Just for Howard, who lamented the lack of MTT runs on the Mac (today’s master):
+-+-+-+--+--+--+--+--+--+
| Phase | Section | MPI Version | Dura
We talked about the Travis testing on the call today. Here's the short version
of what Travis is doing:
- it's a free, cloud-based CI infrastructure
- they provide Ubuntu VMs and containers for testing
- they also provide OS X VMs for testing
- the testing we're doing for OMPI is basically:
f
In today's telcom we decided to skip next week's meeting.
https://github.com/open-mpi/ompi/wiki/WeeklyTelcon_20151215
Also, reminder, NO meeting next week 12/22/2015.
On Dec 15, 2015, at 5:18 AM, Justin Cinkelj wrote:
>
> Second question.
> With two threads in one process, I got attempt to
> opal_shmem_segment_dettach() and munmap() on same mmap-ed address, from both
> threads. I 'fixed' that by replacing "ds_buf->seg_cpid = getpid()" with
> gettid(), and t
Looks like there is a missing conditional in
mca_btl_vader_component_close(). Will add it and PR to 1.10 and 2.x.
-Nathan
On Tue, Dec 15, 2015 at 11:18:11AM +0100, Justin Cinkelj wrote:
> I'm trying to port Open MPI to OS with threads instead of processes.
> Currently, during MPI_Finalize, I get
Justin,
at first glance, vader should be symmetric (e.g.
call opal_shmem_segment_dettach() instead of munmap()
Nathan, can you please comment ?
using tid instead of pid should also do the trick
that being said, a more elegant approach would be to create a new module in
the shmem framework
basica
I'm trying to port Open MPI to OS with threads instead of processes.
Currently, during MPI_Finalize, I get attempt to call munmap first with
address of 0x20c0 and later 0x20c8.
mca_btl_vader_component_close():
munmap (mca_btl_vader_component.my_segment,
mca_btl_vader_component.
