From 6a212a7fb8ac2abe96a8b447d3778d6ebc97fcf9 Mon Sep 17 00:00:00 2001 From: Tom Eccles Date: Tue, 15 Apr 2025 17:57:26 +0100 Subject: [PATCH] [Fortran] disable pointer_check_11.f90 which exhibits UB (#232) This test calls an empty subroutine with a null pointer as an argument. In gfortran it is hoped that a flag will detect this and produce an error. Flang does not perform any such analysis, and so on some machines when optimization is enabled, this test never terminates. I don't think this is a flang bug as such because the input code is doing undefined behaviour. --- Fortran/gfortran/regression/DisabledFiles.cmake | 1 + 1 file changed, 1 insertion(+) diff --git a/Fortran/gfortran/regression/DisabledFiles.cmake b/Fortran/gfortran/regression/DisabledFiles.cmake index 158d03b963..59fc4f5e67 100644 --- a/Fortran/gfortran/regression/DisabledFiles.cmake +++ b/Fortran/gfortran/regression/DisabledFiles.cmake @@ -1042,6 +1042,7 @@ file(GLOB FAILING_FILES CONFIGURE_DEPENDS namelist_96.f90 # real data for integer NAMELIST input no_unit_error_1.f90 pointer_check_10.f90 + pointer_check_11.f90 # test exhibits UB, sometimes at O3 it hangs forever pointer_remapping_6.f08 unpack_bounds_1.f90