From afdd1d0992900ea09074e2afed47c18831bdbdde Mon Sep 17 00:00:00 2001 From: leavelet Date: Wed, 10 Jun 2026 10:22:35 +0000 Subject: [PATCH] Demote per-request load-back log to debug init_load_back logged 'loading back N tokens for node M' at INFO for every request triggering an L2->L1 HiCache load-back (hi_mamba already had it at debug). Demote and switch both to lazy %-formatting so the f-string is not built when debug is off. Co-Authored-By: Claude Fable 5 --- python/sglang/srt/mem_cache/hi_mamba_radix_cache.py | 4 +++- python/sglang/srt/mem_cache/hiradix_cache.py | 6 ++++-- 2 files changed, 7 insertions(+), 3 deletions(-) diff --git a/python/sglang/srt/mem_cache/hi_mamba_radix_cache.py b/python/sglang/srt/mem_cache/hi_mamba_radix_cache.py index 84ae861f6..f3107dc6e 100644 --- a/python/sglang/srt/mem_cache/hi_mamba_radix_cache.py +++ b/python/sglang/srt/mem_cache/hi_mamba_radix_cache.py @@ -265,7 +265,9 @@ class HiMambaRadixCache(MambaRadixCache): loading_values = self.load_back(last_node, mem_quota) if loading_values is not None: logger.debug( - f"loading back {len(loading_values)} tokens for node {last_node.id}" + "loading back %d tokens for node %d", + len(loading_values), + last_node.id, ) return loading_values, last_node diff --git a/python/sglang/srt/mem_cache/hiradix_cache.py b/python/sglang/srt/mem_cache/hiradix_cache.py index f67ac56aa..b90377c98 100644 --- a/python/sglang/srt/mem_cache/hiradix_cache.py +++ b/python/sglang/srt/mem_cache/hiradix_cache.py @@ -3737,8 +3737,10 @@ class HiRadixCache(RadixCache): if last_node.evicted: loading_values = self.load_back(last_node, mem_quota) if loading_values is not None: - logger.info( - f"loading back {len(loading_values)} tokens for node {last_node.id}" + logger.debug( + "loading back %d tokens for node %d", + len(loading_values), + last_node.id, ) return loading_values, last_node