diff --git a/csrc/engine/scheduler.hpp b/csrc/engine/scheduler.hpp
index b24fe19..e82cbb7 100644
--- a/csrc/engine/scheduler.hpp
+++ b/csrc/engine/scheduler.hpp
@@ -310,7 +310,11 @@ struct Scheduler {
             if (!is_any_point && !is_all_point) {
                 return _schedule_decode(acquire);
             }
-            bool want = (get_gpu_cache_usage() <= GPU_CACHE_PREFILL_WATERMARK) && _has_prefill_waiting();
+            // A prompt prefilled in chunks reserves the blocks of its whole prompt with its first chunk, and only its remaining chunks can
+            // give them back. Holding it to the watermark would stall it for as long as its own blocks keep the usage above it, which is
+            // for good if it needs more than that share of the cache by itself.
+            const bool continues_chunked_prefill = !waiting.empty() && waiting.front()->schedule.chunked_rest_len > 0;
+            bool want = _has_prefill_waiting() && (continues_chunked_prefill || get_gpu_cache_usage() <= GPU_CACHE_PREFILL_WATERMARK);
             bool do_prefill;
             if (is_any_point) {
                 do_prefill = (_dp_allreduce(want ? 0 : 1, c10d::ReduceOp::MIN) == 0); // ANY：任一 rank 想做即触发
diff --git a/tests/engine/test_scheduler_resources.py b/tests/engine/test_scheduler_resources.py
index 02a1b16..b2fe054 100644
--- a/tests/engine/test_scheduler_resources.py
+++ b/tests/engine/test_scheduler_resources.py
@@ -256,3 +256,20 @@ def test_preempted_request_is_not_queued_behind_a_sequence_that_cannot_be_schedu
     engine.run()
     assert multimodal.status.name == "FINISHED_LENGTH_CAPPED"
     engine.assert_all_resources_returned()
+
+
+def test_prompt_that_takes_nearly_the_whole_cache_is_prefilled_to_the_end(
+    cpu_group: dist.ProcessGroup, monkeypatch: pytest.MonkeyPatch
+) -> None:
+    # 10 usable blocks of 4 tokens. The first chunk reserves all of them, far above the usage at which new prompts are admitted.
+    engine = _Engine(cpu_group, monkeypatch, num_blocks=11, block_size=4, max_num_batched_tokens=8)
+    prompt = engine.add_request(prompt_len=37, max_tokens=2)  # 5 chunks of at most 8 tokens
+
+    engine.step()
+    assert prompt.status.name == "WAITING"
+    assert engine.scheduler.get_num_free_gpu_blocks() == 0
+
+    engine.run()
+
+    assert prompt.status.name == "FINISHED_LENGTH_CAPPED"
+    engine.assert_all_resources_returned()