Expand description
High-performance LRU (Least Recently Used) cache with capacity management.
This module provides an arena-based doubly-linked-list LRU cache that avoids
heap allocations per node by storing all nodes in a Vec<Option<LruNode>>
and threading prev/next indices through them. Eviction, insertion, and
lookup are all O(1) amortised.
Structsยง
- Cache
Stats - Snapshot of cache statistics.
- LruCache
- High-performance LRU cache backed by an arena of
Option<LruNode>slots.