[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: 2 non-libtool libs want to start
From: |
Ross Boylan |
Subject: |
Re: 2 non-libtool libs want to start |
Date: |
Sun, 19 Dec 2004 16:50:53 -0800 |
On Sat, 2004-12-11 at 07:58, Peter O'Gorman wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Ross Boylan wrote:
>
> | I verified with nm that both the R and boost test libraries define
> | (not just refer to) the symbol _main.
>
> Hmm, that is a little strange, I just built R (the fink package for 2.0.1)
> and don't have a _main function defined in or referenced from the R framework.
On further digging, the problem is specific to R 1.9 (specifically, I
looked at 1.9.1). The link line for the library includes all the files
under system/, and that means system.c, which has main, is included.
R 2.0 library doesn't have main on any of the systems I've checked.
>
> | I'm stunned it works at all. For example, the documentation on the
> | loader on OS X seems to imply it will barf on doubly defined symbols.
> This totally depends on how the library was linked.
>
>
> | On linux, when I changed the order of the libs, the boost test suite
> | took control and ran all the tests I had specified. That was the
> | behavior I wanted.
>
> This should also work on darwin/Mac OS X, the linker will (normally) use the
> first definition it sees. Please let me know how to reproduce your setup and
> I'll try to have something by year-end.
Let me know if you still want to pursue this and I can help. For me,
going to 2.0 looks as if it will solve the problem. For you, perhaps
the oddball linking problem this raises is still of interest.
Ross