New AIX libstdc++ testsuite regression 20_util/allocator_members

[Mark Mitchell]

> I'm not sure this really should be catagorized as a "high" priority
> regression then, since this underlying behavior has been an issue for
> 3.0.x compilers on AIX as well. (We just didn't know about it.)

This may be true, but we have to be a little bit careful with this way
of thinking.

The compiler's never "broken" or "not broken" -- it's always somewhat
broken.  We want to keep it from getting more broken; hopefully, we're
making it less broken.

I don't mind punting if we have to, but if we can do something we
should.  I'll try to figure out why you're seeing different test results
than me.  (I didn't remove the linker map bits, so that's likely it --
but I'm not sure why that would affect the static link.)

--
Mark Mitchell                   mark@codesourcery.com
CodeSourcery, LLC               http://www.codesourcery.com