[Gretl-users] Re: MPI problems
On Tue, 10 Sep 2019, Sven Schreiber wrote:
Hi,
I'm having trouble using MPI parallelization with a August 15th snapshot
on Windows 10, i7 CPU. Gretl apparently freezes, "never" comes back.
This is with the parallel_specs package (sample script), or with the
bootstrap illustration code in the gretl+MPI manual.
I will update the snapshot and see what happens, but is there some other
kind of testing code available which would give some useful output or
even diagnosis? When I start gretl, 'eval $sysinfo.mpi' tells me 1
(yes), and I have freshly installed Microsoft's MPI on this machine. So
most recent version there, 10 I think.
I see what you mean. Gretl was hanging, waiting for output from
gretlmpi after it had already exited. This was a side effect of some
changes I made in June/July, trying to get output from gretlmpi to
appear in the GUI (script output window) in real time -- something
that is quite difficult to arrange on Windows.
There was a race thing going on here: it could work OK depending on
how much printing was done in the mpi block.
Anyway, I've attempted a fix in git and today's snapshots. If the
problem is not fully solved I might have to give up on real-time
output on Windows -- waiting for gretlmpi to finish is safer.
Allin