refactor: add a check before layer_sharding logging (#7186)

### What this PR does / why we need it?
We should only display this log message when layer_sharding is enabled.
- vLLM version: v0.16.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/4034c3d32e30d01639459edd3ab486f56993876d

Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>
This commit is contained in:
Qiu authored and GitHub committed 2026-03-12 11:56:04 +08:00
1 parent 5f3826b093
commit aa0143e55d
1 file changed
+6 -5
+6 -5
View File
@@ -52,11 +52,12 @@ class AscendConfig:
self.dump_config_path = additional_config.get("dump_config_path", None)
self._construct_weight_prefetch_config(additional_config)
self.layer_sharding = additional_config.get("layer_sharding", None)
logger.info_once(
f"Linear layer sharding enabled with config: {self.layer_sharding}. "
"Note: This feature works optimally with FLASHCOMM2 and DSA-CP enabled; "
"using it without these features may result in significant performance degradation."
)
if self.layer_sharding:
logger.info_once(
f"Linear layer sharding enabled with config: {self.layer_sharding}. "
"Note: This feature works optimally with FLASHCOMM2 and DSA-CP enabled; "
"using it without these features may result in significant performance degradation."
)
self.enable_shared_expert_dp = (
additional_config.get("enable_shared_expert_dp", False)