[AMD] Scope the EAGLE greedy-verify TP broadcast to ROCm only (#35195)

python/sglang/srt/speculative/eagle_utils.py

@@ -737,6 +737,21 @@ def eagle_sample(
             target_predict=target_predict,
             topk=verify_input.tree_topk,
         )
+
+        if _is_hip:
+            # On ROCm, the per-rank draft tokens can differ, so ranks accept a
+            # different number of drafts, desynchronize the committed seq_lens, and
+            # deadlock the next TP collective. Broadcast from rank 0 to ensure
+            # consistency, the same way the sampling branch below does.
+            tp_group = (
+                get_parallel().attn_tp_group
+                if is_dp_attention_enabled()
+                else get_tp_group()
+            )
+            if tp_group.world_size > 1:
+                tp_group.broadcast(predict, src=0)
+                tp_group.broadcast(accept_index, src=0)
+                tp_group.broadcast(num_correct_drafts, src=0)
     else:
         from sgl_kernel import (
             top_k_renorm_prob,
@@ -824,17 +839,19 @@ def eagle_sample(
             deterministic=True,
         )
 
-    # Sync the verify decision across TP ranks: small per-rank differences in the
+        # Sync sampling results across TP ranks: different GPUs may
-    # tensors feeding this point can make one rank accept a different number of
+        # produce slightly different target_probs due to floating-point
-    # drafts, which desynchronizes the committed seq_lens and deadlocks the next
+        # non-determinism in softmax/top_k/top_p, causing different
-    # TP collective. Broadcast from rank 0 to ensure consistency.
+        # sampled tokens. Broadcast from rank 0 to ensure consistency.
-    tp_group = (
+        tp_group = (
-        get_parallel().attn_tp_group if is_dp_attention_enabled() else get_tp_group()
+            get_parallel().attn_tp_group
-    )
+            if is_dp_attention_enabled()
-    if tp_group.world_size > 1:
+            else get_tp_group()
-        tp_group.broadcast(predict, src=0)
+        )
-        tp_group.broadcast(accept_index, src=0)
+        if tp_group.world_size > 1:
-        tp_group.broadcast(num_correct_drafts, src=0)
+            tp_group.broadcast(predict, src=0)
+            tp_group.broadcast(accept_index, src=0)
+            tp_group.broadcast(num_correct_drafts, src=0)
 
     if SIMULATE_ACC_LEN > 0:
         # Do simulation. The helper builds (and returns) a replacement