Okay, this "feature" has now been added to the devel trunk with r22329.
Feel free to test it and let me know of problems. I have a test code for it in
orte/test/mpi/early_abort.c
On Dec 16, 2009, at 11:27 AM, Jeff Squyres wrote:
> I think I understand you're saying:
>
> - it's ok to abort duri
Ralph,
There seems to be some problems after this commit. The hello_world application
(the MPI flavor) complete, I get all the output but in addition I have a nice
message stating that my MPI application didn't call MPI_Init.
[bosilca@dancer c]$ mpirun -np 8 --mca pml ob1 ./hello
Hello, world,
Well that stinks! I'll take another look at it - was working for me, but...been
there before!
On Dec 17, 2009, at 1:13 PM, George Bosilca wrote:
> Ralph,
>
> There seems to be some problems after this commit. The hello_world
> application (the MPI flavor) complete, I get all the output but in
I found it - one silly missing line/typo. Sorry about the problem - thanks for
pointing it out!
Fix coming in a second.
On Dec 17, 2009, at 1:13 PM, George Bosilca wrote:
> Ralph,
>
> There seems to be some problems after this commit. The hello_world
> application (the MPI flavor) complete, I
Should have it now in r22332.
Thanks again
Ralph
On Dec 17, 2009, at 1:13 PM, George Bosilca wrote:
> Ralph,
>
> There seems to be some problems after this commit. The hello_world
> application (the MPI flavor) complete, I get all the output but in addition I
> have a nice message stating tha
