building 6.4 on sunos 4.1.4 - Mailing list pgsql-hackers

From Stephen Kogge
Subject building 6.4 on sunos 4.1.4
Date
Msg-id 199811171854.NAA18221@tuna.uimage.com
Whole thread Raw
Responses Re: [HACKERS] building 6.4 on sunos 4.1.4
List pgsql-hackers
Hello
I am the fellow trying to build 6.4 for sunos - it seems changes somewhere
in the past have messed up the sunos4.1.4 build.
I have a list to keep -Wall from complaining (just warnings) and have
added  the needed #defines back for the port.
I am down to the build in 
...src/interfaces/libpq
where it tries to do

...
ar crs libpq.a `lorder fe-auth.o fe-connect.o fe-exec.o fe-misc.o fe-print.o 
fe-
lobj.o dllist.o pqsignal.o | tsort`
tsort: cycle in data
tsort: fe-connect.o
tsort: fe-auth.o
tsort: cycle in data
tsort: fe-exec.o
tsort: fe-connect.o
ar: bad option `s'
gmake[2]: *** [libpq.a] Error 1
...
since sunos does not have a 's' option for ar I have traced this back
to Makefile.global which sets

AROPT=crs
The override for sunos (cr) is not invoked. HUMMMM
The 's' option not available means an explicit ranlib is required.
If whoever thinks they understand  configure can come up with a patch  so
sunos sets AROPT=cr and does the ranlib I will apply it and pass back the other
patches for sunos - really not many. It just takes a real long time to build on
a IPC (SS1+). 
If no-one speaks up I will see if I can come up with a configure patch, but
if it is built from another tool then that tool needs to be patched.

-- Stephen N. Koggestevek@uimage.comhttp://www.uimage.com




pgsql-hackers by date:

Previous
From: Vince Vielhaber
Date:
Subject: Re: [HACKERS] New to PostgreSQL, is this a DoS?
Next
From: Peter T Mount
Date:
Subject: Re: [HACKERS] PREPARE