On Mon, Nov 24, 2014 at 11:45:08PM -0500, Theodore Ts'o wrote:

On Tue, Nov 25, 2014 at 12:53:32PM +1100, Dave Chinner wrote:

quoted

On Fri, Nov 21, 2014 at 02:59:23PM -0500, Theodore Ts'o wrote:

quoted

Guarantee that the on-disk timestamps will be no more than 24 hours
stale.

Signed-off-by: Theodore Ts'o <tytso@mit.edu>

If we put these inodes on the dirty inode list with at writeback
time of 24 hours, this is completely unnecessary.

What do you mean by "a writeback time of 24 hours"?  Do you mean
creating a new field in the inode which specifies when the writeback
should happen? 

No.

I still worry about the dirty inode list getting
somewhat long large in the strictatime && lazytime case, and the inode
bloat nazi's coming after us for adding a new field to struct inode
structure.

Use another pure inode time dirty list, and move the inode to the
existing dirty list when it gets marked I_DIRTY.

Or do you mean trying to abuse the dirtied_when field in some way?

No abuse necessary at all. Just a different inode_dirtied_after()
check is requires if the inode is on the time dirty list in
move_expired_inodes().

Cheers,

Dave.
-- 
Dave Chinner
david@fromorbit.com