GCC 4.1.1 m68k-rtems generating invalid code for -m5307 proce ssor ?

Kirspel, Kevin kevin.kirspel at optimedical.com
Tue Jun 26 14:21:57 UTC 2007


I am using this as a learning experience. At first I just assumed that when
you built the m68k-rtems target that it would generate support for all m68k
processors that gcc supported. After going through this exercise, I learned
a little about cross-compiling gcc.

-----Original Message-----
From: Ralf Corsepius [mailto:ralf.corsepius at rtems.org] 
Sent: Tuesday, June 26, 2007 10:14 AM
To: Kirspel, Kevin
Cc: 'rtems-users at rtems.org'
Subject: RE: GCC 4.1.1 m68k-rtems generating invalid code for -m5307 proce
ssor ?

On Tue, 2007-06-26 at 08:42 -0400, Kirspel, Kevin wrote:
> True.

And what is the reason for you not to do so?


> -----Original Message-----
> From: Ralf Corsepius [mailto:ralf.corsepius at rtems.org] 
> Sent: Tuesday, June 26, 2007 8:41 AM
> To: Kirspel, Kevin
> Cc: 'Chris Johns'; Ian Caddy; 'rtems-users at rtems.org'
> Subject: RE: GCC 4.1.1 m68k-rtems generating invalid code for -m5307 proce
> ssor ?
> 
> On Tue, 2007-06-26 at 08:26 -0400, Kirspel, Kevin wrote:
> > The issue is that when I compile the rtems toolchains for m68k, they do
> not
> > generate the libraries for the m5307 processor. I am currently trying to
> see
> > how I can get these libraries generated.  
> Why do you need such library? Wanting to have one for each and every
> variant of a cpu would be pretty silly and inapplicable.
> 
> You'd normally want to chose a "cpu variant" from a set of cpu-variants
> a toolchain supports that "works reasonably" for you, not the one that
> "squeezes out the max" from a cpu.




More information about the users mailing list