Re: A proposed patch for undefined_symbol ()

[Ian Lance Taylor <ian at airs dot com>]

   Date: 1 Mar 2000 13:42:04 -0500
   From: Ian Lance Taylor <ian@zembu.com>

      Date: Wed, 1 Mar 2000 10:35:53 -0800
      From: "H . J . Lu" <hjl@valinux.com>

      > undefined symbols unless there is an option in the Linker to force a fatal error if
      > any undefined symbols remain  at

      My patch doesn't change the undefined symbol warning with -Bsymbolic
      when building DSO. What it does is not to abort if undefined symbol
      is ok.

      >  the end of the link. So I feel  we need to have a separate option in the Linker to
      > address this.
      > 

      We have --noinhibit-exec to ignore all errors. But I don't
      think it is appropriate for -Bsymbolic.

   I agree with Koundinya, though; sometimes, when using -Bsymbolic, you
   do want any undefined symbol to be an error.  On Solaris, I think this
   is some -z option.  On some other systems, I think it is a --assert
   option.  I think it's OK for the default to be a warning, but we might
   to make sure that people can still get an error if they want one.

Actually, sorry, I see we already have the --no-undefined option which
accomplishes what I want.  Never mind.  We should implement -z and
--assert to use that for compatibility, if appropriate.

Ian