RFC: Removing OLD_EXCEPTIONS powerpc multilib variants

[Till Straumann]

I don't understand the need for a toolchain variant
for the exception model in the first place -- what
different features does OLD_EXCEPTIONS trigger in
libgcc/newlib/... ??

I believe this to be pure bloat anyways. It only would
matter if you intend to create RTEMS libcpu variants
- one for the old and one for the 'new' exception model.

-- Till

Ralf Corsepius wrote:
> Hi,
> 
> Till's question has caused me to look into the powerpc multilibs, again.
> 
> In RTEMS-CVS/trunk I see that only 2 cpu-variants still apply old-
> exception-processing:
> 
> dmv177.cfg:CPU_CFLAGS = -mcpu=603e -D_OLD_EXCEPTIONS -Dppc603e
> gen405.cfg:CPU_CFLAGS = -mcpu=403 -D_OLD_EXCEPTIONS -Dppc405
> helas403.cfg:CPU_CFLAGS = -mcpu=403 -D_OLD_EXCEPTIONS -Dppc403
> ppcn_60x.cfg:CPU_CFLAGS = -mcpu=603e -D_OLD_EXCEPTIONS -Dppc603e
> score603e.cfg:CPU_CFLAGS = -mcpu=603e -D_OLD_EXCEPTIONS -Dppc603e
> 
> Current GCC-4.0 however still supports these old-exception-processing
> based multilib-variants:
> m403/roe
> m601/roe
> m601/roe/nof
> m602/roe
> m602/roe/nof
> m603/roe
> m603/roe/nof
> m603e/roe
> m603e/roe/nof
> 
> In my understanding, this means, all but
> m403/roe
> m603e/roe
> old-exception-processing multilib variants can be removed [1].
> 
> Comments, opinions.
> 
> If you should maintain/use an old exception-based BSP not using one of
> these 2 variants, which is not in RTEMS main source-tree speak up NOW or
> it will be too late :-)
> 
> Ralf
> 
> [1] The gen405 config above looks suspicious to me -mcpu=405 would be
> different ISA than -mcpu=403. Comments?
> I am inclined to think a m405/roe multilib is missing.
> 
> 
>