https://gcc.gnu.org/bugzilla/show_bug.cgi?id=117763
--- Comment #13 from Jürgen Reuter <juergen.reuter at desy dot de> --- (In reply to Paul Thomas from comment #12) > Created attachment 59694 [details] > Fix for this PR > > Hi Juergen and Harald, > > Mea culpa once again! @Harald: Your pin pointing with the .diff allowed me > to pinpoint the offending patch immediately. Thanks. > > @Juergen: Are you in a position to apply the patch and to test the full code? > Yes, I can, the full patch, or just the lines below? I will do it right after I provide the reproducer for the ICE issue. > The attached is a partial reversion of the patch for PR109345. > > - else if (sym && sym->ts.type == BT_CLASS && sym->attr.dummy) > + else if (sym && sym->ts.type == BT_CLASS > + && expr->ref->type == REF_COMPONENT > + && expr->ref->next->type == REF_ARRAY > + && expr->ref->next->next == NULL > + && CLASS_DATA (sym)->attr.dimension) > > Is what does the job. The rest of the patch is a bit of a tidy up and the > first bit is a "just in case..." > > I am regression testing once again and should be in a position to push at > lunchtime. > > Regards > > Paul
