From e46bc2e7eb90a370bc27fa2fd98cb8251e7da1ec Mon Sep 17 00:00:00 2001 From: Pedro Falcato Date: Wed, 7 Aug 2024 18:33:35 +0100 Subject: [PATCH 01/16] mseal: fix is_madv_discard() is_madv_discard did its check wrong. MADV_ flags are not bitwise, they're normal sequential numbers. So, for instance: behavior & (/* ... */ | MADV_REMOVE) tagged both MADV_REMOVE and MADV_RANDOM (bit 0 set) as discard operations. As a result the kernel could erroneously block certain madvises (e.g MADV_RANDOM or MADV_HUGEPAGE) on sealed VMAs due to them sharing bits with blocked MADV operations (e.g REMOVE or WIPEONFORK). This is obviously incorrect, so use a switch statement instead. Link: https://lkml.kernel.org/r/20240807173336.2523757-1-pedro.falcato@gmail.com Link: https://lkml.kernel.org/r/20240807173336.2523757-2-pedro.falcato@gmail.com Fixes: 8be7258aad44 ("mseal: add mseal syscall") Signed-off-by: Pedro Falcato Tested-by: Jeff Xu Reviewed-by: Jeff Xu Cc: Kees Cook Cc: Liam R. Howlett Cc: Shuah Khan Cc: Signed-off-by: Andrew Morton --- mm/mseal.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/mm/mseal.c b/mm/mseal.c index bf783bba8ed0..15bba28acc00 100644 --- a/mm/mseal.c +++ b/mm/mseal.c @@ -40,9 +40,17 @@ static bool can_modify_vma(struct vm_area_struct *vma) static bool is_madv_discard(int behavior) { - return behavior & - (MADV_FREE | MADV_DONTNEED | MADV_DONTNEED_LOCKED | - MADV_REMOVE | MADV_DONTFORK | MADV_WIPEONFORK); + switch (behavior) { + case MADV_FREE: + case MADV_DONTNEED: + case MADV_DONTNEED_LOCKED: + case MADV_REMOVE: + case MADV_DONTFORK: + case MADV_WIPEONFORK: + return true; + } + + return false; } static bool is_ro_anon(struct vm_area_struct *vma) From 5f75cfbd6bb02295ddaed48adf667b6c828ce07b Mon Sep 17 00:00:00 2001 From: David Hildenbrand Date: Thu, 1 Aug 2024 22:47:48 +0200 Subject: [PATCH 02/16] mm/hugetlb: fix hugetlb vs. core-mm PT locking We recently made GUP's common page table walking code to also walk hugetlb VMAs without most hugetlb special-casing, preparing for the future of having less hugetlb-specific page table walking code in the codebase. Turns out that we missed one page table locking detail: page table locking for hugetlb folios that are not mapped using a single PMD/PUD. Assume we have hugetlb folio that spans multiple PTEs (e.g., 64 KiB hugetlb folios on arm64 with 4 KiB base page size). GUP, as it walks the page tables, will perform a pte_offset_map_lock() to grab the PTE table lock. However, hugetlb that concurrently modifies these page tables would actually grab the mm->page_table_lock: with USE_SPLIT_PTE_PTLOCKS, the locks would differ. Something similar can happen right now with hugetlb folios that span multiple PMDs when USE_SPLIT_PMD_PTLOCKS. This issue can be reproduced [1], for example triggering: [ 3105.936100] ------------[ cut here ]------------ [ 3105.939323] WARNING: CPU: 31 PID: 2732 at mm/gup.c:142 try_grab_folio+0x11c/0x188 [ 3105.944634] Modules linked in: [...] [ 3105.974841] CPU: 31 PID: 2732 Comm: reproducer Not tainted 6.10.0-64.eln141.aarch64 #1 [ 3105.980406] Hardware name: QEMU KVM Virtual Machine, BIOS edk2-20240524-4.fc40 05/24/2024 [ 3105.986185] pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 3105.991108] pc : try_grab_folio+0x11c/0x188 [ 3105.994013] lr : follow_page_pte+0xd8/0x430 [ 3105.996986] sp : ffff80008eafb8f0 [ 3105.999346] x29: ffff80008eafb900 x28: ffffffe8d481f380 x27: 00f80001207cff43 [ 3106.004414] x26: 0000000000000001 x25: 0000000000000000 x24: ffff80008eafba48 [ 3106.009520] x23: 0000ffff9372f000 x22: ffff7a54459e2000 x21: ffff7a546c1aa978 [ 3106.014529] x20: ffffffe8d481f3c0 x19: 0000000000610041 x18: 0000000000000001 [ 3106.019506] x17: 0000000000000001 x16: ffffffffffffffff x15: 0000000000000000 [ 3106.024494] x14: ffffb85477fdfe08 x13: 0000ffff9372ffff x12: 0000000000000000 [ 3106.029469] x11: 1fffef4a88a96be1 x10: ffff7a54454b5f0c x9 : ffffb854771b12f0 [ 3106.034324] x8 : 0008000000000000 x7 : ffff7a546c1aa980 x6 : 0008000000000080 [ 3106.038902] x5 : 00000000001207cf x4 : 0000ffff9372f000 x3 : ffffffe8d481f000 [ 3106.043420] x2 : 0000000000610041 x1 : 0000000000000001 x0 : 0000000000000000 [ 3106.047957] Call trace: [ 3106.049522] try_grab_folio+0x11c/0x188 [ 3106.051996] follow_pmd_mask.constprop.0.isra.0+0x150/0x2e0 [ 3106.055527] follow_page_mask+0x1a0/0x2b8 [ 3106.058118] __get_user_pages+0xf0/0x348 [ 3106.060647] faultin_page_range+0xb0/0x360 [ 3106.063651] do_madvise+0x340/0x598 Let's make huge_pte_lockptr() effectively use the same PT locks as any core-mm page table walker would. Add ptep_lockptr() to obtain the PTE page table lock using a pte pointer -- unfortunately we cannot convert pte_lockptr() because virt_to_page() doesn't work with kmap'ed page tables we can have with CONFIG_HIGHPTE. Handle CONFIG_PGTABLE_LEVELS correctly by checking in reverse order, such that when e.g., CONFIG_PGTABLE_LEVELS==2 with PGDIR_SIZE==P4D_SIZE==PUD_SIZE==PMD_SIZE will work as expected. Document why that works. There is one ugly case: powerpc 8xx, whereby we have an 8 MiB hugetlb folio being mapped using two PTE page tables. While hugetlb wants to take the PMD table lock, core-mm would grab the PTE table lock of one of both PTE page tables. In such corner cases, we have to make sure that both locks match, which is (fortunately!) currently guaranteed for 8xx as it does not support SMP and consequently doesn't use split PT locks. [1] https://lore.kernel.org/all/1bbfcc7f-f222-45a5-ac44-c5a1381c596d@redhat.com/ Link: https://lkml.kernel.org/r/20240801204748.99107-1-david@redhat.com Fixes: 9cb28da54643 ("mm/gup: handle hugetlb in the generic follow_page_mask code") Signed-off-by: David Hildenbrand Acked-by: Peter Xu Reviewed-by: Baolin Wang Tested-by: Baolin Wang Cc: Peter Xu Cc: Oscar Salvador Cc: Muchun Song Cc: Signed-off-by: Andrew Morton --- include/linux/hugetlb.h | 33 ++++++++++++++++++++++++++++++--- include/linux/mm.h | 11 +++++++++++ 2 files changed, 41 insertions(+), 3 deletions(-) diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index c9bf68c239a0..45bf05ad5c53 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -944,10 +944,37 @@ static inline bool htlb_allow_alloc_fallback(int reason) static inline spinlock_t *huge_pte_lockptr(struct hstate *h, struct mm_struct *mm, pte_t *pte) { - if (huge_page_size(h) == PMD_SIZE) + const unsigned long size = huge_page_size(h); + + VM_WARN_ON(size == PAGE_SIZE); + + /* + * hugetlb must use the exact same PT locks as core-mm page table + * walkers would. When modifying a PTE table, hugetlb must take the + * PTE PT lock, when modifying a PMD table, hugetlb must take the PMD + * PT lock etc. + * + * The expectation is that any hugetlb folio smaller than a PMD is + * always mapped into a single PTE table and that any hugetlb folio + * smaller than a PUD (but at least as big as a PMD) is always mapped + * into a single PMD table. + * + * If that does not hold for an architecture, then that architecture + * must disable split PT locks such that all *_lockptr() functions + * will give us the same result: the per-MM PT lock. + * + * Note that with e.g., CONFIG_PGTABLE_LEVELS=2 where + * PGDIR_SIZE==P4D_SIZE==PUD_SIZE==PMD_SIZE, we'd use pud_lockptr() + * and core-mm would use pmd_lockptr(). However, in such configurations + * split PMD locks are disabled -- they don't make sense on a single + * PGDIR page table -- and the end result is the same. + */ + if (size >= PUD_SIZE) + return pud_lockptr(mm, (pud_t *) pte); + else if (size >= PMD_SIZE || IS_ENABLED(CONFIG_HIGHPTE)) return pmd_lockptr(mm, (pmd_t *) pte); - VM_BUG_ON(huge_page_size(h) == PAGE_SIZE); - return &mm->page_table_lock; + /* pte_alloc_huge() only applies with !CONFIG_HIGHPTE */ + return ptep_lockptr(mm, pte); } #ifndef hugepages_supported diff --git a/include/linux/mm.h b/include/linux/mm.h index c4b238a20b76..6549d0979b28 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -2920,6 +2920,13 @@ static inline spinlock_t *pte_lockptr(struct mm_struct *mm, pmd_t *pmd) return ptlock_ptr(page_ptdesc(pmd_page(*pmd))); } +static inline spinlock_t *ptep_lockptr(struct mm_struct *mm, pte_t *pte) +{ + BUILD_BUG_ON(IS_ENABLED(CONFIG_HIGHPTE)); + BUILD_BUG_ON(MAX_PTRS_PER_PTE * sizeof(pte_t) > PAGE_SIZE); + return ptlock_ptr(virt_to_ptdesc(pte)); +} + static inline bool ptlock_init(struct ptdesc *ptdesc) { /* @@ -2944,6 +2951,10 @@ static inline spinlock_t *pte_lockptr(struct mm_struct *mm, pmd_t *pmd) { return &mm->page_table_lock; } +static inline spinlock_t *ptep_lockptr(struct mm_struct *mm, pte_t *pte) +{ + return &mm->page_table_lock; +} static inline void ptlock_cache_init(void) {} static inline bool ptlock_init(struct ptdesc *ptdesc) { return true; } static inline void ptlock_free(struct ptdesc *ptdesc) {} From ace0741a55e453c265cbf3d965eea7f687cd6d45 Mon Sep 17 00:00:00 2001 From: Pasha Tatashin Date: Thu, 8 Aug 2024 21:34:34 +0000 Subject: [PATCH 03/16] mm: don't account memmap on failure Patch series "Fixes for memmap accounting", v4. Memmap accounting provides us with observability of how much memory is used for per-page metadata: i.e. "struct page"'s and "struct page_ext". It also provides with information of how much was allocated using boot allocator (i.e. not part of MemTotal), and how much was allocated using buddy allocated (i.e. part of MemTotal). This small series fixes a few problems that were discovered with the original patch. This patch (of 3): When we fail to allocate the mmemmap in alloc_vmemmap_page_list(), do not account any already-allocated pages: we're going to free all them before we return from the function. Link: https://lkml.kernel.org/r/20240809191020.1142142-1-pasha.tatashin@soleen.com Link: https://lkml.kernel.org/r/20240808213437.682006-1-pasha.tatashin@soleen.com Link: https://lkml.kernel.org/r/20240808213437.682006-2-pasha.tatashin@soleen.com Fixes: 15995a352474 ("mm: report per-page metadata information") Signed-off-by: Pasha Tatashin Reviewed-by: Fan Ni Reviewed-by: Yosry Ahmed Acked-by: David Hildenbrand Tested-by: Alison Schofield Reviewed-by: Muchun Song Acked-by: David Rientjes Cc: Dan Williams Cc: Domenico Cerasuolo Cc: Joel Granados Cc: Johannes Weiner Cc: Li Zhijian Cc: Matthew Wilcox (Oracle) Cc: Mike Rapoport Cc: Nhat Pham Cc: Sourav Panda Cc: Vlastimil Babka Cc: Yi Zhang Signed-off-by: Andrew Morton --- mm/hugetlb_vmemmap.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/mm/hugetlb_vmemmap.c b/mm/hugetlb_vmemmap.c index 829112b0a914..4f51e0596197 100644 --- a/mm/hugetlb_vmemmap.c +++ b/mm/hugetlb_vmemmap.c @@ -392,13 +392,10 @@ static int alloc_vmemmap_page_list(unsigned long start, unsigned long end, for (i = 0; i < nr_pages; i++) { page = alloc_pages_node(nid, gfp_mask, 0); - if (!page) { - mod_node_page_state(NODE_DATA(nid), NR_MEMMAP, i); + if (!page) goto out; - } list_add(&page->lru, list); } - mod_node_page_state(NODE_DATA(nid), NR_MEMMAP, nr_pages); return 0; From f4cb78af91e3b2b7aa76dbf8213b898fa8811b12 Mon Sep 17 00:00:00 2001 From: Pasha Tatashin Date: Thu, 8 Aug 2024 21:34:35 +0000 Subject: [PATCH 04/16] mm: add system wide stats items category /proc/vmstat contains events and stats, events can only grow, but stats can grow and shrink. vmstat has the following: ------------------------- NR_VM_ZONE_STAT_ITEMS: per-zone stats NR_VM_NUMA_EVENT_ITEMS: per-numa events NR_VM_NODE_STAT_ITEMS: per-numa stats NR_VM_WRITEBACK_STAT_ITEMS: system-wide background-writeback and dirty-throttling tresholds. NR_VM_EVENT_ITEMS: system-wide events ------------------------- Rename NR_VM_WRITEBACK_STAT_ITEMS to NR_VM_STAT_ITEMS, to track the system-wide stats, we are going to add per-page metadata stats to this category in the next patch. Also delete unused writeback_stat_name(). Link: https://lkml.kernel.org/r/20240809191020.1142142-2-pasha.tatashin@soleen.com Link: https://lkml.kernel.org/r/20240808213437.682006-3-pasha.tatashin@soleen.com Fixes: 15995a352474 ("mm: report per-page metadata information") Signed-off-by: Pasha Tatashin Suggested-by: Yosry Ahmed Tested-by: Alison Schofield Acked-by: David Hildenbrand Acked-by: David Rientjes Cc: Dan Williams Cc: Domenico Cerasuolo Cc: Joel Granados Cc: Johannes Weiner Cc: Li Zhijian Cc: Matthew Wilcox (Oracle) Cc: Mike Rapoport Cc: Muchun Song Cc: Nhat Pham Cc: Sourav Panda Cc: Vlastimil Babka Cc: Yi Zhang Cc: Fan Ni Signed-off-by: Andrew Morton --- include/linux/vmstat.h | 15 ++++----------- mm/vmstat.c | 6 +++--- 2 files changed, 7 insertions(+), 14 deletions(-) diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h index 23cd17942036..9ab4fa5e09b5 100644 --- a/include/linux/vmstat.h +++ b/include/linux/vmstat.h @@ -34,10 +34,11 @@ struct reclaim_stat { unsigned nr_lazyfree_fail; }; -enum writeback_stat_item { +/* Stat data for system wide items */ +enum vm_stat_item { NR_DIRTY_THRESHOLD, NR_DIRTY_BG_THRESHOLD, - NR_VM_WRITEBACK_STAT_ITEMS, + NR_VM_STAT_ITEMS, }; #ifdef CONFIG_VM_EVENT_COUNTERS @@ -514,21 +515,13 @@ static inline const char *lru_list_name(enum lru_list lru) return node_stat_name(NR_LRU_BASE + lru) + 3; // skip "nr_" } -static inline const char *writeback_stat_name(enum writeback_stat_item item) -{ - return vmstat_text[NR_VM_ZONE_STAT_ITEMS + - NR_VM_NUMA_EVENT_ITEMS + - NR_VM_NODE_STAT_ITEMS + - item]; -} - #if defined(CONFIG_VM_EVENT_COUNTERS) || defined(CONFIG_MEMCG) static inline const char *vm_event_name(enum vm_event_item item) { return vmstat_text[NR_VM_ZONE_STAT_ITEMS + NR_VM_NUMA_EVENT_ITEMS + NR_VM_NODE_STAT_ITEMS + - NR_VM_WRITEBACK_STAT_ITEMS + + NR_VM_STAT_ITEMS + item]; } #endif /* CONFIG_VM_EVENT_COUNTERS || CONFIG_MEMCG */ diff --git a/mm/vmstat.c b/mm/vmstat.c index 04a1cb6cc636..6f8aa4766f16 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -1257,7 +1257,7 @@ const char * const vmstat_text[] = { "pgdemote_khugepaged", "nr_memmap", "nr_memmap_boot", - /* enum writeback_stat_item counters */ + /* system-wide enum vm_stat_item counters */ "nr_dirty_threshold", "nr_dirty_background_threshold", @@ -1790,7 +1790,7 @@ static const struct seq_operations zoneinfo_op = { #define NR_VMSTAT_ITEMS (NR_VM_ZONE_STAT_ITEMS + \ NR_VM_NUMA_EVENT_ITEMS + \ NR_VM_NODE_STAT_ITEMS + \ - NR_VM_WRITEBACK_STAT_ITEMS + \ + NR_VM_STAT_ITEMS + \ (IS_ENABLED(CONFIG_VM_EVENT_COUNTERS) ? \ NR_VM_EVENT_ITEMS : 0)) @@ -1827,7 +1827,7 @@ static void *vmstat_start(struct seq_file *m, loff_t *pos) global_dirty_limits(v + NR_DIRTY_BG_THRESHOLD, v + NR_DIRTY_THRESHOLD); - v += NR_VM_WRITEBACK_STAT_ITEMS; + v += NR_VM_STAT_ITEMS; #ifdef CONFIG_VM_EVENT_COUNTERS all_vm_events(v); From 9d85731110241fb8ca9445ea4177d816041a8825 Mon Sep 17 00:00:00 2001 From: Pasha Tatashin Date: Thu, 8 Aug 2024 21:34:36 +0000 Subject: [PATCH 05/16] mm: don't account memmap per-node Fix invalid access to pgdat during hot-remove operation: ndctl users reported a GPF when trying to destroy a namespace: $ ndctl destroy-namespace all -r all -f Segmentation fault dmesg: Oops: general protection fault, probably for non-canonical address 0xdffffc0000005650: 0000 [#1] PREEMPT SMP KASAN PTI KASAN: probably user-memory-access in range [0x000000000002b280-0x000000000002b287] CPU: 26 UID: 0 PID: 1868 Comm: ndctl Not tainted 6.11.0-rc1 #1 Hardware name: Dell Inc. PowerEdge R640/08HT8T, BIOS 2.20.1 09/13/2023 RIP: 0010:mod_node_page_state+0x2a/0x110 cxl-test users report a GPF when trying to unload the test module: $ modrpobe -r cxl-test dmesg BUG: unable to handle page fault for address: 0000000000004200 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 0 P4D 0 Oops: Oops: 0000 [#1] PREEMPT SMP PTI CPU: 0 UID: 0 PID: 1076 Comm: modprobe Tainted: G O N 6.11.0-rc1 #197 Tainted: [O]=OOT_MODULE, [N]=TEST Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 02/06/15 RIP: 0010:mod_node_page_state+0x6/0x90 Currently, when memory is hot-plugged or hot-removed the accounting is done based on the assumption that memmap is allocated from the same node as the hot-plugged/hot-removed memory, which is not always the case. In addition, there are challenges with keeping the node id of the memory that is being remove to the time when memmap accounting is actually performed: since this is done after remove_pfn_range_from_zone(), and also after remove_memory_block_devices(). Meaning that we cannot use pgdat nor walking though memblocks to get the nid. Given all of that, account the memmap overhead system wide instead. For this we are going to be using global atomic counters, but given that memmap size is rarely modified, and normally is only modified either during early boot when there is only one CPU, or under a hotplug global mutex lock, therefore there is no need for per-cpu optimizations. Also, while we are here rename nr_memmap to nr_memmap_pages, and nr_memmap_boot to nr_memmap_boot_pages to be self explanatory that the units are in page count. [pasha.tatashin@soleen.com: address a few nits from David Hildenbrand] Link: https://lkml.kernel.org/r/20240809191020.1142142-4-pasha.tatashin@soleen.com Link: https://lkml.kernel.org/r/20240809191020.1142142-4-pasha.tatashin@soleen.com Link: https://lkml.kernel.org/r/20240808213437.682006-4-pasha.tatashin@soleen.com Fixes: 15995a352474 ("mm: report per-page metadata information") Signed-off-by: Pasha Tatashin Reported-by: Yi Zhang Closes: https://lore.kernel.org/linux-cxl/CAHj4cs9Ax1=CoJkgBGP_+sNu6-6=6v=_L-ZBZY0bVLD3wUWZQg@mail.gmail.com Reported-by: Alison Schofield Closes: https://lore.kernel.org/linux-mm/Zq0tPd2h6alFz8XF@aschofie-mobl2/#t Tested-by: Dan Williams Tested-by: Alison Schofield Acked-by: David Hildenbrand Acked-by: David Rientjes Tested-by: Yi Zhang Cc: Domenico Cerasuolo Cc: Fan Ni Cc: Joel Granados Cc: Johannes Weiner Cc: Li Zhijian Cc: Matthew Wilcox (Oracle) Cc: Mike Rapoport Cc: Muchun Song Cc: Nhat Pham Cc: Sourav Panda Cc: Vlastimil Babka Cc: Yosry Ahmed Signed-off-by: Andrew Morton --- include/linux/mmzone.h | 2 -- include/linux/vmstat.h | 7 ++++--- mm/hugetlb_vmemmap.c | 8 ++++---- mm/mm_init.c | 3 +-- mm/page_alloc.c | 1 - mm/page_ext.c | 18 ++++------------- mm/sparse-vmemmap.c | 11 ++++------ mm/sparse.c | 5 ++--- mm/vmstat.c | 46 ++++++++++++++++++++---------------------- 9 files changed, 41 insertions(+), 60 deletions(-) diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 41458892bc8a..1dc6248feb83 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -220,8 +220,6 @@ enum node_stat_item { PGDEMOTE_KSWAPD, PGDEMOTE_DIRECT, PGDEMOTE_KHUGEPAGED, - NR_MEMMAP, /* page metadata allocated through buddy allocator */ - NR_MEMMAP_BOOT, /* page metadata allocated through boot allocator */ NR_VM_NODE_STAT_ITEMS }; diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h index 9ab4fa5e09b5..9eb77c9007e6 100644 --- a/include/linux/vmstat.h +++ b/include/linux/vmstat.h @@ -38,6 +38,8 @@ struct reclaim_stat { enum vm_stat_item { NR_DIRTY_THRESHOLD, NR_DIRTY_BG_THRESHOLD, + NR_MEMMAP_PAGES, /* page metadata allocated through buddy allocator */ + NR_MEMMAP_BOOT_PAGES, /* page metadata allocated through boot allocator */ NR_VM_STAT_ITEMS, }; @@ -618,7 +620,6 @@ static inline void lruvec_stat_sub_folio(struct folio *folio, lruvec_stat_mod_folio(folio, idx, -folio_nr_pages(folio)); } -void __meminit mod_node_early_perpage_metadata(int nid, long delta); -void __meminit store_early_perpage_metadata(void); - +void memmap_boot_pages_add(long delta); +void memmap_pages_add(long delta); #endif /* _LINUX_VMSTAT_H */ diff --git a/mm/hugetlb_vmemmap.c b/mm/hugetlb_vmemmap.c index 4f51e0596197..0c3f56b3578e 100644 --- a/mm/hugetlb_vmemmap.c +++ b/mm/hugetlb_vmemmap.c @@ -185,11 +185,11 @@ static int vmemmap_remap_range(unsigned long start, unsigned long end, static inline void free_vmemmap_page(struct page *page) { if (PageReserved(page)) { + memmap_boot_pages_add(-1); free_bootmem_page(page); - mod_node_page_state(page_pgdat(page), NR_MEMMAP_BOOT, -1); } else { + memmap_pages_add(-1); __free_page(page); - mod_node_page_state(page_pgdat(page), NR_MEMMAP, -1); } } @@ -341,7 +341,7 @@ static int vmemmap_remap_free(unsigned long start, unsigned long end, copy_page(page_to_virt(walk.reuse_page), (void *)walk.reuse_addr); list_add(&walk.reuse_page->lru, vmemmap_pages); - mod_node_page_state(NODE_DATA(nid), NR_MEMMAP, 1); + memmap_pages_add(1); } /* @@ -396,7 +396,7 @@ static int alloc_vmemmap_page_list(unsigned long start, unsigned long end, goto out; list_add(&page->lru, list); } - mod_node_page_state(NODE_DATA(nid), NR_MEMMAP, nr_pages); + memmap_pages_add(nr_pages); return 0; out: diff --git a/mm/mm_init.c b/mm/mm_init.c index 75c3bd42799b..f9a60ffc5532 100644 --- a/mm/mm_init.c +++ b/mm/mm_init.c @@ -1623,8 +1623,7 @@ static void __init alloc_node_mem_map(struct pglist_data *pgdat) panic("Failed to allocate %ld bytes for node %d memory map\n", size, pgdat->node_id); pgdat->node_mem_map = map + offset; - mod_node_early_perpage_metadata(pgdat->node_id, - DIV_ROUND_UP(size, PAGE_SIZE)); + memmap_boot_pages_add(DIV_ROUND_UP(size, PAGE_SIZE)); pr_debug("%s: node %d, pgdat %08lx, node_mem_map %08lx\n", __func__, pgdat->node_id, (unsigned long)pgdat, (unsigned long)pgdat->node_mem_map); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 28f80daf5c04..875d76e8684a 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -5755,7 +5755,6 @@ void __init setup_per_cpu_pageset(void) for_each_online_pgdat(pgdat) pgdat->per_cpu_nodestats = alloc_percpu(struct per_cpu_nodestat); - store_early_perpage_metadata(); } __meminit void zone_pcp_init(struct zone *zone) diff --git a/mm/page_ext.c b/mm/page_ext.c index c191e490c401..641d93f6af4c 100644 --- a/mm/page_ext.c +++ b/mm/page_ext.c @@ -214,8 +214,7 @@ static int __init alloc_node_page_ext(int nid) return -ENOMEM; NODE_DATA(nid)->node_page_ext = base; total_usage += table_size; - mod_node_page_state(NODE_DATA(nid), NR_MEMMAP_BOOT, - DIV_ROUND_UP(table_size, PAGE_SIZE)); + memmap_boot_pages_add(DIV_ROUND_UP(table_size, PAGE_SIZE)); return 0; } @@ -275,10 +274,8 @@ static void *__meminit alloc_page_ext(size_t size, int nid) else addr = vzalloc_node(size, nid); - if (addr) { - mod_node_page_state(NODE_DATA(nid), NR_MEMMAP, - DIV_ROUND_UP(size, PAGE_SIZE)); - } + if (addr) + memmap_pages_add(DIV_ROUND_UP(size, PAGE_SIZE)); return addr; } @@ -323,25 +320,18 @@ static void free_page_ext(void *addr) { size_t table_size; struct page *page; - struct pglist_data *pgdat; table_size = page_ext_size * PAGES_PER_SECTION; + memmap_pages_add(-1L * (DIV_ROUND_UP(table_size, PAGE_SIZE))); if (is_vmalloc_addr(addr)) { - page = vmalloc_to_page(addr); - pgdat = page_pgdat(page); vfree(addr); } else { page = virt_to_page(addr); - pgdat = page_pgdat(page); BUG_ON(PageReserved(page)); kmemleak_free(addr); free_pages_exact(addr, table_size); } - - mod_node_page_state(pgdat, NR_MEMMAP, - -1L * (DIV_ROUND_UP(table_size, PAGE_SIZE))); - } static void __free_page_ext(unsigned long pfn) diff --git a/mm/sparse-vmemmap.c b/mm/sparse-vmemmap.c index 1dda6c53370b..edcc7a6b0f6f 100644 --- a/mm/sparse-vmemmap.c +++ b/mm/sparse-vmemmap.c @@ -469,13 +469,10 @@ struct page * __meminit __populate_section_memmap(unsigned long pfn, if (r < 0) return NULL; - if (system_state == SYSTEM_BOOTING) { - mod_node_early_perpage_metadata(nid, DIV_ROUND_UP(end - start, - PAGE_SIZE)); - } else { - mod_node_page_state(NODE_DATA(nid), NR_MEMMAP, - DIV_ROUND_UP(end - start, PAGE_SIZE)); - } + if (system_state == SYSTEM_BOOTING) + memmap_boot_pages_add(DIV_ROUND_UP(end - start, PAGE_SIZE)); + else + memmap_pages_add(DIV_ROUND_UP(end - start, PAGE_SIZE)); return pfn_to_page(pfn); } diff --git a/mm/sparse.c b/mm/sparse.c index e4b830091d13..0f018c6f9ec5 100644 --- a/mm/sparse.c +++ b/mm/sparse.c @@ -463,7 +463,7 @@ static void __init sparse_buffer_init(unsigned long size, int nid) sparsemap_buf = memmap_alloc(size, section_map_size(), addr, nid, true); sparsemap_buf_end = sparsemap_buf + size; #ifndef CONFIG_SPARSEMEM_VMEMMAP - mod_node_early_perpage_metadata(nid, DIV_ROUND_UP(size, PAGE_SIZE)); + memmap_boot_pages_add(DIV_ROUND_UP(size, PAGE_SIZE)); #endif } @@ -643,8 +643,7 @@ static void depopulate_section_memmap(unsigned long pfn, unsigned long nr_pages, unsigned long start = (unsigned long) pfn_to_page(pfn); unsigned long end = start + nr_pages * sizeof(struct page); - mod_node_page_state(page_pgdat(pfn_to_page(pfn)), NR_MEMMAP, - -1L * (DIV_ROUND_UP(end - start, PAGE_SIZE))); + memmap_pages_add(-1L * (DIV_ROUND_UP(end - start, PAGE_SIZE))); vmemmap_free(start, end, altmap); } static void free_map_bootmem(struct page *memmap) diff --git a/mm/vmstat.c b/mm/vmstat.c index 6f8aa4766f16..e875f2a4915f 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -1033,6 +1033,24 @@ unsigned long node_page_state(struct pglist_data *pgdat, } #endif +/* + * Count number of pages "struct page" and "struct page_ext" consume. + * nr_memmap_boot_pages: # of pages allocated by boot allocator + * nr_memmap_pages: # of pages that were allocated by buddy allocator + */ +static atomic_long_t nr_memmap_boot_pages = ATOMIC_LONG_INIT(0); +static atomic_long_t nr_memmap_pages = ATOMIC_LONG_INIT(0); + +void memmap_boot_pages_add(long delta) +{ + atomic_long_add(delta, &nr_memmap_boot_pages); +} + +void memmap_pages_add(long delta) +{ + atomic_long_add(delta, &nr_memmap_pages); +} + #ifdef CONFIG_COMPACTION struct contig_page_info { @@ -1255,11 +1273,11 @@ const char * const vmstat_text[] = { "pgdemote_kswapd", "pgdemote_direct", "pgdemote_khugepaged", - "nr_memmap", - "nr_memmap_boot", /* system-wide enum vm_stat_item counters */ "nr_dirty_threshold", "nr_dirty_background_threshold", + "nr_memmap_pages", + "nr_memmap_boot_pages", #if defined(CONFIG_VM_EVENT_COUNTERS) || defined(CONFIG_MEMCG) /* enum vm_event_item counters */ @@ -1827,6 +1845,8 @@ static void *vmstat_start(struct seq_file *m, loff_t *pos) global_dirty_limits(v + NR_DIRTY_BG_THRESHOLD, v + NR_DIRTY_THRESHOLD); + v[NR_MEMMAP_PAGES] = atomic_long_read(&nr_memmap_pages); + v[NR_MEMMAP_BOOT_PAGES] = atomic_long_read(&nr_memmap_boot_pages); v += NR_VM_STAT_ITEMS; #ifdef CONFIG_VM_EVENT_COUNTERS @@ -2285,25 +2305,3 @@ static int __init extfrag_debug_init(void) module_init(extfrag_debug_init); #endif - -/* - * Page metadata size (struct page and page_ext) in pages - */ -static unsigned long early_perpage_metadata[MAX_NUMNODES] __meminitdata; - -void __meminit mod_node_early_perpage_metadata(int nid, long delta) -{ - early_perpage_metadata[nid] += delta; -} - -void __meminit store_early_perpage_metadata(void) -{ - int nid; - struct pglist_data *pgdat; - - for_each_online_pgdat(pgdat) { - nid = pgdat->node_id; - mod_node_page_state(NODE_DATA(nid), NR_MEMMAP_BOOT, - early_perpage_metadata[nid]); - } -} From d75abd0d0bc29e6ebfebbf76d11b4067b35844af Mon Sep 17 00:00:00 2001 From: Waiman Long Date: Tue, 6 Aug 2024 12:41:07 -0400 Subject: [PATCH 06/16] mm/memory-failure: use raw_spinlock_t in struct memory_failure_cpu The memory_failure_cpu structure is a per-cpu structure. Access to its content requires the use of get_cpu_var() to lock in the current CPU and disable preemption. The use of a regular spinlock_t for locking purpose is fine for a non-RT kernel. Since the integration of RT spinlock support into the v5.15 kernel, a spinlock_t in a RT kernel becomes a sleeping lock and taking a sleeping lock in a preemption disabled context is illegal resulting in the following kind of warning. [12135.732244] BUG: sleeping function called from invalid context at kernel/locking/spinlock_rt.c:48 [12135.732248] in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 270076, name: kworker/0:0 [12135.732252] preempt_count: 1, expected: 0 [12135.732255] RCU nest depth: 2, expected: 2 : [12135.732420] Hardware name: Dell Inc. PowerEdge R640/0HG0J8, BIOS 2.10.2 02/24/2021 [12135.732423] Workqueue: kacpi_notify acpi_os_execute_deferred [12135.732433] Call Trace: [12135.732436] [12135.732450] dump_stack_lvl+0x57/0x81 [12135.732461] __might_resched.cold+0xf4/0x12f [12135.732479] rt_spin_lock+0x4c/0x100 [12135.732491] memory_failure_queue+0x40/0xe0 [12135.732503] ghes_do_memory_failure+0x53/0x390 [12135.732516] ghes_do_proc.constprop.0+0x229/0x3e0 [12135.732575] ghes_proc+0xf9/0x1a0 [12135.732591] ghes_notify_hed+0x6a/0x150 [12135.732602] notifier_call_chain+0x43/0xb0 [12135.732626] blocking_notifier_call_chain+0x43/0x60 [12135.732637] acpi_ev_notify_dispatch+0x47/0x70 [12135.732648] acpi_os_execute_deferred+0x13/0x20 [12135.732654] process_one_work+0x41f/0x500 [12135.732695] worker_thread+0x192/0x360 [12135.732715] kthread+0x111/0x140 [12135.732733] ret_from_fork+0x29/0x50 [12135.732779] Fix it by using a raw_spinlock_t for locking instead. Also move the pr_err() out of the lock critical section and after put_cpu_ptr() to avoid indeterminate latency and the possibility of sleep with this call. [longman@redhat.com: don't hold percpu ref across pr_err(), per Miaohe] Link: https://lkml.kernel.org/r/20240807181130.1122660-1-longman@redhat.com Link: https://lkml.kernel.org/r/20240806164107.1044956-1-longman@redhat.com Fixes: 0f383b6dc96e ("locking/spinlock: Provide RT variant") Signed-off-by: Waiman Long Acked-by: Miaohe Lin Cc: "Huang, Ying" Cc: Juri Lelli Cc: Len Brown Cc: Naoya Horiguchi Cc: Signed-off-by: Andrew Morton --- mm/memory-failure.c | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 581d3e5c9117..7066fc84f351 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -2417,7 +2417,7 @@ struct memory_failure_entry { struct memory_failure_cpu { DECLARE_KFIFO(fifo, struct memory_failure_entry, MEMORY_FAILURE_FIFO_SIZE); - spinlock_t lock; + raw_spinlock_t lock; struct work_struct work; }; @@ -2443,20 +2443,22 @@ void memory_failure_queue(unsigned long pfn, int flags) { struct memory_failure_cpu *mf_cpu; unsigned long proc_flags; + bool buffer_overflow; struct memory_failure_entry entry = { .pfn = pfn, .flags = flags, }; mf_cpu = &get_cpu_var(memory_failure_cpu); - spin_lock_irqsave(&mf_cpu->lock, proc_flags); - if (kfifo_put(&mf_cpu->fifo, entry)) + raw_spin_lock_irqsave(&mf_cpu->lock, proc_flags); + buffer_overflow = !kfifo_put(&mf_cpu->fifo, entry); + if (!buffer_overflow) schedule_work_on(smp_processor_id(), &mf_cpu->work); - else + raw_spin_unlock_irqrestore(&mf_cpu->lock, proc_flags); + put_cpu_var(memory_failure_cpu); + if (buffer_overflow) pr_err("buffer overflow when queuing memory failure at %#lx\n", pfn); - spin_unlock_irqrestore(&mf_cpu->lock, proc_flags); - put_cpu_var(memory_failure_cpu); } EXPORT_SYMBOL_GPL(memory_failure_queue); @@ -2469,9 +2471,9 @@ static void memory_failure_work_func(struct work_struct *work) mf_cpu = container_of(work, struct memory_failure_cpu, work); for (;;) { - spin_lock_irqsave(&mf_cpu->lock, proc_flags); + raw_spin_lock_irqsave(&mf_cpu->lock, proc_flags); gotten = kfifo_get(&mf_cpu->fifo, &entry); - spin_unlock_irqrestore(&mf_cpu->lock, proc_flags); + raw_spin_unlock_irqrestore(&mf_cpu->lock, proc_flags); if (!gotten) break; if (entry.flags & MF_SOFT_OFFLINE) @@ -2501,7 +2503,7 @@ static int __init memory_failure_init(void) for_each_possible_cpu(cpu) { mf_cpu = &per_cpu(memory_failure_cpu, cpu); - spin_lock_init(&mf_cpu->lock); + raw_spin_lock_init(&mf_cpu->lock); INIT_KFIFO(mf_cpu->fifo); INIT_WORK(&mf_cpu->work, memory_failure_work_func); } From 61ebe5a747da649057c37be1c37eb934b4af79ca Mon Sep 17 00:00:00 2001 From: Hailong Liu Date: Thu, 8 Aug 2024 20:19:56 +0800 Subject: [PATCH 07/16] mm/vmalloc: fix page mapping if vm_area_alloc_pages() with high order fallback to order 0 The __vmap_pages_range_noflush() assumes its argument pages** contains pages with the same page shift. However, since commit e9c3cda4d86e ("mm, vmalloc: fix high order __GFP_NOFAIL allocations"), if gfp_flags includes __GFP_NOFAIL with high order in vm_area_alloc_pages() and page allocation failed for high order, the pages** may contain two different page shifts (high order and order-0). This could lead __vmap_pages_range_noflush() to perform incorrect mappings, potentially resulting in memory corruption. Users might encounter this as follows (vmap_allow_huge = true, 2M is for PMD_SIZE): kvmalloc(2M, __GFP_NOFAIL|GFP_X) __vmalloc_node_range_noprof(vm_flags=VM_ALLOW_HUGE_VMAP) vm_area_alloc_pages(order=9) ---> order-9 allocation failed and fallback to order-0 vmap_pages_range() vmap_pages_range_noflush() __vmap_pages_range_noflush(page_shift = 21) ----> wrong mapping happens We can remove the fallback code because if a high-order allocation fails, __vmalloc_node_range_noprof() will retry with order-0. Therefore, it is unnecessary to fallback to order-0 here. Therefore, fix this by removing the fallback code. Link: https://lkml.kernel.org/r/20240808122019.3361-1-hailong.liu@oppo.com Fixes: e9c3cda4d86e ("mm, vmalloc: fix high order __GFP_NOFAIL allocations") Signed-off-by: Hailong Liu Reported-by: Tangquan Zheng Reviewed-by: Baoquan He Reviewed-by: Uladzislau Rezki (Sony) Acked-by: Barry Song Acked-by: Michal Hocko Cc: Matthew Wilcox Cc: Signed-off-by: Andrew Morton --- mm/vmalloc.c | 11 ++--------- 1 file changed, 2 insertions(+), 9 deletions(-) diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 6b783baf12a1..af2de36549d6 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -3584,15 +3584,8 @@ vm_area_alloc_pages(gfp_t gfp, int nid, page = alloc_pages_noprof(alloc_gfp, order); else page = alloc_pages_node_noprof(nid, alloc_gfp, order); - if (unlikely(!page)) { - if (!nofail) - break; - - /* fall back to the zero order allocations */ - alloc_gfp |= __GFP_NOFAIL; - order = 0; - continue; - } + if (unlikely(!page)) + break; /* * Higher order allocations must be able to be treated as From 40b760cfd44566bca791c80e0720d70d75382b84 Mon Sep 17 00:00:00 2001 From: Zi Yan Date: Fri, 9 Aug 2024 10:59:04 -0400 Subject: [PATCH 08/16] mm/numa: no task_numa_fault() call if PTE is changed When handling a numa page fault, task_numa_fault() should be called by a process that restores the page table of the faulted folio to avoid duplicated stats counting. Commit b99a342d4f11 ("NUMA balancing: reduce TLB flush via delaying mapping on hint page fault") restructured do_numa_page() and did not avoid task_numa_fault() call in the second page table check after a numa migration failure. Fix it by making all !pte_same() return immediately. This issue can cause task_numa_fault() being called more than necessary and lead to unexpected numa balancing results (It is hard to tell whether the issue will cause positive or negative performance impact due to duplicated numa fault counting). Link: https://lkml.kernel.org/r/20240809145906.1513458-2-ziy@nvidia.com Fixes: b99a342d4f11 ("NUMA balancing: reduce TLB flush via delaying mapping on hint page fault") Signed-off-by: Zi Yan Reported-by: "Huang, Ying" Closes: https://lore.kernel.org/linux-mm/87zfqfw0yw.fsf@yhuang6-desk2.ccr.corp.intel.com/ Acked-by: David Hildenbrand Cc: Baolin Wang Cc: Kefeng Wang Cc: Mel Gorman Cc: Yang Shi Cc: Signed-off-by: Andrew Morton --- mm/memory.c | 33 ++++++++++++++++----------------- 1 file changed, 16 insertions(+), 17 deletions(-) diff --git a/mm/memory.c b/mm/memory.c index 34f8402d2046..3c01d68065be 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -5295,7 +5295,7 @@ static vm_fault_t do_numa_page(struct vm_fault *vmf) if (unlikely(!pte_same(old_pte, vmf->orig_pte))) { pte_unmap_unlock(vmf->pte, vmf->ptl); - goto out; + return 0; } pte = pte_modify(old_pte, vma->vm_page_prot); @@ -5358,23 +5358,19 @@ static vm_fault_t do_numa_page(struct vm_fault *vmf) if (!migrate_misplaced_folio(folio, vma, target_nid)) { nid = target_nid; flags |= TNF_MIGRATED; - } else { - flags |= TNF_MIGRATE_FAIL; - vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd, - vmf->address, &vmf->ptl); - if (unlikely(!vmf->pte)) - goto out; - if (unlikely(!pte_same(ptep_get(vmf->pte), vmf->orig_pte))) { - pte_unmap_unlock(vmf->pte, vmf->ptl); - goto out; - } - goto out_map; + task_numa_fault(last_cpupid, nid, nr_pages, flags); + return 0; } -out: - if (nid != NUMA_NO_NODE) - task_numa_fault(last_cpupid, nid, nr_pages, flags); - return 0; + flags |= TNF_MIGRATE_FAIL; + vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd, + vmf->address, &vmf->ptl); + if (unlikely(!vmf->pte)) + return 0; + if (unlikely(!pte_same(ptep_get(vmf->pte), vmf->orig_pte))) { + pte_unmap_unlock(vmf->pte, vmf->ptl); + return 0; + } out_map: /* * Make it present again, depending on how arch implements @@ -5387,7 +5383,10 @@ out_map: numa_rebuild_single_mapping(vmf, vma, vmf->address, vmf->pte, writable); pte_unmap_unlock(vmf->pte, vmf->ptl); - goto out; + + if (nid != NUMA_NO_NODE) + task_numa_fault(last_cpupid, nid, nr_pages, flags); + return 0; } static inline vm_fault_t create_huge_pmd(struct vm_fault *vmf) From fd8c35a92910f4829b7c99841f39b1b952c259d5 Mon Sep 17 00:00:00 2001 From: Zi Yan Date: Fri, 9 Aug 2024 10:59:05 -0400 Subject: [PATCH 09/16] mm/numa: no task_numa_fault() call if PMD is changed When handling a numa page fault, task_numa_fault() should be called by a process that restores the page table of the faulted folio to avoid duplicated stats counting. Commit c5b5a3dd2c1f ("mm: thp: refactor NUMA fault handling") restructured do_huge_pmd_numa_page() and did not avoid task_numa_fault() call in the second page table check after a numa migration failure. Fix it by making all !pmd_same() return immediately. This issue can cause task_numa_fault() being called more than necessary and lead to unexpected numa balancing results (It is hard to tell whether the issue will cause positive or negative performance impact due to duplicated numa fault counting). Link: https://lkml.kernel.org/r/20240809145906.1513458-3-ziy@nvidia.com Fixes: c5b5a3dd2c1f ("mm: thp: refactor NUMA fault handling") Reported-by: "Huang, Ying" Closes: https://lore.kernel.org/linux-mm/87zfqfw0yw.fsf@yhuang6-desk2.ccr.corp.intel.com/ Signed-off-by: Zi Yan Acked-by: David Hildenbrand Cc: Baolin Wang Cc: "Huang, Ying" Cc: Kefeng Wang Cc: Mel Gorman Cc: Yang Shi Cc: Signed-off-by: Andrew Morton --- mm/huge_memory.c | 29 +++++++++++++---------------- 1 file changed, 13 insertions(+), 16 deletions(-) diff --git a/mm/huge_memory.c b/mm/huge_memory.c index f4be468e06a4..67c86a5d64a6 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1685,7 +1685,7 @@ vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf) vmf->ptl = pmd_lock(vma->vm_mm, vmf->pmd); if (unlikely(!pmd_same(oldpmd, *vmf->pmd))) { spin_unlock(vmf->ptl); - goto out; + return 0; } pmd = pmd_modify(oldpmd, vma->vm_page_prot); @@ -1728,22 +1728,16 @@ vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf) if (!migrate_misplaced_folio(folio, vma, target_nid)) { flags |= TNF_MIGRATED; nid = target_nid; - } else { - flags |= TNF_MIGRATE_FAIL; - vmf->ptl = pmd_lock(vma->vm_mm, vmf->pmd); - if (unlikely(!pmd_same(oldpmd, *vmf->pmd))) { - spin_unlock(vmf->ptl); - goto out; - } - goto out_map; + task_numa_fault(last_cpupid, nid, HPAGE_PMD_NR, flags); + return 0; } -out: - if (nid != NUMA_NO_NODE) - task_numa_fault(last_cpupid, nid, HPAGE_PMD_NR, flags); - - return 0; - + flags |= TNF_MIGRATE_FAIL; + vmf->ptl = pmd_lock(vma->vm_mm, vmf->pmd); + if (unlikely(!pmd_same(oldpmd, *vmf->pmd))) { + spin_unlock(vmf->ptl); + return 0; + } out_map: /* Restore the PMD */ pmd = pmd_modify(oldpmd, vma->vm_page_prot); @@ -1753,7 +1747,10 @@ out_map: set_pmd_at(vma->vm_mm, haddr, vmf->pmd, pmd); update_mmu_cache_pmd(vma, vmf->address, vmf->pmd); spin_unlock(vmf->ptl); - goto out; + + if (nid != NUMA_NO_NODE) + task_numa_fault(last_cpupid, nid, HPAGE_PMD_NR, flags); + return 0; } /* From af3b7d09a9934220a8136065a0e6985fe0b67a1b Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 9 Aug 2024 15:32:30 +0300 Subject: [PATCH 10/16] selftests/mm: compaction_test: fix off by one in check_compaction() The "initial_nr_hugepages" variable is unsigned long so it takes up to 20 characters to print, plus 1 more character for the NUL terminator. Unfortunately, this buffer is not quite large enough for the terminator to fit. Also use snprintf() for a belt and suspenders approach. Link: https://lkml.kernel.org/r/87470c06-b45a-4e83-92ff-aac2e7b9c6ba@stanley.mountain Fixes: fb9293b6b015 ("selftests/mm: compaction_test: fix bogus test success and reduce probability of OOM-killer invocation") Signed-off-by: Dan Carpenter Cc: Shuah Khan Signed-off-by: Andrew Morton --- tools/testing/selftests/mm/compaction_test.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/mm/compaction_test.c b/tools/testing/selftests/mm/compaction_test.c index e140558e6f53..2c3a0eb6b22d 100644 --- a/tools/testing/selftests/mm/compaction_test.c +++ b/tools/testing/selftests/mm/compaction_test.c @@ -89,9 +89,10 @@ int check_compaction(unsigned long mem_free, unsigned long hugepage_size, int fd, ret = -1; int compaction_index = 0; char nr_hugepages[20] = {0}; - char init_nr_hugepages[20] = {0}; + char init_nr_hugepages[24] = {0}; - sprintf(init_nr_hugepages, "%lu", initial_nr_hugepages); + snprintf(init_nr_hugepages, sizeof(init_nr_hugepages), + "%lu", initial_nr_hugepages); /* We want to test with 80% of available memory. Else, OOM killer comes in to play */ From 807174a93d24c456503692dc3f5af322ee0b640a Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Fri, 9 Aug 2024 14:48:47 +0300 Subject: [PATCH 11/16] mm: fix endless reclaim on machines with unaccepted memory Unaccepted memory is considered unusable free memory, which is not counted as free on the zone watermark check. This causes get_page_from_freelist() to accept more memory to hit the high watermark, but it creates problems in the reclaim path. The reclaim path encounters a failed zone watermark check and attempts to reclaim memory. This is usually successful, but if there is little or no reclaimable memory, it can result in endless reclaim with little to no progress. This can occur early in the boot process, just after start of the init process when the only reclaimable memory is the page cache of the init executable and its libraries. Make unaccepted memory free from watermark check point of view. This way unaccepted memory will never be the trigger of memory reclaim. Accept more memory in the get_page_from_freelist() if needed. Link: https://lkml.kernel.org/r/20240809114854.3745464-2-kirill.shutemov@linux.intel.com Fixes: dcdfdd40fa82 ("mm: Add support for unaccepted memory") Signed-off-by: Kirill A. Shutemov Reported-by: Jianxiong Gao Acked-by: David Hildenbrand Tested-by: Jianxiong Gao Cc: Borislav Petkov Cc: Johannes Weiner Cc: Kirill A. Shutemov Cc: Matthew Wilcox Cc: Mel Gorman Cc: Mike Rapoport (Microsoft) Cc: Tom Lendacky Cc: Vlastimil Babka Cc: [6.5+] Signed-off-by: Andrew Morton --- mm/page_alloc.c | 42 ++++++++++++++++++++---------------------- 1 file changed, 20 insertions(+), 22 deletions(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 875d76e8684a..8747087acee3 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -287,7 +287,7 @@ EXPORT_SYMBOL(nr_online_nodes); static bool page_contains_unaccepted(struct page *page, unsigned int order); static void accept_page(struct page *page, unsigned int order); -static bool try_to_accept_memory(struct zone *zone, unsigned int order); +static bool cond_accept_memory(struct zone *zone, unsigned int order); static inline bool has_unaccepted_memory(void); static bool __free_unaccepted(struct page *page); @@ -3072,9 +3072,6 @@ static inline long __zone_watermark_unusable_free(struct zone *z, if (!(alloc_flags & ALLOC_CMA)) unusable_free += zone_page_state(z, NR_FREE_CMA_PAGES); #endif -#ifdef CONFIG_UNACCEPTED_MEMORY - unusable_free += zone_page_state(z, NR_UNACCEPTED); -#endif return unusable_free; } @@ -3368,6 +3365,8 @@ retry: } } + cond_accept_memory(zone, order); + /* * Detect whether the number of free pages is below high * watermark. If so, we will decrease pcp->high and free @@ -3393,10 +3392,8 @@ check_alloc_wmark: gfp_mask)) { int ret; - if (has_unaccepted_memory()) { - if (try_to_accept_memory(zone, order)) - goto try_this_zone; - } + if (cond_accept_memory(zone, order)) + goto try_this_zone; #ifdef CONFIG_DEFERRED_STRUCT_PAGE_INIT /* @@ -3450,10 +3447,8 @@ try_this_zone: return page; } else { - if (has_unaccepted_memory()) { - if (try_to_accept_memory(zone, order)) - goto try_this_zone; - } + if (cond_accept_memory(zone, order)) + goto try_this_zone; #ifdef CONFIG_DEFERRED_STRUCT_PAGE_INIT /* Try again if zone has deferred pages */ @@ -6950,9 +6945,6 @@ static bool try_to_accept_memory_one(struct zone *zone) struct page *page; bool last; - if (list_empty(&zone->unaccepted_pages)) - return false; - spin_lock_irqsave(&zone->lock, flags); page = list_first_entry_or_null(&zone->unaccepted_pages, struct page, lru); @@ -6978,23 +6970,29 @@ static bool try_to_accept_memory_one(struct zone *zone) return true; } -static bool try_to_accept_memory(struct zone *zone, unsigned int order) +static bool cond_accept_memory(struct zone *zone, unsigned int order) { long to_accept; - int ret = false; + bool ret = false; + + if (!has_unaccepted_memory()) + return false; + + if (list_empty(&zone->unaccepted_pages)) + return false; /* How much to accept to get to high watermark? */ to_accept = high_wmark_pages(zone) - (zone_page_state(zone, NR_FREE_PAGES) - - __zone_watermark_unusable_free(zone, order, 0)); + __zone_watermark_unusable_free(zone, order, 0) - + zone_page_state(zone, NR_UNACCEPTED)); - /* Accept at least one page */ - do { + while (to_accept > 0) { if (!try_to_accept_memory_one(zone)) break; ret = true; to_accept -= MAX_ORDER_NR_PAGES; - } while (to_accept > 0); + } return ret; } @@ -7037,7 +7035,7 @@ static void accept_page(struct page *page, unsigned int order) { } -static bool try_to_accept_memory(struct zone *zone, unsigned int order) +static bool cond_accept_memory(struct zone *zone, unsigned int order) { return false; } From 7c5e8d212d7d81991a580e7de3904ea213d9a852 Mon Sep 17 00:00:00 2001 From: Muhammad Usama Anjum Date: Fri, 9 Aug 2024 12:56:42 +0500 Subject: [PATCH 12/16] selftests: memfd_secret: don't build memfd_secret test on unsupported arches [1] mentions that memfd_secret is only supported on arm64, riscv, x86 and x86_64 for now. It doesn't support other architectures. I found the build error on arm and decided to send the fix as it was creating noise on KernelCI: memfd_secret.c: In function 'memfd_secret': memfd_secret.c:42:24: error: '__NR_memfd_secret' undeclared (first use in this function); did you mean 'memfd_secret'? 42 | return syscall(__NR_memfd_secret, flags); | ^~~~~~~~~~~~~~~~~ | memfd_secret Hence I'm adding condition that memfd_secret should only be compiled on supported architectures. Also check in run_vmtests script if memfd_secret binary is present before executing it. Link: https://lkml.kernel.org/r/20240812061522.1933054-1-usama.anjum@collabora.com Link: https://lore.kernel.org/all/20210518072034.31572-7-rppt@kernel.org/ [1] Link: https://lkml.kernel.org/r/20240809075642.403247-1-usama.anjum@collabora.com Fixes: 76fe17ef588a ("secretmem: test: add basic selftest for memfd_secret(2)") Signed-off-by: Muhammad Usama Anjum Reviewed-by: Shuah Khan Acked-by: Mike Rapoport (Microsoft) Cc: Albert Ou Cc: James Bottomley Cc: Mike Rapoport (Microsoft) Cc: Palmer Dabbelt Cc: Paul Walmsley Cc: Signed-off-by: Andrew Morton --- tools/testing/selftests/mm/Makefile | 2 ++ tools/testing/selftests/mm/run_vmtests.sh | 3 +++ 2 files changed, 5 insertions(+) diff --git a/tools/testing/selftests/mm/Makefile b/tools/testing/selftests/mm/Makefile index 7b8a5def54a1..cfad627e8d94 100644 --- a/tools/testing/selftests/mm/Makefile +++ b/tools/testing/selftests/mm/Makefile @@ -53,7 +53,9 @@ TEST_GEN_FILES += madv_populate TEST_GEN_FILES += map_fixed_noreplace TEST_GEN_FILES += map_hugetlb TEST_GEN_FILES += map_populate +ifneq (,$(filter $(ARCH),arm64 riscv riscv64 x86 x86_64)) TEST_GEN_FILES += memfd_secret +endif TEST_GEN_FILES += migration TEST_GEN_FILES += mkdirty TEST_GEN_FILES += mlock-random-test diff --git a/tools/testing/selftests/mm/run_vmtests.sh b/tools/testing/selftests/mm/run_vmtests.sh index 03ac4f2e1cce..36045edb10de 100755 --- a/tools/testing/selftests/mm/run_vmtests.sh +++ b/tools/testing/selftests/mm/run_vmtests.sh @@ -374,8 +374,11 @@ CATEGORY="hmm" run_test bash ./test_hmm.sh smoke # MADV_POPULATE_READ and MADV_POPULATE_WRITE tests CATEGORY="madv_populate" run_test ./madv_populate +if [ -x ./memfd_secret ] +then (echo 0 | sudo tee /proc/sys/kernel/yama/ptrace_scope 2>&1) | tap_prefix CATEGORY="memfd_secret" run_test ./memfd_secret +fi # KSM KSM_MERGE_TIME_HUGE_PAGES test with size of 100 CATEGORY="ksm" run_test ./ksm_tests -H -s 100 From edb907a6133323e19311901a39dee68b1c6a2ef8 Mon Sep 17 00:00:00 2001 From: Jinjie Ruan Date: Mon, 12 Aug 2024 14:20:17 +0800 Subject: [PATCH 13/16] crash: fix riscv64 crash memory reserve dead loop On RISCV64 Qemu machine with 512MB memory, cmdline "crashkernel=500M,high" will cause system stall as below: Zone ranges: DMA32 [mem 0x0000000080000000-0x000000009fffffff] Normal empty Movable zone start for each node Early memory node ranges node 0: [mem 0x0000000080000000-0x000000008005ffff] node 0: [mem 0x0000000080060000-0x000000009fffffff] Initmem setup node 0 [mem 0x0000000080000000-0x000000009fffffff] (stall here) commit 5d99cadf1568 ("crash: fix x86_32 crash memory reserve dead loop bug") fix this on 32-bit architecture. However, the problem is not completely solved. If `CRASH_ADDR_LOW_MAX = CRASH_ADDR_HIGH_MAX` on 64-bit architecture, for example, when system memory is equal to CRASH_ADDR_LOW_MAX on RISCV64, the following infinite loop will also occur: -> reserve_crashkernel_generic() and high is true -> alloc at [CRASH_ADDR_LOW_MAX, CRASH_ADDR_HIGH_MAX] fail -> alloc at [0, CRASH_ADDR_LOW_MAX] fail and repeatedly (because CRASH_ADDR_LOW_MAX = CRASH_ADDR_HIGH_MAX). As Catalin suggested, do not remove the ",high" reservation fallback to ",low" logic which will change arm64's kdump behavior, but fix it by skipping the above situation similar to commit d2f32f23190b ("crash: fix x86_32 crash memory reserve dead loop"). After this patch, it print: cannot allocate crashkernel (size:0x1f400000) Link: https://lkml.kernel.org/r/20240812062017.2674441-1-ruanjinjie@huawei.com Signed-off-by: Jinjie Ruan Suggested-by: Catalin Marinas Reviewed-by: Catalin Marinas Acked-by: Baoquan He Cc: Albert Ou Cc: Dave Young Cc: Palmer Dabbelt Cc: Paul Walmsley Cc: Vivek Goyal Signed-off-by: Andrew Morton --- kernel/crash_reserve.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/kernel/crash_reserve.c b/kernel/crash_reserve.c index d3b4cd12bdd1..64d44a52c011 100644 --- a/kernel/crash_reserve.c +++ b/kernel/crash_reserve.c @@ -423,7 +423,8 @@ retry: if (high && search_end == CRASH_ADDR_HIGH_MAX) { search_end = CRASH_ADDR_LOW_MAX; search_base = 0; - goto retry; + if (search_end != CRASH_ADDR_HIGH_MAX) + goto retry; } pr_warn("cannot allocate crashkernel (size:0x%llx)\n", crash_size); From a8fc28dad6d574582cdf2f7e78c73c59c623df30 Mon Sep 17 00:00:00 2001 From: Suren Baghdasaryan Date: Tue, 13 Aug 2024 08:07:56 -0700 Subject: [PATCH 14/16] alloc_tag: introduce clear_page_tag_ref() helper function In several cases we are freeing pages which were not allocated using common page allocators. For such cases, in order to keep allocation accounting correct, we should clear the page tag to indicate that the page being freed is expected to not have a valid allocation tag. Introduce clear_page_tag_ref() helper function to be used for this. Link: https://lkml.kernel.org/r/20240813150758.855881-1-surenb@google.com Fixes: d224eb0287fb ("codetag: debug: mark codetags for reserved pages as empty") Signed-off-by: Suren Baghdasaryan Suggested-by: David Hildenbrand Acked-by: David Hildenbrand Reviewed-by: Pasha Tatashin Cc: Kees Cook Cc: Kent Overstreet Cc: Sourav Panda Cc: Vlastimil Babka Cc: [6.10] Signed-off-by: Andrew Morton --- include/linux/pgalloc_tag.h | 13 +++++++++++++ mm/mm_init.c | 10 +--------- mm/page_alloc.c | 9 +-------- 3 files changed, 15 insertions(+), 17 deletions(-) diff --git a/include/linux/pgalloc_tag.h b/include/linux/pgalloc_tag.h index 18cd0c0c73d9..207f0c83c8e9 100644 --- a/include/linux/pgalloc_tag.h +++ b/include/linux/pgalloc_tag.h @@ -43,6 +43,18 @@ static inline void put_page_tag_ref(union codetag_ref *ref) page_ext_put(page_ext_from_codetag_ref(ref)); } +static inline void clear_page_tag_ref(struct page *page) +{ + if (mem_alloc_profiling_enabled()) { + union codetag_ref *ref = get_page_tag_ref(page); + + if (ref) { + set_codetag_empty(ref); + put_page_tag_ref(ref); + } + } +} + static inline void pgalloc_tag_add(struct page *page, struct task_struct *task, unsigned int nr) { @@ -126,6 +138,7 @@ static inline void pgalloc_tag_sub_pages(struct alloc_tag *tag, unsigned int nr) static inline union codetag_ref *get_page_tag_ref(struct page *page) { return NULL; } static inline void put_page_tag_ref(union codetag_ref *ref) {} +static inline void clear_page_tag_ref(struct page *page) {} static inline void pgalloc_tag_add(struct page *page, struct task_struct *task, unsigned int nr) {} static inline void pgalloc_tag_sub(struct page *page, unsigned int nr) {} diff --git a/mm/mm_init.c b/mm/mm_init.c index f9a60ffc5532..adc3127573cd 100644 --- a/mm/mm_init.c +++ b/mm/mm_init.c @@ -2459,15 +2459,7 @@ void __init memblock_free_pages(struct page *page, unsigned long pfn, } /* pages were reserved and not allocated */ - if (mem_alloc_profiling_enabled()) { - union codetag_ref *ref = get_page_tag_ref(page); - - if (ref) { - set_codetag_empty(ref); - put_page_tag_ref(ref); - } - } - + clear_page_tag_ref(page); __free_pages_core(page, order, MEMINIT_EARLY); } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 8747087acee3..c565de8f48e9 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -5815,14 +5815,7 @@ unsigned long free_reserved_area(void *start, void *end, int poison, const char void free_reserved_page(struct page *page) { - if (mem_alloc_profiling_enabled()) { - union codetag_ref *ref = get_page_tag_ref(page); - - if (ref) { - set_codetag_empty(ref); - put_page_tag_ref(ref); - } - } + clear_page_tag_ref(page); ClearPageReserved(page); init_page_count(page); __free_page(page); From 766c163c2068b45330664fb67df67268e588a22d Mon Sep 17 00:00:00 2001 From: Suren Baghdasaryan Date: Tue, 13 Aug 2024 08:07:57 -0700 Subject: [PATCH 15/16] alloc_tag: mark pages reserved during CMA activation as not tagged During CMA activation, pages in CMA area are prepared and then freed without being allocated. This triggers warnings when memory allocation debug config (CONFIG_MEM_ALLOC_PROFILING_DEBUG) is enabled. Fix this by marking these pages not tagged before freeing them. Link: https://lkml.kernel.org/r/20240813150758.855881-2-surenb@google.com Fixes: d224eb0287fb ("codetag: debug: mark codetags for reserved pages as empty") Signed-off-by: Suren Baghdasaryan Acked-by: David Hildenbrand Cc: Kees Cook Cc: Kent Overstreet Cc: Pasha Tatashin Cc: Sourav Panda Cc: Vlastimil Babka Cc: [6.10] Signed-off-by: Andrew Morton --- mm/mm_init.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/mm/mm_init.c b/mm/mm_init.c index adc3127573cd..51960079875b 100644 --- a/mm/mm_init.c +++ b/mm/mm_init.c @@ -2244,6 +2244,8 @@ void __init init_cma_reserved_pageblock(struct page *page) set_pageblock_migratetype(page, MIGRATE_CMA); set_page_refcounted(page); + /* pages were reserved and not allocated */ + clear_page_tag_ref(page); __free_pages(page, pageblock_order); adjust_managed_page_count(page, pageblock_nr_pages); From 2e6506e1c4eed2676a8412231046f31e10e240da Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Mon, 29 Jul 2024 10:13:06 +0800 Subject: [PATCH 16/16] mm/migrate: fix deadlock in migrate_pages_batch() on large folios Currently, migrate_pages_batch() can lock multiple locked folios with an arbitrary order. Although folio_trylock() is used to avoid deadlock as commit 2ef7dbb26990 ("migrate_pages: try migrate in batch asynchronously firstly") mentioned, it seems try_split_folio() is still missing. It was found by compaction stress test when I explicitly enable EROFS compressed files to use large folios, which case I cannot reproduce with the same workload if large folio support is off (current mainline). Typically, filesystem reads (with locked file-backed folios) could use another bdev/meta inode to load some other I/Os (e.g. inode extent metadata or caching compressed data), so the locking order will be: file-backed folios (A) bdev/meta folios (B) The following calltrace shows the deadlock: Thread 1 takes (B) lock and tries to take folio (A) lock Thread 2 takes (A) lock and tries to take folio (B) lock [Thread 1] INFO: task stress:1824 blocked for more than 30 seconds. Tainted: G OE 6.10.0-rc7+ #6 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. task:stress state:D stack:0 pid:1824 tgid:1824 ppid:1822 flags:0x0000000c Call trace: __switch_to+0xec/0x138 __schedule+0x43c/0xcb0 schedule+0x54/0x198 io_schedule+0x44/0x70 folio_wait_bit_common+0x184/0x3f8 <-- folio mapping ffff00036d69cb18 index 996 (**) __folio_lock+0x24/0x38 migrate_pages_batch+0x77c/0xea0 // try_split_folio (mm/migrate.c:1486:2) // migrate_pages_batch (mm/migrate.c:1734:16) <--- LIST_HEAD(unmap_folios) has .. folio mapping 0xffff0000d184f1d8 index 1711; (*) folio mapping 0xffff0000d184f1d8 index 1712; .. migrate_pages+0xb28/0xe90 compact_zone+0xa08/0x10f0 compact_node+0x9c/0x180 sysctl_compaction_handler+0x8c/0x118 proc_sys_call_handler+0x1a8/0x280 proc_sys_write+0x1c/0x30 vfs_write+0x240/0x380 ksys_write+0x78/0x118 __arm64_sys_write+0x24/0x38 invoke_syscall+0x78/0x108 el0_svc_common.constprop.0+0x48/0xf0 do_el0_svc+0x24/0x38 el0_svc+0x3c/0x148 el0t_64_sync_handler+0x100/0x130 el0t_64_sync+0x190/0x198 [Thread 2] INFO: task stress:1825 blocked for more than 30 seconds. Tainted: G OE 6.10.0-rc7+ #6 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. task:stress state:D stack:0 pid:1825 tgid:1825 ppid:1822 flags:0x0000000c Call trace: __switch_to+0xec/0x138 __schedule+0x43c/0xcb0 schedule+0x54/0x198 io_schedule+0x44/0x70 folio_wait_bit_common+0x184/0x3f8 <-- folio = 0xfffffdffc6b503c0 (mapping == 0xffff0000d184f1d8 index == 1711) (*) __folio_lock+0x24/0x38 z_erofs_runqueue+0x384/0x9c0 [erofs] z_erofs_readahead+0x21c/0x350 [erofs] <-- folio mapping 0xffff00036d69cb18 range from [992, 1024] (**) read_pages+0x74/0x328 page_cache_ra_order+0x26c/0x348 ondemand_readahead+0x1c0/0x3a0 page_cache_sync_ra+0x9c/0xc0 filemap_get_pages+0xc4/0x708 filemap_read+0x104/0x3a8 generic_file_read_iter+0x4c/0x150 vfs_read+0x27c/0x330 ksys_pread64+0x84/0xd0 __arm64_sys_pread64+0x28/0x40 invoke_syscall+0x78/0x108 el0_svc_common.constprop.0+0x48/0xf0 do_el0_svc+0x24/0x38 el0_svc+0x3c/0x148 el0t_64_sync_handler+0x100/0x130 el0t_64_sync+0x190/0x198 Link: https://lkml.kernel.org/r/20240729021306.398286-1-hsiangkao@linux.alibaba.com Fixes: 5dfab109d519 ("migrate_pages: batch _unmap and _move") Signed-off-by: Gao Xiang Reviewed-by: "Huang, Ying" Acked-by: David Hildenbrand Cc: Matthew Wilcox Signed-off-by: Andrew Morton --- mm/migrate.c | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) diff --git a/mm/migrate.c b/mm/migrate.c index e7296c0fb5d5..923ea80ba744 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -1479,11 +1479,17 @@ out: return rc; } -static inline int try_split_folio(struct folio *folio, struct list_head *split_folios) +static inline int try_split_folio(struct folio *folio, struct list_head *split_folios, + enum migrate_mode mode) { int rc; - folio_lock(folio); + if (mode == MIGRATE_ASYNC) { + if (!folio_trylock(folio)) + return -EAGAIN; + } else { + folio_lock(folio); + } rc = split_folio_to_list(folio, split_folios); folio_unlock(folio); if (!rc) @@ -1677,7 +1683,7 @@ static int migrate_pages_batch(struct list_head *from, */ if (nr_pages > 2 && !list_empty(&folio->_deferred_list)) { - if (try_split_folio(folio, split_folios) == 0) { + if (!try_split_folio(folio, split_folios, mode)) { nr_failed++; stats->nr_thp_failed += is_thp; stats->nr_thp_split += is_thp; @@ -1699,7 +1705,7 @@ static int migrate_pages_batch(struct list_head *from, if (!thp_migration_supported() && is_thp) { nr_failed++; stats->nr_thp_failed++; - if (!try_split_folio(folio, split_folios)) { + if (!try_split_folio(folio, split_folios, mode)) { stats->nr_thp_split++; stats->nr_split++; continue; @@ -1731,7 +1737,7 @@ static int migrate_pages_batch(struct list_head *from, stats->nr_thp_failed += is_thp; /* Large folio NUMA faulting doesn't split to retry. */ if (is_large && !nosplit) { - int ret = try_split_folio(folio, split_folios); + int ret = try_split_folio(folio, split_folios, mode); if (!ret) { stats->nr_thp_split += is_thp;