Re: [OMPI devel] "Open MPI"-based MPI library used by K computer

Dear Open MPI community, I'm a member of MPI library development team in Fujitsu, Takahiro Kawashima, who sent mail before, is my colleague. We start to feed back. First, we fixed about MPI_LB/MPI_UB and data packing problem. Program crashes when it meets all of the following conditions: a: The

[OMPI devel] Incorrect and undefined return code/function/data type at C++ header

Dear all, We send next feed back. It's about C++ header file. In ompi/mpi/cxx/*.h, Some definitions of return code, type and function are lacked or incorrect. Attached patch fixes them (This Patch is for V1.4.X). Following list is what is lacked and incorrect. *Undefined return code --

Re: [OMPI devel] Incorrect and undefined return code/function/data type at C++ header

7;t see it in the patch, either). On Dec 4, 2011, at 9:31 PM, Y.MATSUMOTO wrote: Dear all, We send next feed back. It's about C++ header file. In ompi/mpi/cxx/*.h, Some definitions of return code, type and function are lacked or incorrect. Attached patch fixes them (This Patch is for

Re: [OMPI devel] Incorrect and undefined return code/function/data type at C++ header

Dear All, I fixed the patch. (MPI::Fint etc.) So, please replace the patch. Best regards. --- Yuki MATSUMOTO MPI development team, Fujitsu (2011/12/09 11:35), Y.MATSUMOTO wrote: Dear Jeff and all, Thank you for your comment. I'm sorry for not replying sooner. 1:MPI::Fi

[OMPI devel] Gather(linear_sync) is truncated using derived data type

Dear All, Next feed back is about MPI_Gather problem. Gather may be truncated in following condition: 1:ompi_coll_tuned_gather_intra_linear_sync is called. (message size is over 6000B) 2:Either send data type or recv data type is derived type and other data type is predefined data type. Truncat

[OMPI devel] Violating standard in MPI_Close_port

Dear All, Next is question about "MPI_Close_port". According to the MPI-2.2 standard, the "port_name" argument of MPI_Close_port() is marked as 'IN'. But, in Open MPI (both trunk and 1.4.x), the content of "port_name" is updated in MPI_Close_port(). It seems to violate the MPI standard. The foll

[OMPI devel] [PATCH] MPI_FILE_SEEK_SHARED is wrong in Fortran

Dear All, Next is about "MPI_FILE_SEEK_SHARED" in Fortran. When MPI_FILE_SEEK_SHARED is called in Fortran Program, the shared file pointer is not updated. Incorrent function call is the following part: ompi/mpi/f77/file_seek_shared_f.c--- 60 void mpi_file_seek_shared_f(MPI_Fint *fh

[OMPI devel] [PATCH]Some typos in error code, func_name and man

Dear All, We found some typos in error code/func_name/man. Attached three patches fix them(Patch is for in V1.4x). Best regards, Yuki MATSUMOTO MPI development team, Fujitsu Index: ompi/errhandler/errcode-internal.c === --- ompi/err

[OMPI devel] [PATCH]Segmentation Fault occurs when the function called from MPI_Comm_spawn_multiple fails

Dear All, Next feedback is "MPI_Comm_spawn_multiple". When the function called from MPI_Comm_spawn_multiple failed, Segmentation fault occurs. In that condition, "newcomp" sets NULL. But member of "newcomp" is referred at following part. (ompi/mpi/c/comm_spawn_multiple.c) 176 /* set array of

[OMPI devel] [PATCH]Incorrect algorithm choice using coll_tuned_dynamic_rules_filename (over 2GiB message)

Dear All, Next feedback is about "coll_tuned_dynamic_rules_filename". Incorrect algorithm is selected in following conditions: 1:"--mca coll_tuned_use_dynamic_rules 1" is set. 2:"--mca coll_tuned_dynamic_rules_filename" is set. 3: Collective communication which is written in 2, called >= 2GiB com

[OMPI devel] Collective communications may be abend when it use over 2GiB buffer

Dear All, Next feedback is about "collective communications". Collective communication may be abend when it use over 2GiB buffer. This problem occurs following condition: -- communicator_size * count(scount/rcount) >= 2GiB It occurs in even small PC cluster. The following is one of the suspiciou