>(I guess you didn't really type the extra '$' on the end
of $havempi, did you?)

No, I didn't.

With current cvs $havempi returns "46" now. But, running the MLE-example of your MPI-gretl-manual gives the error:

? mpibcast(&param)
The MPI library is not loaded
Error executing script: halting
> mpibcast(&param)

Actually, example 1 (bootstrap) returns:

Error: while reading machinefile 'C:\Program Files\Microsoft MPI\Bin\mpiexec.exe': expecting a positive number of cores following the host name.
Data error
*** error in function B_stats, line 1
> matrix means = meanc(B)
Error executing script: halting
> B_stats(B)

Artur

2014-03-26 18:49 GMT+01:00 Allin Cottrell <cottrell@wfu.edu>:

On Wed, 26 Mar 2014, Artur T. wrote:

> First of all, thank you very much for the implementation of MPI support! I
> am really forward-looking to use it in future for heavy computing tasks.
>
> However, atm I am experiencing some an issue with the installation on a Win
> 7 machine. I have installed MS-MPI and set the directory to:
> "C:\Program Files\Microsoft MPI\Bin\mpiexec.exe"
> but "eval $havempi$" returns zero.

Thanks for testing. I think there may have been an error in the test for
mpiexec.exe on Windows, which could make $havempi fail: can you try with
today's snapshot? (I guess you didn't really type the extra '$' on the end
of $havempi, did you?)

> I use current cvs (32-bit) with no admin rights. Could this be a
> problem?

I don't think so but we may have to explore the rights question if it
still doesn't work with the updated snapshot.

Allin
_______________________________________________
Gretl-devel mailing list
Gretl-devel@lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-devel