This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Patch to configure.in
- To: law at cygnus dot com
- Subject: Re: Patch to configure.in
- From: Manfred Hollstein <manfred at s-direktnet dot de>
- Date: Wed, 29 Jul 1998 08:55:10 +0200 (MET DST)
- Cc: jason at cygnus dot com, egcs-patches at cygnus dot com
- References: <13757.35411.776751.410883@slsvhmt> <5760.901695058@hurl.cygnus.com>
- Reply-To: manfred at s-direktnet dot de, Manfred dot Hollstein at ks dot sel dot alcatel dot de
On Wed, 29 July 1998, 00:50:58, law@hurl.cygnus.com wrote:
>
> In message <13757.35411.776751.410883@slsvhmt>you write:
> > As gxx-include-dir is used e.g. in the libstdc++ and libio directories
> > as well, I suggest moving the check to the toplevel configure script
> > or to add a check there, too. Perhaps we should also add relative path
> > names as "bad values".
> Agreed on adding a check to the toplevel for gxx-include-dir. Can
> you submit a patch for that?
Yes, I'll do.
>
> We probably don't want to add relative names as "bad" since we supposedly
> fix them a while back and it would be nice if they worked :-)
But, how should they work? Being relative would mean, they're
distinctive depending on where you're calling the compiler. Relative
path names inside the build directory should be OK, but not for
directories to be used for installation, shouldn't they?
>
> Thanks,
> jeff
manfred