This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: egcs-199805171100UTC snapshot on m68k-next-nextstep3: strangebehaviour in compiling libiberty ...


  In message <9805171401.AA26942@moene.indiv.nluug.nl> I wrote:

  > Does anyone understand why I get:
  >
  > test x"yes" != xyes || \
  >   gcc -c -DHAVE_CONFIG_H -g -O2 -I.
  > -I../../egcs/libiberty/../include  -fpic
  > ../../egcs/libiberty/cplus-dem.c -o pic/cplus-dem.o
  > /usr/tmp/cc016648.s:2281:"instruction/operands mismatch" --
  > Statement 'jbsr _demangle_template_template_parm,a1' ignored
  > make[1]: *** [cplus-dem.o] Error 1
  > make[1]: Leaving directory
  > `/Users/toon/Unix/compilers/snapshots/obj/libiberty'
  > make: *** [all-libiberty] Error 2

Jeff,

>  Did you configure with --enable-shared?  It looks like
>  it's trying to build a PIC libiberty.

Well, I most certainly did not (if you think a Fortran user has any  
use for shared libraries - like in reducing the memory footprint of  
the entire system by sharing code - think again:  I'm just  
interested in running one big executable at a time :-)

No, I didn't specify --enable-shared.  However, it sure looks like  
the build of libiberty _assumes_ that I did (quoting  
obj-dir/libiberty/Makefile):

# Warning: this fragment is automatically generated
# Following fragment copied from  
../../egcs/libiberty/config/../../config/mh-m68kpic
PICFLAG=-fpic
enable_shared = yes

but I didn't do it, I wasn't there, etc ...

Cheers,
Toon.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]