Re: [RFC PATCH 2/2] mm: Batch page_check_references in shrink_page_list sharing the same i_mmap_mutex

[Tim Chen <tim.c.chen@linux.intel.com>]

On Tue, 2012-09-04 at 08:21 -0700, Tim Chen wrote:
> On Tue, 2012-08-21 at 17:48 -0700, Tim Chen wrote:
> 
> > 
> > Thanks to Matthew's suggestions on improving the patch. Here's the
> > updated version.  It seems to be sane when I booted my machine up.  I
> > will put it through more testing when I get a chance.
> > 
> > Tim
> > 
> 
> Matthew,
> 
> The new patch seems to be causing some of the workloads with mmaped file
> read to seg fault.  Will need to dig further to find out why.
> 
> Tim
> 

Okay, the problem seems to be the code below.  It is too restrictive and
causes some cases where the mutex needs to be taken in try_to_unmap_file
to be missed.

> > +int needs_page_mmap_mutex(struct page *page)
> > +{
> > +	return page->mapping && page_mapped(page) && page_rmapping(page) &&
> > +		!PageKsm(page) && !PageAnon(page);
> > +}
> > +

Changing the check to the following fixes the problem:

@@ -873,8 +873,7 @@ static int page_referenced_file(struct page *page,
 
 int needs_page_mmap_mutex(struct page *page)
 {
-       return page->mapping && page_mapped(page) && page_rmapping(page) &&
-               !PageKsm(page) && !PageAnon(page);
+       return page->mapping && !PageKsm(page) && !PageAnon(page);
 }

I'll do more testing and generate a second version of the patch set with the fixes.

Tim