Thanks, guys.
I will check the code of OB1 more carefully. Thanks.
Best,
Lirong
Message: 7
> Date: Thu, 25 Oct 2012 10:55:51 -0700
> From: Ralph Castain
> Subject: Re: [OMPI devel] NIC Failover and Message Stripping of Open
> MPI.
> To: Open MPI Developers
> Message-ID:
> Content-Type
Okay, 1.7.0rc4 has been posted, with ompi_info fixed :-)
On Oct 25, 2012, at 2:47 PM, Ralph Castain wrote:
> Ah, I see the problem - it is localized to ompi_info and due to the fact that
> we aren't setting things up completely in that code (trying to avoid a
> complete start).
>
> FWIW: you
Le 25/10/2012 23:56, Barrett, Brian W a écrit :
> Hi all -
>
> The MX BTL segfaults during MPI_FINALIZE in the trunk (and did before my
> mpool change in r27485). I'm not really interested in fixing it; the
> problem does not occur with the MX MTL. Does anyone else have interest in
> fixing it?
Hi all -
The MX BTL segfaults during MPI_FINALIZE in the trunk (and did before my
mpool change in r27485). I'm not really interested in fixing it; the
problem does not occur with the MX MTL. Does anyone else have interest in
fixing it? If not, should we remove it from the trunk (we already remo
Ah, I see the problem - it is localized to ompi_info and due to the fact that
we aren't setting things up completely in that code (trying to avoid a complete
start).
FWIW: you are most certainly allowed to amend event callbacks in the base open
function.
I'll fix it. Meantime, the branch seems
Strange - okay, will look.
On Oct 25, 2012, at 2:34 PM, George Bosilca wrote:
> Broken enough that even the ompi_info fails. And for a good reason: one is
> not allowed to amend event callbacks in the component base open function.
>
> Here is the stack:
> 0 0x77a82fc4 in opal_libevent
Broken enough that even the ompi_info fails. And for a good reason: one is not
allowed to amend event callbacks in the component base open function.
Here is the stack:
0 0x77a82fc4 in opal_libevent2019_event_priority_set (ev=0x63b100,
pri=3)
at ../../../../../../ompi/opal/mca/event/
Just an FYI - I asked a similar question recently and got the following answer
from Rolf:
> In my case, it was specific to openib only and it required you to be running
> with two or more IB rails.
> Then, if one of them failed, we just shut it down, and continued with the
> working ones.
> You
On Oct 25, 2012, at 17:54 , Lirong Jian wrote:
> Hi foks,
>
> Sorry to bother you guys, but I have some questions about Open MPI and really
> want your help.
>
> There are some papers (e.g., [1, 2, 3], although they are sort of old-aged)
> mentioning that Open MPI is supporting NIC failover
On Oct 25, 2012, at 1:00 PM, Barrett, Brian W wrote:
> Your first e-mail got eaten by our virus scanner (it doesn't like .bz2
> files),
See http://www.open-mpi.org/community/lists/devel/2012/10/11638.php.
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisc
Your first e-mail got eaten by our virus scanner (it doesn't like .bz2
files), but we could probably only register the libnbc progress function
on first use, but it would slightly slow down all non blocking
collectives. Probably worth it, but not sure I'll have time to add that
code today.
Brian
Something that might not be clear from my initial writeup:
1. I had to go change C code to disable libnbc. Since non-blocking collectives
are part of MPI-3:
a) we have no convenient configure argument to not build the libnbc coll
component (there is a way, but it's laborious), and
b) eve
On Oct 25, 2012, at 12:32 PM, Jeff Squyres (jsquyres) wrote:
> 1. sm NetPipe latencies up to size 150 bytes (run on a Sandy Bride, 2 procs
> same core)
s/core/socket/
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/
Attached are the following graphs:
1. sm NetPipe latencies up to size 150 bytes (run on a Sandy Bride, 2 procs
same core)
2. openib NetPipe latencies up to size 150 bytes (run on 2 old Xeons
[pre-Nehalem] with old Mellanox ConnectX IB HCAs)
3. Same as #1, but all the way up to 8MB
4. Same as #2,
Hi foks,
Sorry to bother you guys, but I have some questions about Open MPI and
really want your help.
There are some papers (e.g., [1, 2, 3], although they are sort of old-aged)
mentioning that Open MPI is supporting NIC failover and message stripping
over multiple NICs. However, when I read the
Hi folks
We have posted the first release candidate for the 1.7.0 release in the usual
place:
http://www.open-mpi.org/software/ompi/v1.7/
Please put it thru the wringer to help us validate it prior to release later
this month. We are still updating the NEWS section, but we REALLY need to
vali
16 matches
Mail list logo
