LKML News v6.14-rc1

[PATCH] mm: Optimize TLB flushes during page reclaim (Vinay Banakar)

https://lore.kernel.org/CALf+9Yc3WjbG89uRWiDC_qYshJ5z_9WCrbEJe42Vbv+gJjs26g@mail.gmail.com

Vinay optimizes shrink_folio_list() to do TLB flush in batch instead of per page being reclaimed.

[LSF/MM/BPF TOPIC] Overhauling hot page detection and promotion based on PTE A bit scanning (Raghavendra K T)

https://lore.kernel.org/20250123105721.424117-1-raghavendra.kt@amd.com

Raghavendra proposes an LSFMMBPF topic for discussions around multiple ways to find hot pages, and possible future composition of those.

[LSF/MM/BPF TOPIC] Enhancements to Page Migration with Multi-threading and Batch Offloading to DMA (Shivank Garg)

https://lore.kernel.org/cf6fc05d-c0b0-4de3-985e-5403977aa3aa@amd.com

Shivank proposes an LSFMMBPF topic for multi-threading and batching based page migration acceleration.

[RFC PATCH] Introduce generalized data temperature estimation framework (Viacheslav Dubeyko)

https://lore.kernel.org/20250123202455.11338-1-slava@dubeyko.com

Viacheslav proposes a generalized data temperature estimation framework that estimates temperature based on how long consistent update is made to a specific file. He also proposes an LSFMMBPF topic for that: https://lore.kernel.org/20250123203319.11420-1-slava@dubeyko.com

[GIT PULL] MM updates for 6.14-rc1 (Andrew Morton)

https://lore.kernel.org/20250124162248.60104eec848619a187242392@linux-foundation.org

Andrew posted the MM pull request for 6.14-rc1. It turned out it has a build error, though. Andrew posted another one after removing the problematic patch, and the second take has merged.

[PATCH 00/27] hugetlb/CMA improvements for large systems (Frank van der Linden)

https://lore.kernel.org/20250127232207.3888640-1-fvdl@google.com

Frank posted a patch series that improves hugetlb and CMA on large systems.

[PATCH v3 0/2] mm/compaction: allow more aggressive proactive compaction (Michal Clapinski)

https://lore.kernel.org/20250127215020.4023545-1-mclapinski@google.com

Michal introduces their system uses free pages reporting-based VM memory management, and posts this patch for making proactive compaction be more aggressive.

[PATCH 0/2] mm: zswap: deprecate zbud and remove z3fold (Yosry Ahmed)

https://lore.kernel.org/Z5gdlO5pOu9XeGce@google.com

Yosry completes deprecation of z3fold by removing it after the two cycles of the deprecation, and start deprecation of zbud.

[LSF/MM/BPF TOPIC] persistent memory, tracing, KASLR, Oh my! (Steven Rostedt)

https://lore.kernel.org/20250127150214.4d715cad@gandalf.local.home

Steven proposes an LSFMMBPF topic for tracing and generic pstore, which is different from the kexec hand over topic.

[LSF/MM/BPF TOPIC] Page allocation for ASI (Brendan Jackman)

https://lore.kernel.org/20250129124034.2612562-1-jackmanb@google.com

Brendan proposes an LSFMMBPF topic for discussing how ASI can be integrated into the page allocator.

LSF/MM/BPF TOPIC] Improve KSM for code patching use-cases (Mathieu Desnoyers)

https://lore.kernel.org/427ca05b-fe39-4365-83c6-a23a0733ee69@efficios.com

Mathiew proposes an LSFMMBPF topic for improving KSM to handle libraries and executable patching use-case for CoW-causing memory usage increase and prevention of code pages sharing across processes.

[LSF/MM/BPF TOPIC] Waste-Free Per-CPU Userspace Memory Allocation (Mathiew Desnoyers)

https://lore.kernel.org/e5bb2190-2688-4491-9d17-574ff925f239@efficios.com

Mathiew proposes yet another LSFMMBPF topic for introducing librseq based per-CPU user-space memory allocator.

[RFC PATCH] mm, memcg: introduce memory.high.throttle (Waiman Long)

https://lore.kernel.org/20250129191204.368199-1-longman@redhat.com

Due to recent change on memory.high memcg file’s throttling behavior, some users prefer to use memory.max instead. Waiman introduces memory.high.throttle to give a control of the throttling behavior to the user-space.

[LSF/MM/BPF TOPIC] Migrating the un-migratable (David Hildenbrand)

https://lore.kernel.org/882b566c-34d6-4e68-9447-6c74a0693f18@redhat.com

David proposes and LSFMMBPF topic to discuss about where unmigratable pages comes, and how we can avoid it.

[LSF/MM/BPF TOPIC] KSM Enhancements: Selective KSM (Sourav Panda)

https://lore.kernel.org/CANruzcR0oN8URqHh86HLuqfiv=pax0-eQ=3_oCK-kX_cuktUGw@mail.gmail.com

Sourav proposes an LSFMMBPF topic for enhancing KSM including a way to use it with security-critical environments.

[LSF/MM/BPF TOPIC] An SRCU that is faster than RCU Tasks Trace (Paul E. McKenney)

https://lore.kernel.org/dab29bdf-1c33-416d-a5d7-fcc6829a4b60@paulmck-laptop

Paul McKenney proposes an LSFMMBPF for discussing about recent srcu change aimed to be used for uretprobe, and could also be used in may places. The aim is to find feedback and potential usages.

[LSF/MM/BPF TOPIC] Guaranteed CMA (Suren Baghadasaryan)

https://lore.kernel.org/CAJuCfpEWVEqsivd7oTvp4foEho_HaD1XNP8KTeKWzG_X2skfGQ@mail.gmail.com

Suren proposes LSFMMBPF topic for discussion of GCMA, which is being used on Android, for upstreaming it.

Linux 6.14-rc1 (Linus Torvalds)

https://lore.kernel.org/CAHk-=wicYNCkEgH06w0mpR+GJhJ_ywe0BLTTFXBAj1+y0dqe4Q@mail.gmail.com

Linus Torvalds closed the mergewindow for Linux 6.14 and released the first candidate. He mentions it was a tiny merge window, maybe due to the holidays, and it is a good thing.