Re: [PATCH 3/3] ext4: update writeback_index based on last page scanned

[Ted Ts'o <tytso@mit.edu>]

On Fri, Oct 22, 2010 at 04:45:17PM -0500, Eric Sandeen wrote:
> As pointed out in a prior patch, updating the mapping's
> writeback_index based on pages written isn't quite right;
> what the writeback index is really supposed to reflect is
> the next page which should be scanned for writeback during
> periodic flush.
> 
> As in write_cache_pages(), write_cache_pages_da() does
> this scanning for us as we assemble the mpd for later
> writeout.  If we keep track of the next page after the
> current scan, we can easily update writeback_index without
> worrying about pages written vs. pages skipped, etc.
> 
> Without this, an fsync will reset writeback_index to
> 0 (its starting index) + however many pages it wrote, which
> can mess up the progress of periodic flush.
> 
> Signed-off-by: Eric Sandeen <sandeen@redhat.com>

Have you done any benchmarks with and without this patch series?

Say, compilebench on a used and mildly fragmented file system?

     		       	    	       - Ted