TensorRT-LLMs/cpp/tensorrt_llm/kernels/decoderMaskedMultiheadAttention
Pengbo Wang 683515b1bd
[None][feat] Use XQA JIT impl by default and mitigate perf loss with sliding window (#10335)
Signed-off-by: Pengbo Wang <221450789+pengbowang-nv@users.noreply.github.com>
2026-01-15 15:47:00 +08:00
..
cubin [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
decoderXQAImplJIT [TRTLLM-10022][feat] Add hopper xqa decode support for skip softmax attention (#10264) 2026-01-11 19:26:10 -05:00
instantiation [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
CMakeLists.txt
copy_cu.py
decoderMaskedMultiheadAttentionLaunch.h [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
decoderMaskedMultiheadAttentionTemplate.h [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
decoderXQAConstants.h [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
decoderXQAImpl.cpp [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
decoderXQAImpl.h [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
decoderXQAImplCommon.cpp [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
decoderXQAImplCommon.h [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
decoderXQAImplPrecompiled.cpp [TRTLLM-10022][feat] Add hopper xqa decode support for skip softmax attention (#10264) 2026-01-11 19:26:10 -05:00
decoderXQAImplPrecompiled.h [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
decoderXQARunner.cpp [None][feat] Use XQA JIT impl by default and mitigate perf loss with sliding window (#10335) 2026-01-15 15:47:00 +08:00
decoderXQARunner.h [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
mmha_notes.md
tensorMapUtils.cpp [TRTLLM-10022][feat] Add hopper xqa decode support for skip softmax attention (#10264) 2026-01-11 19:26:10 -05:00
tensorMapUtils.h [None][fix] Introduce inline namespace to avoid symbol collision (#9541) 2025-12-12 23:32:15 +08:00
xqaParams.h [TRTLLM-10022][feat] Add hopper xqa decode support for skip softmax attention (#10264) 2026-01-11 19:26:10 -05:00