[tree-ssa] gimp vs expand_end_case_type

[Richard Henderson]

On Wed, Sep 03, 2003 at 02:25:33PM -0400, Jason Merrill wrote:
> I would argue that the problem is the C frontend giving an incorrect value
> for TYPE_MAX_VALUE of enums.  The C++ frontend uses the actual range of
> values an object of the type can have; your testcase demonstrates why the C
> practice of only describing the range of named enums is wrong.

I think you're right.  I'll look into this after we get things
back to something approximating stable.  :-/


r~