TensorRT-LLMs/cpp/tensorrt_llm/executor
Robin Kobus ccff86068e
fix: request termination in pipeline parallelism (#3892)
* 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>
2025-05-05 21:51:41 +08:00
..
cache_transmission refactor: Introduce MpiTag enumeration and update MPI function signatures (#3893) 2025-05-04 13:24:29 +02:00
cacheTransceiverConfig.cpp cacheTransceiver buffer manager (#3798) 2025-04-27 11:48:15 +08:00
CMakeLists.txt cacheTransceiver buffer manager (#3798) 2025-04-27 11:48:15 +08:00
contextPhaseParams.cpp Update TensorRT-LLM (#2936) 2025-03-18 21:25:19 +08:00
debugConfig.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
decodingConfig.cpp Feat: Variable-Beam-Width-Search (VBWS) part3 (#3338) 2025-04-08 23:51:27 +08:00
disaggServerUtil.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
dynamicBatchConfig.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
dynamicBatchTuner.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
dynamicBatchTuner.h Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
executor.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
executorConfig.cpp cacheTransceiver buffer manager (#3798) 2025-04-27 11:48:15 +08:00
executorImpl.cpp fix: request termination in pipeline parallelism (#3892) 2025-05-05 21:51:41 +08:00
executorImpl.h refactor: Introduce MpiTag enumeration and update MPI function signatures (#3893) 2025-05-04 13:24:29 +02:00
executorKVCacheEventManager.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
extendedRuntimePerfKnobConfig.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
guidedDecodingConfig.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
guidedDecodingParams.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
intervalSet.h Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
jsonSerialization.cpp feat: Add BW measurement (#3070) 2025-03-28 10:53:00 +08:00
kvCacheConfig.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
kvCacheRetentionConfig.cpp chore: Clean up cpp runtime (#3537) 2025-04-15 16:06:14 +08:00
logitsPostProcessorConfig.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
loraConfig.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
model.h fix: request termination in pipeline parallelism (#3892) 2025-05-05 21:51:41 +08:00
mropeConfig.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
orchestratorConfig.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
orchestratorUtils.h refactor: Introduce MpiTag enumeration and update MPI function signatures (#3893) 2025-05-04 13:24:29 +02:00
outputConfig.cpp feat: Allow individual gatherContext for each additional output (#3374) 2025-04-12 17:00:36 +08:00
parallelConfig.cpp feat: Add numNodes to ParallelConfig (#3346) 2025-04-13 13:55:04 +02:00
peftCacheConfig.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
promptTuningConfig.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
request.cpp feat: Add multimodal embedding field in LlmRequest (#3855) 2025-05-01 12:23:30 +08:00
requestImpl.h feat: Add multimodal embedding field in LlmRequest (#3855) 2025-05-01 12:23:30 +08:00
requestUtils.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
requestUtils.h Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
requestWithId.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
requestWithId.h Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
response.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
responseImpl.h Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
samplingConfig.cpp feat: Variable-Beam-Width-Search (VBWS) Part2 (#3133) 2025-04-02 12:31:28 +08:00
schedulerConfig.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
serialization.cpp feat: Add multimodal embedding field in LlmRequest (#3855) 2025-05-01 12:23:30 +08:00
serializeUtils.h cacheTransceiver buffer manager (#3798) 2025-04-27 11:48:15 +08:00
speculativeDecodingConfig.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
tensor.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
types.cpp Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00