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 15, 02:15 UTC
[Misc] Restrict deep_gemm's log output (#20827)
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
Jee Jee Li committed
Jul 12, 2025 at 11:50 UTC
44d02f54db522fc489d3590d5de9461a52621460
1 file changed
+1
-1
vllm/model_executor/layers/fused_moe/deep_gemm_moe.py
+1
-1
@@ -43,7 +43,7 @@ def _valid_deep_gemm(hidden_states: torch.Tensor, w1: torch.Tensor,
43
aligned by `dg.get_m_alignment_for_contiguous_layout()`.
44
"""
45
if not has_deep_gemm():
46
- logger.debug("DeepGemm disabled: deep_gemm not available.")
46
+ logger.debug_once("DeepGemm disabled: deep_gemm not available.")
47
return False
48
49
M = hidden_states.size(0)