Jeff Hammond

I'll play around with it tomorrow (I'm 10 hours ahead of you). I've spent a lot of time on CFI features lately but haven't tested everything obviously.

This is what I see right now with NVHPC and GCC after applying a trivial patch. Is this success or failure in the test? ``` jhammond@nuclear:~/zfp/gcc$ gfortran ../tests/fortran/testFortran.f90 ../fortran/zfp.f90 -L./lib...

Also, if you rename this file, you no longer need free source form flags. ``` renamed: ../tests/fortran/testFortran.f -> ../tests/fortran/testFortran.f90 ```

Fortran STOP should return numerical values. I think that is what you want for ctest but I haven't verified. I'm not aware of any _relevant_ system where size_t, ptrdiff_t and...

I guess I don't see where the undefined behavior comes in here. Fortran (ISO/IEC DIS 1539-1:2022) says the following: >ISO/IEC 9899:2018 specifies that the representations for nonnegative signed integers are...

Regarding C `enum`, you can exploit the following and define an unused `enum` member that is `INT_MAX` to force the compiler to use a type that is compatible with `int`....

What I tried to say before is, I do not think you can overflow when Fortran passes `INTEGER(kind=c_size_t)` to a `ptrdiff_t` because there is no way for Fortran to access...

While this is user-friendly and I understand it's generous of Intel MPI to do it, it teaches Fortran programmers to do bad things. Real Fortran experts like Bill Long will...

I am working on a new implementation of MPI F08 that is completely separate from the MPI library. It should allow this sort of thing pretty easily. I don't know...

I thought Rob took care of most/all the internal count issues with ROMIO when we wrote the BigMPI paper. One can always test it for internal correctness using the BigMPI...