Re: off-trunk gdb.fortran/dwarf-stride.exp no longer PASSes [Re: [PATCH 0/2] fort_dyn_array: Enable basic Fortran dynamic array support]

[Joel Brobecker <brobecker@adacore.com>]

> Apparently the VLA patchset is not yet checked-in completely, for example the
> off-trunk patchset formerly contained:
> 
>    attr = dwarf2_attr (die, DW_AT_string_length, cu);
>    if (attr)
>      {
> -      length = DW_UNSND (attr);
> +      if (attr_form_is_block (attr))
> +        {
> 
> but current FSF GDB HEAD does not yet recognize attr_form_is_block for
> DW_AT_string_length at all so it just cannot work yet.
> 
> I will have to apply only remaining parts of the former off-trunk patchset but
> that is sure off-topic for upstream GDB.

Just to be sure - did we regress on something in the master branch?
Or is this just a regression on your branch when you only rely on
the changes that were made on master?

The changes that were first submitted early on in the process were
too big, and I suspect some of them were also OBE, so I asked that
the patches be submitted piecemeal, to allow me to understand why
each change was made. This allowed us to eventually get a good chunk
of the work in, but as you can see, some chunks appear to still be
missing...

-- 
Joel