xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-21 04:05:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Log] Optimize kv cache memory log from Bytes to GiB (#25204)

Browse Source 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
This commit is contained in:
Wentao Ye 2025-09-23 12:44:37 -04:00 committed by GitHub
parent 2357480b1a
commit 846197f505
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 6 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
vllm/v1/worker/gpu_worker.py
View File

@ -383,11 +383,13 @@ class Worker(WorkerBase):
f"for non-torch memory, and {GiB(cuda_graph_memory_bytes)} " f"for non-torch memory, and {GiB(cuda_graph_memory_bytes)} "
f"GiB for CUDAGraph memory. Replace gpu_memory_utilization " f"GiB for CUDAGraph memory. Replace gpu_memory_utilization "
f"config with `--kv-cache-memory=" f"config with `--kv-cache-memory="
f"{kv_cache_memory_bytes_to_requested_limit}` to fit into " f"{kv_cache_memory_bytes_to_requested_limit}` "
f"requested memory, or `--kv-cache-memory=" f"({GiB(kv_cache_memory_bytes_to_requested_limit)} GiB) to fit "
f"{kv_cache_memory_bytes_to_gpu_limit}` to fully " f"into requested memory, or `--kv-cache-memory="
f"{kv_cache_memory_bytes_to_gpu_limit}` "
f"({GiB(kv_cache_memory_bytes_to_gpu_limit)} GiB) to fully "
f"utilize gpu memory. Current kv cache memory in use is " f"utilize gpu memory. Current kv cache memory in use is "
f"{int(self.available_kv_cache_memory_bytes)} bytes.") f"{GiB(self.available_kv_cache_memory_bytes)} GiB.")
logger.debug(msg) logger.debug(msg)