Re: GNU/Hurd portability patches - Mailing list pgsql-hackers

From Tom Lane
Subject Re: GNU/Hurd portability patches
Date
Msg-id 3269924.1751401476@sss.pgh.pa.us
Whole thread Raw
In response to Re: GNU/Hurd portability patches  (Michael Banck <mbanck@gmx.net>)
List pgsql-hackers
Michael Banck <mbanck@gmx.net> writes:
> On Tue, Jul 01, 2025 at 12:41:50PM -0400, Tom Lane wrote:
> | * If <limits.h> didn't define IOV_MAX, define our own.  X/Open requires at
> | * least 16.  (GNU Hurd apparently feel that they're not bound by X/Open,
> | * because they don't define this symbol at all.)

> I personally don't care much about those missing limits on the Hurd, but
> Thomas mentioned in
> CA+hUKG+tqFVY7Fi=WBvZ6-UsATjcPNBDtphDm7YLjevm2kxSvw@mail.gmail.com (and
> Samuel Thibault cited the same sentence to me now when I discussed the
> commit with him) that POSIX said "A definition of one of the symbolic
> constants in the following list shall be omitted from <limits.h> on
> specific implementations where the corresponding value is equal to or
> greater than the stated minimum, but is unspecified". So "requires at
> least 16" might be a bit too strong here, AIUI.

Oh, I missed that bit of the spec.  I think "requires at least 16"
is correct anyway, but the parenthetical remark isn't really right.
Not sure if it's worth changing --- the end result is the same in
any case.

            regards, tom lane



pgsql-hackers by date:

Previous
From: Nathan Bossart
Date:
Subject: Re: teach pg_upgrade to handle in-place tablespaces
Next
From: Jacob Champion
Date:
Subject: Re: libpq: Process buffered SSL read bytes to support records >8kB on async API