GitLab

* Implemented deferred initialization via `device='meta'` option for te.Linear and added new PyTorch example to demonstrate its use with FullyShardedDataParallel execution.
Signed-off-by: Alp Dener <adener@nvidia.com>

* correcting Float8Tensor initialization and fixing linting errors
Signed-off-by: Alp Dener <adener@nvidia.com>

* removed duplicate code from upstream rebase, local tests passing
Signed-off-by: Alp Dener <adener@nvidia.com>

* improved comments/documentation for FSDP example
Signed-off-by: Alp Dener <adener@nvidia.com>

* converted reset_parameters() into a base module function
Signed-off-by: Alp Dener <adener@nvidia.com>

* fixed Float8Tensor creation with deferred init, all tests passing locally
Signed-off-by: Alp Dener <adener@nvidia.com>

* extended deferred initialization to all TE modules
Signed-off-by: Alp Dener <adener@nvidia.com>

* fixed linting errors
Signed-off-by: Alp Dener <adener@nvidia.com>

* removed unnecessary reference to the parent module of parameter, added clarifying comments in parameter reset
Signed-off-by: Alp Dener <adener@nvidia.com>

---------
Signed-off-by: Alp Dener <adener@nvidia.com>

* Refactor parameter split in Linear module

Remove module state from noop_cat. Support arbitrary names in parameter split. Handle tensor parallelism.
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Make noop_cat a standalone operation
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Update parameter splits in LayerNormLinear
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Debug case without bias

Fix pylint complaints.
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Remove unused import
Signed-off-by: Tim Moon <tmoon@nvidia.com>

---------
Signed-off-by: Tim Moon <tmoon@nvidia.com>

Signed-off-by: Przemek Tredak <ptredak@nvidia.com>

* Linear and LayerNormLinear weight and bias buffer cleanup at the end of init when there is no parameter split
Signed-off-by: Alp Dener <adener@nvidia.com>

* fixed typo in tensor name
Signed-off-by: Alp Dener <adener@nvidia.com>

* fixed typo in tensor name
Signed-off-by: Alp Dener <adener@nvidia.com>

---------
Signed-off-by: Alp Dener <adener@nvidia.com>

* Disable dynamo for Fused Attention
Signed-off-by: Przemek Tredak <ptredak@nvidia.com>

* Added test
Signed-off-by: Przemek Tredak <ptredak@nvidia.com>

---------
Signed-off-by: Przemek Tredak <ptredak@nvidia.com>
Co-authored-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>

[PyTorch] fix amax calculate during fp8 calibration
Signed-off-by: ladyrick <ladyrick@qq.com>

wgrad should be zero'ed out if a weight parameter is shared among multiple layers
Signed-off-by: Deepak Narayanan <dnarayanan@nvidia.com>

* [PyTorch] Linear: fix computation for wgrad if sequence_parallel=True
Signed-off-by: Markus Schnoes <markus.schnoes@gmx.de>

* Remove buggy gather_along_last_dim
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* [PyTorch] Linear: fix line length
Signed-off-by: Markus Schnoes <markus.schnoes@gmx.de>

* Simplify logic for saving input tensor for Linear backward
Signed-off-by: Tim Moon <tmoon@nvidia.com>

---------
Signed-off-by: Markus Schnoes <markus.schnoes@gmx.de>
Signed-off-by: Tim Moon <tmoon@nvidia.com>
Co-authored-by: Tim Moon <tmoon@nvidia.com>

* Delay caching of transposes
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Review comment
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

---------
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

Improve PyTorch memory usage
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Make user buffer name configurable
Signed-off-by: Sangkug Lym <slym@nvidia.com>

* Apply suggestions from code review
Signed-off-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>

* Fix duplicate argument
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Fix autograd
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

---------
Signed-off-by: Sangkug Lym <slym@nvidia.com>
Signed-off-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Co-authored-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Returning an empty tensor of param dtype for wgrad
Signed-off-by: Selvaraj Anandaraj <selvaraja@computelab-frontend-4-ub22.nvidia.com>

* lint
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

---------
Signed-off-by: Selvaraj Anandaraj <selvaraja@computelab-frontend-4-ub22.nvidia.com>
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Co-authored-by: Selvaraj Anandaraj <selvaraja@computelab-frontend-4-ub22.nvidia.com>
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Experimental FP8 tensor
Co-authored-by: Tim Moon <tmoon@nvidia.com>
Co-authored-by: Sudhakar Singh <sudhakars@nvidia.com>
Co-authored-by: Przemyslaw Tredak <ptrendx@gmail.com>
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Add fp8 tensor to ci test
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* review comments and tests
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Minor changes
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Default to FP8 usage
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Fix docs
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Naming changes
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* minor fix
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Fix transpose caching
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Debug transpose caching

Handle case where transpose cache is updated externally.
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Rename FP8GlobalStateManager.with_fp8_parameters
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* remove Float8Tensor from import API
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Avoid caching FP8 transposes if not required
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Fix import error in FP8 tensor tests
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Fix tranpose caching and checkpointing bug
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Improve caching and fix distopt case
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Update transformer_engine/pytorch/float8_tensor.py
Signed-off-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>

* Remove recursive logic
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Fix cache reset bug
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Store FP8 attributes in dict

Easier for multiple tensors to share, e.g. detached tensors.
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Make sure scale_inv is 1D tensor
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Make sure scale_inv is 1D tensor
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Fixes and detach recipe
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Set default fp8 data type
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

---------
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Signed-off-by: Tim Moon <tmoon@nvidia.com>
Signed-off-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Co-authored-by: Sudhakar Singh <sudhakars@nvidia.com>
Co-authored-by: Przemyslaw Tredak <ptrendx@gmail.com>

Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Initial commit
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Repro for RS output mismatch with Single GEMM + Split pipelined RS
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* minor changes for AG->GEMM pipelined overlap
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Add Atomic Gemm cublasApi attributes and initial implementation of AG->Atomic GEMM
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* AtomicGemm+RS functional with workaround
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* add amax update to layernorm_linear for FP8 unit test accuracy
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Enable reducescatter2_userbuff_strided variants
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Bug fix
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* AG+AtomicGemm overlap functional but gemm doesnt overlap with comm
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Add userbuffers_sendrecv kernel variants
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* TransformerLayer API changes to enable AtomicGemm+RS overlap
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Code cleanup
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Code cleanup2
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* [UB] AllGather Atomic GEMM overlap using userbuffer_sendrecv kernels
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Code cleanup + bug fix for multiatomic sendrecv kernel
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* cleanup
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Bug fixes
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* [UB] Add shuffling for better AG AtomicGEMM overlap
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Bug fix for AG AtomicGemm overlap
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Bug fix for multiAtomicAG and singleAtomicAG
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Use chunk_i+1 as recv_chunk for multiatomic_AG with shuffling
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Launch AtomicGEMM after first-chunk AG
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Rebase to main
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Add FP8 ReduceScatter kernels, AtomicGEMM+FP8 RS not functional
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Revert "Add FP8 ReduceScatter kernels, AtomicGEMM+FP8 RS not functional"

This reverts commit 80a47a76355440cd5fb4314c96fe9fda632d87f9.
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Add support for NVLS-MC and FP8 Reduce Scatter
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Bug fix
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Atomic and Multiatomic FP8 RS functional
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Remove debug print
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* UB comm initialization hang fix
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Code cleanup
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Create new GEMM API for Atomic GEMM
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* CI ready
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* more fixes
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* license
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Bug fix
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Revert NVLS-MC
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Check cu* versions for running atomic gemms
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* lint
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* fixes
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Cleanup
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>

* Add experimental warning
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Better wording
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Add warning to c api
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Fix wording
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

---------
Signed-off-by: Vasudevan Rengasamy <vrengasamy@nvidia.com>
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* wgrad-accumulation patch for legacy megatron-LM
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* fix
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Fix
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Fix fused attention tests
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

---------
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

Change deprecation warnings
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* [PyTorch] Implement GQA based on fused q, k, v projection. Additionally fixes #392
Signed-off-by: Markus Schnoes <markus.schnoes@gmx.de>

* [PyTorch] Extend parameters_split option in Linear and LayerNormLinear to support splitting with different sizes as required by unfused GQA.
Signed-off-by: Markus Schnoes <markus.schnoes@gmx.de>

* fix parameters split
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* fix noop cat to bypass torch.cat and support uneven split
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* fix unit tests
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* fix torch.split args
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* fix cuda graph due to noop_cat
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* fix lint
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* remove the use of enumerate when possible
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

* fix strides in SplitAlongDim
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>

---------
Signed-off-by: Markus Schnoes <markus.schnoes@gmx.de>
Signed-off-by: Charlene Yang <8636796+cyanguwa@users.noreply.github.com>
Co-authored-by: Markus Schnoes <markus.schnoes@gmx.de>

Fix for sequence-parallel
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* CPU initialization
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Fix default value
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Change API and add to RMSNorm
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

---------
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* fix linear sequence parallel when weight is frozen
Signed-off-by: jasonwan <jasonwan@nvidia.com>

* Extend fix to all APIs
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

---------
Signed-off-by: jasonwan <jasonwan@nvidia.com>
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Disable TorchDynamo optimizations in PyTorch modules
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Add test for Torch Dynamo
Signed-off-by: Tim Moon <tmoon@nvidia.com>

* Add torch.dynamo test to qa
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Skip torch.compile test for <v2.0
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

---------
Signed-off-by: Tim Moon <tmoon@nvidia.com>
Signed-off-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Deprecate unused APIs
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* review comments
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Review
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

---------
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

Remove extra buffers
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Better dimension assert for FP8
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* line
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

---------
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Get default dtype from pytorch
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Review comments
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

---------
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* extend fp8 weight placeholders logic for Linear, LNLinear, LNMLP
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* Update transformer_engine/pytorch/module/base.py
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* Update transformer_engine/pytorch/module/base.py
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* Update transformer_engine/pytorch/module/base.py
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* Update transformer_engine/pytorch/module/base.py
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* Update transformer_engine/pytorch/module/base.py
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* Update transformer_engine/pytorch/module/layernorm_linear.py
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* Update transformer_engine/pytorch/module/layernorm_mlp.py
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* Update transformer_engine/pytorch/module/linear.py
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* Update linear.py
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* Update layernorm_linear.py
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* Update layernorm_mlp.py
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>

* lint
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

---------
Signed-off-by: Sudhakar Singh <sudhakars@nvidia.com>
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Clearer error messages for dtype and shape assertions
Signed-off-by: Carlos Mocholí <carlossmocholi@gmail.com>

* Update transformer_engine/pytorch/utils.py
Co-authored-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>
Signed-off-by: Carlos Mocholí <carlossmocholi@gmail.com>

* Update transformer_engine/pytorch/utils.py
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Signed-off-by: Carlos Mocholí <carlossmocholi@gmail.com>

---------
Signed-off-by: Carlos Mocholí <carlossmocholi@gmail.com>
Co-authored-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Initial refactor
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* refactor attention out of transformer.py
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Fix ONNX export
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* linting
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

---------
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>