https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110626
--- Comment #20 from Christopher Albert <albert at tugraz dot at> --- (In reply to Paul Thomas from comment #19) > (In reply to Christopher Albert from comment #18) > > First with your patch plus my patchlet, then your patch alone. > > pault@fedora:~/prs/pdt$ ~/gfc/bin/gfortran -fdump-tree-original > ~/prs/junk.f90;./a.out > non-allocatable > finalize 42 > finalize -1 > assign 0 7 > x 7 y 8 > allocatable > finalize 42 > finalize 7 > assign 0 7 > x 7 y 8 > pault@fedora:~/prs/pdt$ ~/gfc/bin/gfortran -fdump-tree-original > ~/prs/junk.f90;./a.out > non-allocatable > finalize 42 > finalize -1 > assign 0 7 > x 7 y 8 > allocatable > finalize 42 > x 7 y 7 > > As you are aware, Christopher, until the gcc steering committee has done its > thing, we cannot use LLM generated patches. > > I am going to write to Jason Merrill to see if, as an interim, we can use > the fix but not use testcases generated by Claude. I will argue that, if we > do that, the chance that there will be copyright breaches is negligible. > > Cheers > > Paul I see... I was not aware that patch submissions are actively halted now until this decision. This means effectively I could now block gcc development by filing valid patches assisted by Claude that you will not be allowed to pick up so you have to invent entirely different solutions (jk). So until this is settled I will better stop posting.
