On Sat 31-08-19 09:18:16, Kefeng Wang wrote:
> Using %px to show the actual address in print_bad_pte()
> to help us to debug issue.
Yes, those values are of no use when hashed. At least __dump_page prints
mapping directly so there is no reason to differ here. anon_vma doesn't
really disclose much more AFAICS. Printing the addr might disclose
randomization offset for a vma but process usually doesn't live for long
after a bad pte is detected so it should be reasonably safe unless I
miss something
> Signed-off-by: Kefeng Wang <wangkefeng.wang@xxxxxxxxxx>
Acked-by: Michal Hocko <mhocko@xxxxxxxx>
> ---
> mm/memory.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/mm/memory.c b/mm/memory.c
> index e2bb51b6242e..3f0874c9ca38 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -518,7 +518,7 @@ static void print_bad_pte(struct vm_area_struct *vma, unsigned long addr,
> (long long)pte_val(pte), (long long)pmd_val(*pmd));
> if (page)
> dump_page(page, "bad pte");
> - pr_alert("addr:%p vm_flags:%08lx anon_vma:%p mapping:%p index:%lx\n",
> + pr_alert("addr:%px vm_flags:%08lx anon_vma:%px mapping:%px index:%lx\n",
> (void *)addr, vma->vm_flags, vma->anon_vma, mapping, index);
> pr_alert("file:%pD fault:%ps mmap:%ps readpage:%ps\n",
> vma->vm_file,
> --
> 2.20.1
>
--
Michal Hocko
SUSE Labs
