Re: [OMPI users] Deadlock in OpenMPI 1.8.3 and PETSc 3.4.5

Ben, I can't find anything in the MPI standard suggesting that a recursive behavior of the attribute deletion is enforced/supported by the MPI standard. Thus, the current behavior of Open MPI (a single lock for all attributes), while maybe a little strict, is standard compliant (and thus correct).

Re: [OMPI users] OMPI users] OMPI users] OMPI users] OMPI users] MPI inside MPI (still)

We can certainly add an MPI_Info key to redirect stdin, stdout, and stderr. However, that won't happen in the immediate future, nor would it come into the 1.8 series. Meantime, I suspect wrapping these codes in scripts sounds like the way to go. You would call mpirun to start the job in the script

Re: [OMPI users] OMPI users] OMPI users] OMPI users] OMPI users] MPI inside MPI (still)

The option system("env -i ...") has been tested earlier by me and it does work. There is doubt though it would work along with a job scheduler. I will reserve this as a last resort solution. mpi_comm_spawn("/bin/sh","-c","siesta < infile",..) definitely does not work. Patching siesta to start as

Re: [OMPI users] Deadlock in OpenMPI 1.8.3 and PETSc 3.4.5

Hi Ben, Would you mind checking if you still observe this deadlock condition if you use the 1.8.4 rc4 candidate? openmpi-1.8.4rc4.tar.gz I realize the behavior will likely be the same, but this is just to double check

Re: [OMPI users] [EXTERNAL] Re: How to find MPI ranks located in remote nodes?

On Wed, Dec 17, 2014 at 7:29 PM, Jeff Squyres (jsquyres) wrote: > Returning to a super-old thread that was never finished... > > > On Dec 2, 2014, at 6:49 PM, George Bosilca wrote: > > > That's not enough. They will have to check for the right version of Open > MPI and then for the availability

Re: [OMPI users] OMPI users] OpenMPI 1.8.4rc3, 1.6.5 and 1.6.3: segmentation violation in mca_io_romio_dist_MPI_File_close

Eric, As long as lFileNameWithoutTooLongPath length is less than 226 characters and you do not run into some threads related race conditions, that should be just fine, and that roughly covers 99% cases. Thanks for sharing this workaround ! Cheers, Gilles Eric Chamberland wrote: >Hi! > >Here

Re: [OMPI users] OpenMPI 1.8.4rc3, 1.6.5 and 1.6.3: segmentation violation in mca_io_romio_dist_MPI_File_close

Hi! Here is a "poor man's fix" that works for me (the idea is not from me, thanks to Thomas H.): #1- char* lCwd = getcwd(0,0); #2- chdir(lPathToFile); #3- MPI_File_open(...,lFileNameWithoutTooLongPath,...); #4- chdir(lCwd); #5- ... I think there are some limitations but it works very well for

Re: [OMPI users] OMPI users] OMPI users] OMPI users] OMPI users] MPI inside MPI (still)

Alex, You do not want to spawn mpirun. Or if this is really what you want, then just use system("env -i ...") I think what you need is spawn a shell that do the redirection and then invoke your app. This is something like MPI_Comm_spawn("/bin/sh", "-c", "siesta < infile") That being said, i str

Re: [OMPI users] Operators for MPI handles not correctly overloaded with Intel Fortran

Jorg -- I'm sorry for the giant delay in replying; the US holiday and the MPI Forum meeting last week made a disaster out of my already-out-of-control INBOX. :-( Hmm. This almost sounds like a bug in the intel compiler. Do you have the latest version of their compiler, perchance? On Dec 1,

Re: [OMPI users] [EXTERNAL] Re: How to find MPI ranks located in remote nodes?

Returning to a super-old thread that was never finished... On Dec 2, 2014, at 6:49 PM, George Bosilca wrote: > That's not enough. They will have to check for the right version of Open MPI > and then for the availability of the OMPI_ functions. That looks > as having the sabe cost as checking

[OMPI users] Deadlock in OpenMPI 1.8.3 and PETSc 3.4.5

Hi PETSc and OpenMPI teams, I'm running into a deadlock in PETSc 3.4.5 with OpenMPI 1.8.3: 1. PetscCommDestroy calls MPI_Attr_delete 2. MPI_Attr_delete acquires a lock 3. MPI_Attr_delete calls Petsc_DelComm_Outer (through a callback) 4. Petsc_DelComm_Outer calls MPI_Attr_get 5. MPI_Attr_get

Re: [OMPI users] OMPI users] OMPI users] OMPI users] MPI inside MPI (still)

Let me rephrase the previous message: Putting "/bin/sh" in command with info key "ompi_non_mpi" set to ".true." (if command is empty, mpi_comm_spawn tries to execute ' ') of mpi_comm_spawn and "-c" "mpirun -n 1 myapp" in args results in this message:

Re: [OMPI users] OMPI users] OMPI users] OMPI users] MPI inside MPI (still)

Putting "/bin/sh" in command with info key "ompi_non_mpi" set to ".true." (if command is empty, mpi_comm_spawn tries to execute ' ') of mpi_comm_spawn and "-c" "mpirun -n 1 myapp" in args results in this message: /usr/bin/sh: -c: option requires an argument Putting a single string in args as "-

Re: [OMPI users] ERROR: C_FUNLOC function

Siegmar -- I filed https://github.com/open-mpi/ompi/issues/317 and https://github.com/open-mpi/ompi/issues/318. On Dec 17, 2014, at 3:33 PM, Siegmar Gross wrote: > Hi Jeff, > >> This fix was just pushed to the OMPI master. A new master tarball >> should be available shortly (probably with

Re: [OMPI users] OMPI users] OMPI users] OMPI users] MPI inside MPI (still)

I don't think this has any chance of working. The redirection is something interpreted by the shell, and when Open MPI "fork-exec" a process it does not behave as the shell. Thus a potentially non-portable solution would be to instead of launching the mpirun directly to launch it through a shell.

Re: [OMPI users] OMPI users] OMPI users] OMPI users] MPI inside MPI (still)

Ralph, Sorry, "<" as an element of argv to mpi_comm_spawn is interpreted just the same, as another parameter by the spawnee process. But I am confused: wouldn't it be redundant to put "mpirun" "-n" "1" "myapp" as elements of argv, considering role of the other parameters of mpi_comm_spawn like th

Re: [OMPI users] ERROR: C_FUNLOC function

Hi Jeff, > This fix was just pushed to the OMPI master. A new master tarball > should be available shortly (probably within an hour or so -- look > for a tarball dated Dec 17 at http://www.open-mpi.org/nightly/master/). Yes, I could build it now. Thank you very much to everybody who helped to fi

Re: [OMPI users] OMPI users] OMPI users] OMPI users] MPI inside MPI (still)

Have you tried putting the "<" as a separate parameter? In other words, since you are specifying the argv, you have to specify each of them separately. So it looks more like: "mpirun", "-n", "1", "myapp", "<", "stdinfile" Does that work? Ralph On Wed, Dec 17, 2014 at 8:07 AM, Alex A. Schmidt w

Re: [OMPI users] OMPI users] OMPI users] OMPI users] MPI inside MPI (still)

Ralph, I am afraid I will have to insist on i/o redirection matter for the spawnee process. I have a "child" mpi code that do just 2 things: read the 3 parameters passed to it and print them, and then read data from stdin and show it. So, if "stdin_file" is a text file with two lines, say: 10 20

Re: [OMPI users] ERROR: C_FUNLOC function

Siegmar -- This fix was just pushed to the OMPI master. A new master tarball should be available shortly (probably within an hour or so -- look for a tarball dated Dec 17 at http://www.open-mpi.org/nightly/master/). I anticipate that this fix will also make it in for the v1.8.4 release (see h