Jason Evans 6f29a83924 Add rtree lookup path caching.
rtree-based extent lookups remain more expensive than chunk-based run
lookups, but with this optimization the fast path slowdown is ~3 CPU
cycles per metadata lookup (on Intel Core i7-4980HQ), versus ~11 cycles
prior.  The path caching speedup tends to degrade gracefully unless
allocated memory is spread far apart (as is the case when using a
mixture of sbrk() and mmap()).
2016-06-05 20:59:57 -07:00
..
2016-06-05 20:42:23 -07:00
2016-06-05 20:42:23 -07:00
2016-06-05 20:59:57 -07:00
2016-06-05 20:42:23 -07:00
2016-06-03 12:27:41 -07:00
2016-04-17 13:44:59 -07:00
2016-06-03 12:27:41 -07:00
2016-04-12 23:18:25 -07:00
2014-12-08 16:34:04 -08:00
2014-12-08 16:34:04 -08:00
2016-02-23 18:09:25 -08:00
2016-06-05 20:59:57 -07:00
2016-06-05 20:42:24 -07:00
2016-02-19 20:29:06 -08:00
2016-06-05 20:59:57 -07:00