Re: pgsql: Remove dead ports

Remove dead ports

Remove the following ports:

- dgux
- nextstep
- sunos4
- svr4
- ultrix4
- univel

These are obsolete and not worth rescuing. In most cases, there is
circumstantial evidence that they wouldn't work anymore anyway.

Branch
------
master

Details
-------
http://git.postgresql.org/pg/commitdiff/f2f9439fbfba378cb64cd6e5a046e0184cd542c6

Modified Files
--------------
configure | 9 ---
configure.in | 9 ---
doc/src/sgml/installation.sgml | 6 +-
src/Makefile.shlib | 12 ----
src/backend/main/main.c | 4 -
src/backend/port/dynloader/dgux.c | 6 --
src/backend/port/dynloader/dgux.h | 44 -------------
src/backend/port/dynloader/irix.h | 2 +-
src/backend/port/dynloader/nextstep.c | 84 -------------------------
src/backend/port/dynloader/nextstep.h | 26 --------
src/backend/port/dynloader/sunos4.c | 7 --
src/backend/port/dynloader/sunos4.h | 46 --------------
src/backend/port/dynloader/svr4.c | 7 --
src/backend/port/dynloader/svr4.h | 46 --------------
src/backend/port/dynloader/ultrix4.c | 67 --------------------
src/backend/port/dynloader/ultrix4.h | 109 ---------------------------------
src/backend/port/dynloader/univel.c | 7 --
src/backend/port/dynloader/univel.h | 49 ---------------
src/backend/port/dynloader/unixware.h | 2 +-
src/backend/port/nextstep/Makefile | 17 -----
src/backend/port/nextstep/port.c | 62 -------------------
src/include/port/dgux.h | 3 -
src/include/port/nextstep.h | 19 ------
src/include/port/sunos4.h | 7 --
src/include/port/svr4.h | 3 -
src/include/port/ultrix4.h | 57 -----------------
src/include/port/univel.h | 8 ---
src/include/storage/s_lock.h | 13 ----
src/makefiles/Makefile.dgux | 9 ---
src/makefiles/Makefile.sunos4 | 12 ----
src/makefiles/Makefile.svr4 | 15 -----
src/makefiles/Makefile.ultrix4 | 11 ---
src/makefiles/Makefile.univel | 10 ---
src/template/dgux | 1 -
src/template/nextstep | 5 --
src/template/sunos4 | 7 --
src/template/univel | 2 -
37 files changed, 5 insertions(+), 798 deletions(-)

Peter Eisentraut <peter_e(at)gmx(dot)net> writes:
> Remove dead ports

Checking this patch, I noticed that config.guess and config.sub harbor
most of the remaining references to those platforms, which reminded me:
don't we usually update those files from autoconf upstream before beta?

regards, tom lane

On ons, 2012-05-02 at 00:31 -0400, Tom Lane wrote:
> Peter Eisentraut <peter_e(at)gmx(dot)net> writes:
> > Remove dead ports
>
> Checking this patch, I noticed that config.guess and config.sub harbor
> most of the remaining references to those platforms, which reminded me:
> don't we usually update those files from autoconf upstream before beta?

Yes, once we know when beta is, we can move on that. ;-)

Btw., I had intentionally kept the uses in ps_status.c and getrusage.c,
because they remain useful in case someone wants to use these files for
reference. But I guess that's debatable, because we no longer have a
way to prove that those uses actually continue to work.

Peter Eisentraut <peter_e(at)gmx(dot)net> writes:
> On ons, 2012-05-02 at 00:31 -0400, Tom Lane wrote:
>> Checking this patch, I noticed that config.guess and config.sub harbor
>> most of the remaining references to those platforms, which reminded me:
>> don't we usually update those files from autoconf upstream before beta?

> Yes, once we know when beta is, we can move on that. ;-)

Next week, I thought.

regards, tom lane

On Wed, May 02, 2012 at 01:15:48PM -0400, Tom Lane wrote:
> Peter Eisentraut <peter_e(at)gmx(dot)net> writes:
> > On ons, 2012-05-02 at 00:31 -0400, Tom Lane wrote:
> >> Checking this patch, I noticed that config.guess and config.sub harbor
> >> most of the remaining references to those platforms, which reminded me:
> >> don't we usually update those files from autoconf upstream before beta?
>
> > Yes, once we know when beta is, we can move on that. ;-)
>
> Next week, I thought.

How are we handling the Monday release with everyone at PGCon? Was that
resolved?

--
Bruce Momjian <bruce(at)momjian(dot)us> http://momjian.us
EnterpriseDB http://enterprisedb.com

+ It's impossible for everything to be true. +

On Wed, May 02, 2012 at 12:31:10AM -0400, Tom Lane wrote:
> Peter Eisentraut <peter_e(at)gmx(dot)net> writes:
> > Remove dead ports
>
> Checking this patch, I noticed that config.guess and config.sub harbor
> most of the remaining references to those platforms, which reminded me:
> don't we usually update those files from autoconf upstream before beta?

I looked at that because they have 'bsdi' references too. Is that file
auto-generated, and the references are not ours but autoconf's? There
are a lot of platforms there we never supported, e.g.
Night_Hawk:Power_UNIX.

--
Bruce Momjian <bruce(at)momjian(dot)us> http://momjian.us
EnterpriseDB http://enterprisedb.com

+ It's impossible for everything to be true. +

>> Next week, I thought.
>
> How are we handling the Monday release with everyone at PGCon? Was that
> resolved?

I have yet to see a confirmed date, guys. If we expect any support from
the packagers and/or the advocacy volunteers, then people need at least
a week's notice, probably more.

--
Josh Berkus
PostgreSQL Experts Inc.
http://pgexperts.com

Josh Berkus <josh(at)agliodbs(dot)com> writes:
>> How are we handling the Monday release with everyone at PGCon? Was that
>> resolved?

> I have yet to see a confirmed date, guys. If we expect any support from
> the packagers and/or the advocacy volunteers, then people need at least
> a week's notice, probably more.

I haven't seen anybody positively say we can't do a wrap next Thursday
and release Monday, so I've been assuming that's what will happen.
If there are reasons to think it won't work, let's hear 'em now.

regards, tom lane