Re: security_context_t marked as deprecated in libselinux 3.1

Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org> writes:
> On 2020-Aug-14, Michael Paquier wrote:
>> On Thu, Aug 13, 2020 at 08:47:28PM -0400, Tom Lane wrote:
>>> Should we back-patch that? Usually I figure that people might want
>>> to build back PG branches on newer platforms at some point, so that
>>> it's useful to apply portability fixes across-the-board. On the
>>> other hand, since it's only a compiler warning, maybe it's not worth
>>> the trouble.

>> Not sure that's worth the trouble as long as people don't complain
>> about it directly, and it does not prevent the contrib module to
>> work.

> FWIW I just had a CI job fail the "warnings" test because of lacking
> this patch in the back branches :-) What do you think about
> back-patching this to at least 11?

No objection to back-patching from me.

> I would say 10, but since that one
> is going to end soon, it might not be worth much effort. OTOH maybe we
> want to backpatch all the way back to 9.2 given the no-warnings policy
> we recently acquired.

I'm not sure that no-warnings policy extends to stuff as far off the
beaten path as sepgsql. However, I won't stand in the way if you
want to do that. One point though: if you want to touch v10, I'd
suggest waiting till after next week's releases. Unlikely as it
is that this'd break anything, I don't think we should risk it
in the branch's last release.

regards, tom lane