Ben Noordhuis
78ca555845
bench: optimize io.c benchmark
...
Use static buffers. Most clock ticks were spent in malloc() and free() instead
of read() and write().
Fix measurements. Really fast runs would result in bogus results like:
Wrote 1048576000 bytes in -0.731630s using 8192 byte buffers: -1366.811093mB/s
2011-11-10 21:13:37 +01:00
..
2011-08-12 18:42:24 +02:00
2010-09-28 01:14:38 -07:00
2010-09-09 11:03:50 -07:00
2010-09-09 11:03:50 -07:00
2010-09-09 11:03:50 -07:00
2010-09-09 11:03:50 -07:00
2010-10-18 17:48:49 -07:00
2011-11-09 17:34:53 +01:00
2011-10-12 12:29:40 -07:00
2010-06-29 23:59:24 -07:00
2010-10-27 12:09:16 -07:00
2010-10-28 11:42:22 -07:00
2011-11-10 21:13:37 +01:00
2010-10-11 15:21:36 -07:00
2011-10-31 22:02:46 +01:00
2011-01-10 18:00:21 -08:00
2010-10-11 15:21:36 -07:00
2011-02-18 14:01:04 -08:00
2010-10-11 15:21:36 -07:00
2010-10-26 14:53:18 -07:00
2010-09-09 16:22:57 -07:00
2011-11-06 01:25:07 +09:00
2010-08-21 01:24:14 -07:00
2011-10-11 22:40:04 -07:00
2011-10-11 22:40:04 -07:00
2009-07-13 16:38:55 +02:00
2011-02-14 17:26:54 -08:00