Problems trying to compile log4shib

[Jose Lopez Lujan]

Same thing using:

PATH=/usr/sfw/bin:/opt/SUNWspro/bin:/usr/bin:/usr/ccs/bin:/usr/dt/bin:/usr/
openwin/bin:/usr/sbin; export PATH

env     CC="gcc" \
        CXX="gcc" \
        CFLAGS="-m64 " \
        CXXFLAGS="-m64 " \
        LDFLAGS="-L/usr/sfw/lib/sparcv9 -R/usr/sfw/lib/sparcv9" \
        CCFLAGS=-m64 \
        INSTALL=/usr/ucb/install \
        ./configure \
        --disable-static \
        --prefix=/local/sfw \
        --disable-doxygen



gmake
.
.
.
:basic_istringstream<char, std::char_traits<char>, std::allocator<char>
>::basic_istringstream(std::basic_string<char, std::char_traits<char>,
>std::allocator<char> > const&,
>std::_Ios_Openmode)../src/.libs/liblog4shib.so
std::basic_ostream<char, std::char_traits<char> >& std::endl<char,
std::char_traits<char> >(std::basic_ostream<char, std::char_traits<char>
>&)testmain.o
std::basic_ostream<char, std::char_traits<char> >& std::ends<char,
std::char_traits<char> >(std::basic_ostream<char, std::char_traits<char>
>&)../src/.libs/liblog4shib.so
std::basic_ostream<char, std::char_traits<char> >& std::operator<< <char,
std::char_traits<char>, std::allocator<char> >(std::basic_ostream<char,
std::char_traits<char> >&, std::basic_string<char, std::char_traits<char>,
std::allocator<char> > const&)../src/.libs/liblog4shib.so
ld: fatal: symbol referencing errors. No output written to .libs/testmain
collect2: ld returned 1 exit status
gmake[1]: *** [testmain] Error 1
gmake[1]: Leaving directory `/local/sfw/src/log4shib-1.0.4/tests'
gmake: *** [all-recursive] Error 1

The fact is that if I don¹t use "-m64" flag for a 64-bit SPARC
architecture it will compile with any trouble but I'll have a 32-bit MSB
dynamic lib SPARC32PLUS:

# file liblog4shib.so.1.0.4
liblog4shib.so.1.0.4:   ELF 32-bit MSB dynamic lib SPARC32PLUS Version 1,
V8+ Required, dynamically linked, not stripped


Which is not the same as other components that I've successfully compiled
for a 64-bit SPARC architecture, so eventually I'll get either a
ELFCLASS64 or ELFCLASS32 error.

I've uploaded the full make logs and configure scripts
- 32bit: http://goo.gl/fzc0U
- 64bit SPARC: http://goo.gl/lyNl7


Best,
-
Jose Manuel Lopez Lujan
Senior LMS  Coordinator
University of Toronto





On 12-01-13 3:55 PM, "Jose Lopez Lujan" <jm.lopez at utoronto.ca> wrote:

>Hi Scott
>
>Unfortunately I have the same output using gmake:
>
>.
>.
>.
>Making all in tests
>gmake[1]: Entering directory `/local/sfw/src/log4shib-1.0.4/tests'
>/bin/bash ../libtool --mode=link CC -m64  -mt -D_REENTRANT  -m64 -xO2
>-L/usr/sfw/lib/sparcv9 -R/usr/sfw/lib/sparcv9 -o testmain  testmain.o
>../src/liblog4shib.la -lpthread -lnsl -lsocket
>CC -m64 -D_REENTRANT -m64 -xO2 -o .libs/testmain testmain.o  -mt
>-L/usr/sfw/lib/sparcv9 ../src/.libs/liblog4shib.so -lpthread -lnsl
>-lsocket
>ld: fatal: file ../src/.libs/liblog4shib.so: open failed: No such file or
>directory
>ld: fatal: file processing errors. No output written to .libs/testmain
>gmake[1]: *** [testmain] Error 1
>gmake[1]: Leaving directory `/local/sfw/src/log4shib-1.0.4/tests'
>gmake: *** [all-recursive] Error 1
>
>
>
>Best,
>-
>Jose Manuel Lopez Lujan
>Senior LMS  Coordinator
>University of Toronto
>
>
>
>
>
>On 12-01-13 3:47 PM, "Cantor, Scott" <cantor.2 at osu.edu> wrote:
>
>>On 1/13/12 3:39 PM, "Jose Lopez Lujan" <jm.lopez at utoronto.ca> wrote:
>>>
>>>
>>>I searched on the web but didn't find anything relevant so far. It seems
>>>that the script cannot have access to the file
>>>../src/.libs/liblog4shib.so. Looking into the directory, the file
>>>liblog4shib.so.1.0.4 doesn't exist which is the target of the soft link
>>>liblog4shib.so. And there you go, this is my dead-end, I haven't figured
>>>it out why's not compiling the liblog4shib.so.1.0.4 library.
>>
>>I would guess that the answer lies earlier in the build during the link
>>step and that there's a bug in the makefile or something that's causing
>>it
>>to continue with the build. I would guess as an initial possibility that
>>you might switch to GNU make if you're not using it. You'll need it later
>>anyway.
>>
>>-- Scott
>>
>>--
>>To unsubscribe from this list send an email to
>>users-unsubscribe at shibboleth.net
>
>--
>To unsubscribe from this list send an email to
>users-unsubscribe at shibboleth.net