Re: [PATCH v2] fadvise: move active pages to inactive list with POSIX_FADV_DONTNEED

[Andrea Righi <andrea@betterlinux.com>]

On Fri, Jun 24, 2011 at 07:06:42AM +0900, Minchan Kim wrote:
> Hi Andrea,
> Sorry for late response.
> These day, I have no time to see the LKML.
> 
> On Thu, Jun 23, 2011 at 3:36 PM, Andrea Righi <andrea@betterlinux.com> wrote:
> > There were some reported problems in the past about trashing page cache
> > when a backup software (i.e., rsync) touches a huge amount of pages (see
> > for example [1]).
> >
> > This problem has been almost fixed by the Minchan Kim's patch [2] and a
> > proper use of fadvise() in the backup software. For example this patch
> > set [3] has been proposed for inclusion in rsync.
> >
> > However, there can be still other similar trashing problems: when the
> > backup software reads all the source files, some of them may be part of
> > the actual working set of the system. When a
> > posix_fadvise(POSIX_FADV_DONTNEED) is performed _all_ pages are evicted
> > from pagecache, both the working set and the use-once pages touched only
> > by the backup software.
> 
> Agreed. It's rather aggressive.
> 
> >
> > With the following solution when posix_fadvise(POSIX_FADV_DONTNEED) is
> > called for an active page instead of removing it from the page cache it
> > is added to the tail of the inactive list. Otherwise, if it's already in
> > the inactive list the page is removed from the page cache.
> >
> > In this way if the backup was the only user of a page, that page will
> > be immediately removed from the page cache by calling
> > posix_fadvise(POSIX_FADV_DONTNEED). If the page was also touched by
> > other processes it'll be moved to the inactive list, having another
> > chance of being re-added to the working set, or simply reclaimed when
> > memory is needed.
> >
> > Testcase:
> >
> >  - create a 1GB file called "zero"
> >  - run md5sum zero to read all the pages in page cache (this is to
> >    simulate the user activity on this file)
> >  - run "rsync zero zero_copy" (rsync is patched with [3])
> >  - re-run md5sum zero (user activity on the working set) and measure
> >    the time to complete this command
> >
> > The test has been performed using 3.0.0-rc4 vanilla and with this patch
> > applied (3.0.0-rc4-fadvise).
> >
> > Results:
> >                  avg elapsed time      block:block_bio_queue
> >  3.0.0-rc4                  4.127s                      8,214
> >  3.0.0-rc4-fadvise          2.146s                          0
> >
> 
> Great!
> 
> > In the first case the file is evicted from page cache completely and we
> > must re-read it from the disk. In the second case the file is still in
> > page cache (in the inactive list) and we don't need any other additional
> > I/O operation.
> >
> > [1] http://marc.info/?l=rsync&m=128885034930933&w=2
> > [2] https://lkml.org/lkml/2011/2/20/57
> > [3] http://lists.samba.org/archive/rsync/2010-November/025827.html
> >
> > ChangeLog v1 -> v2:
> >  - fix comment in invalidate_mapping_pages()
> >
> > Acked-by: Rik van Riel <riel@redhat.com>
> > Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
> > Signed-off-by: Andrea Righi <andrea@betterlinux.com>
> > ---
> >  mm/swap.c     |    9 +++++----
> >  mm/truncate.c |   10 +++++++---
> >  2 files changed, 12 insertions(+), 7 deletions(-)
> >
> > diff --git a/mm/swap.c b/mm/swap.c
> > index 3a442f1..fc8bb76 100644
> > --- a/mm/swap.c
> > +++ b/mm/swap.c
> > @@ -411,10 +411,11 @@ void add_page_to_unevictable_list(struct page *page)
> >  *
> >  * 1. active, mapped page -> none
> >  * 2. active, dirty/writeback page -> inactive, head, PG_reclaim
> > - * 3. inactive, mapped page -> none
> > - * 4. inactive, dirty/writeback page -> inactive, head, PG_reclaim
> > - * 5. inactive, clean -> inactive, tail
> > - * 6. Others -> none
> > + * 3. active, clean -> inactive, tail
> > + * 4. inactive, mapped page -> none
> > + * 5. inactive, dirty/writeback page -> inactive, head, PG_reclaim
> > + * 6. inactive, clean -> inactive, tail
> > + * 7. Others -> none
> 
> Nitpick.
> I would like to put together them by on line as rather than adding another line.
> 5, [in]active, clean-> inactive, tail.
> I guess it's more easy to understand.

Agreed.

> 
> If you want to put it in another line, please change below comment, too.
> "In 5, why it moves inactive's head.."

Oh right. I'd put both on a single line anyway, as you suggested.

> 
> >  *
> >  * In 4, why it moves inactive's head, the VM expects the page would
> >  * be write it out by flusher threads as this is much more effective
> > diff --git a/mm/truncate.c b/mm/truncate.c
> > index 3a29a61..a36af48 100644
> > --- a/mm/truncate.c
> > +++ b/mm/truncate.c
> > @@ -357,11 +357,15 @@ unsigned long invalidate_mapping_pages(struct address_space *mapping,
> >                        if (lock_failed)
> >                                continue;
> >
> > -                       ret = invalidate_inode_page(page);
> 
> I would like to add comment.
> "Invalidation of active page is rather aggressive as we can't make
> sure it's not a working set of other processes.
> deactivate_page would move it into inactive's tail so the page will
> have a chance to activate again if other processes
> touch it. otherwise, it would be reclaimed simply".

OK.

> 
> > +                       if (PageActive(page))
> > +                               ret = 0;
> > +                       else
> > +                               ret = invalidate_inode_page(page);
> 
> 
> You have to change description of invalidate_mapping_pages.
> 
>  * invalidate_mapping_pages() will not block on IO activity. It will not
>  * invalidate pages which are dirty, locked, under writeback, mapped into
>  * pagetables or on active lru.

Correct.

> 
> >                        unlock_page(page);
> >                        /*
> > -                        * Invalidation is a hint that the page is no longer
> > -                        * of interest and try to speed up its reclaim.
> > +                        * Invalidation of an inactive page is a hint that the
> > +                        * page is no longer of interest and try to speed up
> > +                        * its reclaim.
> >                         */
> >                        if (!ret)
> >                                deactivate_page(page);
> > --
> > 1.7.4.1
> >
> >
> 
> Otherwise, Looks good to me.
> 
> Acked-by: Minchan Kim <minchan.kim@gmail.com>
> 
> -- 
> Kind regards,
> Minchan Kim

Thanks for the review.

I'll add all your comments and post a new version.

-Andrea