Dear all,
We have created a new sub list boost-mpi@lists.boost.org for discussion of
Boost.MPI development. Those interested can visit
http://lists.boost.org/mailman/listinfo.cgi/boost-mpi to sign up.
Best regards
Matthias
___
Boost-mpi mailing lis
Test
___
Boost-mpi mailing list
Boost-mpi@lists.boost.org
http://lists.boost.org/mailman/listinfo.cgi/boost-mpi
On Nov 26, 2012, at 9:03 AM, Alain O Miniussi wrote:
> Hi,
>
> Boost MPI is based on MPI C-bindings and does not use the C++ MPI
> binding at all (and those are going to be deprecated in MPI-3).
>
> Yet, on the platforms I use (and I guess on most platform) Boost MPI is
> likely to bring in a
Hi Alain,
On Nov 27, 2012, at 5:11 PM, Alain O Miniussi wrote:
> Hi,
>
> I have a tentative patch that is supposed to provide binding to the MPI
> thread API. What is the best way to have it reviewed ?
Thank you for your work. I have a few stylistic comments - shall I send them to
you off-l
On Nov 28, 2012, at 10:21 AM, Hal Finkel wrote:
> - Original Message -
>> From: "Matthias Troyer"
>> To: "alain miniussi" , "Discussion of Boost.MPI
>> development"
>> Sent: Wednesday, November 28, 2012 8:25:18 AM
>> Sub
On Nov 28, 2012, at 11:21 AM, Hal Finkel wrote:
> - Original Message -
>> From: "Matthias Troyer"
>> To: "Discussion of Boost.MPI development"
>> Sent: Wednesday, November 28, 2012 9:41:43 AM
>> Subject: Re: [Boost-mpi] Thread API
>
Hi Alain,
You can indeed provide no arg initialization and just fake arguments for MPI-1.
Matthais
On Nov 29, 2012, at 12:13 PM, Alain O Miniussi wrote:
>
> Hi,
>
> While working on the thread issue, a question arises regarding the no
> arg initialization.
>
> I understand that some MPI1.x
On Nov 28, 2012, at 9:22 PM, Hal Finkel wrote:
> - Original Message -
>> From: "Alain O Miniussi"
>> To: "Hal Finkel"
>> Cc: "Discussion of Boost.MPI development"
>> Sent: Wednesday, November 28, 2012 12:14:44 PM
>> Subject: Re: [Boost-mpi] Thread API
>>
>>
>> So, to summarize, we h
On Nov 29, 2012, at 12:06 PM, Alain O Miniussi wrote:
> On Wed, 2012-11-28 at 14:22 -0600, Hal Finkel wrote:
>>> So, to summarize, we have 3 issues:
>>> 1) syntax: lower vs upper case and mt->threading.
>>> 2) throw an exception if the requested level is not available
>>> 3) macro to control av
On Nov 29, 2012, at 1:11 PM, Alain O Miniussi wrote:
>
> That being said, I do not have a strong opinion between:
> a) provide a 2 argument ctor
> b) provide one argument ctor, and test provided level afterward.
>
> I just have a preference for b, because I find it simpler (both in terms
> of A
On Nov 29, 2012, at 5:23 PM, Jeremiah Willcock wrote:
>
> Another issue with MPI versions is that Boost.MPI currently uses functions
> such as MPI_Address that have been removed in MPI 3.0. Is that something
> that should be addressed in the future? I think the replacements for some of
> th
On Nov 28, 2012, at 2:37 PM, joel falcou wrote:
> We're currently having to deal with those and a stduent of mine came up with
> a prototype of future like interface for MPI one sided communications.
>
> Is there interest into digging this a bit more ? OSC are usually very touchy
> to get rig
On Dec 4, 2012, at 10:53 AM, Alain O Miniussi wrote:
> On Thu, 2012-11-29 at 12:55 -0500, Jeremiah Willcock wrote:
>> On Thu, 29 Nov 2012, Matthias Troyer wrote:
>>
>>>
>>> On Nov 29, 2012, at 5:23 PM, Jeremiah Willcock wrote:
>>>>
>>&
since that the minimal level)
>
> If that ok, I'll try to provide a patch accordingly.
>
> Thanks
>
>
> On Thu, 2012-11-29 at 12:04 -0600, Hal Finkel wrote:
>> - Original Message -
>>> From: "Matthias Troyer"
>>> To: "alai
On Dec 4, 2012, at 10:53 AM, Alain O Miniussi wrote:
> On Thu, 2012-11-29 at 12:55 -0500, Jeremiah Willcock wrote:
>> On Thu, 29 Nov 2012, Matthias Troyer wrote:
>>
>>>
>>> On Nov 29, 2012, at 5:23 PM, Jeremiah Willcock wrote:
>>>>
>>&
init.
>
>
> On Mon, 2012-12-31 at 23:05 +0100, Matthias Troyer wrote:
>> On Dec 4, 2012, at 10:53 AM, Alain O Miniussi wrote:
>>
>>> On Thu, 2012-11-29 at 12:55 -0500, Jeremiah Willcock wrote:
>>>> On Thu, 29 Nov 2012, Matthias Troyer wrote:
&g
t 'multiple', but
> will need to deal with the other possibilities anyway, so throwing is
> not a option. But that's just a sample of one).
>
> Regards
>
>
> On Thu, 2013-01-17 at 08:27 -0700, Matthias Troyer wrote:
>> Thank you! I've committe
100, Alain O Miniussi wrote:
>> Hi Matthias,
>>
>> Has it been apply on a branch ? I don't think I see it on trunk.
>>
>> Thanks.
>>
>> Alain
>>
>> On Mon, 2013-01-28 at 23:55 +0100, Matthias Troyer wrote:
>>> Hi Alain,
>>
On 13 Oct 2013, at 10:10, Николай Кинаш wrote:
>
> Hi all.
>
> -bash-4.1$ cat network.cpp
> #include
> #include
> #include
> #include
> #include
> #include
> #include
> #include
> #include
> #include
>
> namespace mpi = boost::mpi;
>
> using namespace boost;
> using namespace graph
On 14 Oct 2013, at 10:53, MM wrote:
> Hello
> In a problem of getting the maximum of a multi-variable function, I would use
> mpi::reduce and provide it with a function object that compares the
> evaluations of this function.
>
> Each process evaluates a part of the space, ie a number of n-tu
xn) gives that optimum evaluation of f.
>
> MM
>
>
> On 14 October 2013 09:57, Matthias Troyer wrote:
>
> On 14 Oct 2013, at 10:53, MM wrote:
>
>> Hello
>> In a problem of getting the maximum of a multi-variable function, I would
>> use mpi::r
Hi,
In orders to use MPI in a multi-threaded environment, even when only one thread
uses MPI, you need to request the necessary level of thread support in the
environment constructor. Then you'd an check whether your MPI implementation
supports multi threading. Note that using MPI the way you d
> On Feb 13, 2014, at 17:44, MM wrote:
>
>> On 13 February 2014 15:33, Matthias Troyer wrote:
>> Hi,
>>
>> In orders to use MPI in a multi-threaded environment, even when only one
>> thread uses MPI, you need to request the necessary level of t
13 February 2014 17:00, Matthias Troyer wrote:
>
>
>
>
>
> On Feb 13, 2014, at 17:44, MM wrote:
>
>> On 13 February 2014 15:33, Matthias Troyer wrote:
>> Hi,
>>
>> In orders to use MPI in a multi-threaded environment, even when only one
>>
Indeed, that is the problem. If we don't want to reserve certain tags for
internal use of Boost.MPI then the only secure way of solving this problem is
to create a copy of the communicator, and send the actual message using a
unique tag in this shadow communicator. We so far hesitated to impleme
y are receiving exactly N messages,
> and want to wait on them simultaneously).
>
> Thanks,
> Walt
>
>
> On Sun, Feb 23, 2014 at 8:18 AM, Matthias Troyer wrote:
> Indeed, that is the problem. If we don't want to reserve certain tags for
> internal use of Boost.MPI then
Hi Ziv,
You can certainly do this with Boost.MPI but it will not be very efficient. The
best strategy would be to create the graphs in the right way to start with.
Matthias
On 06 Feb 2014, at 16:23, Ziv Aginsky wrote:
> I have several graphs that are distributed among several processors. Each
Hi,
You might have more luck asking this on an FFTW mailing list, since your
question has nothing to do with the Boost MPI library.
Best regards
Matthias
On 02 May 2014, at 14:55, Reshma Pawar wrote:
> Respected Sir,
> I am facing problem in using two FFTW plans in two
Hi,
thank you for your interest. There are lots of features that we would like to
implement and your contributions would be welcome
Matthias
On 03 Apr 2014, at 23:56, Nouafel El Bachir wrote:
> Hi,
> I am interested in getting involved with boost.MPI.
> I have around 15 years experience in C+
Hi,
Does this only appear when initializing Boost.MPI from Python or also in pure
C++ programs?
Matthias
> On Sep 29, 2014, at 22:21, Amos Anderson wrote:
>
> Hello --
>
> I'm getting a segmentation fault in my program, which uses boost 1.55. I
> approached the OpenMPI folks, thread sta
Hi Alain,
Thanks for doing this. I already fixed other obsoleted calls in the past and
would thus suggest to apply this patch
Matthias
> On Oct 7, 2014, at 00:58, Alain Miniussi wrote:
>
> Hi
>
> We have a few places where we are using 1.1 stuuf that have been obsoleted in
> favor of 2.
Hi Alain,
it looks good to me
Matthias
> On 28 Oct 2014, at 10:00, Alain Miniussi wrote:
>
>
> Hi,
>
> I just proposed and implementation of send/recv. I'll propose a simplier
> version once cartesian topologies are there but in the meantime I think it's
> relevant to have an equivalent to
Hi Alain,
Do we have any reasons not to pull these to release?
Matthias
> On Nov 5, 2014, at 16:57, Alain Miniussi wrote:
>
>
> Hi,
>
> I noticed that various small but trivial fixes didn't make it from develop to
> realease 1.57
> stuff like:
> https://github.com/boostorg/mpi/pull/12
>
Hi Alain,
Sorry for the late reply, I've been traveling to France and California over the
past two days. Did Noel's reply answer your question? I generally tested it on
one or two local machines before committing to the release branch back when we
used svn.
Matthias
> On 18 Nov 2014, at 21:11
The only reason that keeps me from doing that is problems at work after the SNB
dropped the cap on the Swiss Franc that keep me stuck on administrative issues
and making sure that I have sufficient funding for my research group that
suddenly became much more expensive. Would someone else be able
Hi Alain,
Indeed, since most environments nowadays are homogeneous we can set in on by
default, but should put a warning into the documentation.
Matthias
> On 16 Apr 2015, at 15:53, Alain Miniussi wrote:
>
>
> Hi,
>
> What is the rationale for not having BOOST_MPI_HOMOGENEOUS set by default
36 matches
Mail list logo
