Re: [Patch, fortran] PR34640 - ICE when assigning item of a derived-component to a pointer

[Paul Richard Thomas <paul.richard.thomas@gmail.com>]

Dear All,

We are not quite there yet. Thanks to the posters to the thread on
clf, I have now been convinced that ptr => der_type_array%comp must
return an lbound=1 based descriptor, since the target is not a WHOLE
ARRAY as defined in the standard. In addition, Dominique picked up a
failing testcase with -m32 for which he provided a fix.

I should re-re-re-submit tonight :-)

Paul


On 9 July 2017 at 22:27, Thomas Koenig <tkoenig@netcologne.de> wrote:
> Hi Paul,
>
>> Please find attached what I believe is the final version of the patch.
>
>
> This is looking very good, this does work as advertised and has
> all the corner cases coverd.
>
> I also regtested this on powerpc64-unknown-linux-gnu.
>
>> I took the opportunity of the delay, while the bounds issue was being
>> discussed on clf, to fix class pointer arrays. They now function
>> correctly, as evidenced by pointer_array_8.f90.
>
>
> Excellent!
>
>> A possible final tweak - as asked before, should I bump up the module
>> version number? My inclination is to say that we should.
>
>
> Because we are changing the array descriptor (and thus binary
> compatibility), we have to change the library version in
> libtool-version. Bumping the module version should be done,
> too.
>
>> Bootstrapped and regtested on FC23/x86_64 - OK for trunk?
>
>
> So, OK from my side, with a bump in library and module version.
>
> Maybe you could wait a couple of days before committing to give
> others a chance to also test the patch.
>
> Thanks a lot for finally making gfortran F95-compliant!
>
> Regards
>
>         Thomas
>



-- 
"If you can't explain it simply, you don't understand it well enough"
- Albert Einstein