Re: [rfc][00/37] Eliminate builtin_type_ macros

[Mark Kettenis <mark dot kettenis at xs4all dot nl>]

> Date: Sun, 31 Aug 2008 19:50:45 +0200
> From: uweigand@de.ibm.com
> 
> Hello,
> 
> one of the last remaining "implicit" uses of current_gdbarch is the
> builtin_type_ macros.  This patch set completely removes those macros.
> 
> I've attempted to not simply replace those types with explict uses of
> current_gdbarch, but instead choose the proper architecture to use:
> 
> - I've added a gdbarch pointer to struct expression, and used this
>   per-expression arch throughout expression parsing and evaluation.
> - I've converted some types (like builtin_void_type) back to be
>   platform-neutral, so they can be freely used throughout GDB.
> - In many cases, an existing per-frame or per-objfile arch was available
>   (or could be made available) and should be used.
> - Target-specific code should use target_gdbarch.
> - In a very small number of case, current_gdbarch remains for now.
> 
> Each of the following patches contains a more specific explanation.
> 
> Combination of all 37 patches tested on amd64-linux, s390-ibm-linux,
> s390x-ibm-linux, powerpc-linux, powerpc64-linux, and spu-elf with no
> regressions.  Also, compile-tested with --enable-targets=all (each 
> patch by itself).
> 
> I'd be interested in particular in feedback from the Ada maintainers
> on the Ada-related changes in this patch-set.
> 
> What do you think?

I've probably had one piwo too many at this point, but can we please
stop this Linux [x/zillion] crap?  You can't seriously pretend there
are really 37 independent diffs that people would want to review
and/or test can you?