[PATCH] mm: don't return 0 too early from find_get_pages()

[PATCH] mm: don't return 0 too early from find_get_pages()

[Hugh Dickins]

Callers of find_get_pages(), or its wrapper pagevec_lookup() - notably
truncate_inode_pages_range() - stop looking further when it returns 0.

But if an interrupt comes just after its radix_tree_gang_lookup_slot(),
especially if we have preemptible RCU enabled, isn't it conceivable
that all 14 pages returned could be removed from the page cache by
shrink_page_list(), before find_get_pages() gets to process them?  So
causing it to return 0 although there may be plenty more pages beyond.

Make find_get_pages() and find_get_pages_tag() check for this unlikely
case, and restart should it occur; but callers of find_get_pages_contig()
have no such expectation, it's okay for that to return 0 early.

I have not seen this in practice, just worried by the possibility.

Signed-off-by: Hugh Dickins <hughd@google.com>
---

 mm/filemap.c |   14 ++++++++++++++
 1 file changed, 14 insertions(+)

--- 2.6.38-rc6/mm/filemap.c	2011-01-18 22:04:56.000000000 -0800
+++ linux/mm/filemap.c	2011-02-23 16:06:19.000000000 -0800
@@ -800,6 +800,13 @@ repeat:
 		pages[ret] = page;
 		ret++;
 	}
+
+	/*
+	 * If all entries were removed before we could secure them,
+	 * try again, because callers stop trying once 0 is returned.
+	 */
+	if (unlikely(!ret && nr_found))
+		goto restart;
 	rcu_read_unlock();
 	return ret;
 }
@@ -909,6 +916,13 @@ repeat:
 		pages[ret] = page;
 		ret++;
 	}
+
+	/*
+	 * If all entries were removed before we could secure them,
+	 * try again, because callers stop trying once 0 is returned.
+	 */
+	if (unlikely(!ret && nr_found))
+		goto restart;
 	rcu_read_unlock();
 
 	if (ret)

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] mm: don't return 0 too early from find_get_pages()

[Peter Zijlstra]

On Wed, 2011-02-23 at 21:35 -0800, Hugh Dickins wrote:
> Callers of find_get_pages(), or its wrapper pagevec_lookup() - notably
> truncate_inode_pages_range() - stop looking further when it returns 0.
> 
> But if an interrupt comes just after its radix_tree_gang_lookup_slot(),
> especially if we have preemptible RCU enabled, isn't it conceivable
> that all 14 pages returned could be removed from the page cache by
> shrink_page_list(), before find_get_pages() gets to process them?  So
> causing it to return 0 although there may be plenty more pages beyond.
> 
> Make find_get_pages() and find_get_pages_tag() check for this unlikely
> case, and restart should it occur; but callers of find_get_pages_contig()
> have no such expectation, it's okay for that to return 0 early.
> 
> I have not seen this in practice, just worried by the possibility.
> 
> Signed-off-by: Hugh Dickins <hughd@google.com>

Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>