Re: log_checkpoints, microseconds
| От | Tom Lane |
|---|---|
| Тема | Re: log_checkpoints, microseconds |
| Дата | |
| Msg-id | 20785.1397157615@sss.pgh.pa.us обсуждение исходный текст |
| Ответ на | Re: log_checkpoints, microseconds (Peter Geoghegan <pg@heroku.com>) |
| Ответы |
Re: log_checkpoints, microseconds
Re: log_checkpoints, microseconds |
| Список | pgsql-bugs |
Peter Geoghegan <pg@heroku.com> writes:
> On Thu, Apr 10, 2014 at 11:45 AM, Greg Stark <stark@mit.edu> wrote:
>> I think his point is that to go from microseconds to msec (which I
>> think should just be "ms" btw) you want to multiply by 1000 not
>> divide.
> Right.
I think you're both wrong. 1000 usec = 1 msec, not the other way round.
> Or just use the INSTR_TIME_GET_MILLISEC() macro to begin with,
> and do neither.
The code appears to be trying to track stats at the microsecond
level. The printout is following a policy decision that we prefer
to report units of msec, but that does not mean that we shouldn't
keep microsecond precision internally.
In short, I see nothing that needs fixed here.
regards, tom lane
В списке pgsql-bugs по дате отправления: