mm: page_owner: use new iteration API

The page_ext_next() function assumes that page extension objects for a
page order allocation always reside in the same memory section, which may
not be true and could lead to crashes.  Use the new page_ext iteration API
instead.

Link: https://lkml.kernel.org/r/93c80b040960fa2ebab4a9729073f77a30649862.1741301089.git.luizcap@redhat.com
Fixes: cf54f310d0d3 ("mm/hugetlb: use __GFP_COMP for gigantic folios")
Signed-off-by: Luiz Capitulino <luizcap@redhat.com>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Pasha Tatashin <pasha.tatashin@soleen.com>
Cc: Yu Zhao <yuzhao@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

diff --git a/mm/page_owner.c b/mm/page_owner.c
index a409e2561a8fd91c6c5928cf632ced56a1b3dfa7..849d4a471b6c1b33f7ba83b4e3f8e32a009855cc 100644 (file)
--- a/mm/page_owner.c
+++ b/mm/page_owner.c
@@ -229,17 +229,19 @@ static void dec_stack_record_count(depot_stack_handle_t handle,
                        handle);
 }
 
-static inline void __update_page_owner_handle(struct page_ext *page_ext,
+static inline void __update_page_owner_handle(struct page *page,
                                              depot_stack_handle_t handle,
                                              unsigned short order,
                                              gfp_t gfp_mask,
                                              short last_migrate_reason, u64 ts_nsec,
                                              pid_t pid, pid_t tgid, char *comm)
 {
-       int i;
+       struct page_ext_iter iter;
+       struct page_ext *page_ext;
        struct page_owner *page_owner;
 
-       for (i = 0; i < (1 << order); i++) {
+       rcu_read_lock();
+       for_each_page_ext(page, 1 << order, page_ext, iter) {
                page_owner = get_page_owner(page_ext);
                page_owner->handle = handle;
                page_owner->order = order;
@@ -252,20 +254,22 @@ static inline void __update_page_owner_handle(struct page_ext *page_ext,
                        sizeof(page_owner->comm));
                __set_bit(PAGE_EXT_OWNER, &page_ext->flags);
                __set_bit(PAGE_EXT_OWNER_ALLOCATED, &page_ext->flags);
-               page_ext = page_ext_next(page_ext);
        }
+       rcu_read_unlock();
 }
 
-static inline void __update_page_owner_free_handle(struct page_ext *page_ext,
+static inline void __update_page_owner_free_handle(struct page *page,
                                                   depot_stack_handle_t handle,
                                                   unsigned short order,
                                                   pid_t pid, pid_t tgid,
                                                   u64 free_ts_nsec)
 {
-       int i;
+       struct page_ext_iter iter;
+       struct page_ext *page_ext;
        struct page_owner *page_owner;
 
-       for (i = 0; i < (1 << order); i++) {
+       rcu_read_lock();
+       for_each_page_ext(page, 1 << order, page_ext, iter) {
                page_owner = get_page_owner(page_ext);
                /* Only __reset_page_owner() wants to clear the bit */
                if (handle) {
@@ -275,8 +279,8 @@ static inline void __update_page_owner_free_handle(struct page_ext *page_ext,
                page_owner->free_ts_nsec = free_ts_nsec;
                page_owner->free_pid = current->pid;
                page_owner->free_tgid = current->tgid;
-               page_ext = page_ext_next(page_ext);
        }
+       rcu_read_unlock();
 }
 
 void __reset_page_owner(struct page *page, unsigned short order)
@@ -293,11 +297,11 @@ void __reset_page_owner(struct page *page, unsigned short order)
 
        page_owner = get_page_owner(page_ext);
        alloc_handle = page_owner->handle;
+       page_ext_put(page_ext);
 
        handle = save_stack(GFP_NOWAIT | __GFP_NOWARN);
-       __update_page_owner_free_handle(page_ext, handle, order, current->pid,
+       __update_page_owner_free_handle(page, handle, order, current->pid,
                                        current->tgid, free_ts_nsec);
-       page_ext_put(page_ext);
 
        if (alloc_handle != early_handle)
                /*
@@ -313,19 +317,13 @@ void __reset_page_owner(struct page *page, unsigned short order)
 noinline void __set_page_owner(struct page *page, unsigned short order,
                                        gfp_t gfp_mask)
 {
-       struct page_ext *page_ext;
        u64 ts_nsec = local_clock();
        depot_stack_handle_t handle;
 
        handle = save_stack(gfp_mask);
-
-       page_ext = page_ext_get(page);
-       if (unlikely(!page_ext))
-               return;
-       __update_page_owner_handle(page_ext, handle, order, gfp_mask, -1,
+       __update_page_owner_handle(page, handle, order, gfp_mask, -1,
                                   ts_nsec, current->pid, current->tgid,
                                   current->comm);
-       page_ext_put(page_ext);
        inc_stack_record_count(handle, gfp_mask, 1 << order);
 }
 
@@ -344,44 +342,42 @@ void __set_page_owner_migrate_reason(struct page *page, int reason)
 
 void __split_page_owner(struct page *page, int old_order, int new_order)
 {
-       int i;
-       struct page_ext *page_ext = page_ext_get(page);
+       struct page_ext_iter iter;
+       struct page_ext *page_ext;
        struct page_owner *page_owner;
 
-       if (unlikely(!page_ext))
-               return;
-
-       for (i = 0; i < (1 << old_order); i++) {
+       rcu_read_lock();
+       for_each_page_ext(page, 1 << old_order, page_ext, iter) {
                page_owner = get_page_owner(page_ext);
                page_owner->order = new_order;
-               page_ext = page_ext_next(page_ext);
        }
-       page_ext_put(page_ext);
+       rcu_read_unlock();
 }
 
 void __folio_copy_owner(struct folio *newfolio, struct folio *old)
 {
-       int i;
-       struct page_ext *old_ext;
-       struct page_ext *new_ext;
+       struct page_ext *page_ext;
+       struct page_ext_iter iter;
        struct page_owner *old_page_owner;
        struct page_owner *new_page_owner;
        depot_stack_handle_t migrate_handle;
 
-       old_ext = page_ext_get(&old->page);
-       if (unlikely(!old_ext))
+       page_ext = page_ext_get(&old->page);
+       if (unlikely(!page_ext))
                return;
 
-       new_ext = page_ext_get(&newfolio->page);
-       if (unlikely(!new_ext)) {
-               page_ext_put(old_ext);
+       old_page_owner = get_page_owner(page_ext);
+       page_ext_put(page_ext);
+
+       page_ext = page_ext_get(&newfolio->page);
+       if (unlikely(!page_ext))
                return;
-       }
 
-       old_page_owner = get_page_owner(old_ext);
-       new_page_owner = get_page_owner(new_ext);
+       new_page_owner = get_page_owner(page_ext);
+       page_ext_put(page_ext);
+
        migrate_handle = new_page_owner->handle;
-       __update_page_owner_handle(new_ext, old_page_owner->handle,
+       __update_page_owner_handle(&newfolio->page, old_page_owner->handle,
                                   old_page_owner->order, old_page_owner->gfp_mask,
                                   old_page_owner->last_migrate_reason,
                                   old_page_owner->ts_nsec, old_page_owner->pid,
@@ -391,7 +387,7 @@ void __folio_copy_owner(struct folio *newfolio, struct folio *old)
         * will be freed after migration. Keep them until then as they may be
         * useful.
         */
-       __update_page_owner_free_handle(new_ext, 0, old_page_owner->order,
+       __update_page_owner_free_handle(&newfolio->page, 0, old_page_owner->order,
                                        old_page_owner->free_pid,
                                        old_page_owner->free_tgid,
                                        old_page_owner->free_ts_nsec);
@@ -400,14 +396,12 @@ void __folio_copy_owner(struct folio *newfolio, struct folio *old)
         * for the new one and the old folio otherwise there will be an imbalance
         * when subtracting those pages from the stack.
         */
-       for (i = 0; i < (1 << new_page_owner->order); i++) {
+       rcu_read_lock();
+       for_each_page_ext(&old->page, 1 << new_page_owner->order, page_ext, iter) {
+               old_page_owner = get_page_owner(page_ext);
                old_page_owner->handle = migrate_handle;
-               old_ext = page_ext_next(old_ext);
-               old_page_owner = get_page_owner(old_ext);
        }
-
-       page_ext_put(new_ext);
-       page_ext_put(old_ext);
+       rcu_read_unlock();
 }
 
 void pagetypeinfo_showmixedcount_print(struct seq_file *m,
@@ -813,7 +807,7 @@ static void init_pages_in_zone(pg_data_t *pgdat, struct zone *zone)
                                goto ext_put_continue;
 
                        /* Found early allocated page */
-                       __update_page_owner_handle(page_ext, early_handle, 0, 0,
+                       __update_page_owner_handle(page, early_handle, 0, 0,
                                                   -1, local_clock(), current->pid,
                                                   current->tgid, current->comm);
                        count++;