Skip to content

fix: force gc + clear_cache after KV prefix cache eviction #1832

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
adurham wants to merge 1 commit into
base: main
Choose a base branch
from
+10 −0
Open

fix: force gc + clear_cache after KV prefix cache eviction #1832

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 10 additions & 0 deletions src/exo/worker/engines/mlx/cache.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,3 +1,4 @@
import gc
import os
from copy import deepcopy
from typing import TYPE_CHECKING
Expand Down Expand Up @@ -266,6 +267,7 @@ def _evict_if_needed(self):
if len(self.caches) == 0:
return

evicted_any = False
# Evict LRU entries until below threshold
while (
len(self.caches) > 0
Expand All @@ -278,10 +280,18 @@ def _evict_if_needed(self):
self._snapshots.pop(lru_index)
self._media_regions.pop(lru_index)
self._last_used.pop(lru_index)
evicted_any = True
logger.info(
f"KV cache evicted LRU entry ({evicted_tokens} tokens) due to memory usage"
)

if evicted_any:
# Force Python GC to release array references, then clear Metal buffer cache.
# Without this, evicted MLX arrays stay allocated until the next GC cycle,
# leaking ~3-4 GB between long-context requests.
gc.collect()
mx.clear_cache()

def get_memory_used_percentage(self) -> float:
local_pressure: float = get_memory_used_percentage()

Expand Down