diff options
Diffstat (limited to 'mm/khugepaged.c')
-rw-r--r-- | mm/khugepaged.c | 154 |
1 files changed, 103 insertions, 51 deletions
diff --git a/mm/khugepaged.c b/mm/khugepaged.c index 4734315f7940..5a7d2d5093f9 100644 --- a/mm/khugepaged.c +++ b/mm/khugepaged.c @@ -97,8 +97,8 @@ struct collapse_control { /* Num pages scanned per node */ u32 node_load[MAX_NUMNODES]; - /* Last target selected in hpage_collapse_find_target_node() */ - int last_target_node; + /* nodemask for allocation fallback */ + nodemask_t alloc_nmask; }; /** @@ -734,7 +734,6 @@ static void khugepaged_alloc_sleep(void) struct collapse_control khugepaged_collapse_control = { .is_khugepaged = true, - .last_target_node = NUMA_NO_NODE, }; static bool hpage_collapse_scan_abort(int nid, struct collapse_control *cc) @@ -783,16 +782,11 @@ static int hpage_collapse_find_target_node(struct collapse_control *cc) target_node = nid; } - /* do some balance if several nodes have the same hit record */ - if (target_node <= cc->last_target_node) - for (nid = cc->last_target_node + 1; nid < MAX_NUMNODES; - nid++) - if (max_value == cc->node_load[nid]) { - target_node = nid; - break; - } + for_each_online_node(nid) { + if (max_value == cc->node_load[nid]) + node_set(nid, cc->alloc_nmask); + } - cc->last_target_node = target_node; return target_node; } #else @@ -802,9 +796,10 @@ static int hpage_collapse_find_target_node(struct collapse_control *cc) } #endif -static bool hpage_collapse_alloc_page(struct page **hpage, gfp_t gfp, int node) +static bool hpage_collapse_alloc_page(struct page **hpage, gfp_t gfp, int node, + nodemask_t *nmask) { - *hpage = __alloc_pages_node(node, gfp, HPAGE_PMD_ORDER); + *hpage = __alloc_pages(gfp, HPAGE_PMD_ORDER, node, nmask); if (unlikely(!*hpage)) { count_vm_event(THP_COLLAPSE_ALLOC_FAILED); return false; @@ -955,12 +950,11 @@ static int __collapse_huge_page_swapin(struct mm_struct *mm, static int alloc_charge_hpage(struct page **hpage, struct mm_struct *mm, struct collapse_control *cc) { - /* Only allocate from the target node */ gfp_t gfp = (cc->is_khugepaged ? alloc_hugepage_khugepaged_gfpmask() : - GFP_TRANSHUGE) | __GFP_THISNODE; + GFP_TRANSHUGE); int node = hpage_collapse_find_target_node(cc); - if (!hpage_collapse_alloc_page(hpage, gfp, node)) + if (!hpage_collapse_alloc_page(hpage, gfp, node, &cc->alloc_nmask)) return SCAN_ALLOC_HUGE_PAGE_FAIL; if (unlikely(mem_cgroup_charge(page_folio(*hpage), mm, gfp))) return SCAN_CGROUP_CHARGE_FAIL; @@ -1057,6 +1051,7 @@ static int collapse_huge_page(struct mm_struct *mm, unsigned long address, _pmd = pmdp_collapse_flush(vma, address, pmd); spin_unlock(pmd_ptl); mmu_notifier_invalidate_range_end(&range); + tlb_remove_table_sync_one(); spin_lock(pte_ptl); result = __collapse_huge_page_isolate(vma, address, pte, cc, @@ -1144,6 +1139,7 @@ static int hpage_collapse_scan_pmd(struct mm_struct *mm, goto out; memset(cc->node_load, 0, sizeof(cc->node_load)); + nodes_clear(cc->alloc_nmask); pte = pte_offset_map_lock(mm, pmd, address, &ptl); for (_address = address, _pte = pte; _pte < pte + HPAGE_PMD_NR; _pte++, _address += PAGE_SIZE) { @@ -1242,15 +1238,8 @@ static int hpage_collapse_scan_pmd(struct mm_struct *mm, /* * Check if the page has any GUP (or other external) pins. * - * Here the check is racy it may see total_mapcount > refcount - * in some cases. - * For example, one process with one forked child process. - * The parent has the PMD split due to MADV_DONTNEED, then - * the child is trying unmap the whole PMD, but khugepaged - * may be scanning the parent between the child has - * PageDoubleMap flag cleared and dec the mapcount. So - * khugepaged may see total_mapcount > refcount. - * + * Here the check may be racy: + * it may see total_mapcount > refcount in some cases? * But such case is ephemeral we could always retry collapse * later. However it may report false positive if the page * has excessive GUP pins (i.e. 512). Anyway the same check @@ -1384,16 +1373,43 @@ static int set_huge_pmd(struct vm_area_struct *vma, unsigned long addr, return SCAN_SUCCEED; } +/* + * A note about locking: + * Trying to take the page table spinlocks would be useless here because those + * are only used to synchronize: + * + * - modifying terminal entries (ones that point to a data page, not to another + * page table) + * - installing *new* non-terminal entries + * + * Instead, we need roughly the same kind of protection as free_pgtables() or + * mm_take_all_locks() (but only for a single VMA): + * The mmap lock together with this VMA's rmap locks covers all paths towards + * the page table entries we're messing with here, except for hardware page + * table walks and lockless_pages_from_mm(). + */ static void collapse_and_free_pmd(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long addr, pmd_t *pmdp) { - spinlock_t *ptl; pmd_t pmd; + struct mmu_notifier_range range; mmap_assert_write_locked(mm); - ptl = pmd_lock(vma->vm_mm, pmdp); + if (vma->vm_file) + lockdep_assert_held_write(&vma->vm_file->f_mapping->i_mmap_rwsem); + /* + * All anon_vmas attached to the VMA have the same root and are + * therefore locked by the same lock. + */ + if (vma->anon_vma) + lockdep_assert_held_write(&vma->anon_vma->root->rwsem); + + mmu_notifier_range_init(&range, MMU_NOTIFY_CLEAR, 0, NULL, mm, addr, + addr + HPAGE_PMD_SIZE); + mmu_notifier_invalidate_range_start(&range); pmd = pmdp_collapse_flush(vma, addr, pmdp); - spin_unlock(ptl); + tlb_remove_table_sync_one(); + mmu_notifier_invalidate_range_end(&range); mm_dec_nr_ptes(mm); page_table_check_pte_clear_range(mm, addr, pmd); pte_free(mm, pmd_pgtable(pmd)); @@ -1444,6 +1460,14 @@ int collapse_pte_mapped_thp(struct mm_struct *mm, unsigned long addr, if (!hugepage_vma_check(vma, vma->vm_flags, false, false, false)) return SCAN_VMA_CHECK; + /* + * Symmetry with retract_page_tables(): Exclude MAP_PRIVATE mappings + * that got written to. Without this, we'd have to also lock the + * anon_vma if one exists. + */ + if (vma->anon_vma) + return SCAN_VMA_CHECK; + /* Keep pmd pgtable for uffd-wp; see comment in retract_page_tables() */ if (userfaultfd_wp(vma)) return SCAN_PTE_UFFD_WP; @@ -1477,6 +1501,20 @@ int collapse_pte_mapped_thp(struct mm_struct *mm, unsigned long addr, goto drop_hpage; } + /* + * We need to lock the mapping so that from here on, only GUP-fast and + * hardware page walks can access the parts of the page tables that + * we're operating on. + * See collapse_and_free_pmd(). + */ + i_mmap_lock_write(vma->vm_file->f_mapping); + + /* + * This spinlock should be unnecessary: Nobody else should be accessing + * the page tables under spinlock protection here, only + * lockless_pages_from_mm() and the hardware page walker can access page + * tables while all the high-level locks are held in write mode. + */ start_pte = pte_offset_map_lock(mm, pmd, haddr, &ptl); result = SCAN_FAIL; @@ -1531,6 +1569,8 @@ int collapse_pte_mapped_thp(struct mm_struct *mm, unsigned long addr, /* step 4: remove pte entries */ collapse_and_free_pmd(mm, vma, haddr, pmd); + i_mmap_unlock_write(vma->vm_file->f_mapping); + maybe_install_pmd: /* step 5: install pmd entry */ result = install_pmd @@ -1544,6 +1584,7 @@ drop_hpage: abort: pte_unmap_unlock(start_pte, ptl); + i_mmap_unlock_write(vma->vm_file->f_mapping); goto drop_hpage; } @@ -1600,7 +1641,8 @@ static int retract_page_tables(struct address_space *mapping, pgoff_t pgoff, * An alternative would be drop the check, but check that page * table is clear before calling pmdp_collapse_flush() under * ptl. It has higher chance to recover THP for the VMA, but - * has higher cost too. + * has higher cost too. It would also probably require locking + * the anon_vma. */ if (vma->anon_vma) { result = SCAN_PAGE_ANON; @@ -1702,12 +1744,12 @@ static int collapse_file(struct mm_struct *mm, unsigned long addr, { struct address_space *mapping = file->f_mapping; struct page *hpage; - pgoff_t index, end = start + HPAGE_PMD_NR; + pgoff_t index = 0, end = start + HPAGE_PMD_NR; LIST_HEAD(pagelist); XA_STATE_ORDER(xas, &mapping->i_pages, start, HPAGE_PMD_ORDER); int nr_none = 0, result = SCAN_SUCCEED; bool is_shmem = shmem_file(file); - int nr; + int nr = 0; VM_BUG_ON(!IS_ENABLED(CONFIG_READ_ONLY_THP_FOR_FS) && !is_shmem); VM_BUG_ON(start & (HPAGE_PMD_NR - 1)); @@ -1747,6 +1789,7 @@ static int collapse_file(struct mm_struct *mm, unsigned long addr, xas_set(&xas, start); for (index = start; index < end; index++) { struct page *page = xas_next(&xas); + struct folio *folio; VM_BUG_ON(index != xas.xa_index); if (is_shmem) { @@ -1773,8 +1816,6 @@ static int collapse_file(struct mm_struct *mm, unsigned long addr, } if (xa_is_value(page) || !PageUptodate(page)) { - struct folio *folio; - xas_unlock_irq(&xas); /* swap in or instantiate fallocated page */ if (shmem_get_folio(mapping->host, index, @@ -1862,13 +1903,15 @@ static int collapse_file(struct mm_struct *mm, unsigned long addr, goto out_unlock; } - if (page_mapping(page) != mapping) { + folio = page_folio(page); + + if (folio_mapping(folio) != mapping) { result = SCAN_TRUNCATED; goto out_unlock; } - if (!is_shmem && (PageDirty(page) || - PageWriteback(page))) { + if (!is_shmem && (folio_test_dirty(folio) || + folio_test_writeback(folio))) { /* * khugepaged only works on read-only fd, so this * page is dirty because it hasn't been flushed @@ -1878,20 +1921,20 @@ static int collapse_file(struct mm_struct *mm, unsigned long addr, goto out_unlock; } - if (isolate_lru_page(page)) { + if (folio_isolate_lru(folio)) { result = SCAN_DEL_PAGE_LRU; goto out_unlock; } - if (page_has_private(page) && - !try_to_release_page(page, GFP_KERNEL)) { + if (folio_has_private(folio) && + !filemap_release_folio(folio, GFP_KERNEL)) { result = SCAN_PAGE_HAS_PRIVATE; - putback_lru_page(page); + folio_putback_lru(folio); goto out_unlock; } - if (page_mapped(page)) - try_to_unmap(page_folio(page), + if (folio_mapped(folio)) + try_to_unmap(folio, TTU_IGNORE_MLOCK | TTU_BATCH_FLUSH); xas_lock_irq(&xas); @@ -1970,6 +2013,7 @@ xa_unlocked: if (result == SCAN_SUCCEED) { struct page *page, *tmp; + struct folio *folio; /* * Replacing old pages with new one has succeeded, now we @@ -1997,11 +2041,13 @@ xa_unlocked: index++; } - SetPageUptodate(hpage); - page_ref_add(hpage, HPAGE_PMD_NR - 1); + folio = page_folio(hpage); + folio_mark_uptodate(folio); + folio_ref_add(folio, HPAGE_PMD_NR - 1); + if (is_shmem) - set_page_dirty(hpage); - lru_cache_add(hpage); + folio_mark_dirty(folio); + folio_add_lru(folio); /* * Remove pte page tables, so we can re-fault the page as huge. @@ -2059,7 +2105,8 @@ out: mem_cgroup_uncharge(page_folio(hpage)); put_page(hpage); } - /* TODO: tracepoints */ + + trace_mm_khugepaged_collapse_file(mm, hpage, index, is_shmem, addr, file, nr, result); return result; } @@ -2077,6 +2124,7 @@ static int hpage_collapse_scan_file(struct mm_struct *mm, unsigned long addr, present = 0; swap = 0; memset(cc->node_load, 0, sizeof(cc->node_load)); + nodes_clear(cc->alloc_nmask); rcu_read_lock(); xas_for_each(&xas, page, start + HPAGE_PMD_NR - 1) { if (xas_retry(&xas, page)) @@ -2157,8 +2205,7 @@ static int hpage_collapse_scan_file(struct mm_struct *mm, unsigned long addr, } } - trace_mm_khugepaged_scan_file(mm, page, file->f_path.dentry->d_iname, - present, swap, result); + trace_mm_khugepaged_scan_file(mm, page, file, present, swap, result); return result; } #else @@ -2528,6 +2575,11 @@ void khugepaged_min_free_kbytes_update(void) mutex_unlock(&khugepaged_mutex); } +bool current_is_khugepaged(void) +{ + return kthread_func(current) == khugepaged; +} + static int madvise_collapse_errno(enum scan_result r) { /* @@ -2576,7 +2628,6 @@ int madvise_collapse(struct vm_area_struct *vma, struct vm_area_struct **prev, if (!cc) return -ENOMEM; cc->is_khugepaged = false; - cc->last_target_node = NUMA_NO_NODE; mmgrab(mm); lru_add_drain_all(); @@ -2602,6 +2653,7 @@ int madvise_collapse(struct vm_area_struct *vma, struct vm_area_struct **prev, } mmap_assert_locked(mm); memset(cc->node_load, 0, sizeof(cc->node_load)); + nodes_clear(cc->alloc_nmask); if (IS_ENABLED(CONFIG_SHMEM) && vma->vm_file) { struct file *file = get_file(vma->vm_file); pgoff_t pgoff = linear_page_index(vma, addr); |