Chris Mason dc7fdde39e Btrfs: reduce lock contention during extent insertion
We're spending huge amounts of time on lock contention during
end_io processing because we unconditionally assume we are overwriting
an existing extent in the file for each IO.

This checks to see if we are outside i_size, and if so, it uses a
less expensive readonly search of the btree to look for existing
extents.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
2012-04-27 14:51:05 -04:00
..
2012-01-03 22:55:04 -05:00
2012-01-03 22:54:53 -05:00
2012-01-03 22:54:07 -05:00
2012-01-03 22:54:57 -05:00
2012-01-03 22:55:10 -05:00
2012-02-28 09:52:39 +00:00
2012-01-03 22:54:54 -05:00
2012-01-09 10:48:11 -05:00
2012-02-17 17:34:03 -05:00
2012-01-03 22:55:01 -05:00
2012-02-02 12:55:17 -08:00
2012-01-03 22:55:07 -05:00
2012-01-10 13:45:22 -08:00
2012-03-09 18:59:59 -08:00
2012-01-03 22:55:11 -05:00
2012-01-03 22:54:54 -05:00
2012-02-08 22:07:18 +01:00
2012-01-03 22:54:07 -05:00
2012-03-18 12:25:04 -07:00
2012-01-03 22:52:39 -05:00
2012-03-10 17:07:28 -05:00
2012-01-05 15:40:12 -08:00
2012-01-03 22:54:07 -05:00
2012-03-10 17:05:30 -05:00
2011-10-31 17:30:44 -07:00
2012-01-03 22:52:40 -05:00
2012-01-03 22:54:07 -05:00
2011-11-02 12:53:43 +01:00
2012-01-03 22:53:07 -05:00
2012-01-03 22:54:07 -05:00
2012-01-03 22:52:40 -05:00