jayhshah ed6a82f050 [Cute, Bwd, Sm100] Add varlen for sm100 bwd (#2150) ...

* varlen bwd with rounded padded offsets

* fix mha

* change offset mode to round down multiple

* enable varlen bwd tests

* enable deterministic mode

* fix deadlock and switch mha to no postprocess

* reenable tests

* fix lint error

* use head swizzle/spt for deterministic, update tests

* change padding offset based on arch

* rebase and update interface, tests

* add arch dispatch for padded offset q to postprocess

* address comments

* remove tile sizes from seqlen info class vars

2026-01-09 15:24:29 -08:00

..

benchmark_alibi.py

Add benchmarking code for Alibi (from Sanghun Cho)

2024-01-23 19:00:49 -08:00

benchmark_attn.py

[Cute, Bwd, Sm100] Add varlen for sm100 bwd (#2150)

2026-01-09 15:24:29 -08:00

benchmark_causal.py

Remove old fused softmax kernel from apex/Megatron

2025-08-12 11:30:00 -04:00

benchmark_flash_attention.py

benchmark: qualify all attention backends by methods list (#1881)

2025-09-12 08:00:26 -07:00

benchmark_gemm.py

Use bf16 instead of fp16 in benchmark_gemm.py

2025-02-09 13:02:28 -05:00