This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
RE: egcs-1.0.1 configuration bugs
- To: "'law at cygnus dot com'" <law at cygnus dot com>
- Subject: RE: egcs-1.0.1 configuration bugs
- From: John Hughes <john at AtlanTech dot COM>
- Date: Mon, 26 Jan 1998 09:49:33 +0100
- Cc: "'egcs-bugs at cygnus dot com'" <egcs-bugs at cygnus dot com>
Jeffrey A Law <law@hurl.cygnus.com> Writes:
> In message <01BD27F6.3A254100@malvinas.AtlanTech.COM>you write:
> > 1. In egcs-1.0.1/configure.in I see:
> > mips*-*-* | alpha*-*-osf* | i[3456]86*-*-sysv4* | i[3456]86*-*-unixware*)
> > with_stabs=yes;
> > but the SVR4 and UnixWare "as" can't digest .stabs
> Someone how knows something about those systems will need to respond; I don't
> know enough about them myself.
Well, I can tell you that "as" doesn't know a .stab from a .punch. The debugging
format on SVR4 and children is ELF/DWARF
> > 2. there also seems to be a bug in egcs-1.0.1/gcc/configure:
> > This seems to set "stabs" to "yes" if "--with-stabs=no" is given (the
> > "withval" variable is ignored).
> I suspect you're supposed to use --with-stabs or --without-stabs -- ie they
> are not supposed to have a yes/no flag attached with them.
You may suspect that, but egcs-1.0.1/gcc/configure doesn't, from earlier in the
script:
-without-* | --without-*)
ac_package=`echo $ac_option|sed -e 's/-*without-//'`
# Reject names that are not valid shell variable names.
if test -n "`echo $ac_package| sed 's/[-a-zA-Z0-9_]//g'`"; then
{ echo "configure: error: $ac_package: invalid package name" 1>&2; exit 1;
}
fi
ac_package=`echo $ac_package| sed 's/-/_/g'`
eval "with_${ac_package}=no" ;;
So there is NO difference between "--with-xyzzy=no" and "--without-xyzzy"
And neither of them work.
(Appologies for any pre-coffee growling).
--
John Hughes <john@AtlanTech.COM>,
Atlantic Technologies Inc. Tel: +33-1-43204546
24 rue Montbrun, Fax: +33-1-43204579
75014 PARIS.