[PATCH v6] mm: Proactive compaction (Nitin Gupta) https://lkml.kernel.org/r/20200601194822.30252-1-nigupta@nvidia.com
This is the sixth version of the proactive compaction patchset. It makes compaction to be able to triggered earlier than final memory pressure using knobs. The goal is better THP allocation success.
[PATCH] vmalloc: Convert to XArray (Matthew Wilcox) https://lkml.kernel.org/r/20200603171448.5894-1-willy@infradead.org
This patchset converts the radix tree for vmap blocks into XArray.
incoming (Andrew Morton) https://lkml.kernel.org/r/20200608212922.5b7fa74ca3f4e2444441b7f9@linux-foundation.org
MM-side pull request. It contains the “mmap locking API: initial implementation as rwsem wrappers” patchset.
[PATCH v2 00/16] Introduce kvfree_rcu(1 or 2 arguments) (Uladzislau Rezki) https://lkml.kernel.org/r/20200525214800.93072-1-urezki@gmail.com
This is the second version of the rcu-protected kvfree().
[PATCH v2 0/7] Add histogram measuring mmap_lock contention latency (Axel Rasmussen) https://lkml.kernel.org/r/20200528235238.74233-1-axelrasmussen@google.com
For further analysis of the mmap_sem overhead from both kernel space and user space, this commit adds the latency historgram for mmap_sem acquisition time. Actually, mmap_sem is now renamed into mmap_lock, thanks to Michel’s patch.
[PATCH -V4] swap: Reduce lock contention on swap cache from swap slots allocation (Huang, Ying) https://lkml.
[PATCH v5] mm: Proactive compaction (Nitin Gupta) https://lkml.kernel.org/r/20200518181446.25759-1-nigupta@nvidia.com
The 5th version of the proactive compaction. This patchset make the compaction more proactive to make THP allocation easily success.
[PATCH -V2] swap: Reduce lock contention on swap cache from swap slots allocation (Huang Ying) https://lkml.kernel.org/r/20200520031502.175659-1-ying.huang@intel.com
After swap device is fragmented, there’s no free swap cluster. Therefore, each swap logic of each CPU will linearly scan each swap cluster to find a free slot.
A DAMON showcase website[1] is open. There are
the official documentation of DAMON[2], the heatmap format dynamic access pattern of various realistic workloads for heap area[3], mmap()-ed area[4], and stack[5] area, the dynamic working set size distribution[6] and chronological working set size changes[7], and the latest performance test results[8]. [1] https://damonitor.github.io
[2] https://damonitor.github.io/doc/html/latest
[3] https://damonitor.github.io/test/result/visual/latest/heatmap.0.html
[4] https://damonitor.github.io/test/result/visual/latest/heatmap.1.html
[5] https://damonitor.github.io/test/result/visual/latest/heatmap.2.html
[6] https://damonitor.github.io/test/result/visual/latest/wss_sz.html
[7] https://damonitor.github.io/test/result/visual/latest/wss_time.html
[8] https://damonitor.github.io/test/result/perf/latest/html/index.html
[PATCH v4 00/36] Large pages in the page cache (Matthew Wilcox) https://lkml.kernel.org/r/20200515131656.12890-1-willy@infradead.org
This is the fourth version of the patchset for the huge pages in the page cache. Still it fails xfstests.
Maintainers / Kernel Summit 2020 planning kick-off (Theodore Y. Ts’o) https://lkml.kernel.org/r/20200515163956.GA2158595@mit.edu
Ted annouced that the Maintainers / Kernel Summit will be held in virtual this year, of course due to the COVID19. With this announcement, the CFP for the kernel summit is also opened.
[RFC 00/43] PKRAM: Preserved-over-Kexec RAM (Anthony Yznaga) https://lkml.kernel.org/r/1588812129-8596-1-git-send-email-anthony.yznaga@oracle.com
This patchset introduces special memory called PKRAM. It can be used to preserve VM guest memory across kexec boot for fast restart (VMM Fast Restart). Compared to DRAM based PMEM simulation, PKRAM is more flexible as the memory size is not need to be previously fixed.
[PATCH] mm: Replace zero-length array with flexible-array (Gustavo A. R. Silva) https://lkml.kernel.org/r/20200507185942.GA15193@embeddedor
The zero-length array of C90 are widely used in the kernel code base, but the flexible array member of C99 would be better choice.
[PATCH 0/7] Record the mm_struct in the page table pages (Matthew Wilcox) https://lkml.kernel.org/r/20200428194449.22615-1-willy@infradead.org
Literally, this patchset adds the mm_struct pointer to struct page if it is used for page tables. This will be useful for several cases usch as uncorrectable errors handling and task-page table relation checking, etc.
[PATCH v4] mm: Proactive compaction (Nitin Gupta) https://lkml.kernel.org/r/20200428221055.598-1-nigupta@nvidia.com
This patchset makes the compaction to be more proactively done for better THP allocation success and makes user to be able to control the proactiveness using a sysfs file, /sys/kernel/mm/compaction/proactiveness.
LSFMMBPF 2020 Cancellation announcement (Josef Bacik) https://lkml.kernel.org/r/0b6d3d6f-99de-3603-4b42-c3db5113633d@toxicpanda.com
Due to the COVID-19, LSFMM, which was supposed to be help April, has delayed before, and now it has just canceled. What a shame, but this might be the right thing to do. This mail also announce the schedule of next year LSFMM. It will be held in Palm Springs, on May 12-14, 2021 at the Riviera Palm Springs.
[PATCH 2/2] mm: support vector address ranges for process_madvise (Minchan Kim) https://lkml.
[RFC] autonuma: Support to scan page table asynchronously (Huang Ying) https://lkml.kernel.org/r/20200414081951.297676-1-ying.huang@intel.com
AutoNUMA is currently scanning page tables of the processes in the process context and therefore this can increase the latency of the process. This RFC thus makes the page table scanning asynchronous if there are idle CPUs.
[PATCH v4 00/10] Add a new mmap locking API wrapping mmap_sem calls https://lkml.kernel.org/r/20200415004353.130248-1-walken@google.com
This is just a repost of the patch series rebased on v5.
[GIT PULL] XArray for 5.7-rc1 https://lkml.kernel.org/r/20200331141749.GB21484@bombadil.infradead.org
Matthew’s XArray pull request for 5.7 merge window. Quickly merged by Torvald.
[PATCH v5 00/10] workingset protection/detection on the anonymous LRU list https://lkml.kernel.org/r/1585892447-32059-1-git-send-email-iamjoonsoo.kim@lge.com
The fifth version of the workingset protection.
[PATCH v4 0/2] mm: using CMA for 1 GB hugepages allocation https://lkml.kernel.org/r/20200407010431.1286488-1-guro@fb.com
This patchset make hugetlb to use CMA for 1GB hugepage allocations.
[PATCH RFC] mm: compaction: avoid migrating non-cma pages to a cma area The compaction algorithm doesn’t distinguish CMA from other movable pageblocks.