Complete.Org: Mailing Lists: Archives: freeciv-dev: May 1999:
Re: [Freeciv-Dev] compiling under SunOS 4.1.3
Home

Re: [Freeciv-Dev] compiling under SunOS 4.1.3

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: mstamat@xxxxxxxxxx
Cc: freeciv-dev@xxxxxxxxxxx
Subject: Re: [Freeciv-Dev] compiling under SunOS 4.1.3
From: David Pfitzner <dwp@xxxxxxxxxxxxxx>
Date: Mon, 17 May 1999 22:58:54 +1000 (EST)

Stamatogiannakis Manolis <mstamat@xxxxxxxxxx> wrote:

> I compiled freeciv 1.8.0 on a sparc SunOS 4.1.3 machine,
> using gcc version 2.8.1. There were a few problems I 
> encountered (and solved) and I thought it would be good to report
> them...

Yes, thanks a lot!

If possible, could you please try a recent development
release, either using CVS 
(see http://www.freeciv.org/faq.html#cvs_source)
or snapshots 
(see http://www.freeciv.org/latest.html)

The development release won't (yet) fix your first two problems 
(though your same fixes should still work) but I think it _should_ 
fix all of the realloc(NULL,...) problems.  (Unless there are
any cases where the code has realloc instead of the new fc_realloc,
in which case just change realloc -> fc_realloc; fc_realloc 
explicitly checks whether the first arg is NULL.)

For the first two problems (NULL not in stdlib.h but only stdio.h,
no CLOCKS_PER_SEC) we should probably work out some configure
stuff, though I'm also tempted by quick-and-dirty #ifdef solutions...
  
> #define CLOCKS_PER_SEC 1000000
> 
> (of course the number can be anything positive without causing serious 
> problems, right?)

Yep.  I also seem to recall that some systems may define
CLOCKS_PER_SECOND instead.

[In configure.output]
> checking for strerror... (cached) no

Wow, we actually make use of the configure result for that one!
:-)

Regards,
-- David

[Prev in Thread] Current Thread [Next in Thread]