kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-01-14 06:27:45 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

[None][ci] waive test_mamba2_chunk_scan_combined_prefill_chunking[seqlens1-8] (#7194)

Browse Source 
Signed-off-by: junq <22017000+QiJune@users.noreply.github.com>
This commit is contained in:
QI JUN 2025-08-25 11:52:59 +08:00 committed by GitHub
parent 9c5b464fe0
commit 630e67b845
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 2 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
tests/unittest/_torch/thop/test_mamba2_chunk_ss_update.py
View File

@ -369,6 +369,8 @@ def test_mamba2_chunk_scan_selective_state_update(dim, headdim, ngroups, dstate,
(16, 20),
])
def test_mamba2_chunk_scan_combined_prefill_chunking(mamba_chunk_size, seqlens):
if mamba_chunk_size == 8 and seqlens == (270, 88, 212, 203):
pytest.skip("https://nvbugspro.nvidia.com/bug/5477332")
dim = 1024
headdim = 64
ngroups = 1