A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 16, 07:00 UTC
[ci][distributed] disable hanging tests (#10317)
Signed-off-by: youkaichao <youkaichao@gmail.com>
youkaichao committed
Nov 14, 2024 at 00:23 UTC
29f3ef26a38e5afab529fb9f6098704fd106a779
1 file changed
+1
tests/distributed/test_utils.py
+1
@@ -117,6 +117,7 @@ def allgather_worker(rank, WORLD_SIZE, port1, port2):
117
pg1.barrier()
118
119
120
+@pytest.mark.skip(reason="This test is flaky and prone to hang.")
121
@multi_gpu_test(num_gpus=4)
122
@pytest.mark.parametrize(
123
"worker", [cpu_worker, gpu_worker, broadcast_worker, allgather_worker])