LKML News v6.8-rc3

[GIT PULL] hotfixes for 6.8-rc3 (Andrew Morton)

https://lkml.kernel.org/r/20240128232759.7535e6774bec152556fb9730@linux-foundation.org

Andrew Morton sent a hotfixes pull request to Linus Torvalds.

[LSF/MM/BPF TOPIC] Reclaiming & documenting page flags (Matthew Wilcox)

https://lkml.kernel.org/r/Zbcn-P4QKgBhyxdO@casper.infradead.org

Matthew proposes yet another LSFMM topic for discussing what page flags actually mean altogether, for better documentation.

[PATCH v7 0/1] mm: report per-page metadata information (Sourav Panda)

https://lkml.kernel.org/r/20240129224204.1812062-1-souravpanda@google.com

Sourav from Google posted a patch for adding a new field on /proc/meminfo and /sys/devices/system/node/nodeN/vmstat called PageMetadata and nr_page_metadata respectively, which can helps understand how much memory are being used for page metadata.

[PATCH] mm: memcg: Use larger chunks for proactive reclaim (T.J. Mercier)

https://lkml.kernel.org/r/20240131162442.3487473-1-tjmercier@google.com

There was a change to memory.reclaim memcg file making that to limit the amount of pages to reclaim to maximum of 32 for avoiding over-reclaim. Becase it causes regression due to too frequent reclaim start/stop cycle, this patch remove the limit but adjusts the size proportionally.

[PATCH] mm/slab: Add slabreclaim flag to slabinfo (Fangzheng Zhang)

https://lkml.kernel.org/r/20240131094442.28834-1-fangzheng.zhang@unisoc.com

Fangzheng adds slabreclaim flag to slabinfo for an enhanced slab debugging.

[LSF/MM/BPF TOPIC] Replacing TASK_(UN)INTERRUPTIBLE with regions of uninterruptibility (David Howells)

https://lkml.kernel.org/r/2701318.1706863882@warthog.procyon.org.uk

David suggests a new LSFMM topic for replacing TASK_{INTERRUPTIBLE,KILLABLE,UNINTERRUPTIBLE} with a sort of begin/end_...() functions that defines the uninterruptibility in terms of execution range.

[PATCH v5 00/25] Transparent Contiguous PTEs for User Mappings (Ryan Roberts)

https://lkml.kernel.org/r/20240202080756.1453939-1-ryan.roberts@arm.com

As a part of the variable-sized memory blocks (folios) allocation and mappping performance improvement which aims to achieve 16K kernel performance with 4K kernel, Ryan posted a patchset for opportunistic and transparent contiguous pte mapping usage. This benefits arm64 directly, and provide cleanup/refactoring of x86 and powerpc.

Do we still need SLAB_MEM_SPREAD (and possibly others)? (Steven Rostedt)

https://lkml.kernel.org/r/20240131172027.10f64405@gandalf.local.home

Steven is exploring slab flags and found SLAB_MEM_SPREAD's only use case was removed with commit 16a1d968358a (“mm/slab: remove mm/slab.c and slab_def.h”), so he is asking if the flag is obsolte now.

[LSF/MM/BPF TOPIC] Large folios, swap and fscache (David Howells)

https://lkml.kernel.org/r/2701740.1706864989@warthog.procyon.org.uk

David suggests an LSFMM topic for better handling large folios in swap, since swap is currently pretty simple.

Linux 6.8-rc3 (Linus Torvalds)

https://lkml.kernel.org/r/CAHk-=wisik=He=zySDRHq7fe6k_cOXZeZiCkR41TmbzK2KNZtg@mail.gmail.com

Linus Torvalds released the third release candidate of v6.8. He mentions it is slightly larger than usual, but no real worry at this stage.

Below is the diffstat of the releases in the last two years.

Note that the y-axis is in logarithm. I draw it using https://github.com/sjp38/relstat and https://github.com/sjp38/lazybox using below command:

$ relstat.py --since 2023-02-05 | ~/lazybox/gnuplot/plot.py \
    --data_fmt table --type labeled-lines --xtics_rotate -90 \
    --font "Times New Roman, 5pt" --ylog --pointsize 0.3

And, below is the diffstat of the -rc3 releases in the last two years.