Re: [DynInst_API:] undefined reference to `dlopen' error during parseThat build

[Wei Ming Khoo <weimzz@xxxxxxxxx>]

Yup, that problem is resolved for me.

Just one other thing - This other error may have been brought up before.

Linking test1_18F.mutatee_solo_dynamic_gfortran_32_none_none

gfortran -g -o test1_18F.mutatee_solo_dynamic_gfortran_32_none_none test1_18F_mutatee_solo_gcc_32_none_none.o mutatee_util_gcc_32_none_none.o test1_18F_fortran_gfortran_32_none_none.o mutatee_driver_solo_gcc_32.o    

mutatee_driver_solo_gcc_32.o: In function `main':

/home/wmk26/dyninst-a17e876/testsuite/i386-unknown-linux2.4/../src/mutatee_driver.c:259: multiple definition of `main'

test1_18F_fortran_gfortran_32_none_none.o:/home/wmk26/dyninst-a17e876/testsuite/i386-unknown-linux2.4/../src/dyninst/test1_18F_fortran.F:51: first defined here

collect2: ld returned 1 exit status

make[1]: *** [test1_18F.mutatee_solo_dynamic_gfortran_32_none_none] Error 1

make[1]: Leaving directory `/home/wmk26/dyninst-a17e876/testsuite/i386-unknown-linux2.4'

make: *** [testsuite] Error 2

~/dyninst-a17e876$

My fix was to leave out one of the object files:

gfortran -g -o test1_18F.mutatee_solo_dynamic_gfortran_32_none_none test1_18F_mutatee_solo_gcc_32_none_none.o mutatee_util_gcc_32_none_none.o test1_18F_fortran_gfortran_32_none_none.o

Same for test1_18F.mutatee_solo_dynamic_gfortran_32_pic_none

--wm

On Wed, Nov 21, 2012 at 9:34 PM, Andrew Bernat <bernat@xxxxxxxxxxx> wrote:

I remember what happened. I needed to make this fix on a demo machine and we haven't reintegrated yet. 

If you have a chance, pull down the latest v8.0.1; URL is here: snapshot. It should work out of the box. 

Drew

On Nov 21, 2012, at 11:16 AM, Wei Ming Khoo <weimzz@xxxxxxxxx> wrote:

So e.g. in proccontrol/*/Makefile, LDFLAGS should include -ldl.


- FChE

Yup this worked.

Regards,
--wm

--

Andrew Bernat

Paradyn Project

bernat@xxxxxxxxxxx

http://www.cs.wisc.edu/~bernat