[PATCH] score: PR1607: Add and use CPU_SIZEOF_POINTER

[Ralf Corsepius]

On 11/04/2012 09:41 PM, Joel Sherrill wrote:
> I am pretty sure the h8300 pointer size varies by mulitilib variant.
Yes - It is one of targets, where this applies.

> I am not at home to run checks.

Sebastian's CPU_SIZEOF_POINTER suffers from the same flaws as your 
__RTEMS_SIZEOF_VOID_P__.

They vary between targets and can vary between multilib variants of a 
targets.

I.e. using any of these defines in a global per-target header does not work.

=> This patch is not OK with me.

Ralf