https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69554
--- Comment #8 from Dominique d'Humieres <dominiq at lps dot ens.fr> --- Likely caused by revision r29884. > Please take this as a humble general suggestion: Fortran maintainers > should enforce during patch review that any new diagnostic has > a corresponding testcase triggering it in the testsuite. This seems a lot of work for big patches. Note that this is not specific to gfortran: any one committing a patch should look at the Intel results!
