[$] What is to be done about MGLRU?
The article discusses the challenges and proposed solutions regarding the multi-generational LRU (MGLRU) in Linux memory management. Experts at the 2026 Linux Summit highlighted the need to unify the two existing memory reclaim algorithms, traditional LRU and MGLRU, to improve efficiency and reduce code duplication. A plan was outlined to separate the code bases, evaluate performance, and identify common features between the two implementations.
- ▪MGLRU was introduced to enhance memory reclaim but has complicated the memory management landscape.
- ▪Experts at the Linux Summit emphasized the need to unify the traditional LRU and MGLRU to streamline the code and improve performance.
- ▪The proposed plan includes separating the code bases, defining evaluation workloads, and identifying common features between the two algorithms.
Opening excerpt (first ~120 words) tap to expand
Did you know...? LWN.net is a subscriber-supported publication; we rely on subscribers to keep the entire operation going. Please help out by buying a subscription and keeping LWN on the net. By Jonathan CorbetMay 20, 2026 LSFMM+BPF "Reclaim" is the task of finding memory that can be taken away from its current user and put to better uses within the system; it is a core part of the memory-management picture. The addition of the multi-generational LRU (MGLRU) was meant to provide a better reclaim implementation than the "traditional LRU" that preceded it, but MGLRU has complicated the situation instead.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at LWN.net (Linux Weekly News).