Re: Any one using Compaq(DEC) Tru64 UNIX?

[W . Northcott]

Marcus wrote

>It's possible there's a bug with long long types with HDF5 on your
>system.  I'll try it on an Alpha OSF4 box and see what I get.  Meanwhile,
>you might just configure with --without-hdf5dir.

Unfortunately that might make the libraries pass the tests, but leave my
real problem unfixed.

I am trying to run ArtStkMkt code and this appears to be afflicted with
the same problem of fp data type confusion.  It does not use HDF5.  It
also has no error checking so silly numbers flow around until you get an
overflow at 10e308!

I am fairly sure that you were on the right track with
>GCC has a half-working means by which you can indicate arguments
>and construct a call on the fly.  On some systems it doesn't work,
though, and
>figuring out where values land in a call frame must done by hand.
>You might try the optimistic __builtin_return default and see
>if the tests pass.

I don't sufficiently understand this comment to have a picture of the sort
of C code which would be affected.  What exactly is __builtin_return?  I
can't find it in my documentation of gcc or Swarm.

Can you give me an example of source that might show the error or even
better point me at some calls in the Swarm system?
Now that I have my head around gdb I might manage to find the problem if I
knew better what I was looking for.

Thanks for your help so far
Bill Northcott



                  ==================================
   Swarm-Support is for discussion of the technical details of the day
   to day usage of Swarm.  For list administration needs (esp.
   [un]subscribing), please send a message to <address@hidden>
   with "help" in the body of the message.