Kermit adb6d96516 [Bugfix] Fix GDN KKT precision loss on Hopper GPUs by aligning tl.dot operand layout with WGMMA (#42076) ...

Signed-off-by: kermit <ckeming@outlook.com>

2026-05-09 13:08:46 +00:00

..

__init__.py

[Perf] fuse kernels in gdn (#37813)

2026-04-02 11:52:18 +00:00

chunk_delta_h.py

[CI][AMD][BugFix][Kernel] Cast induction variable to int64 on MI350 for chunk_gated_delta_rule_fwd_kernel_h_blockdim64 to avoid illegal memory access (#39087)

2026-04-08 16:57:18 +08:00

chunk_o.py

[Aiter][ROCm] gdn_linear_attn kernel fusion (#40711)

2026-05-07 23:11:37 -07:00

chunk_scaled_dot_kkt.py

[Bugfix] Fix GDN KKT precision loss on Hopper GPUs by aligning tl.dot operand layout with WGMMA (#42076)

2026-05-09 13:08:46 +00:00

chunk.py

[Aiter][ROCm] gdn_linear_attn kernel fusion (#40711)

2026-05-07 23:11:37 -07:00

cumsum.py

[GDN] Eliminate GPU->CPU sync in prepare_chunk_indices during prefill (#38361)

2026-04-03 13:38:02 +00:00

fused_gdn_prefill_post_conv.py

[Perf] fuse kernels in gdn (#37813)

2026-04-02 11:52:18 +00:00

fused_recurrent.py

[Bugfix] Fix GDN FLA kernel crashes with NULL_BLOCK_ID=0 CUDA graph padding (#39064)

2026-04-11 08:35:19 +00:00

fused_sigmoid_gating.py

[Bugfix] Fix GDN FLA kernel crashes with NULL_BLOCK_ID=0 CUDA graph padding (#39064)

2026-04-11 08:35:19 +00:00

index.py

[Perf] [Bugfix] Fix Triton autotuning in inference for Qwen3.5 (#37338)

2026-03-23 00:37:58 -07:00

kda.py

[Bugfix] correct h matrix layout in chunk_kda output kernel (#40956)

2026-04-30 16:22:41 +08:00

l2norm.py

[Model] Add support for OLMo Hybrid (#32550)

2026-03-05 14:51:06 -05:00

layernorm_guard.py

[Model] Add support for OLMo Hybrid (#32550)

2026-03-05 14:51:06 -05:00

op.py

[Kernel] Accelerate solve_tril with TMA (#26746)

2025-10-20 05:39:02 +00:00

solve_tril.py

[GDN] Eliminate GPU->CPU sync in prepare_chunk_indices during prefill (#38361)

2026-04-03 13:38:02 +00:00

utils.py

[Perf][GDN] Align TMA usage with upstream FLA (#38981)

2026-04-05 00:38:02 +08:00

wy_fast.py

[GDN] Eliminate GPU->CPU sync in prepare_chunk_indices during prefill (#38361)

2026-04-03 13:38:02 +00:00