[PATCH] mm: mmap_lock: fix use-after-free race and css ref leak in tracepoints (Axel Rasmussen) https://email@example.com
This patch fixes a use-after-free bug that reported by syzbot.
[PATCH 0/6] prohibit pinning pages in ZONE_MOVABLE (Pavel Tatashin) https://firstname.lastname@example.org
Pinned pages in ZONE_MOVABLE is a wellknown problem. This patchset fixes the issue by forcing pinning new page to allocated from a zone other than ZONE_MOVABLE and migrating out existing pages from the zone during pinning.
[PATCH v2] mm: memcontrol: account pagetables per node (Shakeel Butt) https://email@example.com
This commit allows user to account memory size used for page tables per node.
[PATCH 0/2] userspace memory reaping using process_madvise (Suren Baghdasaryan) https://firstname.lastname@example.org
This commit allows user-space driven OOM reaping using process_madvise(MADV_DONTNEED).
[PATCH -V6 0/3] autonuma: Migrate on fault among multiple bound nodes (Huang Ying) https://email@example.com
The 6th version for autonuma migration among multiple bound nodes. In this version, more benchmark data and example is added.
[PATCH v9 0/9] mm: introduce memfd_secret system call to create “secret” memory areas (Mike Rapoport) https://firstname.lastname@example.org
The 9th version of the memfd_secret patchset.
[patch V4 0/8] mm/highmem: Preemptible variant of kmap_atomic & friends (Thomas Gleixner) https://email@example.com
As the name suggests, this patchset allow the code using kmap_atomic and similar things to be preemptible. This incur some overhead to highmem, but Thomas claims highmem was always slow.
[PATCH 0/7] Split huge pages to any lower order pages and selftests.
[PATCH -V3 0/2] autonuma: Migrate on fault among multiple bound nodes (Huang Ying) https://firstname.lastname@example.org
This patchset makes AutoNUMA’s cross-socket memory access optimization possible even if the application’s memory is bound to multiple NUMA nodes.
[PATCH] kfence: Avoid stalling work queue task without allocations (Marco Elver) https://email@example.com
This commit makes the toggle allocation gate of kfence to have 1 second timeout, to be prepared for allocation stallings.
[RFC PATCH 0/6] Split huge pages to any lower order pages.
[PATCH v7 0/9] KFENCE: A low-overhead sampling-based memory safety error detector (Marco Elver) https://firstname.lastname@example.org
The 7th version of KFENCE. Now the authors say they believe this is ready to be included in the mm tree. And, it has now merged in the -mm tree.
[PATCH] page_frag: Recover from memory pressure (Matthew Wilcox) https://email@example.com
After an extreme memory pressure, some non-essential packets can be dropped because the ‘page_frag’ allocator do the drops to recover from the memory pressure but it doesn’t change back the behavior even after the recovery.
[PATCH 0/9] More THP fixes (Matthew Wilcox) https://firstname.lastname@example.org
Several THP fixes.
[PATCH v5 0/9] KFENCE: A low-overhead sampling-based memory safety error detector (Marco Elver) https://email@example.com
The fifth version of the KFENCE.
[PATCH -V2 0/2] autonuma: Migrate on fault among multiple bound nodes (Huang Ying) https://firstname.lastname@example.org
This patch is for multiple NUMA nodes bounded application’s AutoNUMA balancing.
[PATCH v6 0/9] KFENCE: A low-overhead sampling-based memory safety error detector (Marco Elver) https://email@example.com
[PATCH] mm: Make allocator take care of memoryless numa node (Xianting Tian) https://firstname.lastname@example.org
CPUs in powerpc like architecture could have no local memory. As the kernel code doesn’t care about it, this commit makes it to handle the case.
incoming (Andrew Morton) https://email@example.com
This is the pull request from Andrew Morton, the mm maintainer, for v5.10.
[PATCH v4 0/5] Speed up mremap on large regions (Kalesh Singh) https://firstname.lastname@example.org)
Repost fo the patch series.
[PATCH RFC v2] Opportunistic memory reclaim (Andrea Righi) https://email@example.com
This patch introduces new file under memcg, ‘memory.swap.reclaim’ that allow users to incur Reclaim even before the memory pressure happens, to help reducing system memory footprint and speed up hibernation/VM migration time, etc.
[PATCH v2 0/2] Add tracepoints around mmap_lock acquisition (Axel Rasmussen) https://firstname.lastname@example.org
This patchset adds tracepoints for mmap_lock to use that for analysis of the contention.
[RFC][PATCH 0/9] [v4][RESEND] Migrate Pages in lieu of discard (Dave Hansen) https://lkml.
[RFC PATCH v2 00/30] 1GB PUD THP support on x86_64 (Zi Yan) https://email@example.com
This patch adds support of 1GB THP using PUD.
[PATCH v4 00/11] KFENCE: A low-overhead sampling-based memory safety error detector (Marco Elver) https://firstname.lastname@example.org
The 4th version of the KFENCE patchset.
[PATCH 0/5] Speed up mremap on large regions (Kalesh Singh) https://email@example.com
This patchset optimizes mremap() by allowing memory regions to be able to mapped at PMD/PUD.
[PATCH] selftests/vm: 10x speedup for hmm-tests https://lkml.
[PATCH v3 00/10] KFENCE: A low-overhead sampling-based memory safety error detector (Marco Elver) https://firstname.lastname@example.org
The third version of KFENCE patchset is uploaded.
Machine lockups on extreme memory pressure (Shakeel Butt) https://lkml.kernel.org/r/CALvZod4FWLsV9byrKQojeus7tMDhHjQHFF5J_JpNsyB0HkaERA@mail.gmail.com
Shakeel reports a machine lockup due to extreme memory pressure. It’s locked up on memcg’s vmpressure sr_lock, which is irq-unsafe. And the lockup made in IRQ. He proposes a solution simply changing the lock with irq-safe one, but he also admits it’s not a maintainable solution.