Hi Riccardo, > Am 12.03.2015 um 14:28 schrieb Riccardo Mottola <[email protected]>: > > Riccardo Mottola wrote: >> Hi Dago, >> >> Dagobert Michelsen wrote: >>>> in config.log I just see: >>>> configure:18550: checking toolchain supports unwind section type >>>> configure:18579: result: no >>>> configure:18589: checking whether .eh_frame section should be read-only >>>> configure:18609: result: no >>>> >>>> no trace of the "expr" error and of the actual test. >>> Take a look in config.log, if in doubt add „set -e“ to configure and >>> store the output. >> I pasted the excerpt of config.log above. >> >> I added set -e (It took me some time to understand I had to add it to >> the script itself).
You must add it to the top of „configure“.
>> I attach it here together with the config.log, see if you get smart
>> out of it! I don't
>
> anybody has a better clue about these two configure checks? I think they
> might be the culprit for the amd64 failure. I don't... I see nothing in the
> logs that helps.
>
> I fear "expr" could be a problem that a test is run with a non-GNU tool and
> it fails, but about unwind... I don't know and it could be the problem.
Look here:
> configure:6217: checking for objdump
> configure:6247: result: no
The missing check for the non-availability of objdump leads to the output
being empty and finally to „expr + 1“ which is invalid:
> + printf '%s\n' 'configure:18591: checking whether .eh_frame section should
> be read-only'
> + printf %s 'checking whether .eh_frame section should be read-only... '
> checking whether .eh_frame section should be read-only... + false
> + libffi_cv_ro_eh_frame=no
> + echo 'extern void foo (void); void bar (void) { foo (); foo (); }'
> + /opt/csw/bin/gcc-4.9 -O2 -pipe -m64 -march=opteron -Wall -fexceptions -c
> -fpic -fexceptions -o conftest.o conftest.c
> + objdump -h conftest.o
> ++ grep -n eh_frame conftest.dump
> ++ cut -d: -f 1
> + libffi_eh_frame_line=
> ++ expr + 1
> expr: syntax error
> + libffi_test_line=p
> + sed -n p conftest.dump
> + grep READONLY conftest.line
> + rm -f conftest.c conftest.dump conftest.line conftest.o
> + printf '%s\n' 'configure:18611: result: no'
> + printf '%s\n' no
> no
I committed an adjusted PATH during configure in r24719. However, the
compilation
still fails with
> libtool: link: /opt/csw/bin/gcc-4.9 -shared -fPIC -DPIC -Wl,-z -Wl,text
> -Wl,-h -Wl,libffi.so.6 -o .libs/libffi.so.6.0.4 src/.libs/prep_cif.o
> src/.libs/types.o src/.libs/raw_api.o src/.libs/java_raw_api.o
> src/.libs/closures.o src/x86/.libs/ffi64.o src/x86/.libs/unix64.o
> src/x86/.libs/ffi.o src/x86/.libs/sysv.o -L/opt/csw/lib/64 -O2 -m64
> -march=opteron -m64 -march=opteron -m64 -march=opteron
> ld: fatal: file src/x86/.libs/unix64.o: section [5].eh_frame: section type is
> SHT_PROGBITS: expected SHT_AMD64_UNWIND
> collect2: error: ld returned 1 exit status
Probably you need to dive deeper in ELF and linking.
Best regards
— Dago
--
"You don't become great by trying to be great, you become great by wanting to
do something,
and then doing it so hard that you become great in the process." - xkcd #896
smime.p7s
Description: S/MIME cryptographic signature
