Skip to content

[https://nvbugs/6627197][fix] Unwaive KV pool rebalance PP loop tests #18038

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
zhaoyuanh-nvidia merged 1 commit into from
Aug 24, 2026
Merged

[https://nvbugs/6627197][fix] Unwaive KV pool rebalance PP loop tests #18038

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
4 changes: 0 additions & 4 deletions tests/integration/test_lists/waives.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -346,10 +346,6 @@ test_e2e.py::test_trtllm_bench_llmapi_launch[pytorch_backend-llama-v3-llama3-8b]
unittest/_torch/attention/sparse/dsa/test_req_idx_per_token.py::test_on_update_kv_lens_rebuilds_stale_map SKIP (https://nvbugs/6574939)
unittest/_torch/attention/test_attention_backends.py::test_attention_backend[deepseekv3_mla-gen-bf16-HND-p32-v1] SKIP (https://nvbugs/6507109)
unittest/_torch/attention/test_attention_backends.py::test_attention_backend[qwen2_0_5b_gqa_hd64-ctx-bf16-HND-p32-v1] SKIP (https://nvbugs/6641268)
unittest/_torch/executor/test_kv_pool_rebalance.py::TestPpLoopDrainWiring::test_drain_length_follows_the_ring_depth SKIP (https://nvbugs/6627197)
unittest/_torch/executor/test_kv_pool_rebalance.py::TestPpLoopDrainWiring::test_drain_suppresses_queueing_then_rebalances_and_resumes SKIP (https://nvbugs/6627197)
unittest/_torch/executor/test_kv_pool_rebalance.py::TestPpLoopDrainWiring::test_loop_queues_normally_when_no_rebalance_is_pending SKIP (https://nvbugs/6627197)
unittest/_torch/executor/test_kv_pool_rebalance.py::TestPpLoopDrainWiring::test_no_drain_is_started_while_one_is_pending SKIP (https://nvbugs/6627197)
unittest/_torch/executor/test_overlap_scheduler.py::test_overlap_scheduler_block_reuse_cache_hit[TorchSampler] SKIP (https://nvbugs/6608387)
unittest/_torch/executor/test_overlap_scheduler.py::test_overlap_scheduler_consistency[no_reuse-cpp_scheduler-TorchSampler] SKIP (https://nvbugs/6561559)
unittest/_torch/modeling/test_gemma4_e2e_dummy.py::test_e2e_text_31b_dummy SKIP (https://nvbugs/6607482)
Expand Down
6 changes: 6 additions & 0 deletions tests/unittest/_torch/executor/test_kv_pool_rebalance.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -907,6 +907,7 @@ def _make_pp_loop_executor(monkeypatch, *, num_micro_batches=2, agreement=(True,
exe.enable_iter_perf_stats = False
exe.iter_counter = 0
exe._mm_encoder_item_scheduling_enabled = False
exe._pending_recompute_pause_ids = set()

# Rebalance state. _uses_kv_manager_v2() reads the explicit flag first.
exe._is_kv_manager_v2 = True
Expand Down Expand Up @@ -988,6 +989,11 @@ def _fetch_new_requests():
# fact and ends the loop.
exe._forward_step_inter_pp = MagicMock(side_effect=_ReachedForward)

# These control-flow tests run in CPU CI; CUDA synchronization is out of scope.
monkeypatch.setattr(
"tensorrt_llm._torch.pyexecutor.py_executor.torch.cuda.current_stream",
MagicMock(),
)
monkeypatch.setattr(
"tensorrt_llm._torch.pyexecutor.py_executor.torch.cuda.set_device", MagicMock()
)
Expand Down
Loading