kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-01-14 06:27:45 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity
8d858f912e
TensorRT-LLMs/cpp/tensorrt_llm/kernels/decoderMaskedMultiheadAttention/decoderXQAImplJIT
History
Pengbo Wang c0e25e5418
[TRTLLM-10022][feat] Add hopper xqa decode support for skip softmax attention (#10264) 
Signed-off-by: Pengbo Wang <221450789+pengbowang-nv@users.noreply.github.com>
2026-01-11 19:26:10 -05:00
..
nvrtcWrapper [TRTLLM-10022][feat] Add hopper xqa decode support for skip softmax attention (#10264) 2026-01-11 19:26:10 -05:00
compileEngine.cpp [TRTLLM-10022][feat] Add hopper xqa decode support for skip softmax attention (#10264) 2026-01-11 19:26:10 -05:00
compileEngine.h [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
cubinObj.cpp [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
cubinObj.h [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
cubinObjRegistry.h [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
decoderXQAImplJIT.cpp [TRTLLM-10022][feat] Add hopper xqa decode support for skip softmax attention (#10264) 2026-01-11 19:26:10 -05:00
decoderXQAImplJIT.h [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
kernelUtils.cpp [TRTLLM-10022][feat] Add hopper xqa decode support for skip softmax attention (#10264) 2026-01-11 19:26:10 -05:00
kernelUtils.h [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
serializationUtils.h [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00