Signed-off-by: Yao Yao <lowsfer@users.noreply.github.com>
Signed-off-by: peaceh <103117813+peaceh-nv@users.noreply.github.com>
Signed-off-by: Jinyang Yuan <154768711+jinyangyuan-nvidia@users.noreply.github.com>
Co-authored-by: Yao Yao <lowsfer@users.noreply.github.com>
Co-authored-by: peaceh-nv <103117813+peaceh-nv@users.noreply.github.com>
Signed-off-by: Barry Kang <43644113+Barry-Delaney@users.noreply.github.com>
Signed-off-by: Omer Ullman Argov <118735753+omera-nv@users.noreply.github.com>
Co-authored-by: Barry Kang <43644113+Barry-Delaney@users.noreply.github.com>
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
Co-authored-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
- Moved sorting related logic to a dedicated function for better clarity and maintainability.
- Enhanced sorting logic to separate finished context requests from ongoing ones before sorting by Lora task ID.
- Updated function documentation to reflect the sorting behavior and its purpose.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
- Only allocate additional outputs on last pipeline parallel rank in trtGptModelInflightBatching and executorImpl.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: CreateNewDecoderRequests
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Consolidate request generation in CreateNewDecoderRequests
- Removed the GenerateRequestOptions class and integrated its functionality into CreateNewDecoderRequests.
- Updated the constructor of CreateNewDecoderRequests to accept parameters for speculative decoding and normalization options.
- Modified the operator() method to handle request generation directly, improving code organization and reducing redundancy.
- Cleaned up associated includes and references throughout the codebase.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Simplify request handling in CreateNewDecoderRequests
- Removed the generateRequestOptions method and integrated its logic directly into the operator() method.
- Updated the request generation process to improve clarity and reduce redundancy.
- Adjusted the return type to streamline the handling of batch slots, decoder requests, and sampling configurations.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Enhance createDecoderRequests method in CreateNewDecoderRequests
- Updated the createDecoderRequests method to include additional parameters for decoder state and CUDA streams, improving flexibility in request handling.
- Removed redundant request generation logic from the operator() method, streamlining the process.
- Adjusted the newRequest method to utilize the updated decoder request structure, enhancing clarity and maintainability.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Use MedusaBuffers instead of RuntimeBuffers in CreateNewDecoderRequests
- Updated references from RuntimeBuffers to MedusaBuffers across the CreateNewDecoderRequests class and its methods, enhancing clarity in buffer management.
- Adjusted method signatures and internal logic to accommodate the new MedusaBuffers type, ensuring compatibility with existing functionality.
- Cleaned up unnecessary includes and improved code organization for better maintainability.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Update CreateNewDecoderRequests to use DecoderState and CudaStream parameters
- Modified method signatures in CreateNewDecoderRequests to replace GptDecoderBatched with runtime::decoder::DecoderState and added a separate CudaStream for the decoder.
- Adjusted the implementation of the operator() method to accommodate the new parameters, enhancing flexibility in request handling.
- Updated associated bindings in the pybind11 interface to reflect the changes in method signatures, ensuring consistency across the codebase.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Update TRTLLMSampler to use refactored create_new_decoder_requests
- Updated the sampler.py to reflect changes in the request handling logic, replacing generate_request_options with create_new_decoder_requests for improved clarity and consistency.
- Updated bindings and method signatures for decoder stream handling.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Update gptDecoderBatchedTest to use CreateNewDecoderRequests::newRequest
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* MoE TRTLLM backend for Qwen3
Signed-off-by: Anthony Chang <anchengc@nvidia.com>
* add extra moe_backend to test
Signed-off-by: Anthony Chang <anchengc@nvidia.com>
* address comments
Signed-off-by: Anthony Chang <anchengc@nvidia.com>
* conditionally compile kernels on newer archs
Signed-off-by: Anthony Chang <anchengc@nvidia.com>
* missing positional arg
Signed-off-by: Anthony Chang <anchengc@nvidia.com>
* Update the routing kernels
Signed-off-by: Christina Zhang <christinaz@nvidia.com>
* Revise usage of TLLM_LOG_ERROR
Signed-off-by: Christina Zhang <christinaz@nvidia.com>
* Add unit test for Qwen3 moe (trtllm_gen backend)
Signed-off-by: Christina Zhang <christinaz@nvidia.com>
* improve weight processing speed of moe_backend=TRTLLM; roughly 2x
Signed-off-by: Anthony Chang <anchengc@nvidia.com>
* tidy and minor fix
Signed-off-by: Anthony Chang <anchengc@nvidia.com>
* temporarily disable accuracy test that has known issue
Signed-off-by: Anthony Chang <anchengc@nvidia.com>
---------
Signed-off-by: Anthony Chang <anchengc@nvidia.com>
Signed-off-by: Christina Zhang <christinaz@nvidia.com>
Co-authored-by: Christina Zhang <christinaz@nvidia.com>
* Add Julien's origina kernel.
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
* Get rid of UpdateKVCache functionality.
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
* Add kernels.
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
* Add torch OP.
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
* Update cmake.
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
* Torch OP must use double as argument dtype.
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
* Add unittest.
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
* Add unittest.
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
* Fix misaligned access when head_dim=64.
In this case, numElemsPerThread=2, numVecPerThread=0. But the store code incorrectly perform vectorized store, some threads (e.g., lane1) issue store to address that is not aligned to 64 bit.
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
* Remove unroll (compiler can do that).
Cleanup code.
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
* Add switch for interleave.
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
* Refactor vectorized load/store.
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
* Implement is_neox. Result not correct yet.
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
* Fix is_neox=True.
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
* Add q_weight and k_weight.
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
---------
Signed-off-by: Bo Li <22713281+bobboli@users.noreply.github.com>
* first commit of cpp moe loadbalance code
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
* add python bindings for moe load balance
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
* add python wrapper, ut and bug fixes
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
* add binding for layerId and update binding test
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
* add host tensor sharing and ut
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
---------
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
This PR adds a customized allreduce to TensorRT-LLM. The new allreduce is used for communication on PCIe-based GPUs via low-precision quantization, which can accelerate the PCIe allreduce process.
Signed-off-by: Hui Kang <hkang@nvidia.com>
Co-authored-by: Hui Kang <hkang@nvidia.com>
* refactor: Copy sequence lengths once in decoder setup
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Update DecoderInputBuffers to remove duplicated buffers
- Renamed and reorganized buffer variables in decoderBuffers.h and decoderBuffers.cpp for better readability.
- Adjusted references in generateRequestOptions.cpp to align with the new buffer structure.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Move getEmbeddingBias to anonymous namespace
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Filter context requests
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: GenerateRequestOptions using more fine-grained functions
- Added a new method `createDecoderRequests` to encapsulate the logic for creating decoder requests from finished context requests.
- Updated the `operator()` method to utilize the new method, improving code clarity and maintainability.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Update TRTLLMDecoder
- Updated the `generate_request_options` call.
- Updated the `make_decoding_batch_input_output` call.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Remove const where we modify input buffers
- Changed `DecoderInputBuffers` parameters from const references to non-const references in multiple functions to allow modifications.
- Updated related function calls to ensure compatibility with the new parameter types.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* fixup! refactor: Copy sequence lengths once in decoder setup
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
- Adds BatchedGemm cubins and the respective call interface from TensorRT-LLM Generator.
- Refactors TRT-LLM Gen MoE runner to call to BMM interface
- The accuracy is verified for DeepSeek R1 FP4
Signed-off-by: Nikita Korobov <nkorobov@nvidia.com>
* Down the gcc toolset version from 13 to 11
Signed-off-by: ZhanruiSunCh <184402041+ZhanruiSunCh@users.noreply.github.com>
* Update rocky8 images
Signed-off-by: ZhanruiSunCh <184402041+ZhanruiSunCh@users.noreply.github.com>
---------
Signed-off-by: ZhanruiSunCh <184402041+ZhanruiSunCh@users.noreply.github.com>
Signed-off-by: Hao Lu <14827759+hlu1@users.noreply.github.com@users.noreply.github.com>
Co-authored-by: Hao Lu <14827759+hlu1@users.noreply.github.com@users.noreply.github.com>
* chore: Remove GptSession/V1 from TRT workflow
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Remove stateful decoders
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Remove GptSession buffers
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Remove GptSession utils
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Remove GptSession kernels
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Remove V1 GPT models from tests
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Remove gptSessionBenchmark from scripts and docs
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Remove gptSession IO classes
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Remove GptSession from test lists
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Remove GptSession from docs
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Remove useless encoder test
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Remove mActualBatchSize from DecoderState
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Remove static batching from ExecutorTest
- Updated `validateContextLogits` and `validateGenerationLogits` functions to remove the `batchingType` parameter.
- Adjusted related test functions to reflect the changes in parameter lists.
- Cleaned up the instantiation of test cases to eliminate unnecessary batchingType references.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
This PR adds a customized allreduce to TensorRT-LLM. The new allreduce is used for communication on PCIe-based GPUs via low-precision quantization, which can accelerate the PCIe allreduce process.
Signed-off-by: Hui Kang <hkang@nvidia.com>
Co-authored-by: Hui Kang <hkang@nvidia.com>
Support DeepSeek-R1 W4A8 on Hopper
Co-authored-by: Barry Kang <43644113+Barry-Delaney@users.noreply.github.com>
Co-authored-by: Jiang Shao <91270701+StudyingShao@users.noreply.github.com>
Signed-off-by: Barry Kang <43644113+Barry-Delaney@users.noreply.github.com>
* feat: Add heuristic for GroupRMSNorm kernel selection.
Implements a logistic regression model to dynamically select between:
- GroupRMSNormBaseKernel: Allocates warps proportional to sum of dimensions
(better SM occupancy in most cases)
- GroupRMSNormLargeBatch: Allocates warps proportional to max dimension
(better block scheduling in large batch scenarios)
Selection heuristic considers batch size, allocated warps, and scheduling
efficiency on the current GPU architecture. Models for Compute Capability
9.x and 10.x are trained base on nsys kernel runtime data.
The default kernel selection is the base kernel.
The python operator group_rms_norm will use the heuristic by default.
User can pick to use the base or large batch kernels as well.
Signed-off-by: Simeng Liu <simengl@nvidia.com>
* Address the comments.
Signed-off-by: Simeng Liu <simengl@nvidia.com>
---------
Signed-off-by: Simeng Liu <simengl@nvidia.com>
* Fallback to NCCL for various patterns when input size is large.
Move the previous implementation to cpp side.
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
* Revising.
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
---------
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
* feat: Reduce branch overhead in groupRMSNorm kernels
* Fix race condition with sm < 90 and avoid all threads in one warp writing to the same shared memory.
Signed-off-by: Simeng Liu <simengl@nvidia.com>
---------
Signed-off-by: Simeng Liu <simengl@nvidia.com>
* Properly get decoding mode according to same logic as cpp.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Cross reference getDecodingMode implementations in pytorch - cpp.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Better bindings for DecodingMode.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Revert to version in main.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Fix.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Revert configuration.py.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
---------
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* disable overlap in encoder
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* feat: invokeGatherBatch
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* feat: overlap same batch
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: add enableTrtOverlap to ExecutorConfig
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* disable overlap for beam search and spec decode
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* skip overlap tests with beam search or speculative decoding
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* moveFinishedContextRequestsToGeneration and skip unfinished requests in updateRequests
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* enable overlap in GptChunkedLongContextTests
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* feat: Enable overlap in gptManagerBenchmark
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* feat: Improve early exit
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Use OptionalRef for newOutputTokens tensor
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* feat: Add overlap scheduling support to TRTLLMDecoder
- Updated TRTLLMDecoder to accept an `enable_overlap_scheduler` parameter.
- Modified the decoder's internal logic to utilize the overlap scheduling feature.
- Adjusted the sequence lengths handling to ensure compatibility with the new scheduling approach.
- Enhanced unit tests to include cases for the overlap scheduler with the TRTLLMDecoder.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* fix: allNewTokens in PP
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* feat: Implement synchronous request termination in batch manager
- Added `terminateRequestSync` method to `TrtEncoderModel` and `TrtGptModelInflightBatching` for handling request termination in the next `forwardSync` call.
- Updated existing request termination logic to utilize the new synchronous method, ensuring generated tokens are cleared appropriately.
- Enhanced logging for clarity in token management during request processing.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* fixup! feat: Implement synchronous request termination in batch manager
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* fix: MockedModelCancelRequest
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* fixup! feat: Implement synchronous request termination in batch manager
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* fix: terminate with timeout
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* fixup! feat: Implement synchronous request termination in batch manager
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* docs: Update doc string for allottedTimeMs
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Move executor recv functions into classes
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Enhance MPI logging and error handling
- Updated MPI logging to include destination and tag information for better traceability during send and receive operations.
- Added error checking for MPI_Wait and MPI_Cancel calls to ensure proper handling of multi-device requests.
- Improved code structure for clarity and maintainability.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Introduce MpiTag enumeration and update MPI function signatures
- Added a new header file `mpiTags.h` to define an enumeration for MPI tags, improving code readability and maintainability.
- Updated function signatures in `mpiUtils.h` and `mpiUtils.cpp` to use the new `MpiTag` type instead of raw integers for tags.
- Refactored various MPI calls across the codebase to utilize the new `MpiTag` enumeration, enhancing type safety and clarity.
- Removed redundant MPI tag constants from several classes, streamlining the code.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* fixup! refactor: Introduce MpiTag enumeration and update MPI function signatures
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Rename tags for consistency
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Move ModelSpec from tests to core library
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Move ModelSpec from runtime to separatedir
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Use new bindings path and clean up
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Updated licenses
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Remove script_dir from path
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* Move all casters to customCasters.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Use customCasters in all bindings.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Added customCasters to userbuffers.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
---------
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* feat: Add group_rms_norm kernel to normalize multiple inputs in a single operator.
Previously, the RMSNorm implementation only supported a single input tensor. With group_rms_norm, multiple tensors can be normalized together:
```python
input_a, input_b, ... = group_rms_norm([input_a, input_b, ...])
```
All input tensors must share the same batch dimension. The kernel partitions work by dynamically assigning warp groups proportional to the last dimension of each input, improving launch efficiency and reducing overhead.
This MR provides two implementations:
GroupRMSNormKernel: Optimized for small-to-medium batch sizes
GroupRMSNormKernelLargeBatch: Contains additional optimizations for large batch sizes
Both kernels are currently exposed as custom PyTorch ops. A future MR will implement heuristic-based kernel selection and expose a unified interface.
Signed-off-by: Simeng Liu <simengl@nvidia.com>
* Resolve comments and fix typo with IS_FLASHINFER_AVAILABLE
Signed-off-by: Simeng Liu <simengl@nvidia.com>
---------
Signed-off-by: Simeng Liu <simengl@nvidia.com>
* support lp in pytorch backend
Signed-off-by: Erin Ho <14718778+hchings@users.noreply.github.com>
* fix tp
Signed-off-by: Erin Ho <14718778+hchings@users.noreply.github.com>
---------
Signed-off-by: Erin Ho <14718778+hchings@users.noreply.github.com>
* support return logprob in llmapi
Signed-off-by: Erin Ho <14718778+hchings@users.noreply.github.com>
update and add test
Signed-off-by: Erin Ho <14718778+hchings@users.noreply.github.com>
stability test
Signed-off-by: Erin Ho <14718778+hchings@users.noreply.github.com>
* revert removal of old flag
Signed-off-by: Erin Ho <erinh@nvidia.com>
Signed-off-by: Erin Ho <14718778+hchings@users.noreply.github.com>
---------
Signed-off-by: Erin Ho <14718778+hchings@users.noreply.github.com>
Signed-off-by: Erin Ho <erinh@nvidia.com>
* Add a new param to LlmRequest and Request to natively support mm
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
* update comment
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
* Update tests to match the new LlmRequest constructor parameters
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
* Modify unitTest and modify mm_embeding's dict name in llama4
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
* Fix based on comments
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
* Fix comment
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
* Fix LlmRequest initialization in kvCacheManagerTest
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
* Clean up code for promt_tuning_config
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
* Clean up prompt_tuning_config in GenerationRequest
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
---------
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
Co-authored-by: Haohang Huang <31998628+symphonylyh@users.noreply.github.com>
* Replace deepseek_allreduce op with the new unified allreduce op and moe_allreduce op.
* Minor revision of moe_allreduce op argument names.
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
Replace libtensorrt_llm_nvrtc_wrapper.so with its source code, which
consists of two parts:
1. NVRTC glue code
2. XQA kernel code
During TensorRT-LLM build, XQA kernel code is embedded as C++ arries via
gen_cpp_header.py and passed to NVRTC for JIT compilation.
Signed-off-by: Ming Wei <2345434+ming-wei@users.noreply.github.com>
* Add mIsGenerationMLA to differentiate ctx and gen MLA in AttentionOp.
For Generation MLA, if FlashMLA is used, do not check the existence of FMHA based MLA kernel.
Signed-off-by: Bo Li <bobboli0202@gmail.com>
* Run pre-commit.
Signed-off-by: Bo Li <bobboli0202@gmail.com>
* Fix compile error.
Signed-off-by: Bo Li <bobboli0202@gmail.com>
---------
Signed-off-by: Bo Li <bobboli0202@gmail.com>
test: add test cases for 0.19 release (#3608)
* fix test name
* add quickstart test for nemotron-ultra
* add rcca multi-node test case for deepseek-v3
* add rcca info
---------
squash (#3642)
fix: nvbugs/5187237: fix deterministic mode crash (#3448)
* nvbugs/5187237 nvbugs/5112075: fix deterministic mode error
* remove waive
* Revert "remove waive"
This reverts commit 0bf5486d19906d692bfb7a6262333c296b0087ac.
* revert ar fusion
---------
update fp8 doc (#3647)
tests: change qa perf test to trtllm-bench (#3619)
fix: FP8 quantized lm_head (NvBug 5214229) (#3567)
infra: Add PR approval protection for the release branch (#3634)
fix: nvbugs/5231298: pytorch allreduce issue (#3673)
Fix: nvbugs/5222698 variable not defined (#3630)
* Fix: nvbugs/5222698 variable not defined
* Tidy code
---------
test:sync waives.txt from main branch by disabling test_perf/gpt_350m-cppmanager case (#3685)
test:restore fp8 kv cache testing for L0 (#3671)
doc: Update DeepSeek perf docs (#3693)
* Update DeepSeek perf docs
* update
* Apply suggestions from code review
---------
tests: waive test_llm_multi_node (#3664)
fix: update test_user_buffers_mm_add_prologue atol (#3711)
Fix: cherry-pick hmac encryption from main branch (#3635)
* security fix cherry-pick changes from main
* fix hmac in remote mpi session (#3649)
---------
Un-waive DS-V3-Lite tests. (#3621)
fix: FP8 kv accuracy (#3675)
* fix FP8 kv accuracy
* update doc
---------
Fix script options for engines. (#3622)
unwaive multi-node test (#3721)
chore : Split more tests out of gpt tests (#3524) (#3674)
doc:add torch examples link into torch backend documentation (#3749)
test: Get Eagle tests working (#3593) (#3722)
Waive L0 test (#3756)
waive failed case in perf test, change default max_batch_size to 512 and write config.json to output log (#3656)
Update ds v3 parameters in stress test. (#3676)
waive gemma on L20 (#3766)
https://nvbugs/5141291: Fix convert.py script for Qwen model. (#3758)
Include Qwen2VLDecoderLayer in the smooth_qwen2_model function.
fix: PP4 fixes and cleanup (#3688)
remove benchmark test list (#3643)
skip disagg deepseek test if sm!=90 (#3720)
test: skip failed cases on B200 (#3710)
* add skip condition to tests
* fix error
---------
test: [nvbug: 5234494] skip_pre_ada for fp8 cases (#3718)
* skip_pre_ada for fp8 cases
* update
* update after rebase
---------
add know issue to deepseek doc. (#3800)
Fix ModelOpt Mixtral AWQ OOM (#3714) (#3761)
Waive L0 tests (#3826)
fix: Reduce memory usage in fused moe op associated with AutoTuning and fix moe fallback issue. (#3793)
* Reduce memory usage in fused moe op associated with AutoTuning.
* Replace pre-defined bucket size strategy with a generating function based on the tune_max_num_tokens.
* Add free_memory logic of workspace in min_latency_mode fused moe path.
* Fix fused_moe fallback issue. (#3652)
min_latency_mode is only set to False during warmup phase. Thus when it becomes true during inference, all tactics fall back to the default one and thus cause perf regression.
---------
[doc] Better document for Draft-Target-Model (DTM) speculative decoding (#3797)
Fix pre-commit
Fix again
Address some review comments for the MI
Signed-off-by: Dom Brown <3886319+DomBrown@users.noreply.github.com>
Co-authored-by: Zhanrui Sun <184402041+ZhanruiSunCh@users.noreply.github.com>
* refactor: Fix headsize 72 attention error for TRTLLM attn backend in PyTorch workflow
- Remove the head size pre-check logic in AttentionOp because head size 72 can be supported with fmha kernels.
- Added support for head size 72 in unfused attention kernels(QKVPreprocessing).
- Enhanced unit tests by introducing a scenario generation function for better test coverage of attention configurations(include head size 72).
Signed-off-by: qixiang-99 <203170375+qixiang-99@users.noreply.github.com>
* update: Waive head_dim=72 test cases and enhance test representation
- Added a waiver for head_dim=72 cases on post sm100 in the test suite to address known issues.
- Introduced a custom __repr__ method in the Scenario class for pytest substring match.
Signed-off-by: qixiang-99 <203170375+qixiang-99@users.noreply.github.com>
---------
Signed-off-by: qixiang-99 <203170375+qixiang-99@users.noreply.github.com>
* add MNNVL memory mapping support
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
* add more MPI environment for trtllm-llmapi-launch
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
* add MoE communication and prepare kernels
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
* add MNNVL AlltoAll support for DeepSeekV3
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
* add output dump for throughput benchmark
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
* support dynamic kernel launch grid
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
* address review comments
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
* address review comments #2
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
---------
Signed-off-by: Dongxu Yang <78518666+dongxuy04@users.noreply.github.com>
- Removed kernel under test call, as it was not needed
- Removed kernel itself
- Removed kernel tests
- Removed other unused kernels and their tests
- Some static analysis clean up
* Use updateDecoderBuffers in python decoder.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Fix synchronize in trtllm decoder.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Enable by default.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Use guided_decoder to setup seqslots and free them.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Use always decode_async and update_requests.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Update decoder buffers.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Fix speculative decoding tests.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Send new_tensors_host instead of assuming dict.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Make default False in enable_trtllm_decoder.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Partially fix mtp, partially fix py_executor.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Update request states before sending disagg ctx cache.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Fix disagg test for torch decoder.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Make isend_tensor_list and recv_tensor_list for sending the tensors_host.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Formatting.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Fix rebase.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Add disagg serving case to guided decoder.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Get overlap scheduling to work.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Update cutlass to main.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Update after rebasing.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Formatting.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Update to use decode async and update requests.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Properly pass information to update_requests
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Formatting.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Make disaggregated serving a step closer to working.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Fix rebase.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Fix rebase and format.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Copy new device tokens more pythonic.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Restore MTP add dummy reqs.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Add ordereddict import to py_executor.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Formatting.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Added seq slot manager. Add test.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Use transmission for single tensor except when list of tensors is received.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Add TRTLLMDecoder allocation to estimate max kv cache tokens.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Add stream synchronization
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Formatting.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Make memory calculation of decoder adapt to the chosen decoder. Recognize decoder option passed in executorconfig. Make overlap scheduler test run on TinyLlama.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Format
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Add decoder creation to estimate max kv.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Formatting.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Update submodule UCXX inline with main.
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
---------
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
* Rewrite unit test for unified allreduce op. Removing the legacy unit test.
* Revise formats, fusion_op bindings. Put all tensors as optional inputs.
* Move the MoeAllreduceOp to a separate custom op.
* Move all the fusion patterns to the new version of the AllReduce fusion kernel. Remove the AllReduce strategy config. Revise the AllReduce strategies and fusion pattern definitions.
* Add more TODOs, fixing minor bugs, and remove legacy code.
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
* refactor: Restructure DecoderBuffers and DecoderStepAsyncSend
- Move communication logic from `DecoderBuffers` to `DecoderStepAsyncSend`.
- Updated `DecoderStepAsyncSend` constructor to utilize the `DecoderBuffers`, enhancing clarity and reducing parameter complexity.
- Refactored related methods to align with the new class structure, improving maintainability and readability of the code.
These changes streamline the handling of decoding buffers and improve the overall architecture of the batch manager.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Restructure SlotDecoderBuffers and DecoderSlotAsyncSend
- Move communication logic from `SlotDecoderBuffers` to `DecoderSlotAsyncSend`.
- Updated `DecoderSlotAsyncSend` constructor to utilize the `SlotDecoderBuffers`, enhancing clarity and reducing parameter complexity.
- Refactored related methods to align with the new class structure, improving maintainability and readability of the code.
These changes enhance the structure and readability of the batch manager's decoding process.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Log DecodingMode
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Introduce DecoderOutputBuffers and update related classes
- Moved buffers from `DecoderBuffers` to `DecoderOutputBuffers` to better reflect its purpose.
- Updated the `DecoderStepAsyncSend` class to utilize `DecoderOutputBuffers`, enhancing clarity in the communication logic.
- Refactored the constructor and methods in `DecoderBuffers` to accommodate the new structure, improving maintainability.
- Added Python bindings for `DecoderOutputBuffers` to ensure compatibility with existing interfaces.
These changes streamline the handling of output buffers in the decoding process, improving the overall architecture of the batch manager.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Update MPI communicator handling
- Changed the `commSession` parameter type from `std::shared_ptr<mpi::MpiComm>` to `mpi::MpiComm` in `DecoderStepAsyncSend` and `DecoderSlotAsyncSend` classes for improved clarity and reduced complexity.
- Updated related methods and constructors to reflect the new parameter type, enhancing maintainability.
- Refactored the `TrtGptModelInflightBatching` class to accommodate these changes, ensuring consistent usage of `MpiComm`.
These modifications streamline the communication logic in the decoding process, improving the overall architecture of the batch manager.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Replace shared_ptr with unique_ptr for buffer management
- Updated the `TrtGptModelInflightBatching` class to use `std::unique_ptr` instead of `std::shared_ptr` for various buffer types, including `AllReduceBuffers`, `RuntimeBuffers`, `DecoderBuffers`, and `SlotDecoderBuffers`.
- This change enhances memory management and ownership semantics, reducing overhead and improving performance.
These modifications contribute to a cleaner and more efficient architecture in the batch manager.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* Fix hang bug when KV cache is low
Signed-off-by: Iman Tabrizian <itabrizian@nvidia.com>
* Review comments
Signed-off-by: Iman Tabrizian <itabrizian@nvidia.com>
* Fix attentiondp typo
Signed-off-by: Iman Tabrizian <itabrizian@nvidia.com>
* Add CI test for this case
Signed-off-by: Iman Tabrizian <itabrizian@nvidia.com>
* fix: Fix the insertion order for responder futures
Signed-off-by: Iman Tabrizian <10105175+tabrizian@users.noreply.github.com>
* fix: Fix disagg CPP
Signed-off-by: Iman Tabrizian <10105175+tabrizian@users.noreply.github.com>
---------
Signed-off-by: Iman Tabrizian <itabrizian@nvidia.com>
Signed-off-by: Iman Tabrizian <10105175+tabrizian@users.noreply.github.com>
* Feat: Offload ptable to cpu if enable_chunk_context
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
* Feat: offload ptable to cpu for chunk context mode
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
* Fix and add comment
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
* Update Readme for multimodal and add a new param mm_embedding_offloading
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
* fix: Correct prompt table offloading condition in PromptTuningBuffers
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
* Clean up the code
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
* Add commits to explain copy from cpu <-> gpu using pinned memory
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
* Fix namings based on comments
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
* Fix format based on precommit
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
* Modify --mm_embedding_offloading flag
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
---------
Signed-off-by: Kate Cheng <yunhsuanc@nvidia.com>
Co-authored-by: Haohang Huang <31998628+symphonylyh@users.noreply.github.com>
* feat: Integrate GPUDirect Storage (GDS) into Executor API
Squash of several dev commits
Signed-off-by: Dom Brown <3886319+DomBrown@users.noreply.github.com>
* implement variable window attention by breaking the block manager into window block managers per window size
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* revert isCyclic to be true if the min attention window is reached, not per window size
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* add explanatory comment to mCyclicThreshold
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* load correct gemma config
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* don't shadow inputLength in addSequence - it should remain the function scope input length between window size loop iterations
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* fix KVCacheManagerVariableWindowAttentionWithReuseTest for multiple window block managers
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* if TYPE_CHECKING
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* set temp_attention_window_inputs to None explicitly
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* set temp_attention_window_inputs to None explicitly
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* pass dtype as well
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* test_gemma variable sliding window attention
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* allot a fraction of primary/secondaryBlocks to different window size heaps, depending on the window size's total contribution to the kvcache size (i.e., including all layers)
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* remove || mEnableBlockReuse which erroneously triggers beamsearch code for cyclic variable attention window code
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* turn off request delaying for MaxUtil
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* make comments better
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* windowSizesTotalSum using std::accumulate
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* fix error handling of forwardAsync - forwardAsync catch-all catch cleanup code that runs terminateRequest can also fail and must be caught
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* fix comments
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* remove assert that kills disagg tests, since it isn't necessary
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* fix corrupted expression: 'isNewTask && (peftCacheManager ?' -> '(isNewTask && peftCacheManager) ?' which caused boolean algebra. Main is correct
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* add Gemma3 to SUPPORTED_HF_ARCHITECTURES
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* support Gemma3
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* finally fix test_gemma - always spread at least {} into generate_summary_cmd, never None
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* finally fix test_gemma - always spread at least {} into generate_summary_cmd, never None
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* fix kvfactor field for deepseek
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* fix comment
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* fix gemma-3 entries in testlist to include vswa
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* only quantize gemma2 VSWA
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
remove misleading comment
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
fix test_gemma
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* fix test_gemma
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* fix test_gemma
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* in sendRequestInfo, fromOldAllocatedBlockIds->fromOldAllocatedBlockIds, like in main
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
* fix: disable KV cache reuse if using attention sink (#3021)
* fix: disable KV cache reuse if using attention sink
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* fix: disable KV cache reuse if sink bubble
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* add comment
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
Co-authored-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* add space in test output
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* perf: reduce executor lock scope
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Move TokenRangeRetentionConfig implementation to cpp file
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* fix: Improve finished steps handling for external draft tokens
- Fixed a bug where the whole finished steps tensor was being zeroes instead of the slices.
- Replaced the creation of a temporary tensor for finished steps with a direct slice from the input tensor, improving efficiency and readability.
- Updated the tensor management logic to streamline the process of setting zero values for finished steps during batch processing.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Clean up includes
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* add pip scripts dir to path
* move nvrtc_wrapper to conan
* support building nvrtc wrapper from source
---------
Signed-off-by: Tyler Burt <195370667+tburt-nv@users.noreply.github.com>
* fix: Fixing issue with first gen token being returned twice with streaming
Signed-off-by: Patrice Castonguay <55748270+pcastonguay@users.noreply.github.com>
* Fixing not_expectring_strings in test
Signed-off-by: Patrice Castonguay <55748270+pcastonguay@users.noreply.github.com>
---------
Signed-off-by: Patrice Castonguay <55748270+pcastonguay@users.noreply.github.com>
Co-authored-by: QI JUN <22017000+QiJune@users.noreply.github.com>
No change of default value (still ON).
These were hidden cmake vars before that patch.
Fix issue #3289
Signed-off-by: William Tambellini <wtambellini@sdl.com>
Co-authored-by: Yuan Tong <13075180+tongyuantongyu@users.noreply.github.com>
* Add numNodes to ParallelConfig
If not provided, attempt to find the number of nodes by
adding the number of local ranks 0
Update device IDs check accordingly
Signed-off-by: Aurelien Chartier <achartier@nvidia.com>
* Add ParallelConfig pickle test
Signed-off-by: Aurelien Chartier <achartier@nvidia.com>
---------
Signed-off-by: Aurelien Chartier <achartier@nvidia.com>
* refactor: batch slot management in decoder classes
- Changed `forwardBatchSlots` from a single `TensorPtr` to a `std::vector<TensorPtr>` in `decoderBuffers.h` and updated its initialization in `decoderBuffers.cpp`.
- Updated `batchSlots` in `iGptDecoderBatched.h` to a `std::vector<TensorPtr>` for better handling of batch sizes.
- Modified `mBatchSlotsDecoder` in `statefulGptDecoderBatched.h` to use a `std::vector<TensorPtr>` and adjusted its initialization in `statefulGptDecoderBatched.cpp`.
- Ensured proper reshaping of tensors in the setup methods to accommodate the new vector structure.
These changes enhance flexibility in managing tensor buffers across different batch sizes.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Setup batch slots outside of the decoder
- Refactored batch slot management to utilize `makeBatchSlots`, enhancing clarity and functionality in batch processing.
- Introduced `DecoderState` to `MakeDecodingBatchInputOutput` for improved state handling during decoding.
- Updated the `operator()` method to include `decoderState` as a parameter, facilitating better integration with the decoding process.
- Modified related tests to accommodate changes in batch slot handling and ensure proper functionality.
These updates improve the overall structure and efficiency of the decoding process in the batch manager.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Enhance decoder input structure with maxDecodingEngineTokens
- Updated the `Input` class in `iGptDecoderBatched.h` to include a new parameter `maxDecodingEngineTokens` for better control over decoding limits.
- Modified the `MakeDecodingBatchInputOutput` algorithm to compute the maximum number of decoding tokens based on active slots.
- Adjusted the `GptDecoderBatched` class to utilize the new `maxDecodingEngineTokens` parameter, improving clarity in token management during decoding.
- Updated Python bindings to reflect changes in the `Input` class constructor.
- Enhanced tests to ensure proper handling of the new parameter.
These changes improve the flexibility and efficiency of the decoding process in the batch manager.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Streamline decoder input creation and batch slot management
- Introduced a new function `createDecoderInputs` to encapsulate the logic for creating decoder inputs, improving code organization.
- Updated the `operator()` method to utilize the new `createDecoderInputs` function, simplifying the decoding input setup process.
- Removed the `maxOfActiveSlots` template function to streamline the logic for determining the maximum number of active decoding engine tokens.
- Introduced a direct calculation of `maxActiveDecodingEngineTokens` within the `createDecoderInputs` function, enhancing clarity and reducing complexity.
These changes enhance the maintainability and readability of the decoding process in the batch manager.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Update logits handling in decoder batch
- Modified the `decoder_batch::Input` to accept a vector of vectors for logits, enhancing flexibility in tensor management.
- Adjusted the `createDecoderInputs` function to accommodate the new logits structure, ensuring proper batch processing.
- Updated Python bindings to reflect changes in the `Input` class constructor, maintaining compatibility with existing interfaces.
- Refactored the `GptDecoderBatched` and `StatefulGptDecoderBatched` classes to utilize the updated logits structure, improving clarity in tensor slicing and batch size management.
- Enhanced tests to validate the new input structure and ensure correct functionality across various decoding scenarios.
These changes streamline the decoding process and improve the overall maintainability of the codebase.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Rename maxDecodingEngineTokens to maxDecoderSteps
- Updated the `Input` class in `iGptDecoderBatched.h` to rename `maxDecodingEngineTokens` to `maxDecoderSteps` for improved clarity.
- Adjusted the `createDecoderInputs` function to reflect the new naming, ensuring consistency in the decoding process.
- Modified the `GptDecoderBatched` class to utilize `maxDecoderSteps` in its logic, enhancing readability and maintainability.
- Updated Python bindings to expose the renamed parameter, maintaining compatibility with existing interfaces.
These changes enhance the clarity of the decoding parameters and improve the overall structure of the codebase.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: remove usage of `active` vector from prepareForward
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Removed the `active` vector from `decoder_batch::Input`
- Removed the `active` vector from the `Input` class constructor in `iGptDecoderBatched.h`, streamlining the input handling for decoding.
- Updated the `createDecoderInputs` function and related tests to reflect the changes in the `Input` class, ensuring compatibility and maintaining functionality.
- Adjusted Python bindings to accommodate the new constructor signature, enhancing clarity in the interface.
These changes improve the maintainability and readability of the decoding process in the batch manager.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: remove usage of `active` vector from gptDecoderBatchedTest
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Unify the creation of decoder batch inputs in algorithm and tests
- Added a new static method `createDecoderBatchInputs` to streamline the creation of decoder batch inputs, enhancing clarity and maintainability.
- Updated the implementation to utilize active slots directly, simplifying the logic for managing batch slots and logits.
- Refactored the `operator()` method to leverage the new input creation function, ensuring compatibility with existing decoding processes.
- Enhanced tests to validate the new input handling approach, ensuring correct functionality across various scenarios.
These changes improve the overall structure and readability of the decoding process in the batch manager.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: remove usage of active vector from createDecoderBatchInputs
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Update maxDecoderSteps calculation
- Replaced integer division with `common::ceilDiv` for calculating `maxDecoderSteps` and `numDecoderSteps`, ensuring correct handling of token counts.
These changes enhance the robustness of the decoding batch input creation process.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: remove cumLogProbs and logProbs from DecoderBuffers
- Eliminated cumLogProbs and logProbs from DecoderBuffers, streamlining the buffer management.
- Updated related code in decoderBuffers.cpp and bindings.cpp to reflect these changes, ensuring that only host pointers are used for log probabilities.
These modifications enhance code clarity and maintainability by reducing redundancy in buffer management.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: streamline sequence length handling in GptDecoderBatched and StatefulGptDecoderBatched
- Updated GptDecoderBatched to directly use output.sequenceLengths for lengths assignment, removing unnecessary reshaping.
- Adjusted StatefulGptDecoderBatched to ensure sequence lengths are correctly shaped based on actual batch size and max beam width.
These changes enhance clarity and maintainability in the decoding process.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: integrate DecoderState for sequence length management in decoding process
- Updated DecoderBuffers to remove direct handling of sequence lengths, now utilizing DecoderState for this purpose.
- Adjusted MakeDecodingBatchInputOutput to accept DecoderState, enhancing clarity in the decoding input/output management.
- Refactored GptDecoderBatched and StatefulGptDecoderBatched to streamline sequence length handling, ensuring consistency across the decoding workflow.
refactor: update SlotDecoderBuffers to manage sequence lengths directly
- Introduced sequenceLengths and sequenceLengthsHost to SlotDecoderBuffers for better management of sequence lengths.
- Refactored asyncSend and recv methods to utilize the new sequenceLengths member, enhancing clarity and reducing redundancy.
- Updated TrtGptModelInflightBatching to align with the new structure, ensuring consistent handling of sequence lengths across the decoding process.
These changes improve maintainability and streamline the decoding workflow.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Delegate to asyncSend method in SlotDecoderBuffers
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Update ExecutorConfig to use AdditionalModelOutput type
- Changed function signatures and member variables across multiple files to replace std::optional<std::vector<std::string>> with std::optional<std::vector<executor::AdditionalModelOutput>> to include gatherContext flag for each additional output.
- Updated related serialization and deserialization methods to accommodate the new type.
- Adjusted tests to reflect the changes in the output handling structure.
This refactor enhances the flexibility and maintainability of the output configuration in the executor and batch manager components.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Remove equality operator from TrtGptModelOptionalParams
- Deleted the operator== implementation from TrtGptModelOptionalParams to simplify the class.
- Updated the pybind11 bindings to remove the exposure of the equality operator to Python.
This change streamlines the class definition and reduces unnecessary complexity in the bindings.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Enhance copyAdditionalOutputs to utilize AdditionalModelOutput
- Updated the copyAdditionalOutputs function to accept a vector of AdditionalModelOutput, allowing for the inclusion of the gatherContext flag.
- Adjusted the logic to handle context and non-context outputs separately, improving the output handling mechanism.
- Modified related unit tests to incorporate the new gatherContext parameter, ensuring comprehensive testing of the updated functionality.
This refactor improves the flexibility and clarity of output management in the batch processing workflow.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Introduce findOutputTensor utility function for output tensor retrieval
- Added a new utility function, findOutputTensor, to encapsulate the logic for finding output tensors and checking their validity.
- Refactored copyAdditionalOutputs to utilize findOutputTensor, reducing code duplication and improving clarity.
- Enhanced error checking for additional context and generation output tensors.
This change streamlines the output tensor retrieval process, enhancing maintainability and readability in the batch processing workflow.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Check final indices of additional output tensors and update tests
- Added checks to verify the final indices of additional output tensors for context and generation outputs.
- Updated unit tests to verify the changes.
- Add lastTokenIds input tensor to test engines.
- Logits output depends on gatherContextLogits parameter.
- Removed gatherContextOutputs parameter from the validate method in LlmRequest.
- Context outputs do not depend on computeContextLogits parameter.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* fixup! refactor: Check final indices of additional output tensors and update tests
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* fixup! refactor: Update ExecutorConfig to use AdditionalModelOutput type
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* fixup! refactor: Remove equality operator from TrtGptModelOptionalParams
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* docs: Update executor.md
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Clean up includes
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* One of the tactic is not supported during dispatch.
* final_hidden_states should be unpacked if it is not min_latency_mode.
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
* feat: Add NVFP4 UB pattern optimization pass in torch compile
* Add an additional flag for UB fp4 pattern to avoid inverse the scale
* Add NVFP4 related UB patterns
Signed-off-by: Jin Li <59594262+liji-nv@users.noreply.github.com>
* Update atol, some points fails for B200 umbriel.
Signed-off-by: liji-nv <59594262+liji-nv@users.noreply.github.com>
---------
Signed-off-by: Jin Li <59594262+liji-nv@users.noreply.github.com>
Signed-off-by: liji-nv <59594262+liji-nv@users.noreply.github.com>
* feat: trtllm-gen fp4 GEMM
Signed-off-by: Dom Brown <3886319+DomBrown@users.noreply.github.com>
* Clean up
Signed-off-by: Dom Brown <3886319+DomBrown@users.noreply.github.com>
* Remove incorrect header
Signed-off-by: Dom Brown <3886319+DomBrown@users.noreply.github.com>
* Reviewer comment
Signed-off-by: Dom Brown <3886319+DomBrown@users.noreply.github.com>
---------
Signed-off-by: Dom Brown <3886319+DomBrown@users.noreply.github.com>
* Instead of allocating UserBuffers at beginning of runtime, UB buffers
are now managed with global allocator. The allocator will dynamically
assign free UB buffer or allocate new buffer for torch tensor. It makes
userbuffers easier to use.
* In common usecase, the Userbuffers will be allocated correctly during
warm up stage. There is no dynamic allocation during inference.
* UB fusion pattern is rewroten using the new UB Allocator. It contains
following passes:
1. Fuse Quant with allreduce, replace with UB impl, and insert a
copy_to_userbuffers. Currently the normal allreduce still does not
support FP8 quant. So this need to be done in UB pass
2. Convert all supported allreduce with UB and insert copy_to_userbuffers.
3. Fuse op before ar with the copy_to_userbuffers. So the op directly
writes to the userbuffer
4. Remove userbuffers finalize if the output is connect to another UB
allreduce.
Signed-off-by: Jin Li <59594262+liji-nv@users.noreply.github.com>
* Several optimizations and fixings on the Autotuner.
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
* Apply the new Python side Autotuner on current linear for nvFP4 data type.
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
* Apply the new Python side Autotuner on MoE op
* Remove routers from cache key to improve inference perf
* Prevent unnecessary code profiling. Use do_preparation keyword to select which part should be executed during before evaluating any tactic.
* Remove try-catch inside moe profiling process.
* Move default tactic -1 to 0 transforms in cpp runner.
* Revise relavant tests.
* Predefined the bucketizing strategy for fused_moe
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
* Add specific_profile support for AutoTuner to bypass the standard cache search process for perf optimization
* Add specific_profile for moe
* Add specific profile for linear
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
* Fixing and revising according to reviewer's suggestions.
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
* Use lru_cache for inference pref optimization.
* Revert gen_custom_cache_key feature
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
* Replace runner with runner id to achieve a serializable cache.
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
* Code clean up and minor fixings.
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
* Move all tunable runners and custom ops into torch_custom_ops.
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
* Treat min_latency_mode as a independent dynamic tensor. Modify get_valid_tactics to suit for it.
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
---------
Signed-off-by: Yukun He <23156053+hyukn@users.noreply.github.com>
* feat: Add option to run disaggregated serving without ctx servers, to benchmark gen only
Signed-off-by: Patrice Castonguay <55748270+pcastonguay@users.noreply.github.com>
* Fixing comment in sanity check
Signed-off-by: Patrice Castonguay <55748270+pcastonguay@users.noreply.github.com>
---------
Signed-off-by: Patrice Castonguay <55748270+pcastonguay@users.noreply.github.com>
* fp8 kv + bf16 ctx MLA + fp8 gen MLA
Use BF16 for context MLA.
mFP8GenerationMLA and mFP8ContextFMHA shouldn't be enabled together.
Allow mSM==90 for mFP8GenerationMLA==true.
For FMHA, dataTypeKv should be FP8.
For FP8 MLA generation, the output is still in BF16.
Refine debug info for FMHA kernel metadata.
Use inputType, outputType, SM together to hash kernel list.
Add FP8 MLA generation FMHA kernel.
Special WAR of NUM_COMPUTE_GROUPS for MLA generation kernel.
Separate the implementation of fused_multihead_attention_v2.h to CPP and print some debug info if checkIfKernelExist fails.
Refine debug info in fused_multihead_attention_v2.cpp
Correct FP8 MLA metadata.
New kernel provided by Yuxin, which outputs BF16.
smem size is not set correctly, which will lead to illegal mem access.
Yuxin fixed the error in FMHA MLA kernel: previously the BF16 isn't correctly written: some parts are repeatedly written, while some others are untouched.
There are two bmm1 scales that should be set correctly.
New kernel generated by Yuxin.
Modificatiosn to common/attentionOp for FP8 MLA on Hopper using FMHA.
Not necessary. If mFP8GenerationMLA, is_fp8_out is false, so mFP8ContextFMHA is false.
Skip a check in fmhaDispatcher.
Modifications in fmhaRunner:
- Debug dump.
- if (!isFP8GenerationMLA) skips a lot of flag setting.
- TMA descriptor modification for qo (by Yuxin).
Cleanup debug output.
Clean up o tma descriptor modifications.
Signed-off-by: Bo Li <bobboli0202@gmail.com>
* Resolve conflicts.
Signed-off-by: Bo Li <bobboli0202@gmail.com>
* Apply the patch of FP8 FlashMLA and resolve conflicts.
Signed-off-by: Bo Li <bobboli0202@gmail.com>
* Fix compilation error.
Signed-off-by: Bo Li <bobboli0202@gmail.com>
* Fix compile error.
Signed-off-by: Bo Li <bobboli0202@gmail.com>
* pick blackwell support
Signed-off-by: Dylan Chen <191843203+DylanChen-NV@users.noreply.github.com>
* Add copyright notice to fused_multihead_attention_v2.cpp.
Signed-off-by: Bo Li <bobboli0202@gmail.com>
* Add license.
Signed-off-by: Bo Li <bobboli0202@gmail.com>
* Add missing license.
Signed-off-by: Bo Li <bobboli0202@gmail.com>
* Exclude building flashMLA kernels under sm90.
Signed-off-by: Bo Li <bobboli0202@gmail.com>
* Revert "Exclude building flashMLA kernels under sm90."
This reverts commit f0c859d459.
Signed-off-by: Bo Li <bobboli0202@gmail.com>
* Use macro to skip compiling FlashMLA for non sm90 targets.
Signed-off-by: Bo Li <bobboli0202@gmail.com>
---------
Signed-off-by: Bo Li <bobboli0202@gmail.com>
Signed-off-by: Dylan Chen <191843203+DylanChen-NV@users.noreply.github.com>
Co-authored-by: Dylan Chen <ziqingc@nvidia.com>
Co-authored-by: Dylan Chen <191843203+DylanChen-NV@users.noreply.github.com>
Co-authored-by: QI JUN <22017000+QiJune@users.noreply.github.com>
* refactor: Expose DecoderState via bindings and integrate in TRTLLMDecoder
- Introduced a new `DecoderState` class in the C++ bindings, encapsulating key functionalities for managing decoding state.
- Adjusted the Python `TRTLLMDecoder` to access properties from `decoder_state`, ensuring consistency and clarity in the decoding process.
These changes streamline the decoder's architecture and enhance maintainability.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: Remove unused new_tokens from DecoderState bindings
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* init trtllm attn no cache
Signed-off-by: Qixiang Lin <qixiangl@nvidia.com>
* fix: fix the seq_len issue and attn metadata prepare for qwen reward model test
fix: fix minor bugs after rebase
Signed-off-by: Qixiang Lin <qixiangl@nvidia.com>
* refactor: remove unnecessary debug logs and clean up commented code
refactor: update max_seq_len documentation and remove max_seq_len for decoder model contructor in PyTorchModelEngine
Signed-off-by: Qixiang Lin <qixiangl@nvidia.com>
* refactor: update calculate_ref_result function to accept tensor inputs and mask type, enhance test_attention_no_cache to support FULL and CAUSAL masks
Signed-off-by: Qixiang Lin <qixiangl@nvidia.com>
* refactor: remove unused BERT attention metadata conversion method and add type assertion for no cache attention in PyTorchModelEngine
Signed-off-by: Qixiang Lin <qixiangl@nvidia.com>
* refactor: remove use_kv_cache parameter from attention function and related classes, update documentation for KV cache handling
Signed-off-by: Qixiang Lin <qixiangl@nvidia.com>
* refactor: implement setAttentionMaskType method for better mask type handling and remove unused conversion function
Signed-off-by: Qixiang Lin <qixiangl@nvidia.com>
* refactor: streamline KV cache handling by replacing direct member access with useKVCache method and simplify token per block assignment
remove Debug code.
Signed-off-by: Qixiang Lin <qixiangl@nvidia.com>
* refactor: Resolve comments for Python code
Simplify no cache attention metadata preparation and streamline related attributes in TrtllmAttentionMetadata
Removed the private method for converting to no cache attention metadata and integrated its logic into the prepare method. Updated the test for BERT sequence classification to reflect these changes and ensure proper handling of attention metadata.
Signed-off-by: Qixiang Lin <qixiangl@nvidia.com>
* docs: Add is_dummy_attention field to attention metadata for simulation operations
Signed-off-by: Qixiang Lin <qixiangl@nvidia.com>
* refactor: add KVCacheParams to attention backend interface and import relevant metadata classes
Updated the attention backend interface to include KVCacheParams and imported TrtllmAttentionMetadata and VanillaAttentionMetadata in model_engine.py for enhanced functionality.
Signed-off-by: Qixiang Lin <qixiangl@nvidia.com>
* fix: fix rebase format issue
Signed-off-by: Qixiang Lin <qixiangl@nvidia.com>
* fix: extend attention mask type handling in MHARunnerFixedParams
Added support for additional attention mask types (BIDIRECTIONAL, BIDIRECTIONALGLM, BLOCKSPARSE) in the MHARunnerFixedParams structure to fix the mapping issue between ContextAttentionMaskType and AttentionMaskType
Signed-off-by: Qixiang Lin <qixiangl@nvidia.com>
* fix: enhance attention mask type handling in TllmGenFmhaRunnerParams
Updated the setAttentionMaskType method to include a switch-case structure for better handling of attention mask types, ensuring proper mapping and error handling for invalid types.
Signed-off-by: Qixiang Lin <qixiangl@nvidia.com>
---------
Signed-off-by: Qixiang Lin <qixiangl@nvidia.com>
* Reapply "refactor: Replace DecoderFinishedEvent with CudaEvent in decoder clas…" (#3183)
This reverts commit 75495730bc.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* fixup! Reapply "refactor: Replace DecoderFinishedEvent with CudaEvent in decoder clas…" (#3183)
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
- Updated the first dimension of additional output tensors to match mMaxNewTokens.
- Copy output of last context token to generation outputs.
- Adjusted the expected output size calculations in unit tests to reflect the correct maximum output length.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* Update fp8 sf layout for blackwell and enable fp8 gemm e2e
* Add test case when m needs to be padded
* Better comment
Signed-off-by: Chang Liu <liuc@nvidia.com>
* Add TODO for fp8 quant kernel
Signed-off-by: Chang Liu <liuc@nvidia.com>
* Enable DCO check
Signed-off-by: Chang Liu <liuc@nvidia.com>
* Fix lint
---------
Signed-off-by: Chang Liu <liuc@nvidia.com>
* refactor: Simplifiy disableLookahead method
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* Update DecoderBuffers comments
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Move numDecodingEngineTokens to DecoderState
This commit introduces new methods in the DecoderState class to manage the number of tokens for each request in a batch. The following changes were made:
- Added `getNumDecodingEngineTokens()` to retrieve the number of tokens for all requests.
- Added `getNumDecodingEngineTokens(SizeType32 batchIdx)` to get the token count for a specific request.
- Added `setNumDecodingEngineTokens(SizeType32 batchIdx, SizeType32 numTokens)` to set the token count for a specific request.
- Updated the setup method to initialize the token count vector based on the maximum batch size.
- Refactored the `CreateNewDecoderRequests` class to utilize the new token management methods, improving clarity and maintainability.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Improve shape variables in DecoderState
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
- Updated the `forwardAsync` method in `GptDecoderBatched` and `iGptDecoderBatched` to return `CudaEvent` instead of `DecoderFinishedEventPtr`, simplifying event handling.
- Removed the `DecoderFinishedEvent` class and its associated usage across various files, streamlining the codebase.
- Adjusted related methods and Python bindings to accommodate the new event structure, ensuring compatibility and maintaining functionality.
These changes enhance the clarity and efficiency of the decoding process in the batch manager.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Update gatherTree function to accept CUDA stream parameter
This commit modifies the gatherTree function signature to include a runtime::CudaStream parameter, enhancing flexibility in stream management. Additionally, it removes unnecessary buffer manager parameters and stream handling from the function, streamlining the code. The finalize method in GptDecoderBatched is also updated to reflect these changes, improving clarity and maintainability in the decoding process.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Update GptDecoderBatched finalize
This commit refactors the GptDecoderBatched class to improve method signatures and reduce code complexity:
- Modified finalize method to accept DecoderState as a parameter
- Updated method signatures to work with the new DecoderState approach
- Improved code organization and readability
The changes continue the ongoing refactoring to centralize decoder state management and simplify the decoder implementation.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* chore: update cutlass to v3.8.0
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: update include directives for consistency and organization in weightOnlyBatchedGemv headers
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* Fix fpA_intB_gemm compilation
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Remove ForwardType enum from GptDecoderBatched
- Remove ForwardType enum from GptDecoderBatched
- Simplify forwardDispatch and forwardDecoder methods
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Remove forwardDecoder method from GptDecoderBatched
- Eliminate the forwardDecoder method to streamline the decoding process.
- Update forwardDispatch to directly call forwardAsync when input batch size is greater than zero.
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
* refactor: Move event handling from forwardDispatch to forwardAsync
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
---------
Signed-off-by: Robin Kobus <19427718+Funatiq@users.noreply.github.com>
Remove extra methods in trtGptModelInflightBatching.h. The methods were moved out of that class during a previous refactoring, but the definitions have been left behind.
Signed-off-by: Aurelien Chartier <achartier@nvidia.com>
