Re: [RTL, ColdFire 24/63] Add support for a MODE_INDEX_REG_CLASS macro

[Mark Mitchell <mark at codesourcery dot com>]

Bernd Schmidt wrote:
> Ian Lance Taylor wrote:
>> Richard Sandiford <richard@codesourcery.com> writes:
>>
>>> @defmac MODE_CODE_BASE_REG_CLASS (@var{mode}, @var{outer_code},
>>> @var{index_code})
>>
>> Oh yeah, MODE_CODE_BASE_REG_CLASS came in here, written and committed
>> by Bernd:
>>     http://gcc.gnu.org/ml/gcc-patches/2005-11/msg01952.html
> 
> FWIW, I think that MODE_INDEX_REG_CLASS is a straightforward
> generalization of what we have now, but I don't want to approve the
> patch over Jeff's head while he has concerns.

I'm not quite sure if I know exactly whose court the ball is in at this
point.  If I understand correctly, Jeff is still unhappy with the patch,
and Richard still thinks it's the right thing.

So, how do we move forward so that we can get this bug fixed, and wrap
up this series of ColdFire patches?  Jeff, do have a specific
alternative in mind?  If I understand correctly, one of the arguments in
favor of this patch is that it's symmetric with some of the other macros
-- with which you're also unhappy.  Perhaps, then, it would be OK to put
this in (arguably compounding the damage) with the theory that if we
want to make some change later, we can do so wholesale without a whole
lot more work than changing the existing bits?

Thanks,

-- 
Mark Mitchell
CodeSourcery
mark@codesourcery.com
(650) 331-3385 x713