Re: BTMaxItemSize seems to be subtly incorrect

Поиск
Список
Период
Сортировка
От Peter Geoghegan
Тема Re: BTMaxItemSize seems to be subtly incorrect
Дата
Msg-id CAH2-Wzm_dbQLOmpxymmTg-CK7wgxPjd7c2hxrZafjtwLHXq8WQ@mail.gmail.com
обсуждение исходный текст
Ответ на Re: BTMaxItemSize seems to be subtly incorrect  (Peter Geoghegan <pg@bowt.ie>)
Список pgsql-hackers
On Fri, Aug 5, 2022 at 10:13 AM Peter Geoghegan <pg@bowt.ie> wrote:
> Update: I discovered that I can get the regression tests to fail (even
> on mainstream 64-bit platforms) by MAXALIGN()'ing the expression that
> we assign to state->maxpostingsize at the top of _bt_dedup_pass().

Looks like this was nothing more than a silly oversight with how the
macro was defined. As written, it would evaluate to the wrong thing at
the same point in nbtdedup.c, just because it was used in an
expression.

Pushed a fix for that just now.
-- 
Peter Geoghegan



В списке pgsql-hackers по дате отправления:

Предыдущее
От: David Zhang
Дата:
Сообщение: Re: Hash index build performance tweak from sorting
Следующее
От: Nathan Bossart
Дата:
Сообщение: Re: optimize lookups in snapshot [sub]xip arrays