I dont't understand what libstdc++ does with gcc 2.7.2.3. this one should use libg++ if I remember well. Are you sure that you're not mixing egcs loader with gcc 2.7 compiler ? I doubt this will work. Cheers Damir ===================================================================== | Damir Buskulic | Universite de Savoie/LAPP | | | Chemin de Bellevue, B.P. 110 | | Tel : +33 (0)450091600 | F-74941 Annecy-le-Vieux Cedex | | e-mail: buskulic@lapp.in2p3.fr | FRANCE | ===================================================================== mailto:buskulic@lapp.in2p3.fr On 8 Jan 1999 lijowski@cosray2.wustl.edu wrote: > > I attempted to recompile the ROOT 2.20/06 source on Solaris 2.5 using > gcc 2.7.2.3 with no luck. > > Below is a Makefile, I have been using for this purpose. > > # Makefile of ROOT for Solaris with gcc > > PLATFORM = solaris > > CXX = g++ > CC = gcc > CXXFLAGS = -fPIC -I/usr/openwin/include > CFLAGS = -fPIC -I/usr/openwin/include > CINTCXXFLAGS = -fPIC -DG__REGEXP1 -DG__UNIX -DG__OSFDLL \ > -DG__SHAREDLIB -DG__ROOT -DG__REDIRECTIO > CINTCFLAGS = -fPIC -DG__REGEXP1 -DG__UNIX -DG__OSFDLL \ > -DG__SHAREDLIB -DG__ROOT -DG__REDIRECTIO > OPT = -g > NOOPT = > LD = $(CXX) > LDFLAGS = $(OPT) > SOFLAGS = -shared > SOEXT = so > SYSLIBS = -lm -ldl > SYSXLIBS = -L/usr/openwin/lib -lX11 -lm -lgen -ldl -lsocket > XLIBS = $(ROOTSYS)/lib/libXpm.a -L/usr/openwin/lib -lX11 -lsunmath > CILIBS = -lm -ltermcap -ldl -L/usr/ccs/lib -lgen > > > ##### MACROS and TARGETS ##### > > include Make-macros > > > ##### DEPENDENCIES ##### > > include Make-depend > > The linking > > g++ -shared -g -o /data/cern/root/lib/libCint.so CINT_G__setup.o CINT_g__cfunc.o CINT_memtest.o CINT_sizeof.o CINT_auxu.o CINT_gcoll.o CINT_new.o CINT_stdstrct.o CINT_cast.o CINT_global1.o CINT_newlink.o CINT_struct.o CINT_debug.o CINT_global2.o CINT_newsos.o CINT_stub.o CINT_decl.o CINT_ifunc.o CINT_oldlink.o CINT_disp.o CINT_inherit.o CINT_opr.o CINT_tmplt.o CINT_init.o CINT_parse.o CINT_typedef.o CINT_dump.o CINT_input.o CINT_pause.o CINT_val2a.o CINT_end.o CINT_intrpt.o CINT_pcode.o CINT_value.o CINT_error.o CINT_loadfile.o CINT_pragma.o CINT_var.o CINT_expr.o CINT_quote.o CINT_fread.o CINT_macro.o CINT_scrupto.o CINT_func.o CINT_malloc.o CINT_shl.o CINT_Api.o CINT_Class.o CINT_Token.o CINT_Apiif.o CINT_DataMbr.o CINT_Type.o CINT_BaseCls.o CINT_Method.o CINT_Typedf.o CINT_CallFunc.o CINT_MethodAr.o CINT_libstrm.o > > produces a huge number of errors like this > > Text relocation remains referenced > against symbol offset in file > _IO_do_write 0x58 /usr/local/lib/libstdc++.a(fileops.o) > _IO_do_write 0x4b4 /usr/local/lib/libstdc++.a(stdiostream.o) > _IO_do_write 0x68c /usr/local/lib/libstdc++.a(fileops.o) > _IO_do_write 0x36c /usr/local/lib/libstdc++.a(stdiostream.o) > _IO_do_write 0xc5c /usr/local/lib/libstdc++.a(fileops.o) > _IO_do_write 0xe8 /usr/local/lib/libstdc++.a(filebuf.o) > _IO_do_write 0x5f8 /usr/local/lib/libstdc++.a(fileops.o) > _IO_do_write 0x100 /usr/local/lib/libstdc++.a(fileops.o) > _IO_do_write 0x64c /usr/local/lib/libstdc++.a(fileops.o) > _IO_do_write 0x5d0 /usr/local/lib/libstdc++.a(fileops.o) > _IO_file_fopen 0x264 /usr/local/lib/libstdc++.a(filebuf.o) > _IO_stdin_buf 0x20 /usr/local/lib/libstdc++.a(stdstreams.o) > _IO_stdin_buf 0x24 /usr/local/lib/libstdc++.a(stdstreams.o) > free 0x95c /usr/local/lib/libstdc++.a(genops.o) > free 0xef4 /usr/local/lib/libstdc++.a(genops.o) > free 0x568 /usr/local/lib/libstdc++.a(genops.o) > free 0x9a0 /usr/local/lib/libstdc++.a(genops.o) > free 0x1f8 /usr/local/lib/libstdc++.a(genops.o) > free 0x84 /usr/local/lib/libstdc++.a(floatconv.o) > free 0x2e4 /usr/local/lib/libstdc++.a(genops.o) > _vt.11fstreambase.3ios 0x10b8 /usr/local/lib/libstdc++.a(fstream.o) > > At the end there are following messages from stderr > > ld: fatal: relocations remain against allocatable but non-writable sections > chmod: /data/cern/root/lib/libCint.so: No such file or directory > ld: fatal: library -lCint: not found > ld: fatal: File processing errors. No output written to /data/cern/root/bin/cint > gmake: *** [/data/cern/root/bin/cint] Error 1 > > I appreciate any hints. > > Michal Lijowski > > >
This archive was generated by hypermail 2b29 : Tue Jan 04 2000 - 00:43:28 MET