diff options
| author | Heikki Linnakangas <heikki.linnakangas@iki.fi> | 2016-09-21 13:14:48 +0300 |
|---|---|---|
| committer | Heikki Linnakangas <heikki.linnakangas@iki.fi> | 2016-09-21 13:16:02 +0300 |
| commit | 93834a20f67ee458c98067e253894457c5a958ad (patch) | |
| tree | 495188d31b5a4138e76a8bb0878bca16401154a1 | |
| parent | 496c90b5e04729916ae9131ac93b2e63b46e3e2f (diff) | |
| download | postgresql-93834a20f67ee458c98067e253894457c5a958ad.tar.gz postgresql-93834a20f67ee458c98067e253894457c5a958ad.zip | |
Fix pgbench's calculation of average latency, when -T is not used.
If the test duration was given in # of transactions (-t or no option),
rather as a duration (-T), the latency average was always printed as 0.
It has been broken ever since the display of latency average was added,
in 9.4.
Fabien Coelho
Discussion: <alpine.DEB.2.20.1607131015370.7486@sto>
| -rw-r--r-- | src/bin/pgbench/pgbench.c | 6 |
1 files changed, 4 insertions, 2 deletions
diff --git a/src/bin/pgbench/pgbench.c b/src/bin/pgbench/pgbench.c index f3d51196843..254ca95f3de 100644 --- a/src/bin/pgbench/pgbench.c +++ b/src/bin/pgbench/pgbench.c @@ -3293,9 +3293,11 @@ printResults(TState *threads, StatsData *total, instr_time total_time, if (throttle_delay || progress || latency_limit) printSimpleStats("latency", &total->latency); else - /* only an average latency computed from the duration is available */ + { + /* no measurement, show average latency computed from run time */ printf("latency average: %.3f ms\n", - 1000.0 * duration * nclients / total->cnt); + 1000.0 * time_include * nclients / total->cnt); + } if (throttle_delay) { |