[m-dev.] /usr/bin/ld: Undefined symbols:_MR_global_pointer _GC_register_dynamic_libraries ___data_start __end _data_start

[Fergus Henderson]

On 01-May-2001, Terrence Brannon <princepawn at earthlink.net> wrote:
> Fergus Henderson writes:
>  > `MR_global_pointer' is supposed to be defined in runtime/mercury_dummy.o,
>  > which should get linked into runtime/libmer_rt.a.
>  > Try `nm' on runtime/mercury_dummy.o.
> 
> Ok, I took a close look at Boehm's changelog and it appears that the
> change to support os x occurred in 5.0. So I got that release and took
> a look at gcconfig.h, put #define macosx at the top and then make test
> worked perfectly.
> 
> so then I copied dyn_load.c and gcconfig.h from the 5.0 distribution
> into the Mercury tree and made it through the compilation of boehm_gc
> directory.

Hmm, OK... but there were some changes to the Mac OS X support in
gc6.0alpha2.  I've merged in gc6.0alpha6 and committed that to the
Mercury cvs repository now.  So you might want to try that.

However, I don't think that has anything to do with your other problem:

> However, the final link is still not finding MR_global_pointer
> 
> nm on libmr_rt.a yields:
> 
> libmer_rt.a(mercury_dummy.o):
> 00000000 T _MR_dummy_function_call
> 00000004 T _MR_dummy_identify_function
> 00000004 C _MR_global_pointer
> 00000004 C _MR_global_pointer_2
> 
> nm on mercury_dummy.o yields:
> 
> 00000000 T _MR_dummy_function_call
> 00000004 T _MR_dummy_identify_function
> 00000004 C _MR_global_pointer
> 00000004 C _MR_global_pointer_2

They both look correct.

> so I don't know how to fix this final problem.

Nor do I.

It might be a linker bug.

Is the libmer_rt.a file being linked in?
You could try putting

	EXTRA_MLFLAGS=-v

or
	EXTRA_MLFLAGS=-v -- -v

in a file named Mmake.params in the top-level mercury directory --
that should give you a bit more verbose information about exactly
how the linker is being invoked.

-- 
Fergus Henderson <fjh at cs.mu.oz.au>  |  "I have always known that the pursuit
                                    |  of excellence is a lethal habit"
WWW: <http://www.cs.mu.oz.au/~fjh>  |     -- the last words of T. S. Garp.
--------------------------------------------------------------------------
mercury-developers mailing list
Post messages to:       mercury-developers at cs.mu.oz.au
Administrative Queries: owner-mercury-developers at cs.mu.oz.au
Subscriptions:          mercury-developers-request at cs.mu.oz.au
--------------------------------------------------------------------------