FWIW: this is on Mac 10.10.2, using clang and clang++

OMPI master - bunch of warnings, but it successfully builds:
In file included from ./libevent/event-internal.h:38:
./libevent/minheap-internal.h:64:6: warning: unused function 'min_heap_ctor' 
[-Wunused-function]
void min_heap_ctor(min_heap_t* s) { s->p = 0; s->n = 0; s->a = 0; }
     ^
./libevent/minheap-internal.h:65:6: warning: unused function 'min_heap_dtor' 
[-Wunused-function]
void min_heap_dtor(min_heap_t* s) { if (s->p) mm_free(s->p); }
     ^
./libevent/minheap-internal.h:66:6: warning: unused function 
'min_heap_elem_init' [-Wunused-function]
void min_heap_elem_init(struct event* e) { e->ev_timeout_pos.min_heap_idx = -1; 
}
     ^
./libevent/minheap-internal.h:67:5: warning: unused function 'min_heap_empty' 
[-Wunused-function]
int min_heap_empty(min_heap_t* s) { return 0u == s->n; }
    ^
./libevent/minheap-internal.h:68:10: warning: unused function 'min_heap_size' 
[-Wunused-function]
unsigned min_heap_size(min_heap_t* s) { return s->n; }
         ^
./libevent/minheap-internal.h:69:15: warning: unused function 'min_heap_top' 
[-Wunused-function]
struct event* min_heap_top(min_heap_t* s) { return s->n ? *s->p : 0; }
              ^
./libevent/minheap-internal.h:71:5: warning: unused function 'min_heap_push' 
[-Wunused-function]
int min_heap_push(min_heap_t* s, struct event* e)
    ^
./libevent/minheap-internal.h:79:15: warning: unused function 'min_heap_pop' 
[-Wunused-function]
struct event* min_heap_pop(min_heap_t* s)
              ^
./libevent/minheap-internal.h:91:5: warning: unused function 
'min_heap_elt_is_top' [-Wunused-function]
int min_heap_elt_is_top(const struct event *e)
    ^
./libevent/minheap-internal.h:96:5: warning: unused function 'min_heap_erase' 
[-Wunused-function]
int min_heap_erase(min_heap_t* s, struct event* e)
    ^
osc_sm_passive_target.c:161:16: warning: variable 'ret' is used uninitialized 
whenever 'if' condition is false [-Wsometimes-uninitialized]
    } else if (module->outstanding_locks[target] == lock_shared) {
               ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
osc_sm_passive_target.c:167:12: note: uninitialized use occurs here
    return ret;
           ^~~
osc_sm_passive_target.c:161:12: note: remove the 'if' if its condition is 
always true
    } else if (module->outstanding_locks[target] == lock_shared) {
           ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
osc_sm_passive_target.c:148:12: note: initialize the variable 'ret' to silence 
this warning
    int ret;
           ^
            = 0
1 warning generated.
pml_cm_recv.c:85:6: warning: no previous prototype for function 
'mca_pml_cm_recv_fast_completion' [-Wmissing-prototypes]
void mca_pml_cm_recv_fast_completion(struct mca_mtl_request_t *mtl_request)
     ^


OMPI v1.8 fails for me at an earlier stage:

make
Making all in config
make[1]: Nothing to be done for `all'.
Making all in contrib
make[1]: Nothing to be done for `all'.
Making all in opal
Making all in include
/Applications/Xcode.app/Contents/Developer/usr/bin/make  all-am
make[3]: Nothing to be done for `all-am'.
Making all in libltdl
CDPATH="${ZSH_VERSION+.}:" && cd . && /bin/sh 
/Users/rhc/openmpi/ompi-release/config/missing aclocal-1.15 -I ../../config
aclocal-1.15: error: ../../config/autogen_found_items.m4:304: file 
'opal/mca/backtrace/configure.m4' does not exist
make[2]: *** [aclocal.m4] Error 1
make[1]: *** [all-recursive] Error 1
make: *** [all-recursive] Error 1

I believe we have seen this before, and it was an issue caused by a change in 
libtool itself. The Mac automatically updated to the new version, which 
triggers the problem.

This is why we embedded ltdl directly into the OMPI master, and is why the 
master is building while the 1.8 branch is not.

I don’t have a ready fix at hand - might have to wait for Jeff to return next 
week to advise.


> On Apr 6, 2015, at 2:41 PM, Paul Hargrove <phhargr...@lbl.gov> wrote:
> 
> On MacOS X 10.8 (where cc and c++ are clang and clang++, and the default ABI 
> is LP64) I've configured the release candidate with
> --prefix=.... --enable-debug CC=cc CXX=c++
> 
> 
> The build fails linking opal_wrapper as shown in the "make V=1" output below.
> 
> Based on the undefined symbol (_lt_libltdlc_LTX_preloaded_symbols) I am 
> guessing this is a libltdl issue.  If relevant, note that configure picks the 
> embedded libltdl:
> 
> *** GNU libltdl setup
> checking location of libltdl... internal copy
> configure: OMPI configuring in opal/libltdl
> 
> Oddly enough, I have no problems with a build configured for the ILP32 ABI as 
> follows:
> --prefix=.... --enable-debug CC=cc CXX=c++ \
> CFLAGS=-m32 --with-wrapper-cflags=-m32 \
> CXXFLAGS=-m32 --with-wrapper-cxxflags=-m32 \
> FCFLAGS=-m32 --with-wrapper-fcflags=-m32
> 
> I can send (off-list) the config.log and configure's complete stdout for both 
> the LP64 and ILP32 builds if requested.
> 
> -Paul
> 
> 
> /bin/sh ../../../libtool  --tag=CC   --mode=link cc  -g -finline-functions 
> -fno-strict-aliasing   -export-dynamic  -Wl,-flat_namespace  -o opal_wrapper 
> opal_wrapper.o ../../../opal/libopen-pal.la <http://libopen-pal.la/> 
> libtool: link: cc -g -finline-functions -fno-strict-aliasing 
> -Wl,-flat_namespace -o .libs/opal_wrapper opal_wrapper.o  
> ../../../opal/.libs/libopen-pal.dylib -lm
> Undefined symbols for architecture x86_64:
>   "_lt_libltdlc_LTX_preloaded_symbols", referenced from:
>       import-atom in libopen-pal.dylib
> ld: symbol(s) not found for architecture x86_64
> clang: error: linker command failed with exit code 1 (use -v to see 
> invocation)
> make[2]: *** [opal_wrapper] Error 1
> make[1]: *** [all-recursive] Error 1
> make: *** [all-recursive] Error 1
> 
> 
> -- 
> Paul H. Hargrove                          phhargr...@lbl.gov 
> <mailto:phhargr...@lbl.gov>
> Computer Languages & Systems Software (CLaSS) Group
> Computer Science Department               Tel: +1-510-495-2352
> Lawrence Berkeley National Laboratory     Fax: +1-510-486-6900
> _______________________________________________
> devel mailing list
> de...@open-mpi.org
> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel
> Link to this post: 
> http://www.open-mpi.org/community/lists/devel/2015/04/17184.php

Reply via email to