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

Settle for /usr symlink (!) (was: Re: [kettenis@delius.kettenis.nl: Hurd bug fix]


Mark filed the following report (thanks, Mark!)

If you look at the report, you will notice that egcs will look in /include
in future.

I don't know if egcs could look in both, /include and /usr/include. I don't
know if this should be done if it is possible.

I just think that such situations will arise, and we should accept Hurd
philosophy here.

At some time we will have to write an amendment to Debian policy to clarify
differences for the Hurd port.

For now, I think I will only support /usr to be a symlink for now.
cross-install will not ask, as there is no option. People who don't use the
symlink will risk to have broken compilations and other stuff.

That also means I will definitely fix the remaining problem with ae package
(/bin/vi). The cpio package should be fixed (not in the archive, probably,
but the maintainer is aware of it).


On Sat, Mar 06, 1999 at 06:14:19PM +0100, Mark Kettenis wrote:
> I just filed the following bug report to the egcs folks.  Is the
> limits.h in the current Debian GNU/Hurd egcs package OK?

(yes, it is)

> ------- Start of forwarded message -------
> Date: 5 Mar 1999 23:32:28 +0100
> From: Mark Kettenis <kettenis@delius.kettenis.nl>
> To: egcs-bugs@egcs.cygnus.com
> Subject: Hurd bug fix
> This fix is necesary for proper `limits.h' generation under the Hurd.
> Mark
> Fri Mar  5 22:40:59 1999  Mark Kettenis  <kettenis@gnu.org>
> 	* config/t-gnu (SYSTEM_HEADER_DIR): New variable.  Set to
> 	`/include' in order to find the system's limits.h.
> - --- /home/kettenis/src/egcs-1.1.1/gcc/config/t-gnu	Mon Aug 11 17:57:15 1997
> +++ egcs-1.1.1/gcc/config/t-gnu	Fri Mar  5 20:22:44 1999
> @@ -1,5 +1,8 @@
> - -LIBGCC1=libgcc1.null
> - -CROSS_LIBGCC1=libgcc1.null
> +# In GNU, "/usr" is a four-letter word.
> +SYSTEM_HEADER_DIR = /include
> +
> +LIBGCC1 = libgcc1.null
> +CROSS_LIBGCC1 = libgcc1.null
>  # The pushl in CTOR initialization interferes with frame pointer elimination.
> ------- End of forwarded message -------

`Rhubarb is no Egyptian god.' Debian http://www.debian.org   finger brinkmd@ 
Marcus Brinkmann              GNU    http://www.gnu.org     master.debian.org
Marcus.Brinkmann@ruhr-uni-bochum.de                        for public  PGP Key
http://homepage.ruhr-uni-bochum.de/Marcus.Brinkmann/       PGP Key ID 36E7CD09

Reply to: