experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Fix 64-bit integer portability problems in TensorFlow kernels.

Browse Source 
Removes reliance on the assumption that tensorflow::int64 is long long. This is intended to eventually enable changing the definition to int64_t from <cstdint>.

PiperOrigin-RevId: 290872365
Change-Id: I18534aeabf153d65c3521599855f8cca279fce51
This commit is contained in:
A. Unique TensorFlower 2020-01-21 19:11:06 -08:00 committed by TensorFlower Gardener
parent 2cbb324ceb
commit a1bc56203f
14 changed files with 22 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
tensorflow/core/kernels/BUILD
View File

@ -5369,6 +5369,7 @@ tf_kernel_library(
"//tensorflow/core:lib",
"//tensorflow/core:lib_internal",
"//third_party/eigen3",
"@com_google_absl//absl/strings:str_format",
],
alwayslink = 1,
)

2
tensorflow/core/kernels/batching_util/serial_device_batch_scheduler.h
View File

@ -432,7 +432,7 @@ void SerialDeviceBatchScheduler<TaskType>::ProcessBatches() {
// the desired target pending.
in_flight_batches_limit_ +=
std::round(options_.target_pending - avg_pending);
in_flight_batches_limit_ = std::max(in_flight_batches_limit_, 1LL);
in_flight_batches_limit_ = std::max(in_flight_batches_limit_, int64{1});
in_flight_batches_limit_ =
std::min(in_flight_batches_limit_, options_.num_batch_threads);
// Add extra processing threads if necessary.

1
tensorflow/core/kernels/data/experimental/BUILD
View File

@ -433,6 +433,7 @@ tf_kernel_library(
"//tensorflow/core/grappler:graph_view",
"//tensorflow/core/kernels/data:dataset_utils",
"//tensorflow/core/profiler/lib:traceme",
"@com_google_absl//absl/strings:str_format",
"@com_google_absl//absl/time",
],
)

5
tensorflow/core/kernels/data/experimental/snapshot_dataset_op.cc
View File

@ -14,6 +14,7 @@ limitations under the License.
==============================================================================*/
#include <random>
#include "absl/strings/str_format.h"
#include "absl/time/clock.h"
#include "tensorflow/core/framework/dataset.h"
#include "tensorflow/core/framework/op_kernel.h"
@ -1420,9 +1421,7 @@ class SnapshotDatasetOp : public UnaryDatasetOpKernel {
string GetSnapshotFilename() {
mutex_lock l(mu_);
string snapshot_data_filename = io::JoinPath(
run_dir_,
absl::StrCat(strings::Printf("%08llu", next_file_index_),
".snapshot"));
run_dir_, absl::StrFormat("%08u.snapshot", next_file_index_));
next_file_index_++;
return snapshot_data_filename;
}

4
tensorflow/core/kernels/data/range_dataset_op.cc
View File

@ -68,9 +68,9 @@ class RangeDatasetOp::Dataset : public DatasetBase {
int64 Cardinality() const override {
if (step_ > 0) {
return std::max(0LL, (stop_ - start_ - 1) / step_ + 1);
return std::max(int64{0}, (stop_ - start_ - 1) / step_ + 1);
} else {
return std::max(0LL, (start_ - stop_ - 1) / -step_ + 1);
return std::max(int64{0}, (start_ - stop_ - 1) / -step_ + 1);
}
}

2
tensorflow/core/kernels/data/skip_dataset_op.cc
View File

@ -72,7 +72,7 @@ class SkipDatasetOp::Dataset : public DatasetBase {
if (n == kInfiniteCardinality || n == kUnknownCardinality) {
return n;
}
return count_ < 0 ? 0 : std::max(0LL, n - count_);
return count_ < 0 ? 0 : std::max(int64{0}, n - count_);
}
Status CheckExternalState() const override {

2
tensorflow/core/kernels/pooling_ops_common.h
View File

@ -607,7 +607,7 @@ void SpatialAvgPool(OpKernelContext* context, Tensor* output,
// so the factor 0.01 (i.e. 1/100) with a max of 10000, was chosen to limit
// the work unit cost to an operating range in which it emperically performed
// best.
const int64 work_unit_cost = std::max(int64{10000}, work_unit_size / 100LL);
const int64 work_unit_cost = std::max(int64{10000}, work_unit_size / 100);
const DeviceBase::CpuWorkerThreads& worker_threads =
*(context->device()->tensorflow_cpu_worker_threads());
Shard(worker_threads.num_threads, worker_threads.workers,

4
tensorflow/core/kernels/quantization_utils.h
View File

@ -718,8 +718,8 @@ inline void RequantizeManyInNewRangeUsingEigen<qint32, quint8>(
.unaryExpr(int64_right_shift_op<32>())) +
(input_offset_fp - output_offset_fp + rounding_delta);
auto intermediate = fp_value.unaryExpr(int64_right_shift_op<fp_shift>());
auto input_requantized = intermediate.cwiseMax(0LL)
.cwiseMin(255LL)
auto input_requantized = intermediate.cwiseMax(int64{0})
.cwiseMin(int64{255})
.template cast<int32>()
.template cast<quint8>();
output->flat<quint8>().device(device) = input_requantized;

2
tensorflow/core/kernels/resize_area_op.cc
View File

@ -275,7 +275,7 @@ class ResizeAreaOp : public OpKernel {
private:
static EIGEN_ALWAYS_INLINE int64 Bound(int64 val, int64 limit) {
return std::min(limit - 1ll, std::max(int64{0}, val));
return std::min(limit - 1, std::max(int64{0}, val));
}
bool align_corners_;

2
tensorflow/core/kernels/resize_bicubic_op.cc
View File

@ -66,7 +66,7 @@ const float* GetCoeffsTable(const bool use_keys_cubic) {
}
inline int64 Bound(int64 val, int64 limit) {
return std::min(limit - 1ll, std::max(int64{0}, val));
return std::min(limit - 1, std::max(int64{0}, val));
}
struct WeightsAndIndices {

2
tensorflow/core/kernels/resize_bicubic_op_test.cc
View File

@ -81,7 +81,7 @@ class ResizeBicubicOpTest : public OpsTestBase {
// Used in the baseline implementation
inline int64 Bound(int64 val, int64 limit) {
return std::min(limit - 1ll, std::max(int64{0}, val));
return std::min(limit - 1, std::max(int64{0}, val));
}
// Used in the baseline implementation

5
tensorflow/core/kernels/sdca_ops.cc
View File

@ -18,6 +18,7 @@ limitations under the License.
#define EIGEN_USE_THREADS
#include <stdint.h>
#include <atomic>
#include <limits>
#include <memory>
@ -25,6 +26,7 @@ limitations under the License.
#include <string>
#include <vector>
#include "absl/strings/str_format.h"
#include "third_party/eigen3/unsupported/Eigen/CXX11/Tensor"
#include "tensorflow/core/framework/device_base.h"
#include "tensorflow/core/framework/kernel_def_builder.h"
@ -47,7 +49,6 @@ limitations under the License.
#include "tensorflow/core/lib/core/status.h"
#include "tensorflow/core/lib/core/stringpiece.h"
#include "tensorflow/core/lib/gtl/inlined_vector.h"
#include "tensorflow/core/lib/strings/stringprintf.h"
#include "tensorflow/core/platform/fingerprint.h"
#include "tensorflow/core/platform/macros.h"
#include "tensorflow/core/platform/mutex.h"
@ -103,7 +104,7 @@ struct ComputeOptions {
static_cast<int64>(num_dense_features) <=
std::numeric_limits<int>::max(),
errors::InvalidArgument(
strings::Printf("Too many feature groups: %lld > %d",
absl::StrFormat("Too many feature groups: %d > %d",
static_cast<int64>(num_sparse_features) +
static_cast<int64>(num_dense_features),
std::numeric_limits<int>::max())));

4
tensorflow/core/kernels/sparse_reduce_op.cc
View File

@ -202,9 +202,9 @@ class SparseReduceOp : public OpKernel {
}
auto CoordinatesToFlatIndex = [](ArraySlice<int64> coords,
ArraySlice<int64> strides) {
ArraySlice<int64> strides) -> int64 {
if (strides.empty()) { // Reduce all.
return 0LL;
return 0;
}
CHECK_EQ(coords.size(), strides.size());
int64 idx = 0;

6
tensorflow/core/kernels/tensor_flag_utils_test.cc
View File

@ -308,15 +308,15 @@ TEST(SparseUtils, FindConfigValueForKey) {
TEST(SparseUtils, GetLinearBucket) {
EXPECT_EQ(11, GetLinearBucket(11, 5));
EXPECT_EQ(11, GetLinearBucket(12, 5));
EXPECT_EQ(1, GetLinearBucket(4ll, 5ll));
EXPECT_EQ(1, GetLinearBucket(int64{4}, int64{5}));
}
TEST(SparseUtils, GetPowerBucket) {
EXPECT_EQ(6, GetPowerBucket(11, 5));
EXPECT_EQ(6, GetPowerBucket(12, 5));
EXPECT_EQ(1332, GetPowerBucket(1335, 11));
EXPECT_EQ(5, GetPowerBucket(5ll, 4ll));
EXPECT_EQ(1, GetPowerBucket(4ll, 1ll));
EXPECT_EQ(5, GetPowerBucket(int64{5}, int64{4}));
EXPECT_EQ(1, GetPowerBucket(int64{4}, int64{1}));
}
} // namespace