Re: [PATCH v2] mm: mprotect: check page dirty when change ptes

[Jerome Glisse <jglisse@redhat.com>]

On Wed, Sep 12, 2018 at 09:03:55AM -0400, Jerome Glisse wrote:
> On Wed, Sep 12, 2018 at 02:49:21PM +0800, Peter Xu wrote:
> > Add an extra check on page dirty bit in change_pte_range() since there
> > might be case where PTE dirty bit is unset but it's actually dirtied.
> > One example is when a huge PMD is splitted after written: the dirty bit
> > will be set on the compound page however we won't have the dirty bit set
> > on each of the small page PTEs.
> > 
> > I noticed this when debugging with a customized kernel that implemented
> > userfaultfd write-protect.  In that case, the dirty bit will be critical
> > since that's required for userspace to handle the write protect page
> > fault (otherwise it'll get a SIGBUS with a loop of page faults).
> > However it should still be good even for upstream Linux to cover more
> > scenarios where we shouldn't need to do extra page faults on the small
> > pages if the previous huge page is already written, so the dirty bit
> > optimization path underneath can cover more.
> > 
> 
> So as said by Kirill NAK you are not looking at the right place for
> your bug please first apply the below patch and read my analysis in
> my last reply.

Just to be clear you are trying to fix a userspace bug that is hidden
for non THP pages by a kernel space bug inside userfaultfd by making
the kernel space bug of userfaultfd buggy for THP too.


> 
> Below patch fix userfaultfd bug. I am not posting it as it is on a
> branch and i am not sure when Andrea plan to post. Andrea feel free
> to squash that fix.
> 
> 
> From 35cdb30afa86424c2b9f23c0982afa6731be961c Mon Sep 17 00:00:00 2001
> From: =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= <jglisse@redhat.com>
> Date: Wed, 12 Sep 2018 08:58:33 -0400
> Subject: [PATCH] userfaultfd: do not set dirty accountable when changing
>  protection
> MIME-Version: 1.0
> Content-Type: text/plain; charset=UTF-8
> Content-Transfer-Encoding: 8bit
> 
> mwriteprotect_range() has nothing to do with the dirty accountable
> optimization so do not set it as it opens a door for userspace to
> unwrite protect pages in a range that is write protected ie the vma
> !(vm_flags & VM_WRITE).
> 
> Signed-off-by: Jerome Glisse <jglisse@redhat.com>
> ---
>  mm/userfaultfd.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/mm/userfaultfd.c b/mm/userfaultfd.c
> index a0379c5ffa7c..59db1ce48fa0 100644
> --- a/mm/userfaultfd.c
> +++ b/mm/userfaultfd.c
> @@ -632,7 +632,7 @@ int mwriteprotect_range(struct mm_struct *dst_mm, unsigned long start,
>  		newprot = vm_get_page_prot(dst_vma->vm_flags);
>  
>  	change_protection(dst_vma, start, start + len, newprot,
> -				!enable_wp, 0);
> +				false, 0);
>  
>  	err = 0;
>  out_unlock:
> -- 
> 2.17.1
>