tensorflow/compiler/jit/kernels/xla_launch_op.cc

/* Copyright 2017 The TensorFlow Authors. All Rights Reserved.

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.
==============================================================================*/

#include "tensorflow/compiler/jit/kernels/xla_launch_op.h"

#include "tensorflow/compiler/jit/defs.h"
#include "tensorflow/compiler/jit/xla_launch_util.h"
#include "tensorflow/compiler/tf2xla/shape_util.h"
#include "tensorflow/compiler/tf2xla/tf2xla_util.h"
#include "tensorflow/compiler/tf2xla/xla_compiler.h"
#include "tensorflow/compiler/tf2xla/xla_op_registry.h"
#include "tensorflow/compiler/xla/client/client_library.h"
#include "tensorflow/compiler/xla/client/local_client.h"
#include "tensorflow/compiler/xla/statusor.h"
#include "tensorflow/core/common_runtime/dma_helper.h"
#include "tensorflow/core/common_runtime/function.h"
#include "tensorflow/core/framework/allocator.h"
#include "tensorflow/core/framework/node_def_util.h"
#include "tensorflow/core/framework/op.h"
#include "tensorflow/core/framework/op_kernel.h"
#include "tensorflow/core/framework/tensor.h"
#include "tensorflow/core/framework/types.h"
#include "tensorflow/core/kernels/variable_ops.h"
#include "tensorflow/core/platform/env.h"
#include "tensorflow/core/platform/stream_executor_no_cuda.h"
#include "tensorflow/core/util/stream_executor_util.h"

namespace tensorflow {

XlaLocalLaunchBase::XlaLocalLaunchBase(OpKernelConstruction* ctx,
                                       const std::vector<int>& constants,
                                       const std::vector<int>& resources,
                                       const NameAttrList& function)
    : OpKernel(ctx),
      constants_(constants),
      resources_(resources),
      device_type_(ctx->device_type()),
      function_(function) {
  if (device_type_ == DeviceType(DEVICE_CPU)) {
    platform_id_ = se::host::kHostPlatformId;
  } else if (device_type_ == DeviceType(DEVICE_GPU)) {
    platform_id_ = ctx->device()
                       ->tensorflow_gpu_device_info()
                       ->stream->parent()
                       ->platform()
                       ->id();
  } else if (XlaDevice::GetMetadata(ctx, &xla_device_metadata_).ok()) {
    use_multiple_streams_ = xla_device_metadata_->UseMultipleStreams();
    platform_id_ = xla_device_metadata_->platform()->id();
  }
}

Status XlaLocalLaunchBase::BuildCompilationCache(OpKernelContext* ctx,
                                                 XlaCompilationCache** cache) {
  if (xla_device_metadata_) {
    *cache = new XlaCompilationCache(xla_device_metadata_->client(),
                                     xla_device_metadata_->jit_device_type());
    return Status::OK();
  }

  auto platform = se::MultiPlatformManager::PlatformWithId(platform_id_);
  if (!platform.ok()) {
    return platform.status();
  }
  xla::LocalClientOptions client_options;
  client_options.set_platform(platform.ValueOrDie());
  client_options.set_intra_op_parallelism_threads(
      ctx->device()->tensorflow_cpu_worker_threads()->num_threads);
  auto client = xla::ClientLibrary::GetOrCreateLocalClient(client_options);
  if (!client.ok()) {
    return client.status();
  }
  const XlaOpRegistry::DeviceRegistration* registration;
  if (!XlaOpRegistry::GetCompilationDevice(device_type_.type(),
                                           &registration)) {
    return errors::InvalidArgument("No JIT device registered for ",
                                   device_type_.type());
  }
  *cache = new XlaCompilationCache(
      client.ValueOrDie(), DeviceType(registration->compilation_device_name));
  return Status::OK();
}

void XlaLocalLaunchBase::Compute(OpKernelContext* ctx) {
  VLOG(1) << "XlaLocalLaunchOpBase::Compute "
          << Canonicalize(function_.name(), AttrSlice(&function_.attr()));
  // We store information about the JIT-compiled XLA computation
  // in the ResourceMgr.
  ResourceMgr* rm = ctx->resource_manager();
  OP_REQUIRES(ctx, rm, errors::Internal("No resource manager."));

  se::Stream* stream =
      ctx->op_device_context() ? ctx->op_device_context()->stream() : nullptr;

  XlaCompilationCache* cache;
  OP_REQUIRES_OK(ctx, rm->LookupOrCreate<XlaCompilationCache>(
                          rm->default_container(), "xla_cache", &cache,
                          [this, ctx](XlaCompilationCache** cache) {
                            return BuildCompilationCache(ctx, cache);
                          }));
  // Hold the reference to the JIT during evaluation. (We could probably
  // free it sooner because the ResourceMgr will retain a reference, but
  // this is more obviously correct.)
  core::ScopedUnref cache_ref(cache);

  std::map<int, OptionalTensor> variables =
      SnapshotResourceVariables(ctx, resources_);

  xla::LocalClient* client = static_cast<xla::LocalClient*>(cache->client());

  XlaAllocator local_xla_allocator(client->backend().platform(),
                                   ctx->device()->GetAllocator({}));
  xla::DeviceMemoryAllocator* xla_allocator;
  // If we are on an XlaDevice, use the underlying XLA platform's allocator
  // directly. We could use the StreamExecutor's allocator which may
  // theoretically be more correct, but XLA returns a nice OOM message in a
  // Status and StreamExecutor does not.
  //
  // Importantly we can't use ctx->device()->GetAllocator() as the allocator
  // (which local_xla_allocator above uses) as on an XlaDevice, this is a
  // dummy allocator that returns XlaTensor objects. The XlaCompiler needs a
  // real allocator to allocate real buffers.
  if (xla_device_metadata_) {
    xla_allocator = client->backend().memory_allocator();
  } else {
    xla_allocator = &local_xla_allocator;
  }

  XlaCompiler::Options options;
  options.client = client;
  if (ctx->op_device_context() != nullptr) {
    options.device_ordinal =
        ctx->op_device_context()->stream()->parent()->device_ordinal();
  }
  options.device_type = cache->device_type();
  options.flib_def = ctx->function_library()->GetFunctionLibraryDefinition();
  options.graph_def_version = ctx->function_library()->graph_def_version();
  options.allow_cpu_custom_calls = (platform_id_ == se::host::kHostPlatformId);
  options.device_allocator = xla_allocator;
  if (xla_device_metadata_) {
    options.shape_representation_fn =
        xla_device_metadata_->shape_representation_fn();
  }

  const XlaCompiler::CompilationResult* kernel;
  xla::LocalExecutable* executable;

  std::map<int, Tensor> constant_args;
  for (int i : constants_) {
    constant_args.insert({i, ctx->input(i)});
  }
  XlaCompiler::CompileOptions compile_options;
  compile_options.is_entry_computation = true;
  // If we resolve constants we never emit them on the device, meaning that if
  // they are needed by a following computation the host has to transfer
  // them. Not resolving constants is expected to be faster than resolving
  // constants.
  compile_options.resolve_compile_time_constants = true;
  // Optimization: where possible, have the computation return a naked array
  // rather than a one-element tuple.
  compile_options.always_return_tuple = false;

  OP_REQUIRES_OK(
      ctx, cache->Compile(options, function_, constant_args, variables, ctx,
                          &kernel, &executable, compile_options));

  VLOG(1) << "Executing XLA Computation...";

  XlaComputationLaunchContext launch_context(
      client, xla_allocator,
      /*allocate_xla_tensors=*/xla_device_metadata_ != nullptr,
      use_multiple_streams_);
  launch_context.PopulateInputs(ctx, kernel, variables);

  // Execute the computation.
  VLOG(2) << "Executing computation.";
  xla::ExecutableRunOptions run_options;
  run_options.set_stream(stream);
  run_options.set_allocator(xla_allocator);
  run_options.set_intra_op_thread_pool(&ctx->eigen_cpu_device());
  run_options.set_rng_seed(GetXLARandomSeed());
  Env* env = Env::Default();
  auto start_time = env->NowMicros();

  auto run_result = executable->Run(launch_context.arguments(), run_options);
  OP_REQUIRES(ctx, run_result.ok(), run_result.status());

  auto elapsed = env->NowMicros() - start_time;
  VLOG(2) << "Elapsed time: " << elapsed << "us";

  OP_REQUIRES_OK(ctx, launch_context.PopulateOutputs(
                          ctx, kernel, run_result.ConsumeValueOrDie()));
  VLOG(1) << "Done";
}

namespace {

// OP_REQUIRES_OK_RETURN is the same as OP_REQUIRES_OK except that
// in error case, it returns RET instead of void.
#define OP_REQUIRES_OK_RETURN(CTX, RET, ...)                \
  do {                                                      \
    ::tensorflow::Status _s(__VA_ARGS__);                   \
    if (!TF_PREDICT_TRUE(_s.ok())) {                        \
      (CTX)->CtxFailureWithWarning(__FILE__, __LINE__, _s); \
      return RET;                                           \
    }                                                       \
  } while (0)

// Helper static functions to construct parameters for
// XlaLocalLaunchBase constructor from OpKernelConstruction.
std::vector<int> ConstantsVector(OpKernelConstruction* ctx) {
  DataTypeVector constant_types;
  OP_REQUIRES_OK_RETURN(ctx, std::vector<int>(),
                        ctx->GetAttr("Tconstants", &constant_types));
  std::vector<int> constants(constant_types.size());
  std::iota(constants.begin(), constants.end(), 0);
  return constants;
}

std::vector<int> ResourcesVector(OpKernelConstruction* ctx) {
  DataTypeVector constant_types;
  OP_REQUIRES_OK_RETURN(ctx, std::vector<int>(),
                        ctx->GetAttr("Tconstants", &constant_types));

  DataTypeVector arg_types;
  OP_REQUIRES_OK_RETURN(ctx, std::vector<int>(),
                        ctx->GetAttr("Targs", &arg_types));

  int num_resources;
  OP_REQUIRES_OK_RETURN(ctx, std::vector<int>(),
                        ctx->GetAttr("Nresources", &num_resources));

  std::vector<int> resources(num_resources);
  std::iota(resources.begin(), resources.end(),
            constant_types.size() + arg_types.size());
  return resources;
}

NameAttrList FunctionAttr(OpKernelConstruction* ctx) {
  const NameAttrList* func;
  OP_REQUIRES_OK_RETURN(ctx, NameAttrList(), ctx->GetAttr("function", &func));
  return *func;
}

#undef OP_REQUIRES_OK_RETURN
}  // namespace

XlaLocalLaunchOp::XlaLocalLaunchOp(OpKernelConstruction* ctx)
    : XlaLocalLaunchBase(ctx, ConstantsVector(ctx), ResourcesVector(ctx),
                         FunctionAttr(ctx)) {}

XlaLocalLaunchOp::~XlaLocalLaunchOp() {
  VLOG(1) << "XlaLocalLaunchOp destroyed";
}

REGISTER_KERNEL_BUILDER(Name("XlaLaunch").Device(DEVICE_CPU), XlaLocalLaunchOp);

REGISTER_KERNEL_BUILDER(Name("XlaLaunch")
                            .Device(DEVICE_GPU)
                            .HostMemory("constants")
                            .HostMemory("resources"),
                        XlaLocalLaunchOp);

}  // namespace tensorflow