[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Problem building from 13027:b9a89ca0fb75
From: |
Tatsuro MATSUOKA |
Subject: |
Re: Problem building from 13027:b9a89ca0fb75 |
Date: |
Tue, 30 Aug 2011 10:18:49 +0900 (JST) |
Hello
This is not also answer.
As far as I remember, the license of the METIS is not compatible to that of
octave so that octave binary with SuiteSparse linked with the METIS cannot be
distributed.
http://octave.1599824.n4.nabble.com/Metis-td1647536.html
So we have to build the SuiteSparse (chlmod) built with -DNPARTITION flag for
the binary distribution.
(For self build, the METIS can be used.)
Regards
Tatsuro
--- On Tue, 2011/8/30, Ben Abbott <address@hidden> wrote:
> I tried to build with tip ...
>
> $ hg tip
> changeset: 13027:b9a89ca0fb75
> tag: tip
> user: John W. Eaton <address@hidden>
> date: Mon Aug 29 13:07:22 2011 -0400
> summary: prevent optimization functions from setting ans in workspace at
> startup
>
> ... and encountered the error below when linking liboctave.
>
> libtool: link: /opt/local/bin/g++-mp-4.4 -dynamiclib -o
> .libs/liboctave.0.dylib [... snip ...]
> ld: warning: directory not found for option
> '-L/opt/local/var/macports/build/_opt_local_var_macports_sources_rsync.macports.org_release_tarballs_ports_lang_gcc44/gcc44/work/build/x86_64-apple-darwin11/libstdc++-v3/src'
> ld: warning: directory not found for option
> '-L/opt/local/var/macports/build/_opt_local_var_macports_sources_rsync.macports.org_release_tarballs_ports_lang_gcc44/gcc44/work/build/x86_64-apple-darwin11/libstdc++-v3/src/.libs'
> Undefined symbols for architecture x86_64:
> "_METIS_NodeComputeSeparator", referenced from:
> _cholmod_metis_bisector in libcholmod.a(cholmod_metis.o)
> "_METIS_NodeND", referenced from:
> _cholmod_metis in libcholmod.a(cholmod_metis.o)
> ld: symbol(s) not found for architecture x86_64
>
> After reverting to 13021 (hg revert --rev 13021 --all), I am able to build
> again.
>
> The "METIS" stuff is due to MacPort's SuiteSparse package I'm linking to.
>
> Is anyone else seeing a problem like this?
>
> In the meantime, I'll try to bisect to isolate the changeset that introduced
> the problem for me.
>
> Ben
>
>
>