PullRequest: 42 Fix the error raising logic and the bug when using sglang with tensor parallel

Merge branch fw/patch20250317-3 of git@code.alipay.com:inclusionAI/AReaL.git into main
https://code.alipay.com/inclusionAI/AReaL/pull_requests/42

Signed-off-by: 郭唯 <kira.gw@antgroup.com>


* .

realhf/experiments/common/common.py

@@ -365,6 +365,10 @@ class CommonExperimentConfig(Experiment):
             ),
         )
 
+    @property
+    def _allocation_mode(self):
+        return AllocationMode.from_str(self.allocation_mode)
+
     def _get_rpc_allocations(self) -> List[RPCAllocation]:
         if self.allocation_mode == "manual" and self.nodelist is None:
             logger.warning(
@@ -377,8 +381,6 @@ class CommonExperimentConfig(Experiment):
 
         self._check_legal_allocation_options()
 
-        self._allocation_mode = AllocationMode.from_str(self.allocation_mode)
-
         rpcs = self.rpcs
         if self.allocation_mode == "search":
             # assert self.mode == "slurm"

realhf/experiments/common/ppo_math_exp.py

@@ -268,12 +268,15 @@ class PPOMATHConfig(CommonExperimentConfig):
     @property
     def rpcs(self):
         if (
-            self.dataset.max_prompt_len + self.ppo.gen.max_new_tokens
+            (self._allocation_mode.is_decoupled_vllm() or self.actor.vllm.hybrid_train)
+            and self.dataset.max_prompt_len + self.ppo.gen.max_new_tokens
             > self.actor.vllm.max_seq_len_to_capture
+            and not self.actor.vllm.enforce_eager
         ):
             raise RuntimeError(
                 f"vllm max seq len to capture {self.actor.vllm.max_seq_len_to_capture} is "
-                f"smaller than the prompt length + generation length {self.dataset.max_prompt_len + self.ppo.gen.max_new_tokens}"
+                f"smaller than the prompt length + generation length "
+                f"{self.dataset.max_prompt_len + self.ppo.gen.max_new_tokens}"
             )
         if not os.path.exists(os.getenv("REAL_MATH_METADATA_PATH")):
             raise RuntimeError(

realhf/impl/model/backend/sglang.py

@@ -369,6 +369,7 @@ class SGLangGenerationEngine(PipelinableEngine):
     def update_weights_from_disk(self, path):
         if constants.model_parallel_rank() != 0:
             dist.barrier(group=constants.model_parallel_group())
+            return
 
         async def _fn():
             async with SGLangAPIClient(