lkml.org 
[lkml]   [2017]   [Jun]   [8]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 2/3] mm/page_ref: Ensure page_ref_unfreeze is ordered against prior accesses
On Thu, Jun 08, 2017 at 11:38:21AM +0200, Vlastimil Babka wrote:
> On 06/06/2017 07:58 PM, Will Deacon wrote:
> > page_ref_freeze and page_ref_unfreeze are designed to be used as a pair,
> > wrapping a critical section where struct pages can be modified without
> > having to worry about consistency for a concurrent fast-GUP.
> >
> > Whilst page_ref_freeze has full barrier semantics due to its use of
> > atomic_cmpxchg, page_ref_unfreeze is implemented using atomic_set, which
> > doesn't provide any barrier semantics and allows the operation to be
> > reordered with respect to page modifications in the critical section.
> >
> > This patch ensures that page_ref_unfreeze is ordered after any critical
> > section updates, by invoking smp_mb__before_atomic() prior to the
> > atomic_set.
> >
> > Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
> > Acked-by: Steve Capper <steve.capper@arm.com>
> > Signed-off-by: Will Deacon <will.deacon@arm.com>
>
> Undecided if it's really needed. This is IMHO not the classical case
> from Documentation/core-api/atomic_ops.rst where we have to make
> modifications visible before we let others see them? Here the one who is
> freezing is doing it so others can't get their page pin and interfere
> with the freezer's work. But maybe there are some (documented or not)
> consistency guarantees to expect once you obtain the pin, that can be
> violated, or they might be added later, so it would be safer to add the
> barrier?

The problem comes if the unfreeze is reordered so that it happens before the
freezer has performed its work. For example, in
migrate_huge_page_move_mapping:


if (!page_ref_freeze(page, expected_count)) {
spin_unlock_irq(&mapping->tree_lock);
return -EAGAIN;
}

newpage->index = page->index;
newpage->mapping = page->mapping;

get_page(newpage);

radix_tree_replace_slot(&mapping->page_tree, pslot, newpage);

page_ref_unfreeze(page, expected_count - 1);


then there's nothing stopping the CPU (and potentially the compiler) from
reordering the unfreeze call so that it effectively becomes:


if (!page_ref_freeze(page, expected_count)) {
spin_unlock_irq(&mapping->tree_lock);
return -EAGAIN;
}

page_ref_unfreeze(page, expected_count - 1);

newpage->index = page->index;
newpage->mapping = page->mapping;

get_page(newpage);

radix_tree_replace_slot(&mapping->page_tree, pslot, newpage);


which then means that the freezer's work is carried out without the page
being frozen.

Will

\
 
 \ /
  Last update: 2017-06-12 01:26    [W:0.071 / U:0.340 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site