Re: Add 64-bit XIDs into PostgreSQL 15

HI Maxim
> The aim of this patch is to make Postgres support 64-bit XIDs.
> This is why the TransactionID type size increases from 4 to 8 bytes.
>It also has an effect on the proc array, allowing two transactions that
> that are more than 2 billion XIDs apart to run at the same time.

> You couldn't store tuples that were more than 2 billion XIDs apart
> on a single heap page. That is correct. However, this annoying
> limitation comes only from the page format. Moreover, it looks like
> as long as we have a page format with a base, we will not be able
> to bypass this limitation. Yet, running transactions far apart is
> totally accepted.
Yes ，Furthermore, this approach is not unprecedented. Several
PostgreSQL-derived systems have already adopted 64-bit transaction IDs

Thanks

On Tue, Feb 10, 2026 at 2:19 PM Maxim Orlov <orlovmg(at)gmail(dot)com> wrote:

>
> On Mon, 9 Feb 2026 at 18:03, Heikki Linnakangas <hlinnaka(at)iki(dot)fi> wrote:
>
>>
>> The point is that we still do not want to use FullTransactionID
>> everywhere. Only in some places related to visibility checks that need
>> to deal with XIDs stored on sidk, like heapam_visibility.c and clog.c,
>> and it will probably spill over to some other places. But things like
>> the proc array can continue to use 32-bit XIDs.
>>
>> We will still have the limitation that you cannot have two transactions
>> *running* that are more than 2 billion XIDs apart. I think that's fine,
>> and we should not try to lift that limitation as part of this patch.
>>
>> The aim of this patch is to make Postgres support 64-bit XIDs.
> This is why the TransactionID type size increases from 4 to 8 bytes.
> It also has an effect on the proc array, allowing two transactions that
> that are more than 2 billion XIDs apart to run at the same time.
>
> You couldn't store tuples that were more than 2 billion XIDs apart
> on a single heap page. That is correct. However, this annoying
> limitation comes only from the page format. Moreover, it looks like
> as long as we have a page format with a base, we will not be able
> to bypass this limitation. Yet, running transactions far apart is
> totally accepted.
>
>
> --
> Best regards,
> Maxim Orlov.
>