GitLab

* Add rng_state output for cross fused attention
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add rng_state and output for the flash attention backward
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add bias for the jax cross attn API
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Fix a minor bug
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add bias in the backward for the arbitrary fused attn backend
Signed-off-by: Reese Wang <rewang@nvidia.com>

---------
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Refactor sharding.py for the further custom_partitioning migration
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Migrating both FWD and BWD of LayerNorm/RMSNorm from xmap to custom_partitioning.
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Migrating both FWD and BWD of all kinds of softmax from xmap to custom_partitioning.
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Fix the wrong order of parameters to LN/RMSN bwd in ln_mlp_fp8.
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* WAR to LN/RMSN_fp8 before migrating to CP.
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Fix the wrong order of parameters of bwd of LN/RMSN_fp8.
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Following review feedback to modify
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Force the hidden dim in Norm ops to no sharding and add warning msg.
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Reuse fwd_rule in VJP functions
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Migrating both FWD and BWD of self-fused-attn from xmap to custom_partitioning.
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Migrating both FWD and BWD of cross-fused-attn from xmap to custom_partitioning.
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* add gelu and dgelu.
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Reuse fwd_rule in VJP functions for attentions
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Apply native FP8 Dtypes to fp8.py
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Migrating cast_and_transpose from xmap to custom_partitioning
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Migrating transpose from xmap to custom_partitioning
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Apply XLA pattern match to perform FP8 GEMM.
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* migrate layernorm_fp8 to custom_partitioning.
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Unify code style
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Extend supported of Transpose with FP8
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Implementing layernorm_fp8_dot based on migrated custom calls.
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Renaming variables and publish NVTE_FP8_COLLECTION_NAME
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Replace Q/DQ custom calls with native XLA implementations
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* migrate gelu_fp to custom_partitioning.
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Miner fix
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Support custom calls with mutli-dims
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Support gerneral dot indices in _fp8_dot_impl
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Implementing layernrom_geglu_fp8_mlp
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Remove GEMM custom calls
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Remove xmap related code
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Fix typo and add query-function to FP8MetaPackage
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Fix some bugs of custom calls
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Fix CT's bugs
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Update UTs/eaxmaples to adapt to the API changes.
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Unify kernel initilization in MLP.
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Modifing with code review's feedback
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Update README and Add deprecating warning to *ShardingType
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>

* Canonicalize the dtype
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Adding assertion for non-supported batch dims.
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Adding doc/examples to _multidim_transpose
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Set FP8 meta as WeightHParamsCollection.OVERWRITE_WITH_GRADIENT in Praxis modules.
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Set FP8 meta as WeightHParamsCollection.OVERWRITE_WITH_GRADIENT in Praxis modules.
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Apply dtype-based rtol/atol to UTs
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Deprecate QKV_INTERLEAVED enum
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Skip test_distributed_custom_ops.py
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Fix the wrong sharding of bias in SelfAttn
Signed-off-by: Ming Huang <mingh@nvidia.com>

* WAR to fix the wrong cu_seqlen of MHA when DP/FSDP enabled
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Adding distributed ops unit-tests
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Adding license to test_distributed_*
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Follow review feedback to modify
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Use total bytes involved in collective ops as criteria.
Signed-off-by: Ming Huang <mingh@nvidia.com>

---------
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Ming-Xu Huang <mingh@nvidia.com>
Co-authored-by: Donglin Yang <dongliny@nvidia.com>

[C/JAX] Support more mask types for the arbitrary seqlen kernels and minor changes of JAX bias (#469)

* Move bias to float32
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Enable varlen
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Increase neg infinity abs values
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Enable varlen tests
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Remove unnecessary code
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Fix lint
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Support variable sequence length after cuDNN 8.9.6
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Use unique_ptr instead of shared_ptr
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add a new mask type: PADDING_CAUSAL_MASK
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Support flash padding mask after 8.9.6
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Enhance the Max512 handling for causal masking and add the related tests
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Update the fused attn support lists
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Remove padding_aware from the caching
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Fix libtransformer.so issue
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Reduce the pad ratio tests
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Fix a bug with cuDNN 8.9.5
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Release backend resource after the module level unit test
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Clean the jax live arrays before running the unit tests
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Fix too-few-public-methods lint
Signed-off-by: Reese Wang <rewang@nvidia.com>

---------
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Deprecate QKV_INTERLEAVED use in JAX
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Deprecate QKV_INTERLEAVED use in Paddle
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Enhance qkv enum mappings
Signed-off-by: rewang <rewang@nvidia.com>

* Fix LD_LIBRARY_PATH issue
Signed-off-by: rewang <rewang@nvidia.com>

* Arbitrary seqlen kernels only support self attention currently
Signed-off-by: rewang <rewang@nvidia.com>

---------
Signed-off-by: Reese Wang <rewang@nvidia.com>
Signed-off-by: rewang <rewang@nvidia.com>

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

* [JAX] Enhance Dropout in TransformerLayer.

1. Fixed missing setup of dropout RNG key in TransformerLayer and
   LayerNormMLP.
2. Allowing seperated dropout rate for FC1's output and other hiddens.
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Fix wrong fp8 scale in _update_fp8_metas_impl
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Fix typo
Signed-off-by: Ming Huang <mingh@nvidia.com>

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

* Fused attention kernel only supports sm80 and sm90
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Update transformer_engine/jax/csrc/modules.cpp
Co-authored-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>
Signed-off-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* arbitary fused kernel supports sm86/sm89 after 8.9.3
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Skip sm70
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Forward is_fused_attn_kernel_available to cpp backend
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Remove cpp is_fused_attn_available API
Signed-off-by: Reese Wang <rewang@nvidia.com>

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

* Initially commit for FSDP
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Adding support to fsdp xmap sharding
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Specify WeightHParamsCollection of fp8 meta.
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Support partial FP8 custom calls with FSDP.
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Adding amax reduction on the fsdp mesh dim.
Signed-off-by: Ming Huang <mingh@nvidia.com>

* clean code
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Fix the wrong batch axis in logic_axis_rules and add sharding_constraint to BMM1
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Support FSDP in fMHA.
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Fix missing all-reduce of wgrads along FSDP axis.
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Change default value of fsdp_axis_name to  for aligning with others
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Fix RuntimeError: with_sharding_constraint requires a non-empty
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Slightly changes (review feedback)
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Removed unnecessary comments
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Mergeing input_dp_dim into weight_fsdp_dim_map
Signed-off-by: Ming Huang <mingh@nvidia.com>

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

---------
Signed-off-by: Ming Huang <mingh@nvidia.com>
Signed-off-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>
Co-authored-by: Tim Moon <4406448+timmoon10@users.noreply.github.com>

* Fix flash attention dropout probability with inference
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add output as the fused attention ctx tensor
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add rng_state as the fused attention ctx tensors
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add flash attention supported lengths to the fused attention
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Refactor attention primitive to reuse abstract shaped array
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Detect backend type to allocate appropriate ctx size
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Skip dropout correctness instead of return success
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Use cudaMemsetAsync and enhance the error handling
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add flash attention kernel elts_per_thread update
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Remove redundant max 512 suffix
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Keep only DType and remove NVTEDType from python
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Fix a float32_attention_logits bugs
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Re-calculate workspace size for self attention
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Enhance bias/dbias shape guard
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Enhance the seed/rng_state checker
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Use jax.core.ShapedArray as jax.abstract_arrays is deprecated
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Enhance the unittest docs
Signed-off-by: Reese Wang <rewang@nvidia.com>

---------
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Fully remove attn_type and set self_attn_mask_type default to 'causal'
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Fix tests with new arguments
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Explicit self_attn_mask_type for examples
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Update transformer_engine/jax/flax/transformer.py
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Signed-off-by: zlsh80826 <rewang@nvidia.com>

* Update transformer_engine/jax/flax/transformer.py
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>
Signed-off-by: zlsh80826 <rewang@nvidia.com>

---------
Signed-off-by: Reese Wang <rewang@nvidia.com>
Signed-off-by: zlsh80826 <rewang@nvidia.com>
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com>

* Enable fused attention dropout
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Cast the uint32 key/counter to int64
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Update dropout support in fused attention docs
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Revise devPtrCuSeqlen* to align the naming
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Support different Jax PRNG impls
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Revert CastAsync since it is not used
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Implement is_training for 16-bit fused attn
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add fused attn with dropout sanity unit tests
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Enhance the comments readability and rng_state checker
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Change the attention dropout shape to align other frameworks
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Make encoder tests deterministic
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Change the default seed for the jax encoder tests
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Maintain offset in TE
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Enhance the resource safety
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Revert rng_state type to allow only i64
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Handle the corner case for elts_per_threads calculation
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Populate rng state by kernels
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Rename rng_state as seed in cpp_extensions
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Update the attention dropout comment
Signed-off-by: Reese Wang <rewang@nvidia.com>

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

* Add self_attn_mask_type and replace attn_type
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Refine the keyword style for the better readability
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Replace attn_type with attn_mask_type in praxis transformer
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Fix typos
Signed-off-by: Reese Wang <rewang@nvidia.com>

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

Signed-off-by: Ming Huang <mingh@nvidia.com>

* Unfused scale+softmax if bias is present
Signed-off-by: Reese Wang <rewang@nvidia.com>

* WAR a causal masking + no_bias bug and add the unittests
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Fix the optional args (bias) sharding
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Disable fused attn in JAX by default, enable it with NVTE_USE_FUSED_ATTN
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add thread local for the plan cache
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Rename dbeta to dbias for the readability
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add scaled softmax with dropout test cases
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Updated NVTE_FUSED_ATTN variable name
Signed-off-by: Reese Wang <rewang@nvidia.com>

---------
Signed-off-by: Reese Wang <rewang@nvidia.com>

[JAX] Fix missing axes parameters in TransformerLayer and the wrong shape of bias in LayerNormMLP (#196)

Fixed missing axes and wrong shape of bias in LayerNormMLP
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Add fused attention unit tests
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Use NVTE_* enums
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Use NVTE_Mask_Type and remove FMHADescriptor
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Move common functions to utils
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Change namespace to fused_attn
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Move fused_attn_max_512_fwd_qkvpacked under the general APIs
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add fused_attn_max_512_bwd_qkvpacked
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Move fused_attn_max_512_bwd_qkvpacked under the general APIs
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Remove redundant blank line
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Fix a potential bug for cu_seqlen converter
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Reformat fused_attn_max_512
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Refine the unfused attention warning message
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Rename to fused_attn_max_512
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Remove the deprecated header
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Fix flax import
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Rename to fused attn
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add attention related mask
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add attn_mask_type and attn_bias_type
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Refactor jax primitive API
* Merge q_cu_seqlen and kv_cu_seqlen
* Remove is_causal_masking
* Replace seed with rng_state
* Add is_training argument
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Remove dsoftmax from the customcall
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add None guard for bias and dropout_rng
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add version guard
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add is_fused_attn_kernel_available() to correctly dispatch the attention impl
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Fix the merge conflict
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Adjust the code style
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add the missing blank lines
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Change the order of FADescriptor members
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Enhance the readability of fused_attn_max_512.cu
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Generalize the input dimension unpacking
Signed-off-by: Reese Wang <rewang@nvidia.com>

* 16 bits fused attention requires 8.9.1
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Update fused attention support matrix
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Handle None type when sharding
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Change to the padding ratio
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Performance optimization for non-bias cases
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Revert the cudnn-frontend PRIVATE keyword which was used for debugging
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Revert "Update fused attention support matrix"

This reverts commit 4effe67d0f08f733919a329ce5ab421958740f4a.
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Treat b * s as total_seqs to align ragged cases
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add FP16/BF16 max_seqlen <= 512 fused attention to the support matrix
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Refine test_fused_attn.py

* Replace reference code with flax.linen
* Remove unnecessary comments
* Use AttnMaskType
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Unify the cuDNN compile version
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add dropout to the support matrix
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Slightly adjust the headers
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Typo fix: remove redundant either
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Consolidating fused attention requirements
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Replace cudnn_frontend::throw_if with NVTE_CHECK for the better error line report
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Rename to fused_attn_fp16_bf16_max_seqlen_512 for the better readability
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Remove CUDNN_FRONTEND_UNUSED
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Add more annotations to the custom calls
Signed-off-by: Reese Wang <rewang@nvidia.com>

---------
Signed-off-by: Reese Wang <rewang@nvidia.com>

* Adjust Module Structure.

1. Collect Flax related modules to a sub-folder, flax.
2. Add a function to unify scale_init for zero-centered-gamma LN.
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Make changes be compatible to previous versions.
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Adapt jax/examples to the new module structure.
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Update jax/docs and Add deprecated warning.
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Update README
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Adding deprecated_wrapper
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Adding deprecated warning to flax modules which imported via transformer_engine.jax
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Fix CI errors and update docs.
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Removing unnecessary deprecated warning in docs.
Signed-off-by: Ming Huang <mingh@nvidia.com>

* Implementing __iter__ to DeprecatedEnum.
Signed-off-by: Ming Huang <mingh@nvidia.com>

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