Update combine config

deep_ep/buffer.py

@@ -247,7 +247,7 @@ class Buffer:
             2: Config(Buffer.num_sms, 10, 256, 6, 128),
             4: Config(Buffer.num_sms, 9, 256, 6, 128),
             8: Config(Buffer.num_sms, 4, 256, 6, 128),
-            16: Config(Buffer.num_sms, 4, 288, 16, 128),
+            16: Config(Buffer.num_sms, 4, 288, 12, 128),
             24: Config(Buffer.num_sms, 1, 288, 8, 128),
             32: Config(Buffer.num_sms, 1, 288, 8, 128),
             64: Config(Buffer.num_sms, 1, 288, 20, 128),

tests/test_internode.py

@@ -209,8 +209,8 @@ def test_main(args: argparse.Namespace, num_sms: int,
 
     # Tune combine performance
     best_time, best_results = 1e10, None
-    for nvl_chunk_size in range(1, 13, 1):
-        for rdma_chunk_size in range(8, 33, 4):
+    for nvl_chunk_size in range(1, 8, 1):
+        for rdma_chunk_size in range(12 if num_nodes == 2 else 8, 33, 4):
             config = deep_ep.Config(num_sms, nvl_chunk_size, nvl_buffer_size, rdma_chunk_size, rdma_buffer_size)
             tune_args = {'x': recv_x, 'handle': handle, 'config': config}
             t, notify_t = bench_kineto(lambda: buffer.combine(**tune_args), ('combine', 'notify'))