https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122137
--- Comment #7 from Paul Thomas <pault at gcc dot gnu.org> --- (In reply to Harald Anlauf from comment #4) > https://gcc.gnu.org/pipermail/gcc-testresults/2026-May/876351.html > > We also fail on pdt_77.f03. > > Indeed, with -fsanitize=address,undefined I get: > > % gfc-17 -g pdt_77.f03 -fsanitize=address,undefined -O && ./a.out > > Program received signal SIGBUS: Access to an undefined portion of a memory > object. > > Backtrace for this error: > #0 0x7f8b0de2b655 in ??? > #1 0x7f8b0de2a645 in ??? > #2 0x7f8b0d44708f in ??? > #3 0x7f8b0e2443d2 in ??? > #4 0x7f8b0e3218fc in ??? > #5 0x4041cc in __navier_stokes_type_MOD_alloc_navier_stokes > at /tmp/pdt_77.f03:25 > #6 0x404397 in navier_stokes_test > at /tmp/pdt_77.f03:44 > #7 0x404397 in test > at /tmp/pdt_77.f03:38 > #8 0x404397 in main > at /tmp/pdt_77.f03:39 This seems to have gone from mainline but is still present in 16.1.1 Paul
