Igor -
Sorry for the slow reply; I was on vacation for the last week and a half.
The patch doesn't look quite right to me. If the cm PML is used, the spml
(or something else in the OSHMEM layer) is going to have to call add_procs
on the BML to initialize the procs arrays for the BTLs.
Brian
On
I think the only reason I protected that framework is to reduce the
overhead of an application using a build of Open MPI with CR support, but
no enabling it at runtime. Nothing in the compress framework depends on the
CR infrastructure (although the CR infrastructure can use the compress
framework
+1
On Thu, Dec 19, 2013 at 4:04 PM, Ralph Castain wrote:
> Looks okay to me. On the places where you need to block while waiting for
> an answer, you can use OMPI_WAIT_FOR_COMPLETION - this will spin on
> opal_progress until the condition is met. We use it elsewhere for similar
> purposes.
>
>
(Sorry for the delay, just catching up on email after the holidays)
I think that looks good too.
On Thu, Dec 19, 2013 at 4:01 PM, Ralph Castain wrote:
> +1 from me
>
>
> On Dec 19, 2013, at 12:54 PM, Adrian Reber wrote:
>
> > From: Adrian Reber
> >
> > This patch changes all send/send_buffer
(Sorry for the delay, just catching up on email after the holidays)
I agree with Ralph. You can remove the old function signatures, but keep
the places where you replace a blocking send/recv with a non-blocking
version. Then I think it is good.
Thanks,
Josh
On Wed, Dec 18, 2013 at 9:52 AM, Ral