This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
m68k bfd slow (was: Re: Cleanups for the m68k backend)
- From: Hans-Peter Nilsson <hp at bitrange dot com>
- To: Gunther Nikl <gni at gecko dot de>
- Cc: Marc Espie <espie at quatramaran dot ens dot fr>, <bernie at develer dot com>, <binutils at sources dot redhat dot com>
- Date: Fri, 11 Jul 2003 15:34:55 -0400 (EDT)
- Subject: m68k bfd slow (was: Re: Cleanups for the m68k backend)
(List changed from gcc@gcc.gnu.org to an appropriate one.)
On Fri, 11 Jul 2003, Gunther Nikl wrote:
> On Wed, Jul 09, 2003 at 09:59:03AM -0400, Hans-Peter Nilsson wrote:
> > On Wed, 9 Jul 2003, Gunther Nikl wrote:
> > > On Wed, Jul 09, 2003 at 12:37:09AM +0200, Marc Espie wrote:
> > > > We also have toolchain speed issues. Switching to a brand new toolchain
> > > > (and a slow one) is a hard decision, if it means having build time jump
> > > > from 3 days to one week on a m68k machine.
> > >
> > > Thats what I have observed too: the overhead of BFD-based binutils
> > > is very large :-/ For an already slow platform compared to current
> > > standards thats a pain.
> >
> > Do you mean overhead to build or that recent gas and ld are
> > slow?
>
> Slow building of binutils I could live with since that I would only rarely
> do. No, I meant BFD-based binutils are slow compared with non-BFD versions.
> However, I noticed this with a now outdated BFD-binutils version (2.9?).
> Building time eg. of GCC doubled when using as/ld from 2.9. I reverted to
> my old toolchain. Maybe a newer binutils version is faster than the old
> version that I used but I doubt that. Newer version means more features
> but often it also means slower.
Can you check with a recent version, like 2.14? Maybe there's a
bottleneck that's easy to fix. (Wishful thinking, but hey...)
brgds, H-P