Re: [WIP]Vertical Clustered Index (columnar store extension) - take2
От | Japin Li |
---|---|
Тема | Re: [WIP]Vertical Clustered Index (columnar store extension) - take2 |
Дата | |
Msg-id | ME0P300MB0445CC953E7D94B4A1BC07A4B648A@ME0P300MB0445.AUSP300.PROD.OUTLOOK.COM обсуждение исходный текст |
Ответы |
Re: [WIP]Vertical Clustered Index (columnar store extension) - take2
|
Список | pgsql-hackers |
On Thu, 10 Jul 2025 at 17:51, Peter Smith <smithpb2250@gmail.com> wrote: > On Thu, Jul 10, 2025 at 4:07 PM Japin Li <japinli@hotmail.com> wrote: > ... >> I'm still trying to understand the patches. >> >> diff --git a/src/include/storage/itemptr.h b/src/include/storage/itemptr.h >> index 74b87a9..d97d1c5 100644 >> --- a/src/include/storage/itemptr.h >> +++ b/src/include/storage/itemptr.h >> @@ -46,6 +46,9 @@ typedef struct ItemPointerData >> #endif >> ItemPointerData; >> >> +#define SizeOfIptrData \ >> + (offsetof(ItemPointerData, ip_posid) + sizeof(OffsetNumber)) >> + >> >> I've noticed this macro is currently defined within core; however, I found it only >> used in the VCI extension. >> >> Could you clarify the rationale for its inclusion in the core, and whether it's >> genuinely required there, or if it would be better suited within the extension >> itself? > > Right, this had previously also been reported by Tomas [1]. > > Upon investigation, I found that this was master code from 10 years > ago (back when this VCI patch was implemented). The master code has > moved on since then and removed this macro [2], but this VCI patch did > not... > > I'll try to address this for the next patchset. > > ====== > [1] https://www.postgresql.org/message-id/a748aa6b-c7e6-4d02-a590-ab404d590448%40vondra.me > [2] https://github.com/postgres/postgres/commit/8023b5827fbada6815ce269db4f3373ac77ec7c3 > My apologies, I wasn't following the full email thread. -- Regards, Japin Li
В списке pgsql-hackers по дате отправления: