https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79165
--- Comment #21 from David Malcolm <dmalcolm at gcc dot gnu.org> ---
(In reply to David Malcolm from comment #19)
[...]
> In particular, repeated accesses to the same source line are more expensive
> that they could be: each time it tries to count forwards from the last
> line_record entry, rather than reusing the line information it got last time
> (in input.c:read_line_num). I'll try to fix it.
aermod.f90
With the bad linemap:
get_next_line is 16% of overhead
Without the bad linemap:
get_next_line is 0.4% of overhead
tramp3d without the bad linemap, with -Wall shows get_next_line is only 0.35%
of overhead
Not sure how much of a win this would be, and it doesn't seem like gcc 7
material at this stage.
