https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91039
--- Comment #2 from urbanjost at comcast dot net --- After posting this for comment on the Fortran newsgroup I realize this is not technically a bug, but I would like it to remain as an enhancement request that this type of bug be reported when "...-fbounds-check" is used on the compile. urba...@comcast.net <urba...@comcast.net> schrieb: > I expect the following program to at a minimum fail at run time (and possibly > during compile) where indicated. The program has a bug (as you know), but the error in quesion, a bounds violation, is not a constraint. So, the compiler is free to diagrnose this or not. You could argue that it would be better if a compiler caught this, and I would agree, but it's not a compiler bug.
