[mpich-discuss] Deadlock in MPI_Ibarrier over ch3:sock

[Dave Goodell]

On Jan 23, 2013, at 3:17 PM CST, Jed Brown wrote:

> On Wed, Jan 23, 2013 at 2:44 PM, Dave Goodell <goodell at mcs.anl.gov> wrote:
> 
>> I'm surprised this isn't caught by the "coll/nonblock3" test, although we may just be making too much progress some other way, or we may not have enough coverage of the test itself in that test: http://git.mpich.org/mpich.git/blob/refs/heads/master:/test/mpi/coll/nonblocking3.c
>> 
>> A modest test case would be very helpful.
> 
> <ibarrier.c>

Thanks.  I've created a ticket to track this: http://trac.mpich.org/projects/mpich/ticket/1785

I suspect the code fix and adding your basic regression test will be easy.  I just need to make sure that the nonblocking3 test is doing what we expect it to, which will take a bit more work.

-Dave