Adapt Moss sampling and runner fixtures to vLLM 0.31 interfaces
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
This commit is contained in:
1 parent
0b6591ae0a
commit
b256e1ec4e
6 files changed
+50
-15
No files matched your search
@@ -16,6 +16,7 @@ def test_dispatch_eager_and_manager_paths(monkeypatch, dp_size):
|
||||
runner = object.__new__(OmniGPUModelRunner)
|
||||
runner.cudagraph_manager = None
|
||||
runner.parallel_config = SimpleNamespace(data_parallel_size=dp_size, data_parallel_rank=1)
|
||||
runner.dp_size, runner.dp_rank = dp_size, 1
|
||||
|
||||
import vllm.v1.worker.gpu.dp_utils as dp_utils
|
||||
|
||||
@@ -52,6 +53,7 @@ def test_dispatch_eager_and_manager_paths(monkeypatch, dp_size):
|
||||
|
||||
runner.cudagraph_manager = SimpleNamespace(dispatch=dispatch)
|
||||
runner.parallel_config.data_parallel_size = 1
|
||||
runner.dp_size = 1
|
||||
batch_desc, _ = runner._dispatch_batch_descriptor(
|
||||
num_reqs=1, num_toks=8, uniform_tok_count=8, num_active_loras=0, use_eager=False, max_query_len=8
|
||||
)
|
||||
|
||||
Reference in new issue
Block a user