Yukun He c678774c99 feat: Apply the new torch-flow compatible AutoTuner to both Fused MoE and NVFP4 Linear operators. (#3151) ... * 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>		2025-04-08 14:28:36 +08:00
..
cmake	fix #3109: early exit cmake if find_library() does not find any lib (#3113)	2025-03-29 19:59:03 +08:00
include/tensorrt_llm	feat: use NVRTC for DeepGEMM JIT compilation (#3239)	2025-04-07 20:29:23 +08:00
micro_benchmarks	perf: Add optimizations for deepseek in min latency mode (#3093)	2025-04-02 09:05:24 +08:00
tensorrt_llm	feat: Apply the new torch-flow compatible AutoTuner to both Fused MoE and NVFP4 Linear operators. (#3151)	2025-04-08 14:28:36 +08:00
tests	feat: use cudaMalloc to allocate kvCache (#3303)	2025-04-08 10:59:14 +08:00
CMakeLists.txt	fix: upgrade cmake minimum from 3.18 to 3.27 (#3208)	2025-04-02 15:14:36 +08:00