pgsql/src/makefiles (Makefile.freebsd Makefile.hpux Makefile.irix5 Makefile.linux Makefile.netbsd Makefile.openbsd Makefile.osf Makefile.solaris Makefile.unixware)

Peter,
Be aware that the -R for UnixWare is a *NEW* feature for the 7.1.1b
Feature Supplement. (Should be out by 7.1 REL, but isn't YET).

Larry
* Peter Eisentraut - PostgreSQL <petere@postgresql.org> [001027 21:12]:

Date: Friday, October 27, 2000 @ 19:59:39
Author: petere

Update of /home/projects/pgsql/cvsroot/pgsql/src/makefiles
from hub.org:/home/projects/pgsql/tmp/cvs-serv85608/src/makefiles

Modified Files:
Makefile.freebsd Makefile.hpux Makefile.irix5 Makefile.linux
Makefile.netbsd Makefile.openbsd Makefile.osf Makefile.solaris
Makefile.unixware

----------------------------- Log Message -----------------------------

Make use of system-specific linker option to embed shared library search
path into executables and shared libraries (-rpath or -R for most). Can be
disabled with --disable-rpath, since some binary packaging standards do not
like this option.

--
Larry Rosenman http://www.lerctr.org/~ler
Phone: +1 972-414-9812 (voice) Internet: ler@lerctr.org
US Mail: 1905 Steamboat Springs Drive, Garland, TX 75044-6749

Larry Rosenman writes:

Be aware that the -R for UnixWare is a *NEW* feature for the 7.1.1b
Feature Supplement. (Should be out by 7.1 REL, but isn't YET).

Hmm, do the old linkers bark at this? If so, should we remove it or is it
worth putting in a special check? (And what would that check be?)

--
Peter Eisentraut peter_e@gmx.net http://yi.org/peter-e/

* Peter Eisentraut <peter_e@gmx.net> [001028 09:53]:

Larry Rosenman writes:

Be aware that the -R for UnixWare is a *NEW* feature for the 7.1.1b
Feature Supplement. (Should be out by 7.1 REL, but isn't YET).

Hmm, do the old linkers bark at this? If so, should we remove it or is it
worth putting in a special check? (And what would that check be?)

I think a special check is warranted. We *WANT* the run-path to work.
$ ld -V
UX:ld: INFO: Optimizing C Compilation System (CCS) 4.0 10/07/00 (UDK FS 7.1.1b)

Pre the Feature supplement the (CCS) number was less than 4.0:
UX:ld: INFO: Optimizing C Compilation System (CCS) 3.2 09/28/99 (santamariasbl5.1.rs1ubl7)

I'd use that as a check

Also the __SCO_VERSION__ was less than:
$ more xx.c
#include <stdio.h>
int main(int argc, char **argv)
{
printf("__SCO_VERSION__=%ld\n",__SCO_VERSION__);
}
$ cc -O -o xx xx.c
$ ./xx
__SCO_VERSION__=400200010
$
I'd either parse ld -V or run a quick C program and check the first
digit of __SCO_VERSION__ for 4 or greater (or 3 or less to negate it).

Larry

--
Peter Eisentraut peter_e@gmx.net http://yi.org/peter-e/

--
Larry Rosenman http://www.lerctr.org/~ler
Phone: +1 972-414-9812 (voice) Internet: ler@lerctr.org
US Mail: 1905 Steamboat Springs Drive, Garland, TX 75044-6749