Re: make MaxBackends available in _PG_init

On Tue, Feb 8, 2022 at 9:38 PM Michael Paquier <michael(at)paquier(dot)xyz> wrote:
> On Tue, Feb 08, 2022 at 04:12:26PM -0500, Robert Haas wrote:
> > After some investigation I've determined that it's no longer Friday
> > afternoon. I also spent time investigating whether the patch had
> > problems that would make me uncomfortable with the idea of committing
> > it, and I did not find any. So, I committed it.
>
> @@ -1641,8 +1642,8 @@ SignalBackends(void)
> * XXX in principle these pallocs could fail, which would be bad. Maybe
> * preallocate the arrays? They're not that large, though.
> */
> - pids = (int32 *) palloc(MaxBackends * sizeof(int32));
> - ids = (BackendId *) palloc(MaxBackends * sizeof(BackendId));
> + pids = (int32 *) palloc(GetMaxBackends() * sizeof(int32));
> + ids = (BackendId *) palloc(GetMaxBackends() * sizeof(BackendId));
>
> You could have optimized this one, while on it, as well as the ones in
> pgstat_beinit() and pg_safe_snapshot_blocking_pids(). It is not hot,
> but you did that for all the other callers of GetMaxBackends(). Just
> saying..

Well I didn't do anything myself except review and commit Nathan's
patch, so I suppose you mean he could have done that, but fair enough.
I don't mind if you want to change it around.

--
Robert Haas
EDB: http://www.enterprisedb.com