[ 46/48] mm: Hold a file reference in madvise_remove

[Ben Hutchings]

3.2-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Andy Lutomirski <l...@amacapital.net>

commit 9ab4233dd08036fe34a89c7dc6f47a8bf2eb29eb upstream.

Otherwise the code races with munmap (causing a use-after-free
of the vma) or with close (causing a use-after-free of the struct
file).

The bug was introduced by commit 90ed52ebe481 ("[PATCH] holepunch: fix
mmap_sem i_mutex deadlock")

Cc: Hugh Dickins <h...@veritas.com>
Cc: Miklos Szeredi <mszer...@suse.cz>
Cc: Badari Pulavarty <pbad...@us.ibm.com>
Cc: Nick Piggin <npig...@suse.de>
Signed-off-by: Andy Lutomirski <l...@amacapital.net>
Signed-off-by: Linus Torvalds <torva...@linux-foundation.org>
[bwh: Backported to 3.2:
 - Adjust context
 - madvise_remove() calls vmtruncate_range(), not do_fallocate()]
Signed-off-by: Ben Hutchings <b...@decadent.org.uk>
---
 mm/madvise.c |   18 ++++++++++++++----
 1 file changed, 14 insertions(+), 4 deletions(-)

--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -13,6 +13,7 @@
 #include <linux/hugetlb.h>
 #include <linux/sched.h>
 #include <linux/ksm.h>
+#include <linux/file.h>
 
 /*
  * Any behaviour which results in changes to the vma->vm_flags needs to
@@ -197,14 +198,16 @@ static long madvise_remove(struct vm_are
        struct address_space *mapping;
        loff_t offset, endoff;
        int error;
+       struct file *f;
 
        *prev = NULL;   /* tell sys_madvise we drop mmap_sem */
 
        if (vma->vm_flags & (VM_LOCKED|VM_NONLINEAR|VM_HUGETLB))
                return -EINVAL;
 
-       if (!vma->vm_file || !vma->vm_file->f_mapping
-               || !vma->vm_file->f_mapping->host) {
+       f = vma->vm_file;
+
+       if (!f || !f->f_mapping || !f->f_mapping->host) {
                        return -EINVAL;
        }
 
@@ -218,9 +221,16 @@ static long madvise_remove(struct vm_are
        endoff = (loff_t)(end - vma->vm_start - 1)
                        + ((loff_t)vma->vm_pgoff << PAGE_SHIFT);
 
-       /* vmtruncate_range needs to take i_mutex */
+       /*
+        * vmtruncate_range may need to take i_mutex.  We need to
+        * explicitly grab a reference because the vma (and hence the
+        * vma's reference to the file) can go away as soon as we drop
+        * mmap_sem.
+        */
+       get_file(f);
        up_read(&current->mm->mmap_sem);
        error = vmtruncate_range(mapping->host, offset, endoff);
+       fput(f);
        down_read(&current->mm->mmap_sem);
        return error;
 }


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/