Remove sampling info events and overlap thread file (#11300)

python/sglang/srt/disaggregation/prefill.py

@@ -338,21 +338,8 @@ class SchedulerDisaggregationPrefillMixin:
                 result = self.run_batch(batch)
                 self.result_queue.append((batch.copy(), result))
 
-                if self.last_batch is None:
-                    # Create a dummy first batch to start the pipeline for overlap schedule.
-                    # It is now used for triggering the sampling_info_done event.
-                    tmp_batch = ScheduleBatch(
-                        reqs=None,
-                        forward_mode=ForwardMode.DUMMY_FIRST,
-                        next_batch_sampling_info=self.tp_worker.cur_sampling_info,
-                    )
-                    self.set_next_batch_sampling_info_done(tmp_batch)
-
             if self.last_batch:
                 tmp_batch, tmp_result = self.result_queue.popleft()
-                tmp_batch.next_batch_sampling_info = (
-                    self.tp_worker.cur_sampling_info if batch else None
-                )
                 self.process_batch_result_disagg_prefill(tmp_batch, tmp_result)
 
             if len(self.disagg_prefill_inflight_queue) > 0:
@@ -491,8 +478,6 @@ class SchedulerDisaggregationPrefillMixin:
                 if self.enable_overlap:
                     self.send_kv_chunk(req, last_chunk=False, end_idx=req.tmp_end_idx)
 
-        # We need to remove the sync in the following function for overlap schedule.
-        self.set_next_batch_sampling_info_done(batch)
         self.maybe_send_health_check_signal()
 
     def process_disagg_prefill_inflight_queue(