Fortran regressions

[Thomas Koenig]

Hi Paul,

> We have 91 fortran regressions at present with only 7 of them
> assigned. I am about to make two disappear and will be working on the
> other three carrying my tag.
> 
> Any volunteers given that this is the season for regressions?

I will also be looking at some regressions. One problem is that some
of these are so plain weird (like a combination of an allocatable
class coarray with associate) that I would not even know where
to start looking.

With others, I am not even sure what the correct way of handling things
would be - if we are looking at an ICE on valid or an ICE on invalild,
or vaguely related test cases some of which may be valid and some
of which may be invalid.

Finally, many of these have a subject like "ICE in
<some_function_name>", which (while correct) are not really
informative which area of the compiler this is concerns.

So yes, I'm working on this, but I find it heavy going.

Regards

	Thomas