mm: remove rest usage of VM_NONLINEAR and pte_file()

Hello Kirill A. Shutemov,

The patch 05864bbd92f9: "mm: remove rest usage of VM_NONLINEAR and pte_file()" from Jan 17, 2015, leads to the following static checker warning:

	mm/memcontrol.c:4794 mc_handle_file_pte()
	warn: passing uninitialized 'pgoff'

mm/memcontrol.c
4774 static struct page *mc_handle_file_pte(struct vm_area_struct *vma,
4775 unsigned long addr, pte_t ptent, swp_entry_t *entry)
4776 {

  4777          struct page *page = NULL;
  4778          struct address_space *mapping;
  4779          pgoff_t pgoff;

4780
4781 if (!vma->vm_file) /* anonymous vma */
4782 return NULL;
4783 if (!(mc.flags & MOVE_FILE))
4784 return NULL;
4785
4786 mapping = vma->vm_file->f_mapping;
4787 if (pte_none(ptent))
4788 pgoff = linear_page_index(vma, addr);

We used to have an "else pgoff = pte_to_pgoff(ptent);" but now it's just uninitialized data.

4789
4790 /* page is moved even if it's not RSS of this task(page-faulted). */
4791 #ifdef CONFIG_SWAP
4792 /* shmem/tmpfs may report page out on swap: account for that too. */
4793 if (shmem_mapping(mapping)) {

  4794                  page = find_get_entry(mapping, pgoff);
                                                       ^^^^^

Used here.

4795 if (radix_tree_exceptional_entry(page)) {
4796 swp_entry_t swp = radix_to_swp_entry(page);
4797 if (do_swap_account)
4798 *entry = swp;
4799 page = find_get_page(swap_address_space(swp), swp.val);
4800 }
4801 } else
4802 page = find_get_page(mapping, pgoff);
4803 #else

  4804          page = find_get_page(mapping, pgoff);
                                              ^^^^^

And here.

4805 #endif
4806 return page;
4807 }

regards,
dan carpenter

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Please test the patch below.