Re: XID formatting and SLRU refactorings (was: Add 64-bit XIDs into PostgreSQL 15)
От | Heikki Linnakangas |
---|---|
Тема | Re: XID formatting and SLRU refactorings (was: Add 64-bit XIDs into PostgreSQL 15) |
Дата | |
Msg-id | 4dd273c5-f594-1c04-3710-5811e62b18db@iki.fi обсуждение исходный текст |
Ответ на | Re: XID formatting and SLRU refactorings (was: Add 64-bit XIDs into PostgreSQL 15) (Maxim Orlov <orlovmg@gmail.com>) |
Ответы |
Re: XID formatting and SLRU refactorings (was: Add 64-bit XIDs into PostgreSQL 15)
Re: XID formatting and SLRU refactorings (was: Add 64-bit XIDs into PostgreSQL 15) |
Список | pgsql-hackers |
On 07/03/2023 13:38, Maxim Orlov wrote: > As for making pg_multixact 64 bit, I spend the last couple of days to > make proper pg_upgrade for pg_multixact's and for pg_xact's > with wraparound and I've understood, that it is not a simple task > compare to pg_xact's. The problem is, we do not have epoch for > multixacts, so we do not have ability to "overcome" wraparound. The > solution may be adding some kind of epoch for multixacts or > make them 64 bit in "main" 64-xid patch, but in perspective of this > thread, in my view, this should be last in line here. That is true for pg_multixact/offsets. We will indeed need to add an epoch and introduce the concept of FullMultiXactIds for that. However, we can change pg_multixact/members independently of that. We can extend MultiXactOffset from 32 to 64 bits, and eliminate pg_multixact/members wraparound, while keeping multi-xids 32 bits wide. - Heikki
В списке pgsql-hackers по дате отправления: