Skip to content

[None][ci] Waive blackwell test on spec gate. #9502

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
zheyuf merged 2 commits into from
Nov 26, 2025
Merged

[None][ci] Waive blackwell test on spec gate. #9502

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
a9d257a
Waive blackwell test on spec gate.
zheyuf Nov 26, 2025
2adee6a
Add nvbugs link
zheyuf Nov 26, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion tests/unittest/_torch/speculative/test_spec_gate.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@
import pytest
import torch
from utils.llm_data import llm_models_root
from utils.util import similar
from utils.util import similar, skip_blackwell

from tensorrt_llm import LLM, SamplingParams
from tensorrt_llm._torch.speculative.speculation_gate import SpeculationGate
Expand All @@ -20,6 +20,7 @@
# It is set with acceptance window and acceptance threshold in spec_config.
# This test set the max_concurrency to a large value to prevent spec decode turned off due to number of effective requests > max_concurrency,
# So that we can only focus on the turning off effect from the SpeculationGate.
@skip_blackwell # TODO: Remove after fixing TRTLLM-GEN FMHA segfault on Blackwell. NVBugs: https://nvbugspro.nvidia.com/bug/5698292
@pytest.mark.high_cuda_memory
def test_spec_gate_e2e():
total_mem_gb = torch.cuda.get_device_properties(0).total_memory / 1e9
Expand Down