Actually,

        I got curious and checked it myself, and to my surprise the $31 had 
disappeared , so i searched a bit and realized that I need to add escape 
characters to $ because of m4, and so I did:

                    OMPI_GCC_INLINE_ASSIGN='"bis [$]31,[$]31,%0" : "=&r"(ret)'


both to alpha-*)  and  alphaev6-*)

                but now, although configure reports that gcc supports inline 
assembly:

checking if gcc supports GCC inline assembly... yes   <----------
checking if gcc supports DEC inline assembly... no
checking if gcc supports XLC inline assembly... no
checking if g++ supports GCC inline assembly... yes <----------

                it says that it cannot generate atomic operations, since it 
lacks a 
ALPHA.asm:

checking for pre-built assembly file... no (not in asm-data)
checking whether possible to generate assembly file... failed
configure: WARNING: Could not build atomic operations assembly file.
configure: WARNING: There will be no atomic operations for this build.
checking for atomic assembly filename... none

        but it continues the configure process. 
        I start compiling and I get another error:


al_object.c  -fPIC -DPIC -o class/.libs/opal_object.o
opal_object.s: Assembler messages:
opal_object.s:445: Error: syntax error
opal_object.s:446: Error: bad expression
opal_object.s:446: Error: syntax error
opal_object.s:447: Error: bad expression
opal_object.s:447: Error: syntax error
opal_object.s:448: Error: bad expression
opal_object.s:448: Error: syntax error
opal_object.s:449: Error: syntax error
opal_object.s:450: Error: bad expression
opal_object.s:450: Error: syntax error
opal_object.s:451: Error: junk at end of line, first unrecognized character is 
`/'
opal_object.s:452: Error: bad expression
opal_object.s:452: Error: syntax error
opal_object.s:522: Error: syntax error
opal_object.s:523: Error: bad expression
opal_object.s:523: Error: syntax error
opal_object.s:524: Error: bad expression
opal_object.s:524: Error: syntax error
opal_object.s:525: Error: bad expression
opal_object.s:525: Error: syntax error
opal_object.s:526: Error: syntax error
opal_object.s:527: Error: bad expression
opal_object.s:527: Error: syntax error
opal_object.s:528: Error: junk at end of line, first unrecognized character is 
`/'
opal_object.s:529: Error: bad expression
opal_object.s:529: Error: syntax error
make[2]: *** [class/opal_object.lo] Error 1
make[2]: Leaving directory `/usr/src/mpi/openmpi-1.1.1/opal'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/usr/src/mpi/openmpi-1.1.1/opal'
make: *** [all-recursive] Error 1
(...)


        , but beware that I'm using the gcc flag '-save-temps' to check where 
the 
problem was:

        make CFLAGS=-save-temps

        so I started searching where that macro might come from and ended up in
        opal/include/opal/sys/alpha/atomic.h
        I took out the extra //  comments in the asm code from
         'opal_atomic_cmpset_32' and opal_atomic_cmpset_64'  which seemed to 
fix that 
problem and restarted the compilation and after some more compiling it bombed 
out in:

(...)
then mv -f "$depbase.Tpo" "$depbase.Plo"; else rm -f "$depbase.Tpo"; exit 1; 
fi
gcc -DHAVE_CONFIG_H -I. -I. -I../opal/include -I../orte/include 
-I../ompi/include -I../ompi/include -I.. -save-temps -MT 
class/opal_object.lo -MD -MP -MF class/.deps/opal_object.Tpo -c 
class/opal_object.c  -fPIC -DPIC -o class/.libs/opal_object.o
opal_object.s: Assembler messages:
opal_object.s:734: Error: can't resolve `.rodata' {.rodata section} - `L0' 
{.text section}
opal_object.s:734: Error: can't resolve `.text' {.text section} - `L0' 
{.rodata section}
make[2]: *** [class/opal_object.lo] Error 1
make[2]: Leaving directory `/usr/src/mpi/openmpi-1.1.1/opal'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/usr/src/mpi/openmpi-1.1.1/opal'
make: *** [all-recursive] Error 1
(...)

        and here I got stuck :)

        I send my config.log in the attachment, with just the bis [$]31 
modification

                                        Regards,
                                                                                
Nuno




On Sunday 17 September 2006 19:01, Brian Barrett wrote:
> > (...)
> > gcc -O3 -DNDEBUG -fno-strict-aliasing -pthread -o .libs/opal_wrapper
> > opal_wrapper.o -Wl,--export-dynamic  ../../../opal/.libs/libopal.so
> > -ldl -lnsl -lutil -lm -Wl,--rpath -Wl,/opt/openmpi-1.1.1/lib
> > ../../../opal/.libs/libopal.so: undefined reference to
> > `opal_atomic_cmpset_acq_32'
> > ../../../opal/.libs/libopal.so: undefined reference to
> > `opal_atomic_cmpset_32'
> > (...)
>
> Can you send the config.log file generated by Open MPI's configure,
> with your bis $31,$31 change?

-- 
http://aeminium.org/slug/

*****************************************************************************
**                                                                         **
** WARNING:  This email contains an attachment of a very suspicious type.  **
** You are urged NOT to open this attachment unless you are absolutely     **
** sure it is legitimate.  Opening this attachment may cause irreparable    **
** damage to your computer and your files.  If you have any questions      **
** about the validity of this message, PLEASE SEEK HELP BEFORE OPENING IT. **
**                                                                         **
** This warning was added by the IU Computer Science Dept. mail scanner.   **
*****************************************************************************

<<attachment: bis-alpha-config.log.zip>>

Reply via email to