lkml news

LKML News v5.19-rc6

[PATCH v9 0/9] x86: Show in sysfs if a memory node is able to do encryption (Martin Fernandez) https://lkml.kernel.org/r/20220704135833.1496303-1-martin.fernandez@eclypsium.com This patchset makes each memory descriptor node to show if it has the EFI_MEMORY_CPU_CRYPRO attribute. [PATCH v3] arm64: enable THP_SWAP for arm64 (Barry Song) https://lkml.kernel.org/r/20220706072707.114376-1-21cnbao@gmail.com THP_SWAP, which dealys the split of THP for swap, includes swap throughput and architecture independent, but enabled on x86_64 only. This patch enables it on arm64.

LKML News v5.19-rc5

[PATCH v3] docs: rename Documentation/vm to Documentation/mm (Mike Rapoport) https://lkml.kernel.org/r/20220627060026.180439-1-rppt@kernel.org As the subject says, this commit renames the Documentation/vm directory to Documentation/mm. [PATCH v6 0/4] KVM: mm: count KVM mmu usage in memory stats (Yosry Ahmed) https://lkml.kernel.org/r/20220628220938.3657876-1-yosryahmed@google.com This patchset allows users account KVM mmu usages with NR_SECONDARY_PAGETABLE memory stat. [RFC PATCH] mm/slub: enable debugging memory wasting of kmalloc (Feng Tang) https://lkml.kernel.org/r/20220630014715.73330-1-feng.tang@intel.com As kmalloc allocates memory in power of 2 granularity, it could waste a lot of memory.

LKML News v5.19-rc4

[PATCH Part2 v6 00/49] Add AMD Secure Nested Paging (SEV-SNP) (Ashsh Kalra) https://lkml.kernel.org/r/cover.1655761627.git.ashish.kalra@amd.com This is second part patchset of Secure Encrypted Paging (SEV-SNP) work, which focuses on the changes required in a host OS for the feature support. [PATCH v7 00/12] mm/demotion: Memory tiers and demotion (Aneesh Kumar K.V) https://lkml.kernel.org/r/20220622082513.467538-1-aneesh.kumar@linux.ibm.com This patchset improves memory tiers construction to be more explicit and flexible, and memory allocation for demotion to have no change.

LKML News v5.19-rc3

[PATCH 0/8] Extend DAMOS for Proactive LRU-lists Sorting (SeongJae Park) https://lkml.kernel.org/r/20220613192301.8817-1-sj@kernel.org After the RFC, official patchset for DAMON-based LRU-lists sorting has posted. This immediately merged in the -mm tree. [PATCH v12 00/14] Multi-Gen LRU Framework (Yu Zhao) https://lkml.kernel.org/r/20220614071650.206064-1-yuzhao@google.com A new version of MGLRU has posted. Compared to the previous version, this version fixes a bug and nits, and contains new test results based on v5.19-rc1. Maintainer’s / Kernel Summit 2022 CFP (Theodore Ts’o) https://lkml.

LKML News v5.19-rc2

[PATCH v3 0/6] introduce mirrored memory support for arm64 (Wupeng Ma) https://lkml.kernel.org/r/20220607093805.1354256-1-mawupeng1@huawei.com This patchset adds memory mirroring support, which relies on UEFI and used to separate memory into two separate channels, to arm64. [PATCH] mm: mempolicy: N:M interleave policy for tiered memory nodes (Johannes Weiner) https://lkml.kernel.org/r/20220607171949.85796-1-hannes@cmpxchg.org Current pages NUMA-interleaving policy is 1:1 mapping, 1 for CPU node, 1 for remote note. As more types of tiered memory systems having CPU-less memory nodes are coming, this patchset introduces an N:M interleave policy, which N pages are allocated from the top-tier node while M pages are allocated from low-tier nodes.

LKML News v5.19-rc1

[PATCH] arm64: enable THP_SWAP for arm64 (Barry Song) https://lkml.kernel.org/r/20220524071403.128644-1-21cnbao@gmail.com This patch enables THP_SWAP, which can help swap performance on systems having fast swap storages, for arm64. [CFP LPC 2022] Kernel Memory Management Microconference (Vlastimil Babka) https://lkml.kernel.org/r/d24e9ac9-0903-3c15-c446-2962f44a360f@suse.cz This year, LPC will have kernel memory management microconference. It will supplement the LSF/MM and lead by Matthew Wilcox and Vlastimil Babka. [PATCH v2] arm64: enable THP_SWAP for arm64 (Barry Song) https://lkml.kernel.org/r/20220527100644.293717-1-21cnbao@gmail.com Second version of the THP_SWAP enablement for arm64 has posted.

LKML News v5.18

[RFC] Add swappiness argument to memory.reclaim (Yosry Ahmed) https://lkml.kernel.org/r/CAJD7tkbDpyoODveCsnaqBBMZEkDvshXJmNdbk51yKSNgD7aGdg@mail.gmail.com Yosry is proposing to add swappiness argument to memory.reclaim cgroup file to provide a way to enforce user policy on a stateless per-reclaim basis. [PATCH v11 00/14] Multi-Gen LRU Framework (Yu Zhao) https://lkml.kernel.org/r/20220518014632.922072-1-yuzhao@google.com The eleventh version of the patchset for MGLRU has posted. Maybe this has revised based on the comments from LSFMM. [PATCH] memcg: provide reclaim stats via ‘memory.reclaim’ (Vaibhav Jain) https://lkml.

LKML News v5.18-rc7

[PATCH v3 0/6] mm: introduce shrinker debugfs interface (Roman Gushchin) https://lkml.kernel.org/r/20220509183820.573666-1-roman.gushchin@linux.dev This patchset introduces a descent debugfs interface for debugging the shrinker mechanism. [PATCH v3 0/3] Fix CONT-PTE/PMD size hugetlb issue when unmapping or migrating (Baolin Wang) https://lkml.kernel.org/r/cover.1652147571.git.baolin.wang@linux.alibaba.com For PTE-cont or PMD-cont size hugetlb pages, migration or unmapping of poisoned hugetlb pages which are using ptep_clear_flush() and set_pte_at() can cause data consistency issue. This patchset tries to fix the issue.

LKML News v5.18-rc6

Memory allocation on speculative fastpaths (Paul E. McKenney) https://lkml.kernel.org/r/20220503155913.GA1187610@paulmck-ThinkPad-P17-Gen-1 There was a discussion about allocating memory inside RCU read critical section with GFP_KERNEL during LSFMM. Paul is summing up his points about it in this mail. [PATCH] mm,doc: Add new documentation structure (Matthew Wilcox) https://lkml.kernel.org/r/20220503063032.3680308-1-willy@infradead.org There was a session for MM documentation in the LSFMM. Matthew is restructuring the Documentation files to be more similar to Mel Gorman’s “Understanding the Linux Virtual Memory Manager”.

LKML News v5.18-rc5

[PATCH v5 0/4] memcg: introduce per-memcg proactive reclaim (Yosry Ahmed) https://lkml.kernel.org/r/20220425190040.2475377-1-yosryahmed@google.com The fifth version of the patchset for adding per-memch proactive reclamation trigger interface. mmotm 2022-04-25-17-59 uploaded (Andrew Morton) https://lkml.kernel.org/r/20220426005932.848CBC385A4@smtp.kernel.org The mmotm has updated. It has rebased on 5.18-rc4, added 62 new patches, modified 8 patches, and dropped 19 patches. In total, 334 patches in the queue. [PATCH -V2 0/3] memory tiering: hot page selection (Huang Ying) https://lkml.kernel.org/r/20220426085105.60822-1-ying.huang@intel.com This patchset adds hot pages selection to the tiered memory promotion mechanism for better performance.