diff options
Diffstat (limited to 'mm/huge_memory.c')
| -rw-r--r-- | mm/huge_memory.c | 150 |
1 files changed, 92 insertions, 58 deletions
diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 874000f97b..3f50578eb9 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -37,6 +37,7 @@ #include <linux/page_owner.h> #include <linux/sched/sysctl.h> #include <linux/memory-tiers.h> +#include <linux/compat.h> #include <asm/tlb.h> #include <asm/pgalloc.h> @@ -65,7 +66,11 @@ unsigned long transparent_hugepage_flags __read_mostly = (1<<TRANSPARENT_HUGEPAGE_DEFRAG_KHUGEPAGED_FLAG)| (1<<TRANSPARENT_HUGEPAGE_USE_ZERO_PAGE_FLAG); -static struct shrinker deferred_split_shrinker; +static struct shrinker *deferred_split_shrinker; +static unsigned long deferred_split_count(struct shrinker *shrink, + struct shrink_control *sc); +static unsigned long deferred_split_scan(struct shrinker *shrink, + struct shrink_control *sc); static atomic_t huge_zero_refcount; struct page *huge_zero_page __read_mostly; @@ -96,11 +101,11 @@ bool hugepage_vma_check(struct vm_area_struct *vma, unsigned long vm_flags, return in_pf; /* - * Special VMA and hugetlb VMA. + * khugepaged special VMA and hugetlb VMA. * Must be checked after dax since some dax mappings may have * VM_MIXEDMAP set. */ - if (vm_flags & VM_NO_KHUGEPAGED) + if (!in_pf && !smaps && (vm_flags & VM_NO_KHUGEPAGED)) return false; /* @@ -128,12 +133,18 @@ bool hugepage_vma_check(struct vm_area_struct *vma, unsigned long vm_flags, !hugepage_flags_always()))) return false; - /* Only regular file is valid */ - if (!in_pf && file_thp_enabled(vma)) - return true; - - if (!vma_is_anonymous(vma)) + if (!vma_is_anonymous(vma)) { + /* + * Trust that ->huge_fault() handlers know what they are doing + * in fault path. + */ + if (((in_pf || smaps)) && vma->vm_ops->huge_fault) + return true; + /* Only regular file is valid in collapse path */ + if (((!in_pf || smaps)) && file_thp_enabled(vma)) + return true; return false; + } if (vma_is_temporary_stack(vma)) return false; @@ -229,11 +240,7 @@ static unsigned long shrink_huge_zero_page_scan(struct shrinker *shrink, return 0; } -static struct shrinker huge_zero_page_shrinker = { - .count_objects = shrink_huge_zero_page_count, - .scan_objects = shrink_huge_zero_page_scan, - .seeks = DEFAULT_SEEKS, -}; +static struct shrinker *huge_zero_page_shrinker; #ifdef CONFIG_SYSFS static ssize_t enabled_show(struct kobject *kobj, @@ -454,6 +461,38 @@ static inline void hugepage_exit_sysfs(struct kobject *hugepage_kobj) } #endif /* CONFIG_SYSFS */ +static int __init thp_shrinker_init(void) +{ + huge_zero_page_shrinker = shrinker_alloc(0, "thp-zero"); + if (!huge_zero_page_shrinker) + return -ENOMEM; + + deferred_split_shrinker = shrinker_alloc(SHRINKER_NUMA_AWARE | + SHRINKER_MEMCG_AWARE | + SHRINKER_NONSLAB, + "thp-deferred_split"); + if (!deferred_split_shrinker) { + shrinker_free(huge_zero_page_shrinker); + return -ENOMEM; + } + + huge_zero_page_shrinker->count_objects = shrink_huge_zero_page_count; + huge_zero_page_shrinker->scan_objects = shrink_huge_zero_page_scan; + shrinker_register(huge_zero_page_shrinker); + + deferred_split_shrinker->count_objects = deferred_split_count; + deferred_split_shrinker->scan_objects = deferred_split_scan; + shrinker_register(deferred_split_shrinker); + + return 0; +} + +static void __init thp_shrinker_exit(void) +{ + shrinker_free(huge_zero_page_shrinker); + shrinker_free(deferred_split_shrinker); +} + static int __init hugepage_init(void) { int err; @@ -482,12 +521,9 @@ static int __init hugepage_init(void) if (err) goto err_slab; - err = register_shrinker(&huge_zero_page_shrinker, "thp-zero"); - if (err) - goto err_hzp_shrinker; - err = register_shrinker(&deferred_split_shrinker, "thp-deferred_split"); + err = thp_shrinker_init(); if (err) - goto err_split_shrinker; + goto err_shrinker; /* * By default disable transparent hugepages on smaller systems, @@ -505,10 +541,8 @@ static int __init hugepage_init(void) return 0; err_khugepaged: - unregister_shrinker(&deferred_split_shrinker); -err_split_shrinker: - unregister_shrinker(&huge_zero_page_shrinker); -err_hzp_shrinker: + thp_shrinker_exit(); +err_shrinker: khugepaged_destroy(); err_slab: hugepage_exit_sysfs(hugepage_kobj); @@ -599,7 +633,10 @@ static unsigned long __thp_get_unmapped_area(struct file *filp, { loff_t off_end = off + len; loff_t off_align = round_up(off, size); - unsigned long len_pad, ret; + unsigned long len_pad, ret, off_sub; + + if (IS_ENABLED(CONFIG_32BIT) || in_compat_syscall()) + return 0; if (off_end <= off_align || (off_end - off_align) < size) return 0; @@ -625,7 +662,13 @@ static unsigned long __thp_get_unmapped_area(struct file *filp, if (ret == addr) return addr; - ret += (off - ret) & (size - 1); + off_sub = (off - ret) & (size - 1); + + if (current->mm->get_unmapped_area == arch_get_unmapped_area_topdown && + !off_sub) + return ret + size; + + ret += off_sub; return ret; } @@ -1349,7 +1392,8 @@ vm_fault_t do_huge_pmd_wp_page(struct vm_fault *vmf) if (folio_ref_count(folio) == 1) { pmd_t entry; - page_move_anon_rmap(page, vma); + folio_move_anon_rmap(folio, vma); + SetPageAnonExclusive(page); folio_unlock(folio); reuse: if (unlikely(unshare)) { @@ -1490,9 +1534,9 @@ vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf) struct vm_area_struct *vma = vmf->vma; pmd_t oldpmd = vmf->orig_pmd; pmd_t pmd; - struct page *page; + struct folio *folio; unsigned long haddr = vmf->address & HPAGE_PMD_MASK; - int page_nid = NUMA_NO_NODE; + int nid = NUMA_NO_NODE; int target_nid, last_cpupid = (-1 & LAST_CPUPID_MASK); bool migrated = false, writable = false; int flags = 0; @@ -1514,36 +1558,34 @@ vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf) can_change_pmd_writable(vma, vmf->address, pmd)) writable = true; - page = vm_normal_page_pmd(vma, haddr, pmd); - if (!page) + folio = vm_normal_folio_pmd(vma, haddr, pmd); + if (!folio) goto out_map; /* See similar comment in do_numa_page for explanation */ if (!writable) flags |= TNF_NO_GROUP; - page_nid = page_to_nid(page); + nid = folio_nid(folio); /* * For memory tiering mode, cpupid of slow memory page is used * to record page access time. So use default value. */ - if (node_is_toptier(page_nid)) - last_cpupid = page_cpupid_last(page); - target_nid = numa_migrate_prep(page, vma, haddr, page_nid, - &flags); - + if (node_is_toptier(nid)) + last_cpupid = folio_last_cpupid(folio); + target_nid = numa_migrate_prep(folio, vma, haddr, nid, &flags); if (target_nid == NUMA_NO_NODE) { - put_page(page); + folio_put(folio); goto out_map; } spin_unlock(vmf->ptl); writable = false; - migrated = migrate_misplaced_page(page, vma, target_nid); + migrated = migrate_misplaced_folio(folio, vma, target_nid); if (migrated) { flags |= TNF_MIGRATED; - page_nid = target_nid; + nid = target_nid; } else { flags |= TNF_MIGRATE_FAIL; vmf->ptl = pmd_lock(vma->vm_mm, vmf->pmd); @@ -1555,9 +1597,8 @@ vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf) } out: - if (page_nid != NUMA_NO_NODE) - task_numa_fault(last_cpupid, page_nid, HPAGE_PMD_NR, - flags); + if (nid != NUMA_NO_NODE) + task_numa_fault(last_cpupid, nid, HPAGE_PMD_NR, flags); return 0; @@ -1825,7 +1866,7 @@ int change_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma, #ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION if (is_swap_pmd(*pmd)) { swp_entry_t entry = pmd_to_swp_entry(*pmd); - struct page *page = pfn_swap_entry_to_page(entry); + struct folio *folio = page_folio(pfn_swap_entry_to_page(entry)); pmd_t newpmd; VM_BUG_ON(!is_pmd_migration_entry(*pmd)); @@ -1834,7 +1875,7 @@ int change_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma, * A protection check is difficult so * just be safe and disable write */ - if (PageAnon(page)) + if (folio_test_anon(folio)) entry = make_readable_exclusive_migration_entry(swp_offset(entry)); else entry = make_readable_migration_entry(swp_offset(entry)); @@ -1856,7 +1897,7 @@ int change_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma, #endif if (prot_numa) { - struct page *page; + struct folio *folio; bool toptier; /* * Avoid trapping faults against the zero page. The read-only @@ -1869,8 +1910,8 @@ int change_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma, if (pmd_protnone(*pmd)) goto unlock; - page = pmd_page(*pmd); - toptier = node_is_toptier(page_to_nid(page)); + folio = page_folio(pmd_page(*pmd)); + toptier = node_is_toptier(folio_nid(folio)); /* * Skip scanning top tier node if normal numa * balancing is disabled @@ -1881,7 +1922,8 @@ int change_huge_pmd(struct mmu_gather *tlb, struct vm_area_struct *vma, if (sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING && !toptier) - xchg_page_access_time(page, jiffies_to_msecs(jiffies)); + folio_xchg_access_time(folio, + jiffies_to_msecs(jiffies)); } /* * In case prot_numa, we are under mmap_read_lock(mm). It's critical @@ -2483,7 +2525,7 @@ static void __split_huge_page_tail(struct folio *folio, int tail, if (page_is_idle(head)) set_page_idle(page_tail); - page_cpupid_xchg_last(page_tail, page_cpupid_last(head)); + folio_xchg_last_cpupid(new_folio, folio_last_cpupid(folio)); /* * always add to the tail because some iterators expect new @@ -2791,7 +2833,7 @@ void folio_undo_large_rmappable(struct folio *folio) spin_lock_irqsave(&ds_queue->split_queue_lock, flags); if (!list_empty(&folio->_deferred_list)) { ds_queue->split_queue_len--; - list_del(&folio->_deferred_list); + list_del_init(&folio->_deferred_list); } spin_unlock_irqrestore(&ds_queue->split_queue_lock, flags); } @@ -2830,7 +2872,7 @@ void deferred_split_folio(struct folio *folio) #ifdef CONFIG_MEMCG if (memcg) set_shrinker_bit(memcg, folio_nid(folio), - deferred_split_shrinker.id); + deferred_split_shrinker->id); #endif } spin_unlock_irqrestore(&ds_queue->split_queue_lock, flags); @@ -2904,14 +2946,6 @@ next: return split; } -static struct shrinker deferred_split_shrinker = { - .count_objects = deferred_split_count, - .scan_objects = deferred_split_scan, - .seeks = DEFAULT_SEEKS, - .flags = SHRINKER_NUMA_AWARE | SHRINKER_MEMCG_AWARE | - SHRINKER_NONSLAB, -}; - #ifdef CONFIG_DEBUG_FS static void split_huge_pages_all(void) { |