Commit 08142579 authored by Jan Kara's avatar Jan Kara Committed by Linus Torvalds
Browse files

mm: fix assertion mapping->nrpages == 0 in end_writeback() 



Under heavy memory and filesystem load, users observe the assertion
mapping->nrpages == 0 in end_writeback() trigger.  This can be caused by
page reclaim reclaiming the last page from a mapping in the following
race:

	CPU0				CPU1
  ...
  shrink_page_list()
    __remove_mapping()
      __delete_from_page_cache()
        radix_tree_delete()
					evict_inode()
					  truncate_inode_pages()
					    truncate_inode_pages_range()
					      pagevec_lookup() - finds nothing
					  end_writeback()
					    mapping->nrpages != 0 -> BUG
        page->mapping = NULL
        mapping->nrpages--

Fix the problem by doing a reliable check of mapping->nrpages under
mapping->tree_lock in end_writeback().

Analyzed by Jay <jinshan.xiong@whamcloud.com>, lost in LKML, and dug out
by Miklos Szeredi <mszeredi@suse.de>.

Cc: Jay <jinshan.xiong@whamcloud.com>
Cc: Miklos Szeredi <mszeredi@suse.de>
Signed-off-by: default avatarJan Kara <jack@suse.cz>
Cc: <stable@kernel.org>
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
parent 9b679320

fs/inode.c

+7 −0
Original line number Diff line number Diff line
@@ -423,7 +423,14 @@ EXPORT_SYMBOL(remove_inode_hash); 
void end_writeback(struct inode *inode)

{

	might_sleep();

	/*

	 * We have to cycle tree_lock here because reclaim can be still in the

	 * process of removing the last page (in __delete_from_page_cache())

	 * and we must not free mapping under it.

	 */

	spin_lock_irq(&inode->i_data.tree_lock);

	BUG_ON(inode->i_data.nrpages);

	spin_unlock_irq(&inode->i_data.tree_lock);

	BUG_ON(!list_empty(&inode->i_data.private_list));

	BUG_ON(!(inode->i_state & I_FREEING));

	BUG_ON(inode->i_state & I_CLEAR);

include/linux/fs.h

+1 −0
Original line number Diff line number Diff line
@@ -639,6 +639,7 @@ struct address_space { 
	struct prio_tree_root	i_mmap;		/* tree of private and shared mappings */

	struct list_head	i_mmap_nonlinear;/*list VM_NONLINEAR mappings */

	struct mutex		i_mmap_mutex;	/* protect tree, count, list */

	/* Protected by tree_lock together with the radix tree */

	unsigned long		nrpages;	/* number of total pages */

	pgoff_t			writeback_index;/* writeback starts here */

	const struct address_space_operations *a_ops;	/* methods */

mm/truncate.c

+5 −0
Original line number Diff line number Diff line
@@ -304,6 +304,11 @@ EXPORT_SYMBOL(truncate_inode_pages_range); 
 * @lstart: offset from which to truncate

 *

 * Called under (and serialised by) inode->i_mutex.

 *

 * Note: When this function returns, there can be a page in the process of

 * deletion (inside __delete_from_page_cache()) in the specified range.  Thus

 * mapping->nrpages can be non-zero when this function returns even after

 * truncation of the whole mapping.

 */

void truncate_inode_pages(struct address_space *mapping, loff_t lstart)

{

CRA Git | Maintained and supported by SUSTech CRA and CCSE