Open MPI logo

Open MPI User's Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Open MPI User's mailing list

From: Matthew Knepley (knepley_at_[hidden])
Date: 2005-06-16 15:58:44


Ben Allan <baallan_at_[hidden]> writes:
> I tried that just now and it doesn't look different. Will check your
> other mail (which apparently i haven't reached yet in the mail reader).
> I take that back, apparently -parseable isn't recognized but -parsable
> is. A :-separated output results. tolerable. :)
> Kudos to you all, actually. I haven't seen anything this useful
>>From the mpich team yet. {now matt can correct me...}
>
> My users expect to combine c,c++,fortran,python,java(!),and fortran-variant-x
> all in the same executable on a diversity of platforms.
> And when it doesn't work, they don't go to you, they
> tell me "hey, make it work, my mpi isn't broken -- it
> runs my vanilla C code all the time." The real issue is,
> of course, the utter insanity of history that is the linker.
> The workaround always involves reverse-engineering the
> compiler wrappers and assembling the link line details
> explicitly. Far better that this kind of insanity be
> testable and the work-arounds picked out by my configure
> scripts than all the users coming back to me for
> individual attention.
>
> The autoconf process, if ompi_info is to be believed,
> checks a lot of machine specific and compiler specific
> things to do with alignment, size, type existence, etc
> and records these assumptions. Throwing a random
> compiler (or worse, the primitive size-changing fortran switches)
> into a compiler wrapper's path is just asking for trouble.
> We need good config info to diagnose these kinds of user idiocy
> efficiently.

  Which is why we NEVER believe anything anyone (or any wrapper) tells us.
I test every include/type/compile/link/run that is necessary. The goal
should be, if it passes configure, the code will definitely work. We are
pretty close in PETSc. You are correct that the MPICH output from 'show'
is not as structured. I don't care as much since we don't use shell/sed/
anything invented before 1980, but it is still a nice feature and step forward.

  One other thing. I don't use the wrappers sometimes when I need to verify
the actual include paths/libraries. I do a lot of code parsing/verifying/generation
and I need this info to do it for C (yuck, I know).

    Thanks,

        Matt

-- 
"Failure has a thousand explanations. Success doesn't need one" -- Sir Alec Guiness