Date: Mon, 27 Feb 2006 18:19:50 +0100 From: Alexander Leidinger <Alexander@Leidinger.net> To: Boris Samorodov <bsam@ipt.ru> Cc: Daniel O'Connor <doconnor@gsoft.com.au>, freebsd-emulation@freebsd.org Subject: Re: linux_base and libreadline.so symlink (was: Re: Matlab) Message-ID: <20060227181950.079e44f3@Magellan.Leidinger.net> In-Reply-To: <16934979_-_@srv.sem.ipt.ru> References: <20060227111759.GC19451@math.jussieu.fr> <4402E4E1.1010208@andric.com> <20060227121512.GA23561@math.jussieu.fr> <200602272329.47792.doconnor@gsoft.com.au> <16934979_-_@srv.sem.ipt.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
Am Mon, 27 Feb 2006 16:26:20 +0300
schrieb Boris Samorodov <bsam@ipt.ru>:
> > Unfortunately it does not just ignore the fact it's a FreeBSD binary - it
> > blows up.
>
> > You can try adding a manual symlink - I have libreadline.so.4 from linux_base
> > but not libreadline.so, ie I suggest you do this..
> > cd /compat/linux/usr/lib
> > ln -s libreadline.so.4 libreadline.so
>
> Why didn't linux distribution do this symlink?
> Should our linux_base port do it (as well as for libtermcap.so and
> maybe others)?
The dynamic linker should have references to specific versions of a
library. The symlinks are typically only needed when compiling (more
correctly: linking) a program.
It may be the case that matlab opens the lib via dlopen(3), but I would
expect that other libs are dlopen'ed then too, not only libreadline.
I suggest to ktrace(1) or truss(1) matlab to determine what it tries to
do.
Bye,
Alexander.
--
"One world, one web, one program" -- Microsoft promotional ad
"Ein Volk, ein Reich, ein Fuehrer" -- Adolf Hitler
http://www.Leidinger.net Alexander @ Leidinger.net
GPG fingerprint = C518 BC70 E67F 143F BE91 3365 79E2 9C60 B006 3FE7
WL http://www.amazon.de/exec/obidos/registry/1FZ4DTHQE9PQ8/ref=wl_em_to/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060227181950.079e44f3>