vm: fix mlock() on stack guard page

Commit 53a7706d5ed8 ("mlock: do not hold mmap_sem for extended periods
of time") changed mlock() to care about the exact number of pages that
__get_user_pages() had brought it.  Before, it would only care about
errors.

And that doesn't work, because we also handled one page specially in
__mlock_vma_pages_range(), namely the stack guard page.  So when that
case was handled, the number of pages that the function returned was off
by one.  In particular, it could be zero, and then the caller would end
up not making any progress at all.

Rather than try to fix up that off-by-one error for the mlock case
specially, this just moves the logic to handle the stack guard page
into__get_user_pages() itself, thus making all the counts come out
right automatically.

Reported-by: Robert Święcki <robert@swiecki.net>
Cc: Hugh Dickins <hughd@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

diff --git a/mm/memory.c b/mm/memory.c
index 9da8cab1b1b0abceae9569a794447f3ddb0c1134..b623a249918c95a819e4b3f1b7c0814d24f78a03 100644 (file)
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1410,6 +1410,13 @@ no_page_table:
        return page;
 }
 
+static inline int stack_guard_page(struct vm_area_struct *vma, unsigned long addr)
+{
+       return (vma->vm_flags & VM_GROWSDOWN) &&
+               (vma->vm_start == addr) &&
+               !vma_stack_continue(vma->vm_prev, addr);
+}
+
 /**
  * __get_user_pages() - pin user pages in memory
  * @tsk:       task_struct of target task
@@ -1488,7 +1495,6 @@ int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
                vma = find_extend_vma(mm, start);
                if (!vma && in_gate_area(mm, start)) {
                        unsigned long pg = start & PAGE_MASK;
-                       struct vm_area_struct *gate_vma = get_gate_vma(mm);
                        pgd_t *pgd;
                        pud_t *pud;
                        pmd_t *pmd;
@@ -1513,10 +1519,11 @@ int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
                                pte_unmap(pte);
                                return i ? : -EFAULT;
                        }
+                       vma = get_gate_vma(mm);
                        if (pages) {
                                struct page *page;
 
-                               page = vm_normal_page(gate_vma, start, *pte);
+                               page = vm_normal_page(vma, start, *pte);
                                if (!page) {
                                        if (!(gup_flags & FOLL_DUMP) &&
                                             is_zero_pfn(pte_pfn(*pte)))
@@ -1530,12 +1537,7 @@ int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
                                get_page(page);
                        }
                        pte_unmap(pte);
-                       if (vmas)
-                               vmas[i] = gate_vma;
-                       i++;
-                       start += PAGE_SIZE;
-                       nr_pages--;
-                       continue;
+                       goto next_page;
                }
 
                if (!vma ||
@@ -1549,6 +1551,13 @@ int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
                        continue;
                }
 
+               /*
+                * If we don't actually want the page itself,
+                * and it's the stack guard page, just skip it.
+                */
+               if (!pages && stack_guard_page(vma, start))
+                       goto next_page;
+
                do {
                        struct page *page;
                        unsigned int foll_flags = gup_flags;
@@ -1631,6 +1640,7 @@ int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
                                flush_anon_page(vma, page, start);
                                flush_dcache_page(page);
                        }
+next_page:
                        if (vmas)
                                vmas[i] = vma;
                        i++;

diff --git a/mm/mlock.c b/mm/mlock.c
index 2689a08c79affabda1e46b65271c60e3ba653ab7..6b55e3efe0df4ebd9aafd71e9317bc664956006f 100644 (file)
--- a/mm/mlock.c
+++ b/mm/mlock.c
@@ -135,13 +135,6 @@ void munlock_vma_page(struct page *page)
        }
 }
 
-static inline int stack_guard_page(struct vm_area_struct *vma, unsigned long addr)
-{
-       return (vma->vm_flags & VM_GROWSDOWN) &&
-               (vma->vm_start == addr) &&
-               !vma_stack_continue(vma->vm_prev, addr);
-}
-
 /**
  * __mlock_vma_pages_range() -  mlock a range of pages in the vma.
  * @vma:   target vma
@@ -188,12 +181,6 @@ static long __mlock_vma_pages_range(struct vm_area_struct *vma,
        if (vma->vm_flags & VM_LOCKED)
                gup_flags |= FOLL_MLOCK;
 
-       /* We don't try to access the guard page of a stack vma */
-       if (stack_guard_page(vma, start)) {
-               addr += PAGE_SIZE;
-               nr_pages--;
-       }
-
        return __get_user_pages(current, mm, addr, nr_pages, gup_flags,
                                NULL, NULL, nonblocking);
 }