Re: -O switch - Mailing list pgsql-hackers

From Tom Lane
Subject Re: -O switch
Date
Msg-id 686614.1604339880@sss.pgh.pa.us
Whole thread Raw
In response to Re: -O switch  (Magnus Hagander <magnus@hagander.net>)
Responses Re: -O switch
List pgsql-hackers
Magnus Hagander <magnus@hagander.net> writes:
> PFA a patch to do this.

One thing you missed is that the getopt() calls in both postmaster.c
and postgres.c have 'o:' entries that should be removed.  Also IIRC
there is a "case 'o'" in postgres.c to go along with that.

> Initially I kept the dynamic argv/argc in even though it's now
> hardcoded, in case we wanted to add something back. But given the way
> it looks now, perhaps we should just get rid of BackendRun()
> completely and directly call PostgresMain()? Or keep BackendRun() with
> just setting the TopMemoryContext, but removing the dynamic parts?

I'd be inclined to keep it as-is for now.  It's not adding any significant
amount of cycles compared to the process fork, so we might as well
preserve flexibility.

Is it really possible to not include miscadmin.h in postmaster.c?
I find that a bit surprising.

            regards, tom lane



pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: Collation versioning
Next
From: Stephen Frost
Date:
Subject: Re: Disable WAL logging to speed up data loading