Newsgroups: comp.lang.prolog
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!hookup!news.moneng.mei.com!howland.reston.ans.net!news.sprintlink.net!EU.net!sun4nl!freya.let.rug.nl!vannoord
From: vannoord@let.rug.nl (Gertjan van Noord)
Subject: Re: ProTcl on HPUX and Sicstus2.1#9
Sender: news@let.rug.nl (News system at let.rug.nl)
Message-ID: <1994Oct31.145325.5582@let.rug.nl>
Date: Mon, 31 Oct 1994 14:53:25 GMT
References: <1994Oct26.124040.5053@let.rug.nl> <CyBqAF.BII@ecrc.de>
Nntp-Posting-Host: vidar.let.rug.nl
Organization: Faculteit der Letteren, Rijksuniversiteit Groningen, NL
Lines: 36

In article <CyBqAF.BII@ecrc.de>, micha@ecrc.de (Micha Meier) writes:
> In article 5053@let.rug.nl,  vannoord@let.rug.nl (Gertjan van Noord) writes:
> >
> >I have been trying to get ProTcl 1.3 to work with Sicstus 2.1 
> >patch level 9, on a HP machine (HPUX 9.0, series 700).
> >
> >I was succesful in getting the following combinations to work:
> >- Linux + Sicstus2.1 patchlevels 8 and 9, by linking the ProTcl
> >  files statically
> >- HPUX + Sicstus2.1 patchlevel 8, by linking both the ProTcl
> >  and library(charsio.o) statically.
> >
> >Using dynamic loading I get a couple of the following warnings:
> >
> >ld: (Warning) Inter-quadrant branch in /usr/lib/X11R5.supplement/libtcl.a(tclExpr.o)
> >
> 
> This was the reason why I couldn't get it right with dynamic loading on eclipse.
> The reason is that tcl makes a table of pointers to math functions and the
> dynamic loader somehow manages to get the addresses right, but destroys
> their arguments. I'm surprized you get more with dynamic than with
> static loading, is there anything particular in the patch #9 ?
> 
> --Micha

Nothing in the CHANGES file seems to be related.. The strange thing still is
that I have no problems on my Linux box (one difference between the tcl/tk installation
on Linux and on HPUX is that in the former case I use dynamic tk and tcl libraries
whereas on HPUX I use static tcl tk libraries).

Yes I am surprised too that I get more with dynamic than with static loading. At
SICS they say this is unexpected also. 

Anything else I might try to do?


