https://gcc.gnu.org/bugzilla/show_bug.cgi?id=125051
--- Comment #6 from Andre Vehreschild <vehre at gcc dot gnu.org> --- Hi Paul, I have checked your changes, but still get regressions in libgomp: FAIL: libgomp.fortran/pr29629.f90 -O0 (internal compiler error: in gfc_trans_assignment_1, at fortran/trans-expr.cc:13307) FAIL: libgomp.fortran/pr29629.f90 -O0 (test for excess errors) FAIL: libgomp.fortran/pr29629.f90 -O1 (internal compiler error: in gfc_trans_assignment_1, at fortran/trans-expr.cc:13307) FAIL: libgomp.fortran/pr29629.f90 -O1 (test for excess errors) ... I am still confused why gfc_current_ns is not set to the function/subroutines ns when its body is translated from the beginning. From my point of view it makes sense to set it now, but I would rather learn why it was not done from the beginning (or possibly removed later). Any one any clue? - Andre
