* Move allocators to SessionState so they're decoupled from ExecutionProviders
- when looking up an allocator it's based on OrtMemoryInfo not the EP so SessionState is a more natural place for that infromation to be stored
- add device based lookup
- simplifies logic for copying feeds/fetches across devices
Cleanup SessionState and SessionStateInitializer
- provide more things to SessionState at construction time so we don't construct and instance and immediately after call a bunch of setters
- simplify SessionStateInitializer
- reduced down to FinalizeSessionState method
Provide alternative std::mutex implementation on Windows. OrtMutex is no longer an alias of std::mutex.
We do it because:
1. This new thing is faster and much much simpler.
2. Static constructors are considered harmful. We should avoid such thing as possible as we can.
Rework TensorSeq in a manner consistent with Tensor and SparseTensor
in terms of type system setup.
Reduce templating. Introduce helpers to ensure the same
data type.
Make OrtValue __dtor not virtual.
Introduce ContainerChecker
* Add CUDA If operator.
Uses CPU operator for implementation.
By adding a CUDA version the inputs/outputs (with the exception of the 'cond' input) stay on GPU, and no other logic is required to avoid a copy to CPU across the control flow node.
* dump cuda tensor
* move data_type definition
* Dump cuda tensors for cuda build only.
Output tensor location (if it is not in CPU or pinned)
* update for cuda build
* Update for code review feedback
* update for CR feedback
* use data transfer manager for tensor copy
* Introduce execution mode for clarity and extensibility; Change Python APIs accordingly; Replace DisableSequentialExecution API with EnableParallelExecution for clarity.
* Fix cuda build
* Modify the test slightly
* Make C and C# APIs consistent with Python.
Remove gsl subodule and replace with a local copy of gsl-lite
Refactor for onnxruntime::make_unique
gsl::span size and index are now size_t
Remove lambda auto argument type detection.
Remove constexpr from fail_fast in gsl due to Linux not being happy.
Comment out std::stream support due to MacOS std lib broken.
Move make_unique into include/core/common so it is accessible for server builds.
Relax requirements for onnxruntime/test/providers/cpu/ml/write_scores_test.cc
due to x86 build.
Add ONNXRUNTIME_ROOT to Server Lib includes so gsl is recognized
* Don't return shape for non-const initializer in InferenceContextImpl::getInputType
Don't return initializer for non-const initializer in InferenceContextImpl::getInputData
Update graph_utils to support these scenarios
- fix GetConstantInitializer to make sure a name is for an outer scope value before checking a parent graph, as local name could shadow an outer scope initializer.
* Rework the feed/fetch copy setup so that it can be calculated upfront by the control flow nodes. Also simplifies how it all works.
Update the control flow nodes to do the calculation prior to graph execution.
Description: make default CPU allocator to use MLAS preferred alignment
Motivation and Context
This is needed for C API to have an aligned default CPU allocator, the same as the one in CPU provider
* remove memory copy between CUDA and TRT
* add info to RegisterExecutionProvider input
* use new IDeviceAllocator for trt allocator
* remove SetDefaultInputsMemoryType from TRT EP
* remove onnx-tensorrt 5.0
* add submodule onnx-tensorrt branch 5.1
* remove redundancy
* Update transformer_memcpy.cc
* Update tensorrt_execution_provider.cc
* switch to TensorRT 5.1.5.0
* update python binding
* disable failed test case on TensorRT
* Update activation_op_test.cc
* upgrade to TensorRT container 19.06
* update according to feedback
* add comments
* remove tensorrt allocator and use cuda(gpu) allocator
* update onnx-tensorrt submodule
* change ci build cuda directory name
Address #1155
Add debug helper methods to be able to dump input name and shape information for node inputs, and the data from node outputs.
As the input data comes from graph inputs, initializers or node outputs we don't dump it.
Must be manually enabled by building with '--cmake_extra_defines onnxruntime_DEBUG_NODE_INPUTS_OUTPUTS=ON'
* updated cmake files for trt
* added trt execution provider
* added trt basic test
* removed trt_path action attribute
* Add files via upload
* Update build.py
* Update trt_allocator.h
* fixed issues found by reviewers
* changed cast operator
* added comment for custom kernel implementation
* changed auto to auto&
* changed to function compile APIs for TRT execution provider
* changed to function compile APIs for TRT execution provider
* added new DType DInt64
* adapted to the changes of onnxruntime_c_api
* removed trt kernel (use function compile instead)
* updated onnx-tensorrt submodule
* set default memory type to TRT fused kernel
* resolve merge conflict
* fixed the issue that USE_CUDA conflicts with USE_TRT
* construct graph by adding nodes in topological order
* made changes for Windows
* change buffers type
* bypass HasImplementationOf check for TRT XP because TRT kernel is not registered
* added domain to version info in rebuilt model proto
* added trt to test option list
* added DomainToVersionMap() to GraphViewer
* removed Copy()
* fixed broken code
* format the code to clang format
* used local reference to the frequently used values
* fixed a couple of issues according to reviewers feedback
* fixed a couple of issues according to reviewers feedback
* added python binding for TRT and enable use_cuda when use_trt is on
* fixed a redefinition issue
* changed shared_ptr to unique_ptr on trt engines, and made a few changes required by reviewers
* enabled trtexecution provider for unit tests
* renamed trt to tensorrt
* added tesorrt to python binding
* update submodule onnx and onnx-tensorrt
* made a couple of minor changes based on reviewer's feedback
* added CUDA_CHECK
* removed test code
* fixed broken code after merge
* updated onnx-tensorrt submodule
* added post processing to align trt inputs/outputs with graph inputs/outputs
* updated onnx submodule
* added CUDA fallback for TensorRT and fixed TensorRT cmake issue
* added ci pipeline for tensorrt and removed some redundent code from trt xp
* fixed syntax issue
* updated onnx-tensorrt submodule
* fix trt build problem by: (#602)
1. Add additional /wd for debug build
2. Add io.h for additional targets
3. Bring back mb version of getopt
* Update install_ubuntu.sh
* Update linux-gpu-tensorrt-ci-pipeline.yml
* Update linux-gpu-tensorrt-ci-pipeline.yml
* Update run_build.sh
* Update run_build.sh
* Update run_build.sh
* Update run_build.sh
* fixed the issue that GetKernelRegistry returns nullptr
* merged master to this branch
* moved some data types to private
* fixed tensorrt CI pipeline issue
* customized test data for TensorRT pipeline
* added onnx-tensorrt in json file and fixed an issue in ci script
* added comments
1. Support the new external data extension in ONNX 1.4 onnx/onnx#678
2. Enable onnxruntime_perf_test in Mac Build
3. move path_lib.h from onnx_test_runner source dir to onnxruntime_framework
4. Enable memory planner for string tensors
5. Make memory planner always enabled, to simplify model loading logic
6. Delete some duplicated code between onnxruntime_perf_test and onnx_test_runner
7. Delete win_getopt_mb lib.
8. Remove the dependency on Pathcch lib, which is only available on Windows 8 and newer.
* Various optimizations to reduce the setup and execution cost.
Cache information about the feeds and fetches, and any device copies required to execute the graph so we minimize checking for later calls to ExecuteGraph using the same input/output.
- enable use of caching in Loop and Scan
- make use of caching optional for InferenceSession::Run
- handle calls to Run with different feeds and fetches to support scenarios where there may be a truncated sequence in some calls
Take the feed names and MLValue instances as vectors so the order is deterministic.
Add unit tests
Update onnxruntime_perf_test to enable caching.
* Couple of tweaks.
Fix shared library unit test failure.
Attempt to workaround MacOS build failure due to VC++ bug around including reaching scope values in a lambda automatically.
* Rework order of init in Run so we get nice error messages about invalid feed/output names.
* Refine logic around copying MLValue using execution provider so common code can be used. Simplify the logic due to this change.
Split the paths for executing with/without cached info so we can be more const correct with how FeedsFetchesManager is passed in. This makes it clearer when a shared instance can be used due to it being const.
Cache the FeedsFetchesManager instances in the control flow nodes. They can be re-used across calls to Compute.
* Removed unused local variable to fix some builds.
* Fix build issue by cleaning up some more unused params.
* Check names when using cache entry from SessionState. Add unit test.
* Separate out the NodeArg index information from ExecutionFrame so it is only calculated once.
* Skip copy to/from device if only CPU execution provider is registered.
Cleanups.
* Address PR comments.
Clean up a few areas.
* Fix Linux build error
* Add the ability to use a custom allocator for fetches.
Allows control flow nodes to forward the allocation to the control flow op and avoid an unnecessary copy when the subgraph output has a symbolic dimension.
Update Scan and If to use custom allocators when applicable.
* Remove unnecessary forward declaration
* Fix Mac build warnings
Applies to all public headers and macros, plus many internal ones. There are still some internal things with OnnxRuntime in the name, but this fixes all public functions & macros.
