mm: fix BUG in __split_huge_page_pmd
[linux-3.10.git] / mm / huge_memory.c
index 6049376..0164b09 100644 (file)
@@ -163,35 +163,34 @@ static int start_khugepaged(void)
 }
 
 static atomic_t huge_zero_refcount;
-static unsigned long huge_zero_pfn __read_mostly;
+static struct page *huge_zero_page __read_mostly;
 
-static inline bool is_huge_zero_pfn(unsigned long pfn)
+static inline bool is_huge_zero_page(struct page *page)
 {
-       unsigned long zero_pfn = ACCESS_ONCE(huge_zero_pfn);
-       return zero_pfn && pfn == zero_pfn;
+       return ACCESS_ONCE(huge_zero_page) == page;
 }
 
 static inline bool is_huge_zero_pmd(pmd_t pmd)
 {
-       return is_huge_zero_pfn(pmd_pfn(pmd));
+       return is_huge_zero_page(pmd_page(pmd));
 }
 
-static unsigned long get_huge_zero_page(void)
+static struct page *get_huge_zero_page(void)
 {
        struct page *zero_page;
 retry:
        if (likely(atomic_inc_not_zero(&huge_zero_refcount)))
-               return ACCESS_ONCE(huge_zero_pfn);
+               return ACCESS_ONCE(huge_zero_page);
 
        zero_page = alloc_pages((GFP_TRANSHUGE | __GFP_ZERO) & ~__GFP_MOVABLE,
                        HPAGE_PMD_ORDER);
        if (!zero_page) {
                count_vm_event(THP_ZERO_PAGE_ALLOC_FAILED);
-               return 0;
+               return NULL;
        }
        count_vm_event(THP_ZERO_PAGE_ALLOC);
        preempt_disable();
-       if (cmpxchg(&huge_zero_pfn, 0, page_to_pfn(zero_page))) {
+       if (cmpxchg(&huge_zero_page, NULL, zero_page)) {
                preempt_enable();
                __free_page(zero_page);
                goto retry;
@@ -200,7 +199,7 @@ retry:
        /* We take additional reference here. It will be put back by shrinker */
        atomic_set(&huge_zero_refcount, 2);
        preempt_enable();
-       return ACCESS_ONCE(huge_zero_pfn);
+       return ACCESS_ONCE(huge_zero_page);
 }
 
 static void put_huge_zero_page(void)
@@ -220,9 +219,9 @@ static int shrink_huge_zero_page(struct shrinker *shrink,
                return atomic_read(&huge_zero_refcount) == 1 ? HPAGE_PMD_NR : 0;
 
        if (atomic_cmpxchg(&huge_zero_refcount, 1, 0) == 1) {
-               unsigned long zero_pfn = xchg(&huge_zero_pfn, 0);
-               BUG_ON(zero_pfn == 0);
-               __free_page(__pfn_to_page(zero_pfn));
+               struct page *zero_page = xchg(&huge_zero_page, NULL);
+               BUG_ON(zero_page == NULL);
+               __free_page(zero_page);
        }
 
        return 0;
@@ -713,6 +712,11 @@ static int __do_huge_pmd_anonymous_page(struct mm_struct *mm,
                return VM_FAULT_OOM;
 
        clear_huge_page(page, haddr, HPAGE_PMD_NR);
+       /*
+        * The memory barrier inside __SetPageUptodate makes sure that
+        * clear_huge_page writes become visible before the set_pmd_at()
+        * write.
+        */
        __SetPageUptodate(page);
 
        spin_lock(&mm->page_table_lock);
@@ -724,12 +728,6 @@ static int __do_huge_pmd_anonymous_page(struct mm_struct *mm,
        } else {
                pmd_t entry;
                entry = mk_huge_pmd(page, vma);
-               /*
-                * The spinlocking to take the lru_lock inside
-                * page_add_new_anon_rmap() acts as a full memory
-                * barrier to be sure clear_huge_page writes become
-                * visible after the set_pmd_at() write.
-                */
                page_add_new_anon_rmap(page, vma, haddr);
                set_pmd_at(mm, haddr, pmd, entry);
                pgtable_trans_huge_deposit(mm, pgtable);
@@ -765,12 +763,12 @@ static inline struct page *alloc_hugepage(int defrag)
 
 static bool set_huge_zero_page(pgtable_t pgtable, struct mm_struct *mm,
                struct vm_area_struct *vma, unsigned long haddr, pmd_t *pmd,
-               unsigned long zero_pfn)
+               struct page *zero_page)
 {
        pmd_t entry;
        if (!pmd_none(*pmd))
                return false;
-       entry = pfn_pmd(zero_pfn, vma->vm_page_prot);
+       entry = mk_pmd(zero_page, vma->vm_page_prot);
        entry = pmd_wrprotect(entry);
        entry = pmd_mkhuge(entry);
        set_pmd_at(mm, haddr, pmd, entry);
@@ -795,20 +793,20 @@ int do_huge_pmd_anonymous_page(struct mm_struct *mm, struct vm_area_struct *vma,
                if (!(flags & FAULT_FLAG_WRITE) &&
                                transparent_hugepage_use_zero_page()) {
                        pgtable_t pgtable;
-                       unsigned long zero_pfn;
+                       struct page *zero_page;
                        bool set;
                        pgtable = pte_alloc_one(mm, haddr);
                        if (unlikely(!pgtable))
                                return VM_FAULT_OOM;
-                       zero_pfn = get_huge_zero_page();
-                       if (unlikely(!zero_pfn)) {
+                       zero_page = get_huge_zero_page();
+                       if (unlikely(!zero_page)) {
                                pte_free(mm, pgtable);
                                count_vm_event(THP_FAULT_FALLBACK);
                                goto out;
                        }
                        spin_lock(&mm->page_table_lock);
                        set = set_huge_zero_page(pgtable, mm, vma, haddr, pmd,
-                                       zero_pfn);
+                                       zero_page);
                        spin_unlock(&mm->page_table_lock);
                        if (!set) {
                                pte_free(mm, pgtable);
@@ -887,16 +885,16 @@ int copy_huge_pmd(struct mm_struct *dst_mm, struct mm_struct *src_mm,
         * a page table.
         */
        if (is_huge_zero_pmd(pmd)) {
-               unsigned long zero_pfn;
+               struct page *zero_page;
                bool set;
                /*
                 * get_huge_zero_page() will never allocate a new page here,
                 * since we already have a zero page to copy. It just takes a
                 * reference.
                 */
-               zero_pfn = get_huge_zero_page();
+               zero_page = get_huge_zero_page();
                set = set_huge_zero_page(pgtable, dst_mm, vma, addr, dst_pmd,
-                               zero_pfn);
+                               zero_page);
                BUG_ON(!set); /* unexpected !pmd_none(dst_pmd) */
                ret = 0;
                goto out_unlock;
@@ -1560,7 +1558,8 @@ static int __split_huge_page_splitting(struct page *page,
        return ret;
 }
 
-static void __split_huge_page_refcount(struct page *page)
+static void __split_huge_page_refcount(struct page *page,
+                                      struct list_head *list)
 {
        int i;
        struct zone *zone = page_zone(page);
@@ -1646,7 +1645,7 @@ static void __split_huge_page_refcount(struct page *page)
                BUG_ON(!PageDirty(page_tail));
                BUG_ON(!PageSwapBacked(page_tail));
 
-               lru_add_page_tail(page, page_tail, lruvec);
+               lru_add_page_tail(page, page_tail, lruvec, list);
        }
        atomic_sub(tail_count, &page->_count);
        BUG_ON(atomic_read(&page->_count) <= 0);
@@ -1753,7 +1752,8 @@ static int __split_huge_page_map(struct page *page,
 
 /* must be called with anon_vma->root->rwsem held */
 static void __split_huge_page(struct page *page,
-                             struct anon_vma *anon_vma)
+                             struct anon_vma *anon_vma,
+                             struct list_head *list)
 {
        int mapcount, mapcount2;
        pgoff_t pgoff = page->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT);
@@ -1784,7 +1784,7 @@ static void __split_huge_page(struct page *page,
                       mapcount, page_mapcount(page));
        BUG_ON(mapcount != page_mapcount(page));
 
-       __split_huge_page_refcount(page);
+       __split_huge_page_refcount(page, list);
 
        mapcount2 = 0;
        anon_vma_interval_tree_foreach(avc, &anon_vma->rb_root, pgoff, pgoff) {
@@ -1799,12 +1799,19 @@ static void __split_huge_page(struct page *page,
        BUG_ON(mapcount != mapcount2);
 }
 
-int split_huge_page(struct page *page)
+/*
+ * Split a hugepage into normal pages. This doesn't change the position of head
+ * page. If @list is null, tail pages will be added to LRU list, otherwise, to
+ * @list. Both head page and tail pages will inherit mapping, flags, and so on
+ * from the hugepage.
+ * Return 0 if the hugepage is split successfully otherwise return 1.
+ */
+int split_huge_page_to_list(struct page *page, struct list_head *list)
 {
        struct anon_vma *anon_vma;
        int ret = 1;
 
-       BUG_ON(is_huge_zero_pfn(page_to_pfn(page)));
+       BUG_ON(is_huge_zero_page(page));
        BUG_ON(!PageAnon(page));
 
        /*
@@ -1824,7 +1831,7 @@ int split_huge_page(struct page *page)
                goto out_unlock;
 
        BUG_ON(!PageSwapBacked(page));
-       __split_huge_page(page, anon_vma);
+       __split_huge_page(page, anon_vma, list);
        count_vm_event(THP_SPLIT);
 
        BUG_ON(PageCompound(page));
@@ -1906,9 +1913,8 @@ static inline void free_mm_slot(struct mm_slot *mm_slot)
 static struct mm_slot *get_mm_slot(struct mm_struct *mm)
 {
        struct mm_slot *mm_slot;
-       struct hlist_node *node;
 
-       hash_for_each_possible(mm_slots_hash, mm_slot, node, hash, (unsigned long)mm)
+       hash_for_each_possible(mm_slots_hash, mm_slot, hash, (unsigned long)mm)
                if (mm == mm_slot->mm)
                        return mm_slot;
 
@@ -2280,6 +2286,8 @@ static void collapse_huge_page(struct mm_struct *mm,
                goto out;
 
        vma = find_vma(mm, address);
+       if (!vma)
+               goto out;
        hstart = (vma->vm_start + ~HPAGE_PMD_MASK) & HPAGE_PMD_MASK;
        hend = vma->vm_end & HPAGE_PMD_MASK;
        if (address < hstart || address + HPAGE_PMD_SIZE > hend)
@@ -2319,7 +2327,12 @@ static void collapse_huge_page(struct mm_struct *mm,
                pte_unmap(pte);
                spin_lock(&mm->page_table_lock);
                BUG_ON(!pmd_none(*pmd));
-               set_pmd_at(mm, address, pmd, _pmd);
+               /*
+                * We can only use set_pmd_at when establishing
+                * hugepmds and never for establishing regular pmds that
+                * points to regular pagetables. Use pmd_populate for that
+                */
+               pmd_populate(mm, pmd, pmd_pgtable(_pmd));
                spin_unlock(&mm->page_table_lock);
                anon_vma_unlock_write(vma->anon_vma);
                goto out;
@@ -2376,7 +2389,7 @@ static int khugepaged_scan_pmd(struct mm_struct *mm,
        struct page *page;
        unsigned long _address;
        spinlock_t *ptl;
-       int node = -1;
+       int node = NUMA_NO_NODE;
 
        VM_BUG_ON(address & ~HPAGE_PMD_MASK);
 
@@ -2406,7 +2419,7 @@ static int khugepaged_scan_pmd(struct mm_struct *mm,
                 * be more sophisticated and look at more pages,
                 * but isn't for now.
                 */
-               if (node == -1)
+               if (node == NUMA_NO_NODE)
                        node = page_to_nid(page);
                VM_BUG_ON(PageCompound(page));
                if (!PageLRU(page) || PageLocked(page) || !PageAnon(page))
@@ -2686,6 +2699,7 @@ void __split_huge_page_pmd(struct vm_area_struct *vma, unsigned long address,
 
        mmun_start = haddr;
        mmun_end   = haddr + HPAGE_PMD_SIZE;
+again:
        mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
        spin_lock(&mm->page_table_lock);
        if (unlikely(!pmd_trans_huge(*pmd))) {
@@ -2708,7 +2722,14 @@ void __split_huge_page_pmd(struct vm_area_struct *vma, unsigned long address,
        split_huge_page(page);
 
        put_page(page);
-       BUG_ON(pmd_trans_huge(*pmd));
+
+       /*
+        * We don't always have down_write of mmap_sem here: a racing
+        * do_huge_pmd_wp_page() might have copied-on-write to another
+        * huge page before our split_huge_page() got the anon_vma lock.
+        */
+       if (unlikely(pmd_trans_huge(*pmd)))
+               goto again;
 }
 
 void split_huge_page_pmd_mm(struct mm_struct *mm, unsigned long address,