[OMPI users] jemalloc and crash in opal_memory_linux_ptmalloc2_open

Hi all - I'm trying to use jemalloc with my project, but I get a crash in opal_memory_linux_ptmalloc2_open when jemalloc is linked. If I use tcmalloc, this does not happen. Any ideas? Is there a sanctioned way to override malloc libraries in conjunction with openmpi? Thanks, Brian

Re: [OMPI users] MPI_THREAD_FUNNELED and enable-mpi-thread-multiple

. Is this still the case in 1.6 and 1.7 series? Thanks, Brian On Mon, Feb 4, 2013 at 9:09 PM, Roland Schulz <rol...@utk.edu> wrote: > > > > On Mon, Jan 28, 2013 at 9:20 PM, Brian Budge <brian.bu...@gmail.com> wrote: >> >> I believe that yes, you have t

Re: [OMPI users] Unexpected Behavior with C++ New Memory Allocation and MPI IO

You appear to be using new and delete[] together. Instead you should be using new[] and delete[] and new and delete together. Brian On Wed, Jun 12, 2013 at 4:44 PM, Corey Allen wrote: > I have done a search on this and I haven't found an explanation. I am not a

[OMPI users] design advice for multi-threaded load balancing application

Hi all - I have an application where the master node will spawn slaves to perform computation (using the singleton Comm_spawn_multiple paradigm available in OpenMPI) . The master will only decide the work to do, and also provide data common to all the computations. The slaves are

Re: [OMPI users] MPI_THREAD_FUNNELED and enable-mpi-thread-multiple

I believe that yes, you have to compile enable-mpi-thread-multiple to get anything other than SINGLE. Brian On Tue, Jan 22, 2013 at 12:56 PM, Roland Schulz wrote: > Hi, > > compiling 1.6.1 or 1.6.2 without enable-mpi-thread-multiple returns from > MPI_Init_thread as provided

Re: [OMPI users] MPI_Isend/MPI_Recv problem in a multi-thread program

Did you build openmpi with multithreading enabled? If not then that could be the problem. Brian On Dec 5, 2012 3:20 AM, "赵印" wrote: > Hi all, > > I have a MPI_Isend/MPI_Recv problem in a multi-thread program. > > In the program: > *The first machine* has one thread

Re: [OMPI users] valgrind slaves in singleton mode

> > mpirun -mca orte_fork_agent "valgrind " ./my_app > > We will execute "valgrind ./my_app" whenever we start one of > your processes. This includes process launched via comm_spawn. > > HTH > Ralph > > On Nov 16, 2012, at 4:38 PM, Brian Budge <br

Re: [OMPI users] valgrind slaves in singleton mode

Thanks very much Ralph. Silly me I thought it might actually be some effort :) Brian On Fri, Nov 16, 2012 at 4:04 PM, Ralph Castain wrote: > Easiest solution: just add valgrind into the cmd line > > mpirun valgrind ./my_app > > > On Nov 16, 2012, at 3:37 PM, "Tom Bryan

[OMPI users] valgrind slaves in singleton mode

Hi all - I'm using openmpi to spawn child processes in singleton mode. If I use mpirun, I can just run > mpirun valgrind myprog With spawn, it is expected that the spawned process will call mpi_init(_thread). If I want to run valgrind on my processes, what steps should be taken? I'm currently

Re: [OMPI users] MPI_Recv MPI_THREAD_MULTIPLE

On Tue, Nov 13, 2012 at 1:56 AM, 赵印 wrote: > I have a problem here. > > My program runs perfectly in MPI version 1.6 series, but it would run into > some problem in MPI version 1.4x series. *Does MPI 1.4x version have a > bug related in MPI_Recv.* > > The log in Node[1] says

Re: [OMPI users] singleton spawn regression

h didn't get into the 1.6.2 > official release, and the version numbers in the repo didn't immediately get > updated - so the nightly build was still labeled as 1.6.2 even after the > official release came out. > > > On Oct 16, 2012, at 10:46 AM, Brian Budge <brian.bu

[OMPI users] singleton spawn regression

Hi all - There was a bug in version 1.6.1 that caused singleton spawn not to work correctly with multi-machine configurations. I verified that a nightly build of 1.6.2 fixed this issue, in particular 1.6.2a1r27234 works. I just grabbed the 1.6.2 official release, and it appears that somehow the

Re: [OMPI users] EXTERNAL: Re: unacceptable latency in gathering process

Hi Ralph - Is this really true? I've been using thread_multiple in my openmpi programs for quite some time... There may be known cases where it will not work, but for vanilla MPI use, it seems good to go. That's not to say that you can't create your own deadlock if you're not careful, but they

Re: [OMPI users] MPI_Spawn and process allocation policy

On Mon, Oct 1, 2012 at 10:33 AM, Ralph Castain wrote: > Yes, that is the expected behavior as you describe it. > > If you want to run on hosts that are not already provided (via hostfile in > the environment or on the command line), then you need to use the "add-host" > or

Re: [OMPI users] MPI_Spawn and process allocation policy

On Wed, Sep 12, 2012 at 10:23 AM, Ralph Castain <r...@open-mpi.org> wrote: > > On Sep 12, 2012, at 9:55 AM, Brian Budge <brian.bu...@gmail.com> wrote: > >> On Wed, Aug 17, 2011 at 12:05 AM, Simone Pellegrini >> <spellegr...@dps.uibk.ac.at> wrote: >>

Re: [OMPI users] trouble_MPI

On Tue, Sep 18, 2012 at 2:14 PM, Alidoust wrote: > > Dear Madam/Sir, > > > I have a serial Fortran code (f90), dealing with matrix diagonalizing > subroutines, and recently got its parallel version to be faster in some > unfeasible parts via the serial program. > I have

Re: [OMPI users] MPI_Spawn and process allocation policy

On Wed, Aug 17, 2011 at 12:05 AM, Simone Pellegrini wrote: > On 08/16/2011 11:15 PM, Ralph Castain wrote: >> >> I'm not finding a bug - the code looks clean. If I send you a patch, could >> you apply it, rebuild, and send me the resulting debug output? > > yes, I could

Re: [OMPI users] MPI::Intracomm::Spawn and cluster configuration

e (singleton comm_spawn is so rarely > used that it can easily be overlooked for some time). > > Thx > Ralph > > > > > On Aug 31, 2012, at 3:32 PM, Brian Budge <brian.bu...@gmail.com> wrote: > >> Thanks, much appreciated. >> >> On Fri, Aug 31, 2

Re: [OMPI users] MPI::Intracomm::Spawn and cluster configuration

lease is issued, if that helps. > > > On Aug 31, 2012, at 2:33 PM, Brian Budge <brian.bu...@gmail.com> wrote: > >> Hi Ralph - >> >> This is true, but we may not know until well into the process whether >> we need MPI at all. We have an SMP/NUMA mode that is designed

Re: [OMPI users] MPI::Intracomm::Spawn and cluster configuration

aemon running the job - only difference is in the number of > characters the user types to start it. > > > On Aug 30, 2012, at 8:44 AM, Brian Budge <brian.bu...@gmail.com> wrote: > >> In the event that I need to get this up-and-running soon (I do need >> somethin

[OMPI users] valgrind/memory leaks from spawn

Hi all - I'm writing a program which will start in a single process. This program will call init (THREAD_MULTIPLE), and finalize. In between, it will call spawn an unknown number of times (think of the program as a daemon that launches jobs over and over again). I'm running a simple example

Re: [OMPI users] MPI::Intracomm::Spawn and cluster configuration

In the event that I need to get this up-and-running soon (I do need something working within 2 weeks), can you recommend an older version where this is expected to work? Thanks, Brian On Tue, Aug 28, 2012 at 4:58 PM, Brian Budge <brian.bu...@gmail.com> wrote: > Thanks! > > On Tu

Re: [OMPI users] MPI::Intracomm::Spawn and cluster configuration

Thanks! On Tue, Aug 28, 2012 at 4:57 PM, Ralph Castain <r...@open-mpi.org> wrote: > Yeah, I'm seeing the hang as well when running across multiple machines. Let > me dig a little and get this fixed. > > Thanks > Ralph > > On Aug 28, 2012, at 4:51 PM, Brian Budge &l

Re: [OMPI users] MPI::Intracomm::Spawn and cluster configuration

e set - does it work okay? > > It works fine for me, hence the question. > > Also, what OMPI version are you using? > > On Aug 28, 2012, at 4:25 PM, Brian Budge <brian.bu...@gmail.com> wrote: > >> I see. Okay. So, I just tried removing the check for universe siz

Re: [OMPI users] MPI::Intracomm::Spawn and cluster configuration

(not the original singleton!) reads the > hostfile to find out how many nodes are around, and then does the launch. > > You are trying to check the number of nodes from within the singleton, which > won't work - it has no way of discovering that info. > > > > > On Aug 28,

Re: [OMPI users] MPI::Intracomm::Spawn and cluster configuration

>echo hostsfile localhost budgeb-sandybridge Thanks, Brian On Tue, Aug 28, 2012 at 2:36 PM, Ralph Castain <r...@open-mpi.org> wrote: > Hmmm...what is in your "hostsfile"? > > On Aug 28, 2012, at 2:33 PM, Brian Budge <brian.bu...@gmail.com> wrote: > >

Re: [OMPI users] MPI::Intracomm::Spawn and cluster configuration

< size << std::endl; } std::cerr << "slave responding..." << std::endl; MPI_Finalize(); return 0; } Any ideas? Thanks for any help. Brian On Wed, Aug 22, 2012 at 9:03 AM, Ralph Castain <r...@open-mpi.org> wrote: > It really is just

Re: [OMPI users] MPI::Intracomm::Spawn and cluster configuration

Hi. I know this is an old thread, but I'm curious if there are any tutorials describing how to set this up? Is this still available on newer open mpi versions? Thanks, Brian On Fri, Jan 4, 2008 at 7:57 AM, Ralph Castain wrote: > Hi Elena > > I'm copying this to the user list

Re: [OMPI users] busy wait in MPI_Recv

d if there is nothing else important for the processor to turn to, a fast MPI_Recv is what matters. Dick Treumann - MPI Team IBM Systems & Technology Group Dept X2ZA / MS P963 -- 2455 South Road -- Poughkeepsie, NY 12601 Tele (845) 433-7846 Fax (845) 433-8363 From: Brian Budge <bri

[OMPI users] busy wait in MPI_Recv

Hi all - I just ran a small test to find out the overhead of an MPI_Recv call when no communication is occurring. It seems quite high. I noticed during my google excursions that openmpi does busy waiting. I also noticed that the option to -mca mpi_yield_when_idle seems not to help much (in

Re: [OMPI users] my leak or OpenMPI's leak?

yes, sorry. I did mean 1.5. In my case, going back to 1.43 solved my oom problem. On Sun, Oct 17, 2010 at 4:57 PM, Ralph Castain <r...@open-mpi.org> wrote: > There is no OMPI 2.5 - do you mean 1.5? > > On Oct 17, 2010, at 4:11 PM, Brian Budge wrote: > >> Hi Jody - &g

Re: [OMPI users] my leak or OpenMPI's leak?

Hi Jody - I noticed this exact same thing the other day when I used OpenMPI v 2.5 built with valgrind support. I actually ran out of memory due to this. When I went back to v 2.43, my program worked fine. Are you also using 2.5? Brian On Wed, Oct 6, 2010 at 4:32 AM, jody

Re: [OMPI users] Dynamic process tutorials?

org> wrote: > > On Jul 12, 2010, at 11:12 AM, Brian Budge wrote: > >> HI Ralph - >> >> Thanks for the reply.  I think this patch sounds great!  The idea in >> our software is that it won't be known until after the program is >> running whet

Re: [OMPI users] Dynamic process tutorials?

ite some graphic tool > which will call mpirun or mpiexec. But somewhere you have to tell OpenMPI > what to run on how many processors etc. > > I'd suggest you take a look at the "MPI-The Complete Reference" Vol I and II > > Jody > > On Mon, Jul 12, 2010 at

Re: [OMPI users] Dynamic process tutorials?

an external program like mpirun. Is there a plan for this to enter the mainline? Brian On Mon, Jul 12, 2010 at 8:29 AM, Ralph Castain <r...@open-mpi.org> wrote: > > On Jul 12, 2010, at 9:07 AM, Brian Budge wrote: > >> Hi Jody - >> >> Thanks for the reply.  is there a w

Re: [OMPI users] Dynamic process tutorials?

rguments > will be set to an intercommunicator of thes spawner and the spawnees. > You can use this intercommunicator as the communicator argument > in the MPI_functions. > > Jody > On Fri, Jul 9, 2010 at 5:56 PM, Brian Budge <brian.bu...@gmail.com> wrote: >> Hi all - >> >> I'v

Re: [OMPI users] MPI_Init() and MPI_Init_thread()

I believe that it specifies the *minimum* threading model supported. If I recall, opmi is already funnel safe even in single mode. However, if mpi calls are made from outside the main thread, you should specify funneled for portability Brian On Mar 2, 2010 11:59 PM, "Terry Frankcombe"

Re: [OMPI users] Sending relatively large messages with high frequency

Is your code multithreaded? On Feb 25, 2010 12:56 AM, "Amr Hassan" wrote: Thanks alot for your reply, I'm using blocking Send and Receive. All the clients are sending data and the server is receive the messages from the clients with MPI_ANY_SOURCE as the sender. Do you

Re: [OMPI users] Sending relatively large messages with high frequency

We've seen similar things in our code. In our case it is probably due to a race condition. Try running the segv'ing process in a debugger, and it will likely show you a bug in your code On Feb 24, 2010 9:36 PM, "Amr Hassan" wrote: Hi All, I'm facing a strange problem

[OMPI users] MPI_ERR_TRUNCATE returned from MPI_Test

Hi all - We are receiving an error of MPI_ERR_TRUNCATE from MPI_Test (after enabling the RETURN error handler). I'm confused as to what might cause this, as I was assuming that this generally resulted from a recv call being made requesting fewer bytes than were sent. Can anyone shed some light

Re: [OMPI users] MPI-2 Supported on Open MPI 1.2.5?

One small (or to some, not so small) note is that full multi-threading with OpenMPI is very unlikely to work with infiniband right now. Brian On Mon, Mar 10, 2008 at 6:24 AM, Michael wrote: > Quick answer, till you get a complete answer, Yes, OpenMPI has long > supported most

Re: [OMPI users] new to MPI+infiniband

: > > > Brian, > >Here is how I do it: > > > > ./configure --prefix /opt/openmpi-1.2.4 --with-openib=/usr/local/ > > ofed \ > > --without-tm CC=icc CXX=icpc F77=ifort FC=ifort \ > > --with-threads=posix --enable-mpi-threads > > > > > >

[OMPI users] new to MPI+infiniband

Hi all - I have been using OpenMPI for quite a while now, and its working out great. I was looking at the FAQ and trying to figure out how to configure OpenMPI with infiniband. It shows how to enable IB pointing to the OFED directory. I have infiniband built into the kernel, along with IP over

Re: [OMPI users] Run a process double

Henry - OpenMP and OpenMPI are two different things. OpenMP is a way to automatically (in limited situations) parallelize your code using a threading model. OpenMPI is an MPI implementation. MPI is a message passing standard, which usually parallelizes computation on a process basis. Brian

Re: [OMPI users] ld_library_path not being updated

Thanks! That appears to have done it. Brian On 1/17/07, Scott Atchley <atch...@myri.com> wrote: On Jan 17, 2007, at 10:45 AM, Brian Budge wrote: > Hi Adrian - > > Thanks for the reply. I have been investigating this further. It > appears that ssh isn't start

Re: [OMPI users] ld_library_path not being updated

ue, Jan 16, 2007 at 05:22:35PM -0800, Brian Budge wrote: > Hi all - Hi! > If I run from host-0: > > mpirun -np 4 -host host-0 myprogram > > I have no problems, but if I run > >mpirun -np 4 -host host-1 myprogram > error while loading shared libraries: libSGUL.so: can

[OMPI users] ld_library_path not being updated

Hi all - I'm having a bit of an issue with my library paths and mpi that I can't quite seem to resolve. If I run from host-0: mpirun -np 4 -host host-0 myprogram I have no problems, but if I run mpirun -np 4 -host host-1 myprogram I get an error like this: error while loading shared

Re: [OMPI users] dma using infiniband protocol

OMPI's PML for some other project? Or are you writing MPI applications? Or ...? On Nov 2, 2006, at 2:22 PM, Brian Budge wrote: > Thanks for the pointer, it was a very interesting read. > > It seems that by default OpenMPI uses the nifty pipelining trick > with pinning pages while tra

Re: [OMPI users] dma using infiniband protocol

order to register a different memory segment for another memory transfer. Brian On Nov 2, 2006, at 12:22 PM, Brian Budge wrote: > Thanks for the pointer, it was a very interesting read. > > It seems that by default OpenMPI uses the nifty pipelining trick > with pinning pages while tra

Re: [OMPI users] dma using infiniband protocol

searched around, but it's possible I just used the wrong search terms. Thanks, Brian On 11/2/06, Jeff Squyres <jsquy...@cisco.com> wrote: This paper explains it pretty well: http://www.open-mpi.org/papers/euro-pvmmpi-2006-hpc-protocols/ On Nov 2, 2006, at 1:37 PM, Brian Budge wrote:

[OMPI users] dma using infiniband protocol

Hi all - I'm wondering how DMA is handled in OpenMPI when using the infiniband protocol. In particular, will I get a speed gain if my read/write buffers are already pinned via mlock? Thanks, Brian