Re: [MiNT] MiNT lib 55

["Guido Flohr" <gufl0000@stud.uni-sb.de>]

On Mon, May 15, 2000 at 07:38:39AM +0300, Martin-Éric Racine wrote:
> On 2000-5-14, Guido Flohr <gufl0000@stud.uni-sb.de> wrote:
> 
> > > > > > I doubt that you will gain a lot by only giving -m68020-60.  
> > > 
> > > Well, Frank's specs transform that into 020-040 plus FPU.
> > 
> > And the implied -m68881 is meaningless for the kernel, see below.
> 
> There is a life beyond library and kernel building.  It's called
> applications.

This is really getting a nuisance ...
_You_ were talking about kernel building (although the subject of this
thread suggests something different).

As it comes to applications, simply read the gcc docs, and you will learn
that -m68020-60 benefits from most (if not all
relevant) additional features of all processors >= 68020.  It also enables
code generation for m68881, including those instructions that are emulated
on the 68040.  Read gcc.info, Invoking GCC->Submodel Options->M680x0
Options.  You will notice that a lot of the statements made here that you
care to contradict are quoted verbatim from this source which is certainly
trustworthy when it comes to gcc optimization.

> > And you have been taught countless times that the kernel
> > doesn't do any floating point arithmetics.
> 
> Spell:   A P P L I C A T I O N S.  
> 
> There have been at least 3 threads where I state that an SSH
> binary without FPU is significantly slower.

Then compile and link it with -m68020-60 or -m68881.

> > Anyway, you shouldn't make such a statement if you don't even
> > know the effect of the most common optimizations like
> > -fomit-frame-pointer.
> 
> If it really was such a common optimization, I would have ran
> into it ages ago.  

I couldn't have proven any better that you still have a lot to learn about
software engineering.
-- 
http://www.stud.uni-saarland.de/
Send your spam to president@whitehouse.gov and your replies to
mailto:guido at freemint dot de