Re: default CPU target for ix86 based ports
On Thu, Aug 07, 2003 at 06:43:35AM +0200, Jan-Benedict Glaw wrote:
> On Wed, 2003-08-06 23:08:22 +0200, Matthias Klose <doko@cs.tu-berlin.de>
> wrote in message <[🔎] 16177.28230.634867.251269@gargle.gargle.HOWL>:
> > Jan-Benedict Glaw writes:
> > > i386 seems to die, sun4m also does have servere problems... Where does
> > > this lead to? All these seem to arise from doing optimization which
> > > hasn't been proved to (really) make things better... Everything I see is
> > > that it's breaking stuff.
> >
> > the ix86 change was for _compatibility_ reasons, not for
> > _performance_.
>
> I know. Compatibility. To whom? Compatibility to allow someone to copy a
> eg. SuSE C++ binary over to a debian box. Why may it crash (or work in
> some undefined way)? Because it was optimized for i486+, or did I get
> the whole thing wrong?
>
> Am I wrong or did we, "forced" because we wanted to be binary compatible
> to some major distributions, just follow others and doing optimization
> just as they did?
>
> See, I'm not ranting over this one special "bug" introduced into
> libstdc++5, I'm not ranting on hwmath on sparc (esp. in kernel:), but
> I'm ranting on the common movement to optimize a case and break another.
>
> MfG, JBG
Suggest you read the list archives before raising this discussion
again. It had nothing at all to do with optimization, either ours or
theirs.
--
Daniel Jacobowitz
MontaVista Software Debian GNU/Linux Developer
Reply to: