[Fwd: Unsupported targets slated for removal]

[Ralf Corsepius]

On Wed, 2005-03-23 at 09:02 -0600, Joel Sherrill  wrote:
> Hi,
> 
> Just back from time off for a sick kid and some spring holiday.
> 
> I saw this on the binutils list.  It includes some RTEMS targets
> so I wanted to pass it along.   These are the RTEMS targets on
> the list:
> 
> 
>      h8300-*-rtems*
>      h8500-*-rtems*
>      i960-*-rtems*
>      m68k-*-rtemscoff*
>      sh-*-rtemscoff*
> 
> My thoughts:
> 
> h8300 -- already removed from gdb.
Grumble ...

I don't know about its production status, but from the toolchains' POV
this is one of the most interesting target.

I presume, this is why they want to kill it ;)
It trips each and every design flaw everywhere.

>   Is it time to schedule
>    obsoletion in RTEMS?

> h8500 -- port never really existed
> i960 -- already dead
> m68k-rtemscoff -- m68k long ago switched to ELF, so OK
These are OK with me.

> sh-rtemscoff -- default is now elf.  Historically COFF was
>    more stable than ELF for sh but we are already switched
>    to ELF for sh-rtems so this shouldn't be an issue.
Hmm, this will probably kill the sh1 and sh2, otherwise killing sh-
rtemscoff is OK with me.

> Any comments from anyone out there?
I increasingly feel GCC and GDB are increasingly playing it nice to the
"big ones" and are gradually abandoning the embedded CPUs.

A very unfortunate tendency ... which should not be in our interest.

> >     openrisc-*-*
> >     or32-*-*
You forgot this one - It's the #1 on my personal deprecation list.

Ralf