unwaive test, a previous PR may have resolved hang

Signed-off-by: Ludwig Schneider <[email protected]>

Author: nv-lschneider

tensorrt_llm/_torch/custom_ops/torch_custom_ops.py

@@ -1690,7 +1690,8 @@ def get_valid_tactics(
         **kwargs,
     ) -> List[int]:
         valid_strategies = [
-            AllReduceStrategy.NCCL_SYMMETRIC.value,
+            # TODO: NCCL_SYMMETRIC will cause hang during tuning process
+            # AllReduceStrategy.NCCL_SYMMETRIC.value,
             AllReduceStrategy.NCCL.value,
         ]
         # Fallback in allreduceOp is set to NCCL_SYMMETRIC as default
@@ -1719,7 +1720,7 @@ def forward(
         input, residual, norm_weight, scale, bias, workspace = inputs
         if tactic == -1:
             # TODO: Use NCCL instead of NCCL_SYMMETRIC to avoid hanging during tuning process
-            tactic = AllReduceStrategy.NCCL_SYMMETRIC.value
+            tactic = AllReduceStrategy.NCCL.value
 
         return torch.ops.trtllm.allreduce(
             input,

tests/integration/test_lists/waives.txt

@@ -269,7 +269,6 @@ full:sm89/accuracy/test_disaggregated_serving.py::TestLlama3_1_8BInstruct::test_
 full:sm89/accuracy/test_disaggregated_serving.py::TestLlama3_1_8BInstruct::test_tp_pp_symmetric[MMLU-tp2pp2] SKIP (https://nvbugs/5596337)
 accuracy/test_llm_api_pytorch.py::TestQwen3_235B_A22B::test_nvfp4[latency_moe_trtllm] SKIP (https://nvbugs/5721672)
 accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8_4gpus[tp4-fp8kv=True-attn_backend=FLASHINFER-torch_compile=True] SKIP (https://nvbugs/5741304)
-unittest/_torch/multi_gpu/test_allreduce.py::test_allreduce_fusion_patterns[2-residual_rms_norm_out_quant_fp8-hidden:7168-seqlen:8192] SKIP (https://nvbugs/5741392)
 unittest/executor/test_rpc.py::TestRpcCorrectness::test_incremental_task_async SKIP (https://nvbugs/5741476)
 accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8_4gpus[pp4-fp8kv=True-attn_backend=TRTLLM-torch_compile=False] SKIP (https://nvbugs/5740377)
 accuracy/test_llm_api_pytorch.py::TestQwen3_30B_A3B::test_nvfp4[dep4_latency_moe_cutlass-torch_compile=False] SKIP (https://nvbugs/5740377)