Solaris make for 0.63 failing
Prof Brian Ripley
Prof Brian Ripley <ripley@stats.ox.ac.uk>
Mon, 16 Nov 1998 17:14:31 +0000 (GMT)
> Date: Mon, 16 Nov 1998 12:04:20 -0500
> From: Paul Gilbert <pgilbert@bank-banque-canada.ca>
> To: Prof Brian Ripley <ripley@stats.ox.ac.uk>, R-devel
<R-devel@stat.math.ethz.ch>
> Subject: Re: Solaris make for 0.63 failing
>
> >You absolutely need GNU make, and on Solaris 2.6 you also need not to
> >use make-3.77 (the current version) as that has a bug which affects
> >this. `They' are using features found in few makes.
>
> I seem to be using GNU make 3.77. Is this a bug that is in both the Sun make
and
> the most recent GNU make, or is the R makefile doing something wrong?
Different causes. However, I get precisely that message from Sun make, and a
rather different one from GNU make 3.77 (about a garbled up file name).
The problems are:
(i) the Makefiles use GNU makefile syntax, and so are not compatible
with Sun's make. A feature, I think.
(ii) make 3.77 is set up to be part compiled for 64-bit file systems
and part not. This matters if you have complicated wildcards, only,
on Solaris 2.6 and other 64-bit-file-system aware systems, only.
A bug! If you want to fix it, rebuild 3.77 and use the same flags in
the Makefile for glob as in the main directory.
As to whether `the R makefile is doing something wrong', that is a matter of
opinion, and Theirs counts.
Brian
--
Brian D. Ripley, ripley@stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272860 (secr)
Oxford OX1 3TG, UK Fax: +44 1865 272595
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._