Hi Tobias, This looks good to me - OK for master.
Thanks for the patch Paul On Wed, 30 Sep 2020 at 09:59, Tobias Burnus <tob...@codesourcery.com> wrote: > The non-contiguous had both check false positive and false > negative results. Some more refinements > are surely possible, but hopefully there are no longer > false positives. > > I also now used this check for pointer assignments where the > LHS pointer has the contiguous attribute. > > In the non-contiguous-check function: > - for 'dt(i)%array' it returned true due to dt(i) but that's > an element, which is contiguous. > - ref_size (which is a size) is compared with 'arr_size' calculated > via dep_difference,, which returns upper-lower but array size is > (upper-lower)+1. > - fixed a memory leak. > > OK? > > Tobias > > ----------------- > Mentor Graphics (Deutschland) GmbH, Arnulfstraße 201, 80634 München / > Germany > Registergericht München HRB 106955, Geschäftsführer: Thomas Heurung, > Alexander Walter > -- "If you can't explain it simply, you don't understand it well enough" - Albert Einstein
