zhhuang-nv 97bc680cd8 feat: support kv cache reuse for MLA (#3571) ... * support kv cache reuse for MLA load compressed_kv and k_pe and do up-projection use 192/128 head size MLA context kernel support Blackwell and Hopper now Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com> * add CI test Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com> * fix: set k_pe head_num to 1 for kernel 2 and kernel 2V2 Signed-off-by: Mingyang Jiang <13463932+jmydurant@users.noreply.github.com> * resolve comments Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com> * use GPTJ style RoPE for MLA Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com> * fix rebase error and some docs Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com> * fix kv_lens Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com> * tiny fix Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com> * fix torch compile Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com> * fix: use normal device memory instead of pinned memory for unit test Signed-off-by: Mingyang Jiang <13463932+jmydurant@users.noreply.github.com> * fix L0 tests Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com> * fix torch compile after rebase Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com> * resolve comments Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com> * resolve comments again Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com> --------- Signed-off-by: Zhen Huang <145532724+zhhuang-nv@users.noreply.github.com> Signed-off-by: Mingyang Jiang <13463932+jmydurant@users.noreply.github.com> Signed-off-by: zhhuang-nv <145532724+zhhuang-nv@users.noreply.github.com> Co-authored-by: Mingyang Jiang <13463932+jmydurant@users.noreply.github.com>		2025-05-15 15:22:21 +08:00
..
__init__.py	Update TensorRT-LLM (#2755)	2025-02-11 03:01:00 +00:00
_util.py	feat: support kv cache reuse for MLA (#3571)	2025-05-15 15:22:21 +08:00
config_utils.py	feat: support kv cache reuse for MLA (#3571)	2025-05-15 15:22:21 +08:00
config.py	Breaking change: perf: Enable scheduling overlap by default (#4174)	2025-05-15 14:27:36 +08:00
cuda_graph_runner.py	[TRTLLM-3105][feat] Add Piecewise CUDA Graph Support (#3804)	2025-05-09 11:04:01 +08:00
decoder.py	Breaking change: perf: Enable scheduling overlap by default (#4174)	2025-05-15 14:27:36 +08:00
guided_decoder.py	feat: Support the Structural Tag in guided decoding (#4066)	2025-05-12 17:24:50 +08:00
kv_cache_transceiver.py	cacheTransceiver buffer manager (#3798)	2025-04-27 11:48:15 +08:00
layerwise_nvtx_marker.py	Update TensorRT-LLM (#2849)	2025-03-04 18:44:00 +08:00
llm_request.py	feat: adopt new logprob definition in PyTorch flow (#4057)	2025-05-08 20:16:40 +08:00
model_engine.py	feat: support kv cache reuse for MLA (#3571)	2025-05-15 15:22:21 +08:00
py_executor_creator.py	feat: support kv cache reuse for MLA (#3571)	2025-05-15 15:22:21 +08:00
py_executor.py	Breaking change: perf: Enable scheduling overlap by default (#4174)	2025-05-15 14:27:36 +08:00
resource_manager.py	[fix] Fix add_dummy_requests for spec decoding cases (#4084)	2025-05-09 16:52:51 +08:00
scheduler.py	refactor: collect executor and decoder states into dataclass (#3234)	2025-04-15 16:31:45 +08:00
seq_slot_manager.py	fix: skip add new slot if request has slot 0 (#3991)	2025-05-06 07:46:39 +02:00