Re: Allow logical failover slots to wait on synchronous replication
От | John H |
---|---|
Тема | Re: Allow logical failover slots to wait on synchronous replication |
Дата | |
Msg-id | CA+-JvFvJG=j4BwXecCBx_ZTLML11NiG53_QOtZPMWFC=osdoNg@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: Allow logical failover slots to wait on synchronous replication (Bertrand Drouvot <bertranddrouvot.pg@gmail.com>) |
Ответы |
Re: Allow logical failover slots to wait on synchronous replication
|
Список | pgsql-hackers |
Hi Bertrand, > 1 === > ... > That's worth additional comments in the code. There's this comment already about caching the value already, not sure if you prefer something more? /* Cache values to reduce contention on lock */ > 2 === > ... > Looks like setting initialized to true is missing once done. Thanks, will update. > 3 === > ... > NUM_SYNC_REP_WAIT_MODE is small but as the goal is the keep the lock time as > short as possible I wonder if it wouldn't be better to use memcpy() here instead > of this for loop. > It results in a "Wdiscarded-qualifiers" which is safe given we take the lock, but adds noise? What do you think? "slot.c:2756:46: warning: passing argument 2 of ‘memcpy’ discards ‘volatile’ qualifier from pointer target type [-Wdiscarded-qualifiers]" Thanks, -- John Hsu - Amazon Web Services
В списке pgsql-hackers по дате отправления: