kernel/linux.git/include/linux/memcontrol.h, branch v6.6.132

mm: memcg: restore subtree stats flushing

2025-11-24T09:30:13+00:00

[ Upstream commit 7d7ef0a4686abe43cd76a141b340a348f45ecdf2 ] Stats flushing for memcg currently follows the following rules: - Always flush the entire memcg hierarchy (i.e. flush the root). - Only one flusher is allowed at a time. If someone else tries to flush concurrently, they skip and return immediately. - A periodic flusher flushes all the stats every 2 seconds. The reason this approach is followed is because all flushes are serialized by a global rstat spinlock. On the memcg side, flushing is invoked from userspace reads as well as in-kernel flushers (e.g. reclaim, refault, etc). This approach aims to avoid serializing all flushers on the global lock, which can cause a significant performance hit under high concurrency. This approach has the following problems: - Occasionally a userspace read of the stats of a non-root cgroup will be too expensive as it has to flush the entire hierarchy [1]. - Sometimes the stats accuracy are compromised if there is an ongoing flush, and we skip and return before the subtree of interest is actually flushed, yielding stale stats (by up to 2s due to periodic flushing). This is more visible when reading stats from userspace, but can also affect in-kernel flushers. The latter problem is particulary a concern when userspace reads stats after an event occurs, but gets stats from before the event. Examples: - When memory usage / pressure spikes, a userspace OOM handler may look at the stats of different memcgs to select a victim based on various heuristics (e.g. how much private memory will be freed by killing this). Reading stale stats from before the usage spike in this case may cause a wrongful OOM kill. - A proactive reclaimer may read the stats after writing to memory.reclaim to measure the success of the reclaim operation. Stale stats from before reclaim may give a false negative. - Reading the stats of a parent and a child memcg may be inconsistent (child larger than parent), if the flush doesn't happen when the parent is read, but happens when the child is read. As for in-kernel flushers, they will occasionally get stale stats. No regressions are currently known from this, but if there are regressions, they would be very difficult to debug and link to the source of the problem. This patch aims to fix these problems by restoring subtree flushing, and removing the unified/coalesced flushing logic that skips flushing if there is an ongoing flush. This change would introduce a significant regression with global stats flushing thresholds. With per-memcg stats flushing thresholds, this seems to perform really well. The thresholds protect the underlying lock from unnecessary contention. This patch was tested in two ways to ensure the latency of flushing is up to par, on a machine with 384 cpus: - A synthetic test with 5000 concurrent workers in 500 cgroups doing allocations and reclaim, as well as 1000 readers for memory.stat (variation of [2]). No regressions were noticed in the total runtime. Note that significant regressions in this test are observed with global stats thresholds, but not with per-memcg thresholds. - A synthetic stress test for concurrently reading memcg stats while memory allocation/freeing workers are running in the background, provided by Wei Xu [3]. With 250k threads reading the stats every 100ms in 50k cgroups, 99.9% of reads take <= 50us. Less than 0.01% of reads take more than 1ms, and no reads take more than 100ms. [1] https://lore.kernel.org/lkml/CABWYdi0c6__rh-K7dcM_pkf9BJdTRtAU08M43KO9ME4-dsgfoQ@mail.gmail.com/ [2] https://lore.kernel.org/lkml/CAJD7tka13M-zVZTyQJYL1iUAYvuQ1fcHbCjcOBZcz6POYTV-4g@mail.gmail.com/ [3] https://lore.kernel.org/lkml/CAAPL-u9D2b=iF5Lf_cRnKxUfkiEe0AMDTu6yhrUAzX0b6a6rDg@mail.gmail.com/ [akpm@linux-foundation.org: fix mm/zswap.c] [yosryahmed@google.com: remove stats flushing mutex] Link: https://lkml.kernel.org/r/CAJD7tkZgP3m-VVPn+fF_YuvXeQYK=tZZjJHj=dzD=CcSSpp2qg@mail.gmail.com Link: https://lkml.kernel.org/r/20231129032154.3710765-6-yosryahmed@google.com Signed-off-by: Yosry Ahmed Tested-by: Domenico Cerasuolo Acked-by: Shakeel Butt Cc: Chris Li Cc: Greg Thelen Cc: Ivan Babrou Cc: Johannes Weiner Cc: Michal Hocko Cc: Michal Koutny Cc: Muchun Song Cc: Roman Gushchin Cc: Tejun Heo Cc: Waiman Long Cc: Wei Xu Signed-off-by: Andrew Morton Signed-off-by: Leon Huang Fu Signed-off-by: Greg Kroah-Hartman

mm/khugepaged: convert alloc_charge_hpage() to use folios

2024-11-01T00:58:25+00:00

[ Upstream commit b455f39d228935f88eebcd1f7c1a6981093c6a3b ] Also remove count_memcg_page_event now that its last caller no longer uses it and reword hpage_collapse_alloc_page() to hpage_collapse_alloc_folio(). This removes 1 call to compound_head() and helps convert khugepaged to use folios throughout. Link: https://lkml.kernel.org/r/20231020183331.10770-5-vishal.moola@gmail.com Signed-off-by: Vishal Moola (Oracle) Reviewed-by: Rik van Riel Reviewed-by: Yang Shi Cc: Kefeng Wang Cc: Matthew Wilcox (Oracle) Signed-off-by: Andrew Morton Stable-dep-of: 37f0b47c5143 ("mm: khugepaged: fix the arguments order in khugepaged_collapse_file trace point") Signed-off-by: Sasha Levin

mm: memcontrol: fix GFP_NOFS recursion in memory.high enforcement

2023-09-19T20:21:34+00:00

Breno and Josef report a deadlock scenario from cgroup reclaim re-entering the filesystem: [ 361.546690] ====================================================== [ 361.559210] WARNING: possible circular locking dependency detected [ 361.571703] 6.5.0-0_fbk700_debug_rc0_kbuilder_13159_gbf787a128001 #1 Tainted: G S E [ 361.589704] ------------------------------------------------------ [ 361.602277] find/9315 is trying to acquire lock: [ 361.611625] ffff88837ba140c0 (&delayed_node->mutex){+.+.}-{4:4}, at: __btrfs_release_delayed_node+0x68/0x4f0 [ 361.631437] [ 361.631437] but task is already holding lock: [ 361.643243] ffff8881765b8678 (btrfs-tree-01){++++}-{4:4}, at: btrfs_tree_read_lock+0x1e/0x40 [ 362.904457] mutex_lock_nested+0x1c/0x30 [ 362.912414] __btrfs_release_delayed_node+0x68/0x4f0 [ 362.922460] btrfs_evict_inode+0x301/0x770 [ 362.982726] evict+0x17c/0x380 [ 362.988944] prune_icache_sb+0x100/0x1d0 [ 363.005559] super_cache_scan+0x1f8/0x260 [ 363.013695] do_shrink_slab+0x2a2/0x540 [ 363.021489] shrink_slab_memcg+0x237/0x3d0 [ 363.050606] shrink_slab+0xa7/0x240 [ 363.083382] shrink_node_memcgs+0x262/0x3b0 [ 363.091870] shrink_node+0x1a4/0x720 [ 363.099150] shrink_zones+0x1f6/0x5d0 [ 363.148798] do_try_to_free_pages+0x19b/0x5e0 [ 363.157633] try_to_free_mem_cgroup_pages+0x266/0x370 [ 363.190575] reclaim_high+0x16f/0x1f0 [ 363.208409] mem_cgroup_handle_over_high+0x10b/0x270 [ 363.246678] try_charge_memcg+0xaf2/0xc70 [ 363.304151] charge_memcg+0xf0/0x350 [ 363.320070] __mem_cgroup_charge+0x28/0x40 [ 363.328371] __filemap_add_folio+0x870/0xd50 [ 363.371303] filemap_add_folio+0xdd/0x310 [ 363.399696] __filemap_get_folio+0x2fc/0x7d0 [ 363.419086] pagecache_get_page+0xe/0x30 [ 363.427048] alloc_extent_buffer+0x1cd/0x6a0 [ 363.435704] read_tree_block+0x43/0xc0 [ 363.443316] read_block_for_search+0x361/0x510 [ 363.466690] btrfs_search_slot+0xc8c/0x1520 This is caused by the mem_cgroup_handle_over_high() not respecting the gfp_mask of the allocation context. We used to only call this function on resume to userspace, where no locks were held. But c9afe31ec443 ("memcg: synchronously enforce memory.high for large overcharges") added a call from the allocation context without considering the gfp. Link: https://lkml.kernel.org/r/20230914152139.100822-1-hannes@cmpxchg.org Fixes: c9afe31ec443 ("memcg: synchronously enforce memory.high for large overcharges") Signed-off-by: Johannes Weiner Reported-by: Breno Leitao Reported-by: Josef Bacik Acked-by: Shakeel Butt Acked-by: Michal Hocko Cc: Roman Gushchin Cc: Muchun Song Cc: [5.17+] Signed-off-by: Andrew Morton

Merge tag 'mm-stable-2023-08-28-18-26' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

2023-08-29T21:25:26+00:00

Pull MM updates from Andrew Morton: - Some swap cleanups from Ma Wupeng ("fix WARN_ON in add_to_avail_list") - Peter Xu has a series (mm/gup: Unify hugetlb, speed up thp") which reduces the special-case code for handling hugetlb pages in GUP. It also speeds up GUP handling of transparent hugepages. - Peng Zhang provides some maple tree speedups ("Optimize the fast path of mas_store()"). - Sergey Senozhatsky has improved te performance of zsmalloc during compaction (zsmalloc: small compaction improvements"). - Domenico Cerasuolo has developed additional selftest code for zswap ("selftests: cgroup: add zswap test program"). - xu xin has doe some work on KSM's handling of zero pages. These changes are mainly to enable the user to better understand the effectiveness of KSM's treatment of zero pages ("ksm: support tracking KSM-placed zero-pages"). - Jeff Xu has fixes the behaviour of memfd's MEMFD_NOEXEC_SCOPE_NOEXEC_ENFORCED sysctl ("mm/memfd: fix sysctl MEMFD_NOEXEC_SCOPE_NOEXEC_ENFORCED"). - David Howells has fixed an fscache optimization ("mm, netfs, fscache: Stop read optimisation when folio removed from pagecache"). - Axel Rasmussen has given userfaultfd the ability to simulate memory poisoning ("add UFFDIO_POISON to simulate memory poisoning with UFFD"). - Miaohe Lin has contributed some routine maintenance work on the memory-failure code ("mm: memory-failure: remove unneeded PageHuge() check"). - Peng Zhang has contributed some maintenance work on the maple tree code ("Improve the validation for maple tree and some cleanup"). - Hugh Dickins has optimized the collapsing of shmem or file pages into THPs ("mm: free retracted page table by RCU"). - Jiaqi Yan has a patch series which permits us to use the healthy subpages within a hardware poisoned huge page for general purposes ("Improve hugetlbfs read on HWPOISON hugepages"). - Kemeng Shi has done some maintenance work on the pagetable-check code ("Remove unused parameters in page_table_check"). - More folioification work from Matthew Wilcox ("More filesystem folio conversions for 6.6"), ("Followup folio conversions for zswap"). And from ZhangPeng ("Convert several functions in page_io.c to use a folio"). - page_ext cleanups from Kemeng Shi ("minor cleanups for page_ext"). - Baoquan He has converted some architectures to use the GENERIC_IOREMAP ioremap()/iounmap() code ("mm: ioremap: Convert architectures to take GENERIC_IOREMAP way"). - Anshuman Khandual has optimized arm64 tlb shootdown ("arm64: support batched/deferred tlb shootdown during page reclamation/migration"). - Better maple tree lockdep checking from Liam Howlett ("More strict maple tree lockdep"). Liam also developed some efficiency improvements ("Reduce preallocations for maple tree"). - Cleanup and optimization to the secondary IOMMU TLB invalidation, from Alistair Popple ("Invalidate secondary IOMMU TLB on permission upgrade"). - Ryan Roberts fixes some arm64 MM selftest issues ("selftests/mm fixes for arm64"). - Kemeng Shi provides some maintenance work on the compaction code ("Two minor cleanups for compaction"). - Some reduction in mmap_lock pressure from Matthew Wilcox ("Handle most file-backed faults under the VMA lock"). - Aneesh Kumar contributes code to use the vmemmap optimization for DAX on ppc64, under some circumstances ("Add support for DAX vmemmap optimization for ppc64"). - page-ext cleanups from Kemeng Shi ("add page_ext_data to get client data in page_ext"), ("minor cleanups to page_ext header"). - Some zswap cleanups from Johannes Weiner ("mm: zswap: three cleanups"). - kmsan cleanups from ZhangPeng ("minor cleanups for kmsan"). - VMA handling cleanups from Kefeng Wang ("mm: convert to vma_is_initial_heap/stack()"). - DAMON feature work from SeongJae Park ("mm/damon/sysfs-schemes: implement DAMOS tried total bytes file"), ("Extend DAMOS filters for address ranges and DAMON monitoring targets"). - Compaction work from Kemeng Shi ("Fixes and cleanups to compaction"). - Liam Howlett has improved the maple tree node replacement code ("maple_tree: Change replacement strategy"). - ZhangPeng has a general code cleanup - use the K() macro more widely ("cleanup with helper macro K()"). - Aneesh Kumar brings memmap-on-memory to ppc64 ("Add support for memmap on memory feature on ppc64"). - pagealloc cleanups from Kemeng Shi ("Two minor cleanups for pcp list in page_alloc"), ("Two minor cleanups for get pageblock migratetype"). - Vishal Moola introduces a memory descriptor for page table tracking, "struct ptdesc" ("Split ptdesc from struct page"). - memfd selftest maintenance work from Aleksa Sarai ("memfd: cleanups for vm.memfd_noexec"). - MM include file rationalization from Hugh Dickins ("arch: include asm/cacheflush.h in asm/hugetlb.h"). - THP debug output fixes from Hugh Dickins ("mm,thp: fix sloppy text output"). - kmemleak improvements from Xiaolei Wang ("mm/kmemleak: use object_cache instead of kmemleak_initialized"). - More folio-related cleanups from Matthew Wilcox ("Remove _folio_dtor and _folio_order"). - A VMA locking scalability improvement from Suren Baghdasaryan ("Per-VMA lock support for swap and userfaults"). - pagetable handling cleanups from Matthew Wilcox ("New page table range API"). - A batch of swap/thp cleanups from David Hildenbrand ("mm/swap: stop using page->private on tail pages for THP_SWAP + cleanups"). - Cleanups and speedups to the hugetlb fault handling from Matthew Wilcox ("Change calling convention for ->huge_fault"). - Matthew Wilcox has also done some maintenance work on the MM subsystem documentation ("Improve mm documentation"). * tag 'mm-stable-2023-08-28-18-26' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (489 commits) maple_tree: shrink struct maple_tree maple_tree: clean up mas_wr_append() secretmem: convert page_is_secretmem() to folio_is_secretmem() nios2: fix flush_dcache_page() for usage from irq context hugetlb: add documentation for vma_kernel_pagesize() mm: add orphaned kernel-doc to the rst files. mm: fix clean_record_shared_mapping_range kernel-doc mm: fix get_mctgt_type() kernel-doc mm: fix kernel-doc warning from tlb_flush_rmaps() mm: remove enum page_entry_size mm: allow ->huge_fault() to be called without the mmap_lock held mm: move PMD_ORDER to pgtable.h mm: remove checks for pte_index memcg: remove duplication detection for mem_cgroup_uncharge_swap mm/huge_memory: work on folio->swap instead of page->private when splitting folio mm/swap: inline folio_set_swap_entry() and folio_swap_entry() mm/swap: use dedicated entry for swap in folio mm/swap: stop using page->private on tail pages for THP_SWAP selftests/mm: fix WARNING comparing pointer to 0 selftests: cgroup: fix test_kmem_memcg_deletion kernel mem check ...

mm: memcg: use rstat for non-hierarchical stats

2023-08-24T23:20:18+00:00

Currently, memcg uses rstat to maintain aggregated hierarchical stats. Counters are maintained for hierarchical stats at each memcg. Rstat tracks which cgroups have updates on which cpus to keep those counters fresh on the read-side. Non-hierarchical stats are currently not covered by rstat. Their per-cpu counters are summed up on every read, which is expensive. The original implementation did the same. At some point before rstat, non-hierarchical aggregated counters were introduced by commit a983b5ebee57 ("mm: memcontrol: fix excessive complexity in memory.stat reporting"). However, those counters were updated on the performance critical write-side, which caused regressions, so they were later removed by commit 815744d75152 ("mm: memcontrol: don't batch updates of local VM stats and events"). See [1] for more detailed history. Kernel versions in between a983b5ebee57 & 815744d75152 (a year and a half) enjoyed cheap reads of non-hierarchical stats, specifically on cgroup v1. When moving to more recent kernels, a performance regression for reading non-hierarchical stats is observed. Now that we have rstat, we know exactly which percpu counters have updates for each stat. We can maintain non-hierarchical counters again, making reads much more efficient, without affecting the performance critical write-side. Hence, add non-hierarchical (i.e local) counters for the stats, and extend rstat flushing to keep those up-to-date. A caveat is that we now need a stats flush before reading local/non-hierarchical stats through {memcg/lruvec}_page_state_local() or memcg_events_local(), where we previously only needed a flush to read hierarchical stats. Most contexts reading non-hierarchical stats are already doing a flush, add a flush to the only missing context in count_shadow_nodes(). With this patch, reading memory.stat from 1000 memcgs is 3x faster on a machine with 256 cpus on cgroup v1: # for i in $(seq 1000); do mkdir /sys/fs/cgroup/memory/cg$i; done # time cat /sys/fs/cgroup/memory/cg*/memory.stat > /dev/null real 0m0.125s user 0m0.005s sys 0m0.120s After: real 0m0.032s user 0m0.005s sys 0m0.027s To make sure there are no regressions on cgroup v2, I ran an artificial reclaim/refault stress test [2] that creates (NR_CPUS * 2) cgroups, assigns them limits, runs a worker process in each cgroup that allocates tmpfs memory equal to quadruple the limit (to invoke reclaim continuously), and then reads back the entire file (to invoke refaults). All workers are run in parallel, and zram is used as a swapping backend. Both reclaim and refault have conditional stats flushing. I ran this on a machine with 112 cpus, once on mm-unstable, and once on mm-unstable with this patch reverted. (1) A few runs without this patch: # time ./stress_reclaim_refault.sh real 0m9.949s user 0m0.496s sys 14m44.974s # time ./stress_reclaim_refault.sh real 0m10.049s user 0m0.486s sys 14m55.791s # time ./stress_reclaim_refault.sh real 0m9.984s user 0m0.481s sys 14m53.841s (2) A few runs with this patch: # time ./stress_reclaim_refault.sh real 0m9.885s user 0m0.486s sys 14m48.753s # time ./stress_reclaim_refault.sh real 0m9.903s user 0m0.495s sys 14m48.339s # time ./stress_reclaim_refault.sh real 0m9.861s user 0m0.507s sys 14m49.317s No regressions are observed with this patch. There is actually a very slight improvement. If I have to guess, maybe it's because we avoid the percpu loop in count_shadow_nodes() when calling lruvec_page_state_local(), but I could not prove this using perf, it's probably in the noise. [1] https://lore.kernel.org/lkml/20230725201811.GA1231514@cmpxchg.org/ [2] https://lore.kernel.org/lkml/CAJD7tkb17x=qwoO37uxyYXLEUVp15BQKR+Xfh7Sg9Hx-wTQ_=w@mail.gmail.com/ Link: https://lkml.kernel.org/r/20230803185046.1385770-1-yosryahmed@google.com Link: https://lkml.kernel.org/r/20230726153223.821757-2-yosryahmed@google.com Signed-off-by: Yosry Ahmed Acked-by: Johannes Weiner Acked-by: Roman Gushchin Acked-by: Michal Hocko Cc: Muchun Song Cc: Shakeel Butt Signed-off-by: Andrew Morton

mm/memcg: update obsolete comment above parent_mem_cgroup()

2023-08-21T20:37:32+00:00

Since commit bef8620cd8e0 ("mm: memcg: deprecate the non-hierarchical mode"), use_hierarchy is already deprecated. And it's further removed via commit 9d9d341df4d5 ("cgroup: remove obsoleted broken_hierarchy and warned_broken_hierarchy"). Update corresponding comment. Link: https://lkml.kernel.org/r/20230801124359.2266860-1-linmiaohe@huawei.com Signed-off-by: Miaohe Lin Cc: Michal Hocko Cc: Roman Gushchin Cc: Johannes Weiner Cc: Shakeel Butt Cc: Muchun Song Signed-off-by: Andrew Morton

mm/memcg: fix obsolete function name in mem_cgroup_protection()

2023-08-21T20:37:28+00:00

Commit 45c7f7e1ef17 ("mm, memcg: decouple e{low,min} state mutations from protection checks") changed the function name but not the corresponding comment. Link: https://lkml.kernel.org/r/20230727115934.657787-1-linmiaohe@huawei.com Signed-off-by: Miaohe Lin Cc: Michal Hocko Cc: Roman Gushchin Cc: Johannes Weiner Cc: Shakeel Butt Cc: Muchun Song Signed-off-by: Andrew Morton

memcg: convert get_obj_cgroup_from_page to get_obj_cgroup_from_folio

2023-08-21T20:37:26+00:00

As the one caller now has a folio, pass it in and use it. Removes three calls to compound_head(). Link: https://lkml.kernel.org/r/20230715042343.434588-4-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Cc: Christoph Hellwig Cc: Domenico Cerasuolo Cc: Johannes Weiner Cc: Nhat Pham Cc: Vitaly Wool Cc: Yosry Ahmed Signed-off-by: Andrew Morton

mm/memcg: minor cleanup for MEM_CGROUP_ID_MAX

2023-08-18T17:12:15+00:00

MEM_CGROUP_ID_MAX is only used when CONFIG_MEMCG is configured. So remove unneeded !CONFIG_MEMCG variant. Also it's only used in mem_cgroup_alloc(), so move it from memcontrol.h to memcontrol.c. And further define it as: #define MEM_CGROUP_ID_MAX ((1UL << MEM_CGROUP_ID_SHIFT) - 1) so if someone changes MEM_CGROUP_ID_SHIFT in the future, then MEM_CGROUP_ID_MAX will be updated accordingly, as suggested by Muchun. Link: https://lkml.kernel.org/r/20230708023304.1184111-1-linmiaohe@huawei.com Signed-off-by: Miaohe Lin Reviewed-by: Muchun Song Cc: Michal Hocko Cc: Roman Gushchin Cc: Johannes Weiner Cc: Shakeel Butt Signed-off-by: Andrew Morton

net-memcg: Fix scope of sockmem pressure indicators

2023-08-16T11:21:32+00:00

Now there are two indicators of socket memory pressure sit inside struct mem_cgroup, socket_pressure and tcpmem_pressure, indicating memory reclaim pressure in memcg->memory and ->tcpmem respectively. When in legacy mode (cgroupv1), the socket memory is charged into ->tcpmem which is independent of ->memory, so socket_pressure has nothing to do with socket's pressure at all. Things could be worse by taking socket_pressure into consideration in legacy mode, as a pressure in ->memory can lead to premature reclamation/throttling in socket. While for the default mode (cgroupv2), the socket memory is charged into ->memory, and ->tcpmem/->tcpmem_pressure are simply not used. So {socket,tcpmem}_pressure are only used in default/legacy mode respectively for indicating socket memory pressure. This patch fixes the pieces of code that make mixed use of both. Fixes: 8e8ae645249b ("mm: memcontrol: hook up vmpressure to socket pressure") Signed-off-by: Abel Wu Acked-by: Shakeel Butt Signed-off-by: David S. Miller