A high-throughput and memory-efficient inference and serving engine for LLMs
[CI][BugFix] Fix broken `test_mamba_prefix_cache.py` due to stale mock (#45345)
Signed-off-by: Nick Hill <nickhill123@gmail.com>
Nick Hill committed
Jun 11, 2026 at 20:26 UTC
2263f8a3de64f4cf16488fb43369714c736612a0
1 file changed
+2
tests/v1/e2e/general/test_mamba_prefix_cache.py
+2
@@ -181,6 +181,7 @@ def get_fake_allocate_slots_fn(original_allocate_slots_fn: Callable):
181
num_encoder_tokens: int = 0,
182
full_sequence_must_fit: bool = False,
183
reserved_blocks: int = 0,
184
+ has_scheduled_reqs: bool = True,
185
):
186
ret = original_allocate_slots_fn(
187
self,
@@ -194,6 +195,7 @@ def get_fake_allocate_slots_fn(original_allocate_slots_fn: Callable):
195
num_encoder_tokens,
196
full_sequence_must_fit,
197
reserved_blocks,
198
+ has_scheduled_reqs,
199
)
200
if cur_step_action is not None:
201
cur_block_ids = self.coordinator.single_type_managers[0].req_to_blocks[