Unfortunately, MPI_THREAD_MULTIPLE has never received a lot of
testing in any version of OMPI (including v1.1). Various members
tested the bozo cases (e.g., ensure we don't double lock, etc.), and
periodically tested/debugged simple multi-threaded apps, but not much
more than that.
As such, it's probably pretty lucky that v1.1 works with
THREAD_MULTIPLE. Indeed, v1.2 has changed quite a bit since v1.2,
and only the same level of THREAD_MULTIPLE testing has occurred (so
far) -- checking for the bozo cases.
We do expect to improve this over the next several months; as Tim
mentioned, for the v1.3 timeframe.
On Mar 6, 2007, at 9:06 AM, Bert Wesarg wrote:
> this is realy sad, version 1.1.2 works quite good with threads
> threads which starts mpi requests), only 1 of 10 (or even less)
> kills with
> a SIGSEGV. And this this simple test program works even longer.
> Tim Mattox wrote:
>> Hi Bert Wesarg,
>> Thank you for your quick testing of 1.2rc1. 1.2 is expected to
>> fail when
>> using MPI_THREAD_MULTIPLE. I suspect that a working and tested
>> MPI_THREAD_MULTIPLE will be one of our goals for 1.3.
>> On 3/6/07, Bert Wesarg <wesarg_at_[hidden]> wrote:
>>> I followed the call to test the rc1, but a simple test programm
>>> hangs, but
>>> non deterministic. all but one orted have quit. but no cpu eating
>>> orted or mpirun.
>>> The test system is a xeon cluster with myrinet interconnect.
>>> the outouts are splited into two mails (100kb limit)
>>> Bert Wesarg
>>> devel mailing list
> devel mailing list
Server Virtualization Business Unit