mm/memory.c: properly pte_offset_map_lock/unlock in vm_insert_pages()

Calls to pte_offset_map() in vm_insert_pages() are erroneously not
matched with a call to pte_unmap().  This would cause problems on
architectures where that is not a no-op.

This patch does away with the non-traditional locking in the existing
code, and instead uses pte_offset_map_lock/unlock() as usual,
incrementing PTE as necessary.  The PTE pointer is kept within bounds
since we clamp it with PTRS_PER_PTE.

Link: http://lkml.kernel.org/r/20200618220446.20284-1-arjunroy.kdev@gmail.com
Fixes: 8cd3984d81d5 ("mm/memory.c: add vm_insert_pages()")
Signed-off-by: Arjun Roy <arjunroy@google.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

diff --git a/mm/memory.c b/mm/memory.c
index 1c632faa261169d3e9cd803c1802014b69475648..0e5b25c9b151ff8ba530c0cd745c16b23d81ece8 100644 (file)
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1498,7 +1498,7 @@ out:
 }
 
 #ifdef pte_index
-static int insert_page_in_batch_locked(struct mm_struct *mm, pmd_t *pmd,
+static int insert_page_in_batch_locked(struct mm_struct *mm, pte_t *pte,
                        unsigned long addr, struct page *page, pgprot_t prot)
 {
        int err;
@@ -1506,8 +1506,9 @@ static int insert_page_in_batch_locked(struct mm_struct *mm, pmd_t *pmd,
        if (!page_count(page))
                return -EINVAL;
        err = validate_page_before_insert(page);
-       return err ? err : insert_page_into_pte_locked(
-               mm, pte_offset_map(pmd, addr), addr, page, prot);
+       if (err)
+               return err;
+       return insert_page_into_pte_locked(mm, pte, addr, page, prot);
 }
 
 /* insert_pages() amortizes the cost of spinlock operations
@@ -1517,7 +1518,8 @@ static int insert_pages(struct vm_area_struct *vma, unsigned long addr,
                        struct page **pages, unsigned long *num, pgprot_t prot)
 {
        pmd_t *pmd = NULL;
-       spinlock_t *pte_lock = NULL;
+       pte_t *start_pte, *pte;
+       spinlock_t *pte_lock;
        struct mm_struct *const mm = vma->vm_mm;
        unsigned long curr_page_idx = 0;
        unsigned long remaining_pages_total = *num;
@@ -1536,18 +1538,17 @@ more:
        ret = -ENOMEM;
        if (pte_alloc(mm, pmd))
                goto out;
-       pte_lock = pte_lockptr(mm, pmd);
 
        while (pages_to_write_in_pmd) {
                int pte_idx = 0;
                const int batch_size = min_t(int, pages_to_write_in_pmd, 8);
 
-               spin_lock(pte_lock);
-               for (; pte_idx < batch_size; ++pte_idx) {
-                       int err = insert_page_in_batch_locked(mm, pmd,
+               start_pte = pte_offset_map_lock(mm, pmd, addr, &pte_lock);
+               for (pte = start_pte; pte_idx < batch_size; ++pte, ++pte_idx) {
+                       int err = insert_page_in_batch_locked(mm, pte,
                                addr, pages[curr_page_idx], prot);
                        if (unlikely(err)) {
-                               spin_unlock(pte_lock);
+                               pte_unmap_unlock(start_pte, pte_lock);
                                ret = err;
                                remaining_pages_total -= pte_idx;
                                goto out;
@@ -1555,7 +1556,7 @@ more:
                        addr += PAGE_SIZE;
                        ++curr_page_idx;
                }
-               spin_unlock(pte_lock);
+               pte_unmap_unlock(start_pte, pte_lock);
                pages_to_write_in_pmd -= batch_size;
                remaining_pages_total -= batch_size;
        }