mc 4.5.6




I generated succesfully mc 4.5.6 with egcs 1.1.1 on the
following platforms: 

sparc-sun-solaris2.5.1
sparc-sun-solaris2.6
i686-pc-linux-gnulibc1
i686-pc-linux-gnu
alpha-dec-osf4.0b
hppa1.1-hp-hpux10.20
powerpc-ibm-aix4.3.2.0
mips-sgi-irix6.5

All generation processes were OK. The Makefile seems to have some
nested rules which always make mc/gmc/corba-gmc to re-link at each
make. That is not easy to see if make was correct after all.

- on i686-pc-linux-gnu

 : :ln -s ../edit/libedit.a .
 : :gcc -L/usr/local/lib -o mc dir.o util.o screen.o dialog.o key.o keyxdef.o menu.o file.o win.o color.o help.o find.o profile.o user.o view.o ext.o mouse.o setup.o dlg.o option.o tree.o widget.o chmod.o mad.o wtools.o info.o cons.handler.o chown.o subshell.o terms.o boxes.o hotlist.o achown.o layout.o fsusage.o mountlist.o  regex.o complete.o slint.o command.o cmd.o main.o panelize.o learn.o listmode.o utilunix.o background.o rxvt.o text.o popt.o findme.o poptparse.o poptconfig.o popthelp.o filegui.o filenot.o -L../vfs -L../slang -L../edit -lvfs-mc -lslang -ledit -lgpm  -lintl  -L/usr/local/lib -lglib -lext2fs -lcom_err     
 :*:/usr/local/lib/libcom_err.a(com_err.o): In function `com_err':
 :*:com_err.o(.text+0x170): multiple definition of `com_err'
 :*:../vfs/libvfs-mc.a(undelfs.o)(.text+0x4d4): first defined here
 :*:/usr/bin/ld: Warning: size of symbol `com_err' changed from 211 to 32 in com_err.o
 :*:util.o: In function `get_current_wd':
 :*:util.o(.text+0x1120): the `getwd' function is dangerous and should not be used.
 :*:collect2: ld returned 1 exit status
 :*:gmake[1]: *** [mc] Error 1
 : :gmake[1]: Leaving directory `/scratch/happi/GNU.DESK/mc-4.5.6/src'
 :*:gmake: *** [all] Error 1
 :happi: Entering wait for 11848

  I made util.c use getcwd instead of getwd. 

  Then there was a problem with -lcom_err. I just suppressed it from
  the Make.common file and generation was OK.


- on mips-sgi-irix6.5, the link of gnome/gmc fails as the libraries
  are before the *.o. Solution put $(LDFLAGS) after the .o files in
  the Makefile. The same is true for corba-gmc.
  Then generation ends up succesfully.

- on powerpc-ibm-aix4.3.2.0
: :gcc -c -I..  -I/happi/GNU.DESK/mc-4.5.6/intl -I./..  -I./../vfs -I./../slang -I.. -DBINDIR=\""/usr/local/bin/"\" -DLIBDIR=\""/usr/local/lib/mc/"\" -DICONDIR=\""/usr/local/share/pixmaps/mc/"\"  -DLOCALEDIR=\""/usr/local/share/locale/"\"  -DHAVE_CONFIG_H -O2 -I/usr/local/lib/glib/include -I/usr/local/include fish.c
 :*:/usr/include/arpa/inet.h:152: warning: `struct ether_addr' declared inside parameter list
 :*:/usr/include/arpa/inet.h:152: warning: its scope is only this definition or declaration,
 :*:/usr/include/arpa/inet.h:152: warning: which is probably not what you want.
 :*:/usr/include/arpa/inet.h:153: warning: `struct ether_addr' declared inside parameter list
 :*:/usr/include/arpa/inet.h:154: warning: `struct ether_addr' declared inside parameter list
 :*:/usr/include/arpa/inet.h:159: warning: `struct sockaddr_dl' declared inside parameter list
 :*:/usr/include/arpa/inet.h:160: warning: `struct sockaddr_dl' declared inside parameter list
 :*:/usr/include/arpa/inet.h:161: warning: `struct sockaddr_dl' declared inside parameter list
 :*:fish.c: In function `dir_load':
 :*:fish.c:403: storage size of `tim' isn't known
 :*:fish.c: In function `linear_start':
 :*:fish.c:497: warning: assignment makes pointer from integer without a cast
 : :gmake[1]: Leaving directory `/happi/GNU.DESK/mc-4.5.6/vfs'
 :*:gmake[1]: *** [fish.o] Error 1
 :*:gmake: *** [all] Error 1

  #include <sys/time.h> is missing in vfs.h, solution -> increase domain
  of the #if to include AIX.

  The flag -Wl,-bbigtoc is needed later (big table of contents).

  All is then OK.

Amicalement.
Philippe.
-- 
  ************************************************************************
  *  Philippe Defert: Computing and Networks Division                    *
  *                   CERN,  European Laboratory for Particle Physics    *
  ************************************************************************
  * URL: http://wwwcn.cern.ch/~defert | E*mail: Philippe.Defert@cern.ch  *
  ************************************************************************
  *      Un monde nouveau, tu comprends                          ////\   *
  *      Rien ne sera plus jamais comme avant                    \\\//   *
  *      C'est la fin de l'histoire, le rouge apres le noir       | |    *
  *                                                  J.J. Goldman | |    *  
  ************************************************************************



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]