| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Aleksander Alekseev <aleksander(at)tigerdata(dot)com> |
| Cc: | PostgreSQL Development <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Passing mac-addresses by value? |
| Date: | 2026-02-25 14:40:51 |
| Message-ID: | 167243.1772030451@sss.pgh.pa.us |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Aleksander Alekseev <aleksander(at)tigerdata(dot)com> writes:
> Mac-addresses are passed by reference. Now when Datums are always
> 64-bit values we can pass them by value, similarly to Timestamp[Tz].
> Are there any concears in respect of change like this e.g. backward
> comativility etc?
To pass standard macaddr (6 bytes) by value, we would have to add
tuple pack/unpack code paths to the existing ones for 1/2/4/8 bytes.
I can't see that that's justified. We could convert macaddr8 to be
pass-by-value without such global effects, but it looks like it
would still require wholesale changes to the macaddr8-processing
functions. On the whole I'm doubtful that there's enough gain there
to be worth the work.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Zsolt Parragi | 2026-02-25 15:08:39 | Re: Add GoAway protocol message for graceful but fast server shutdown/switchover |
| Previous Message | jian he | 2026-02-25 14:38:20 | persevere NO INHERIT when Dump not-null constraints on inherited columns |