Re: experimental gcc-2.95.3 (20001222)
- To: Matthias Klose <doko@cs.tu-berlin.de>
- Cc: debian-toolchain@lists.debian.org, bvollins@debian.org, chris@debian.org, cts@debian.org, pb@debian.org
- Subject: Re: experimental gcc-2.95.3 (20001222)
- From: Daniel Jacobowitz <dan@debian.org>
- Date: Sun, 24 Dec 2000 01:38:01 -0500
- Message-id: <[🔎] 20001224013801.A17280@drow.them.org>
- Mail-followup-to: Matthias Klose <doko@cs.tu-berlin.de>, debian-toolchain@lists.debian.org, bvollins@debian.org, chris@debian.org, cts@debian.org, pb@debian.org
- In-reply-to: <[🔎] 14915.43663.613741.476734@gargle.gargle.HOWL>; from doko@cs.tu-berlin.de on Fri, Dec 22, 2000 at 08:25:03PM +0100
- References: <[🔎] 14915.43663.613741.476734@gargle.gargle.HOWL>
On Fri, Dec 22, 2000 at 08:25:03PM +0100, Matthias Klose wrote:
> ... can be found at http://master.debian.org/~doko/gcc. This includes
> Bernd's 20001221 patch kit. The package builds fine for i386, on m68k
> the build fails in libchill according to Christian. The patches for
> arm and powerpc do not apply cleanly anymore. I didn't check for alpha
> and sparc. If I have a wish free for Christmas, then please send me
> updated patches ;-)
I'll try - I'm behind a very slow modem right now, though.
> The packages now installed versioned driver names (gcc-2.95). The
> plain driver name (gcc) is provided by an alternative. Not sure if
> this will stay like this or if the driver names should be hardcoded as
> symbolic links and for each architecture it should be decided which
> compiler version to use as the default version.
>
> advantages for alternatives are:
> - switching between compiler versions is easy.
>
> disadvantages are:
> - you don't know which compiler was used, when looking for bug reports
PLEASE hardcode it! That's a pretty stiff disadvantage.
Dan
/--------------------------------\ /--------------------------------\
| Daniel Jacobowitz |__| SCS Class of 2002 |
| Debian GNU/Linux Developer __ Carnegie Mellon University |
| dan@debian.org | | dmj+@andrew.cmu.edu |
\--------------------------------/ \--------------------------------/
Reply to: