7 +312,7 @@
bool genericActivationPrepare(const Shape& input,
Shape* output) {
- nnOpsCheck(getNumberOfDimensions(input) <= 4);
+ NN_OPS_CHECK(getNumberOfDimensions(input) <= 4);
return SetShape(input, output);
}
@@ -324,9 +326,9 @@
uint32_t num_units = getSizeOfDimension(weights, 0);
uint32_t batch_size = input_size / getSizeOfDimension(weights, 1);
- nnOpsCheck(getSizeOfDimension(bias, 0) == num_units);
- nnOpsCheck(getSizeOfDimension(weights, 1) * batch_size == input_size);
- nnOpsCheck(getNumberOfDimensions(weights) == 2);
+ NN_OPS_CHECK(getSizeOfDimension(bias, 0) == num_units);
+ NN_OPS_CHECK(getSizeOfDimension(weights, 1) * batch_size == input_size);
+ NN_OPS_CHECK(getNumberOfDimensions(weights) == 2);
output->type = input.type;
output->dimensions = {batch_size, num_units};
@@ -342,22 +344,22 @@
OperandType input_type = inputShapes[0].type;
uint32_t num_dimensions = getNumberOfDimensions(inputShapes[0]);
- nnOpsCheck(axis >= 0);
- nnOpsCheck(axis < (int32_t)num_dimensions);
+ NN_OPS_CHECK(axis >= 0);
+ NN_OPS_CHECK(axis < (int32_t)num_dimensions);
int sum_axis = getSizeOfDimension(inputShapes[0], axis);
for (int i = 1; i < num_inputs; ++i) {
- nnOpsCheck(getNumberOfDimensions(inputShapes[i]) == num_dimensions);
- nnOpsCheck(inputShapes[i].type == inputShapes[0].type);
+ NN_OPS_CHECK(getNumberOfDimensions(inputShapes[i]) == num_dimensions);
+ NN_OPS_CHECK(inputShapes[i].type == inputShapes[0].type);
if (input_type == OperandType::TENSOR_QUANT8_ASYMM) {
- nnOpsCheck(inputShapes[0].offset == inputShapes[i].offset);
- nnOpsCheck(inputShapes[0].scale == inputShapes[i].scale);
+ NN_OPS_CHECK(inputShapes[0].offset == inputShapes[i].offset);
+ NN_OPS_CHECK(inputShapes[0].scale == inputShapes[i].scale);
}
for (int d = 0; d < (int32_t)num_dimensions; ++d) {
if (d == axis) {
sum_axis += getSizeOfDimension(inputShapes[i], axis);
} else {
- nnOpsCheck(getSizeOfDimension(inputShapes[0], d) ==
+ NN_OPS_CHECK(getSizeOfDimension(inputShapes[0], d) ==
getSizeOfDimension(inputShapes[i], d));
}
}
@@ -368,8 +370,8 @@
output->dimensions[axis] = sum_axis;
if (input_type == OperandType::TENSOR_QUANT8_ASYMM) {
- nnOpsCheck(inputShapes[0].offset == output->offset);
- nnOpsCheck(inputShapes[0].scale == output->scale);
+ NN_OPS_CHECK(inputShapes[0].offset == output->offset);
+ NN_OPS_CHECK(inputShapes[0].scale == output->scale);
}
return true;
@@ -377,7 +379,7 @@
bool genericNormalizationPrepare(const Shape& input, Shape* output) {
- nnOpsCheck(getNumberOfDimensions(input) == 4);
+ NN_OPS_CHECK(getNumberOfDimensions(input) == 4);
return SetShape(input, output);
}
@@ -397,7 +399,7 @@
for (int32_t i = 0; i < targetDimsSize; ++i) {
int32_t value = targetDims[i];
if (value == -1) {
- nnOpsCheck(strechDim == -1);
+ NN_OPS_CHECK(strechDim == -1);
strechDim = i;
} else {
numOutputElements *= value;
@@ -410,7 +412,7 @@
numOutputElements *= strechValue;
}
- nnOpsCheck(numInputElements == numOutputElements);
+ NN_OPS_CHECK(numInputElements == numOutputElements);
output->type = input.type;
output->dimensions = outDims;
@@ -424,7 +426,7 @@
int32_t width,
int32_t height,
Shape* output) {
- nnOpsCheck(getNumberOfDimensions(input) == 4);
+ NN_OPS_CHECK(getNumberOfDimensions(input) == 4);
uint32_t batches = getSizeOfDimension(input, 0);
uint32_t channels = getSizeOfDimension(input, 3);
@@ -437,15 +439,15 @@
bool depthToSpacePrepare(const Shape& input,
int32_t blockSize,
Shape* output) {
- nnOpsCheck(getNumberOfDimensions(input) == 4);
- nnOpsCheck(blockSize > 0);
+ NN_OPS_CHECK(getNumberOfDimensions(input) == 4);
+ NN_OPS_CHECK(blockSize > 0);
uint32_t batches = getSizeOfDimension(input, 0);
uint32_t height = getSizeOfDimension(input, 1);
uint32_t width = getSizeOfDimension(input, 2);
uint32_t channels = getSizeOfDimension(input, 3);
- nnOpsCheck(channels % (blockSize * blockSize) == 0);
+ NN_OPS_CHECK(channels % (blockSize * blockSize) == 0);
output->type = input.type;
output->dimensions = {batches,
height * blockSize,
@@ -460,16 +462,16 @@
bool spaceToDepthPrepare(const Shape& input,
int32_t blockSize,
Shape* output) {
- nnOpsCheck(getNumberOfDimensions(input) == 4);
- nnOpsCheck(blockSize > 0);
+ NN_OPS_CHECK(getNumberOfDimensions(input) == 4);
+ NN_OPS_CHECK(blockSize > 0);
uint32_t batches = getSizeOfDimension(input, 0);
uint32_t height = getSizeOfDimension(input, 1);
uint32_t width = getSizeOfDimension(input, 2);
uint32_t channels = getSizeOfDimension(input, 3);
- nnOpsCheck(height % blockSize == 0);
- nnOpsCheck(width % blockSize == 0);
+ NN_OPS_CHECK(height % blockSize == 0);
+ NN_OPS_CHECK(width % blockSize == 0);
output->type = input.type;
output->dimensions = {batches,
diff --git a/common/include/OperationsUtils.h b/common/include/OperationsUtils.h
index 88783af..d13e084 100644
--- a/common/include/OperationsUtils.h
+++ b/common/include/OperationsUtils.h
@@ -78,15 +78,18 @@
return (imageSize - filterSize + stride + paddingHead + paddingTail) / stride;
}
-void QuantizeMultiplierSmallerThanOne(double double_multiplier,
+__wur
+bool QuantizeMultiplierSmallerThanOne(double double_multiplier,
int32_t* quantized_multiplier,
int32_t* right_shift);
-void QuantizeMultiplierGreaterThanOne(double double_multiplier,
+__wur
+bool QuantizeMultiplierGreaterThanOne(double double_multiplier,
int32_t* quantized_multiplier,
int* left_shift);
-void GetQuantizedConvolutionMultipler(const Shape& inputShape,
+__wur
+bool GetQuantizedConvolutionMultipler(const Shape& inputShape,
const Shape& filterShape,
const Shape& biasShape,
const Shape& outputShape,
diff --git a/common/operations/Activation.cpp b/common/operations/Activation.cpp
index c649c22..11f54bb 100644
--- a/common/operations/Activation.cpp
+++ b/common/operations/Activation.cpp
@@ -136,9 +136,11 @@
int32_t input_multiplier = 0;
int32_t input_left_shift = 0;
- QuantizeMultiplierGreaterThanOne(input_real_multiplier,
- &input_multiplier,
- &input_left_shift);
+ if (!QuantizeMultiplierGreaterThanOne(input_real_multiplier,
+ &input_multiplier,
+ &input_left_shift)) {
+ return false;
+ }
int32_t input_range_radius =
CalculateInputRadius(kInputIntegerBits, input_left_shift);
@@ -180,9 +182,11 @@
int32_t input_multiplier = 0;
int32_t input_left_shift = 0;
- QuantizeMultiplierGreaterThanOne(input_beta_real_multiplier,
- &input_multiplier,
- &input_left_shift);
+ if (!QuantizeMultiplierGreaterThanOne(input_beta_real_multiplier,
+ &input_multiplier,
+ &input_left_shift)) {
+ return false;
+ }
float diff_min = -1.0f * CalculateInputRadius(kScaledDiffIntegerBits,
input_left_shift);
diff --git a/common/operations/Conv2D.cpp b/common/operations/Conv2D.cpp
index cbdc4f8..149ac8e 100644
--- a/common/operations/Conv2D.cpp
+++ b/common/operations/Conv2D.cpp
@@ -106,10 +106,12 @@
int32_t output_activation_min = 0;
int32_t output_activation_max = 0;
- GetQuantizedConvolutionMultipler(inputShape, filterShape, biasShape,
- outputShape, &real_multiplier);
- QuantizeMultiplierSmallerThanOne(real_multiplier, &output_multiplier,
- &output_shift);
+ if (!GetQuantizedConvolutionMultipler(inputShape, filterShape, biasShape,
+ outputShape, &real_multiplier) ||
+ !QuantizeMultiplierSmallerThanOne(real_multiplier, &output_multiplier,
+ &output_shift)){
+ return false;
+ }
CalculateActivationRangeUint8(activation, outputShape,
&output_activation_min,
&output_activation_max);
diff --git a/common/operations/DepthwiseConv2D.cpp b/common/operations/DepthwiseConv2D.cpp
index 9f43d56..8c4e58c 100644
--- a/common/operations/DepthwiseConv2D.cpp
+++ b/common/operations/DepthwiseConv2D.cpp
@@ -77,10 +77,13 @@
int32_t output_activation_min = 0;
int32_t output_activation_max = 0;
- GetQuantizedConvolutionMultipler(inputShape, filterShape, biasShape,
- outputShape, &real_multiplier);
- QuantizeMultiplierSmallerThanOne(real_multiplier, &output_multiplier,
- &output_shift);
+
+ if (!GetQuantizedConvolutionMultipler(inputShape, filterShape, biasShape,
+ outputShape, &real_multiplier) ||
+ !QuantizeMultiplierSmallerThanOne(real_multiplier, &output_multiplier,
+ &output_shift)) {
+ return false;
+ }
CalculateActivationRangeUint8(activation, outputShape,
&output_activation_min,
&output_activation_max);
diff --git a/common/operations/FullyConnected.cpp b/common/operations/FullyConnected.cpp
index 0cfc9c9..7f4aa6b 100644
--- a/common/operations/FullyConnected.cpp
+++ b/common/operations/FullyConnected.cpp
@@ -57,10 +57,12 @@
int32_t output_activation_min = 0;
int32_t output_activation_max = 0;
- GetQuantizedConvolutionMultipler(inputShape, weightsShape, biasShape,
- outputShape, &real_multiplier);
- QuantizeMultiplierSmallerThanOne(real_multiplier, &output_multiplier,
- &output_shift);
+ if (!GetQuantizedConvolutionMultipler(inputShape, weightsShape, biasShape,
+ outputShape, &real_multiplier) ||
+ !QuantizeMultiplierSmallerThanOne(real_multiplier, &output_multiplier,
+ &output_shift)) {
+ return false;
+ }
CalculateActivationRangeUint8(activation, outputShape,
&output_activation_min,
&output_activation_max);
