[mpich-discuss] mpich2-1.0.8 OpenSolaris

[Rajeev Thakur]

The --enable-sharedlibs option in MPICH2 was not working on Solaris the last
time I looked and was mentioned in the RELEASE_NOTES file. Does it work
without --enable-sharedlibs?

Rajeev

> -----Original Message-----
> From: mpich-discuss-bounces at mcs.anl.gov 
> [mailto:mpich-discuss-bounces at mcs.anl.gov] On Behalf Of Jim Langston
> Sent: Wednesday, February 04, 2009 2:26 PM
> To: mpich-discuss at mcs.anl.gov
> Subject: [mpich-discuss] mpich2-1.0.8 OpenSolaris
> 
> Hi All,
> 
> I'm trying to build 1.0.8 for OpenSolaris (x86) using
> Sun Studio 12, everything builds ok - but when
> I try to build an example:
> 
> langston at alpha:~/MPICH/mpich2-1.0.8/examples$ rm cpi.o
> langston at alpha:~/MPICH/mpich2-1.0.8/examples$ gmake cpi
> ../bin/mpicc -I../src/include -I../src/include -c cpi.c
> ../bin/mpicc   -o cpi cpi.o  -lm
> Undefined            first referenced
>  symbol                  in file
> MPIDU_Process_unlock                
> /export/home/langston/MPICH/mpich2-1.0.8/lib/libmpich.so
> MPIDU_Process_lock                  
> /export/home/langston/MPICH/mpich2-1.0.8/lib/libmpich.so
> MPIDU_Process_lock_init             
> /export/home/langston/MPICH/mpich2-1.0.8/lib/libmpich.so
> ld: fatal: Symbol referencing errors. No output written to cpi
> gmake: *** [cpi] Error 1
> 
> 
> My configuration line is:
> 
> langston at alpha:~/MPICH/mpich2-1.0.8$ head config.log
> This file contains any messages produced by compilers while
> running configure, to aid debugging if configure makes a mistake.
> 
> It was created by configure, which was
> generated by GNU Autoconf 2.62.  Invocation command line was
> 
>   $ ./configure --prefix=/usr/local/mpich2 --enable-threads 
> --with-thread-package=posix --enable-sharedlibs=solaris-x86
> 
> ## --------- ##
> ## Platform. ##
> 
> 
> 
> I added in solaris-x86 into the configure file to recognize I'm on an 
> OpenSolaris system which
> I've included.
> 
> Looking at libmpich.so
> 
> langston at alpha:~/MPICH/mpich2-1.0.8/lib$ nm libmpich.so | grep 
> MPIDU_Process_unlock
>          U MPIDU_Process_unlock
> 
> and
> 
> langston at alpha:~/MPICH/mpich2-1.0.8/src/mpid/common/locks$ nm 
> mpidu_atomic_primitives.o
> 00000000 b Bbss.bss
> 00000000 d Ddata.data
> 00000000 r Drodata.rodata
> 00000000 D emulation_lock
>          U MPID_Abort
> 00000024 T MPIDU_Atomic_add_emulated
> 00000210 T MPIDU_Atomic_cas_aint_emulated
> 00000164 T MPIDU_Atomic_cas_int_emulated
> 000000bc T MPIDU_Atomic_cas_int_ptr_emulated
> 000002bc T MPIDU_Atomic_decr_and_test_emulated
> 00000364 T MPIDU_Atomic_decr_emulated
> 000003f8 T MPIDU_Atomic_fetch_and_add_emulated
> 0000049c T MPIDU_Atomic_fetch_and_decr_emulated
> 0000053c T MPIDU_Atomic_fetch_and_incr_emulated
> 000005e0 T MPIDU_Atomic_incr_emulated
> 000007c8 T MPIDU_Atomic_swap_aint_emulated
> 00000720 T MPIDU_Atomic_swap_int_emulated
> 00000678 T MPIDU_Atomic_swap_int_ptr_emulated
> 00000000 T MPIDU_Interprocess_lock_init
>          U MPIDU_Process_lock
>          U MPIDU_Process_lock_init
>          U MPIDU_Process_unlock
>          U MPIU_Internal_error_printf
> 
> 
> 
> Any thoughts ? hints ?
> 
> thanks,
> 
> Jim
> 
> ///////////////////////////////
> 
> 
> 
> 
> -- 
> /////////////////////////////////////////////
> 
> Jim Langston
> Sun Microsystems, Inc.
> 
> (877) 854-5583 (AccessLine)
> (513) 702-4741 (Cell)
> AIM: jl9594
> jim.langston at sun.com
> 
>