lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date:	Thu, 8 Mar 2007 06:50:04 +0100
From:	Nick Piggin <npiggin@...e.de>
To:	Andrew Morton <akpm@...ux-foundation.org>
Cc:	miklos@...redi.hu,
	Linux Memory Management List <linux-mm@...ck.org>,
	linux-kernel@...r.kernel.org,
	Linus Torvalds <torvalds@...ux-foundation.org>
Subject: Re: [patch 8/6] mm: fix cpdfio vs fault race

On Wed, Mar 07, 2007 at 01:02:14PM -0800, Andrew Morton wrote:
> On Wed, 7 Mar 2007 12:31:21 +0100
> Nick Piggin <npiggin@...e.de> wrote:
> 
> > Index: linux-2.6/mm/memory.c
> > ===================================================================
> > --- linux-2.6.orig/mm/memory.c
> > +++ linux-2.6/mm/memory.c
> > @@ -1664,6 +1664,15 @@ gotten:
> >  unlock:
> >  	pte_unmap_unlock(page_table, ptl);
> >  	if (dirty_page) {
> > +		/*
> > +		 * Yes, Virginia, this is actually required to prevent a race
> > +		 * with clear_page_dirty_for_io() from clearing the page dirty
> > +		 * bit after it clear all dirty ptes, but before a racing
> > +		 * do_wp_page installs a dirty pte.
> > +		 *
> > +		 * do_no_page is protected similarly.
> > +		 */
> > +		wait_on_page_locked(dirty_page);
> >  		set_page_dirty_balance(dirty_page);
> >  		put_page(dirty_page);
> >  	}
> > @@ -2316,6 +2325,7 @@ retry:
> >  unlock:
> >  	pte_unmap_unlock(page_table, ptl);
> >  	if (dirty_page) {
> > +		wait_on_page_locked(dirty_page);
> >  		set_page_dirty_balance(dirty_page);
> >  		put_page(dirty_page);
> >  	}
> > Index: linux-2.6/mm/page-writeback.c
> 
> now that's scary - applying this on top of your
> lock-the-page-in-the-fault-handler patches gives:
> 
> 	if (dirty_page) {
> 		/*
> 		 * Yes, Virginia, this is actually required to prevent a race
> 		 * with clear_page_dirty_for_io() from clearing the page dirty
> 		 * bit after it clear all dirty ptes, but before a racing
> 		 * do_wp_page installs a dirty pte.
> 		 *
> 		 * do_no_page is protected similarly.
> 		 */
> 		wait_on_page_locked(dirty_page);
> 		wait_on_page_locked(dirty_page);
> 		set_page_dirty_balance(dirty_page);
> 		put_page(dirty_page);
> 	}
> 
> One wonders how on earth patch(1) managed to do that.  If it has inserted
> the comment twice as well then it might be explicable..

Ouch ;) Yeah that patch I sent was supposed to apply underneath
the previous ones, sorry I wasn't clear.

> Oh well, let's try this:

Yeah that looks like the correct one for applying on top. Thanks.

> 
> From: Nick Piggin <npiggin@...e.de>
> 
> Fix msync data loss and (less importantly) dirty page accounting
> inaccuracies due to the race remaining in clear_page_dirty_for_io().
> 
> The deleted comment explains what the race was, and the added comments
> explain how it is fixed.
> 
> Signed-off-by: Nick Piggin <npiggin@...e.de>
> Cc: Linus Torvalds <torvalds@...ux-foundation.org>
> Cc: Miklos Szeredi <miklos@...redi.hu>
> Signed-off-by: Andrew Morton <akpm@...l.org>
> ---
> 
>  mm/memory.c         |    9 +++++++++
>  mm/page-writeback.c |   17 ++++++++++++-----
>  2 files changed, 21 insertions(+), 5 deletions(-)
> 
> diff -puN mm/memory.c~mm-fix-cpdfio-vs-fault-race mm/memory.c
> --- a/mm/memory.c~mm-fix-cpdfio-vs-fault-race
> +++ a/mm/memory.c
> @@ -1669,6 +1669,15 @@ gotten:
>  unlock:
>  	pte_unmap_unlock(page_table, ptl);
>  	if (dirty_page) {
> +		/*
> +		 * Yes, Virginia, this is actually required to prevent a race
> +		 * with clear_page_dirty_for_io() from clearing the page dirty
> +		 * bit after it clear all dirty ptes, but before a racing
> +		 * do_wp_page installs a dirty pte.
> +		 *
> +		 * do_no_page is protected similarly.
> +		 */
> +		wait_on_page_locked(dirty_page);
>  		set_page_dirty_balance(dirty_page);
>  		put_page(dirty_page);
>  	}
> diff -puN mm/page-writeback.c~mm-fix-cpdfio-vs-fault-race mm/page-writeback.c
> --- a/mm/page-writeback.c~mm-fix-cpdfio-vs-fault-race
> +++ a/mm/page-writeback.c
> @@ -903,6 +903,8 @@ int clear_page_dirty_for_io(struct page 
>  {
>  	struct address_space *mapping = page_mapping(page);
>  
> +	BUG_ON(!PageLocked(page));
> +
>  	if (mapping && mapping_cap_account_dirty(mapping)) {
>  		/*
>  		 * Yes, Virginia, this is indeed insane.
> @@ -928,14 +930,19 @@ int clear_page_dirty_for_io(struct page 
>  		 * We basically use the page "master dirty bit"
>  		 * as a serialization point for all the different
>  		 * threads doing their things.
> -		 *
> -		 * FIXME! We still have a race here: if somebody
> -		 * adds the page back to the page tables in
> -		 * between the "page_mkclean()" and the "TestClearPageDirty()",
> -		 * we might have it mapped without the dirty bit set.
>  		 */
>  		if (page_mkclean(page))
>  			set_page_dirty(page);
> +		/*
> +		 * We carefully synchronise fault handlers against
> +		 * installing a dirty pte and marking the page dirty
> +		 * at this point. We do this by having them hold the
> +		 * page lock at some point after installing their
> +		 * pte, but before marking the page dirty.
> +		 * Pages are always locked coming in here, so we get
> +		 * the desired exclusion. See mm/memory.c:do_wp_page()
> +		 * for more comments.
> +		 */
>  		if (TestClearPageDirty(page)) {
>  			dec_zone_page_state(page, NR_FILE_DIRTY);
>  			return 1;
> _
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists