EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

[P/D][BugFix][v0.11.0-dev]Fix proxy format processing errors & Layerwise connector performance optimization (#4069)

Browse Source 
### What this PR does / why we need it?
1.Fix proxy format processing errors.
2.Layer-wise connector performance optimization

Signed-off-by: wangxiaoteng <wangxiaoteng@huawei.com>
This commit is contained in:
wangxiaoteng888
2025-11-09 09:55:10 +08:00
committed by GitHub
parent 55e37f5041
commit c2d58c0655
3 changed files with 16 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
tests/ut/kv_connector/test_mooncake_layerwise_connector.py
View File

@@ -32,6 +32,14 @@ class TestKVCacheSendingLayerThread(unittest.TestCase):
self.engine = MagicMock()
self.engine.register_memory.return_value = 0
self.engine.batch_transfer_sync_write.return_value = 1
self._patcher_cs = patch(
'vllm_ascend.distributed.mooncake_layerwise_connector.torch_npu.npu.current_stream'
)
self.mock_current_stream = self._patcher_cs.start()
self.addCleanup(self._patcher_cs.stop)
fake_stream = MagicMock(name="FakeStream")
fake_stream.synchronize = MagicMock()
self.mock_current_stream.return_value = fake_stream
self.first_kv_cache = torch.zeros((2, 2, 2, 8),
dtype=torch.float32,