June 2014 - discuss - mailman3.cels.anl.gov

Re: [mpich-discuss] Internal Error: invalid error code 409e10 (Ring ids do not match)
by Lu, Huiwei 02 Jun '14

02 Jun '14

You can try install MPICH on your user directory by: ./configure —prefix=/your/directory See http://www.mpich.org/static/downloads/3.1/mpich-3.1-installguide.pdf for more information. — Huiwei On Jun 2, 2014, at 5:18 PM, Jakub Łuczyński <doubleloop(a)o2.pl> wrote: > This are Linux platforms, and I do not have administrative rights to change anything in this environment (it is shared, multiuser env, with this version of MPICH installed for quite time i guess). I could try to talk with administrators but i am not 100% sure that the code (posted in this mail) is correct. > > Best regards, > Kuba > > On Tue, Jun 3, 2014 at 12:03 AM, Lu, Huiwei <huiweilu(a)mcs.anl.gov> wrote: > Hi Kuba, > > Since it works with both Open MPI and BGP, it is most likely a problem of your MPICH installation or your platform. > > We have stopped supporting the Windows platform a while ago due to lack of developer resources. Please refer to our FAQ for more information: > http://wiki.mpich.org/mpich/index.php/Frequently_Asked_Questions#Q:_Why_can… > > If it is on windows platform, we recommend you use Microsoft MPI, which can be found here: http://msdn.microsoft.com/en-us/library/bb524831(v=vs.85).aspx > > We also encourage you to use the latest MPICH on Linux or OSX platforms, which can be downloaded here: http://www.mpich.org/downloads/ > > — > Huiwei > > On Jun 2, 2014, at 4:49 PM, Jakub Łuczyński <doubleloop(a)o2.pl> wrote: > > > I wrote my assignment using MPI, and tested it both locally on Open MPI (1.6.5) and on IBM Blue Gene/P (with mpi implementation provided by IBM). Everything worked fine. Turns out that our solutions are tested also in our labs where MPICH, is installed: > > > > $ mpich2version > > MPICH2 Version: 1.4.1p1 > > > > And when I run my solution there I get this strange error: > > $ mpirun -n 2 msp-par.exe 10 10 1 > > Internal Error: invalid error code 409e10 (Ring ids do not match) in MPIR_Reduce_impl:1087 > > Fatal error in PMPI_Reduce: Other MPI error, error stack: > > PMPI_Reduce(1270).....: MPI_Reduce(sbuf=0x7fff693a92e8, rbuf=0x7fff693a9300, count=1, dtype=USER<struct>, op=0x98000000, root=0, MPI_COMM_WORLD) failed > > MPIR_Reduce_impl(1087): > > > > I am literally out of ideas what is wrong! > > > > Below source code fragments (c++): > > > > struct msp_solution > > { > > int x1, y1, x2, y2; > > m_entry_t max_sum; > > msp_solution(); > > msp_solution(const pair<int, int> &c1, const pair<int, int> &c2, int max_sum); > > friend bool operator<(const msp_solution &s1, const msp_solution &s2); > > }; > > > > void max_msp_solution(msp_solution *in, msp_solution *inout, int, MPI_Datatype*) > > { > > *inout = max(*in, *inout); > > } > > > > // somewhere in code > > { > > MPI_Datatype MPI_msp_solution_t; > > MPI_Op max_msp_solution_op; > > > > // create MPI struct from msp_solution > > MPI_Datatype types[] = { MPI_INT, MPI_LONG_LONG_INT }; > > int block_lengths[] = { 4, 2 }; > > MPI_Aint base_addr, x1_addr, max_sum_addr; > > MPI_Get_address(&collected_solution, &base_addr); > > MPI_Get_address(&collected_solution.x1, &x1_addr); > > MPI_Get_address(&collected_solution.max_sum, &max_sum_addr); > > > > MPI_Aint displacements[] = > > { > > x1_addr - base_addr, > > max_sum_addr - base_addr > > }; > > > > MPI_Type_create_struct(2, block_lengths, displacements, types, &MPI_msp_solution_t); > > MPI_Type_commit(&MPI_msp_solution_t); > > > > // max reduction function > > MPI_Op_create((MPI_User_function *) max_msp_solution, 1, &max_msp_solution_op); > > > > ... > > > > msp_solution solution, received_solution; > > MPI_Comm comm, > > ... > > // comm is created using MPI_Comm_split > > // solution is initialized > > MPI_Reduce(&solution, &received_solution, 1, MPI_msp_solution_t, max_msp_solution_op , 0, MPI_COMM_WORLD); > > // ERROR above!!! > > } > > > > > > Is there some error in this? How can I make it run? > > P.S. MPI_Send and MPI_Recv on my struct MPI_msp_solution_t seems to work fine > > > > Thanks in advance! > > Best regards, > > Kuba > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] Internal Error: invalid error code 409e10 (Ring ids do not match)
by Jakub Łuczyński 02 Jun '14

02 Jun '14

This are Linux platforms, and I do not have administrative rights to change anything in this environment (it is shared, multiuser env, with this version of MPICH installed for quite time i guess). I could try to talk with administrators but i am not 100% sure that the code (posted in this mail) is correct. Best regards, Kuba On Tue, Jun 3, 2014 at 12:03 AM, Lu, Huiwei <huiweilu(a)mcs.anl.gov> wrote: > Hi Kuba, > > Since it works with both Open MPI and BGP, it is most likely a problem of > your MPICH installation or your platform. > > We have stopped supporting the Windows platform a while ago due to lack of > developer resources. Please refer to our FAQ for more information: > > http://wiki.mpich.org/mpich/index.php/Frequently_Asked_Questions#Q:_Why_can… > > If it is on windows platform, we recommend you use Microsoft MPI, which > can be found here: > http://msdn.microsoft.com/en-us/library/bb524831(v=vs.85).aspx > > We also encourage you to use the latest MPICH on Linux or OSX platforms, > which can be downloaded here: http://www.mpich.org/downloads/ > > — > Huiwei > > On Jun 2, 2014, at 4:49 PM, Jakub Łuczyński <doubleloop(a)o2.pl> wrote: > > > I wrote my assignment using MPI, and tested it both locally on Open MPI > (1.6.5) and on IBM Blue Gene/P (with mpi implementation provided by IBM). > Everything worked fine. Turns out that our solutions are tested also in our > labs where MPICH, is installed: > > > > $ mpich2version > > MPICH2 Version: 1.4.1p1 > > > > And when I run my solution there I get this strange error: > > $ mpirun -n 2 msp-par.exe 10 10 1 > > Internal Error: invalid error code 409e10 (Ring ids do not match) in > MPIR_Reduce_impl:1087 > > Fatal error in PMPI_Reduce: Other MPI error, error stack: > > PMPI_Reduce(1270).....: MPI_Reduce(sbuf=0x7fff693a92e8, > rbuf=0x7fff693a9300, count=1, dtype=USER<struct>, op=0x98000000, root=0, > MPI_COMM_WORLD) failed > > MPIR_Reduce_impl(1087): > > > > I am literally out of ideas what is wrong! > > > > Below source code fragments (c++): > > > > struct msp_solution > > { > > int x1, y1, x2, y2; > > m_entry_t max_sum; > > msp_solution(); > > msp_solution(const pair<int, int> &c1, const pair<int, int> &c2, int > max_sum); > > friend bool operator<(const msp_solution &s1, const msp_solution &s2); > > }; > > > > void max_msp_solution(msp_solution *in, msp_solution *inout, int, > MPI_Datatype*) > > { > > *inout = max(*in, *inout); > > } > > > > // somewhere in code > > { > > MPI_Datatype MPI_msp_solution_t; > > MPI_Op max_msp_solution_op; > > > > // create MPI struct from msp_solution > > MPI_Datatype types[] = { MPI_INT, MPI_LONG_LONG_INT }; > > int block_lengths[] = { 4, 2 }; > > MPI_Aint base_addr, x1_addr, max_sum_addr; > > MPI_Get_address(&collected_solution, &base_addr); > > MPI_Get_address(&collected_solution.x1, &x1_addr); > > MPI_Get_address(&collected_solution.max_sum, &max_sum_addr); > > > > MPI_Aint displacements[] = > > { > > x1_addr - base_addr, > > max_sum_addr - base_addr > > }; > > > > MPI_Type_create_struct(2, block_lengths, displacements, types, > &MPI_msp_solution_t); > > MPI_Type_commit(&MPI_msp_solution_t); > > > > // max reduction function > > MPI_Op_create((MPI_User_function *) max_msp_solution, 1, > &max_msp_solution_op); > > > > ... > > > > msp_solution solution, received_solution; > > MPI_Comm comm, > > ... > > // comm is created using MPI_Comm_split > > // solution is initialized > > MPI_Reduce(&solution, &received_solution, 1, MPI_msp_solution_t, > max_msp_solution_op , 0, MPI_COMM_WORLD); > > // ERROR above!!! > > } > > > > > > Is there some error in this? How can I make it run? > > P.S. MPI_Send and MPI_Recv on my struct MPI_msp_solution_t seems to work > fine > > > > Thanks in advance! > > Best regards, > > Kuba > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] Internal Error: invalid error code 409e10 (Ring ids do not match)
by Lu, Huiwei 02 Jun '14

02 Jun '14

Hi Kuba, Since it works with both Open MPI and BGP, it is most likely a problem of your MPICH installation or your platform. We have stopped supporting the Windows platform a while ago due to lack of developer resources. Please refer to our FAQ for more information: http://wiki.mpich.org/mpich/index.php/Frequently_Asked_Questions#Q:_Why_can… If it is on windows platform, we recommend you use Microsoft MPI, which can be found here: http://msdn.microsoft.com/en-us/library/bb524831(v=vs.85).aspx We also encourage you to use the latest MPICH on Linux or OSX platforms, which can be downloaded here: http://www.mpich.org/downloads/ — Huiwei On Jun 2, 2014, at 4:49 PM, Jakub Łuczyński <doubleloop(a)o2.pl> wrote: > I wrote my assignment using MPI, and tested it both locally on Open MPI (1.6.5) and on IBM Blue Gene/P (with mpi implementation provided by IBM). Everything worked fine. Turns out that our solutions are tested also in our labs where MPICH, is installed: > > $ mpich2version > MPICH2 Version: 1.4.1p1 > > And when I run my solution there I get this strange error: > $ mpirun -n 2 msp-par.exe 10 10 1 > Internal Error: invalid error code 409e10 (Ring ids do not match) in MPIR_Reduce_impl:1087 > Fatal error in PMPI_Reduce: Other MPI error, error stack: > PMPI_Reduce(1270).....: MPI_Reduce(sbuf=0x7fff693a92e8, rbuf=0x7fff693a9300, count=1, dtype=USER<struct>, op=0x98000000, root=0, MPI_COMM_WORLD) failed > MPIR_Reduce_impl(1087): > > I am literally out of ideas what is wrong! > > Below source code fragments (c++): > > struct msp_solution > { > int x1, y1, x2, y2; > m_entry_t max_sum; > msp_solution(); > msp_solution(const pair<int, int> &c1, const pair<int, int> &c2, int max_sum); > friend bool operator<(const msp_solution &s1, const msp_solution &s2); > }; > > void max_msp_solution(msp_solution *in, msp_solution *inout, int, MPI_Datatype*) > { > *inout = max(*in, *inout); > } > > // somewhere in code > { > MPI_Datatype MPI_msp_solution_t; > MPI_Op max_msp_solution_op; > > // create MPI struct from msp_solution > MPI_Datatype types[] = { MPI_INT, MPI_LONG_LONG_INT }; > int block_lengths[] = { 4, 2 }; > MPI_Aint base_addr, x1_addr, max_sum_addr; > MPI_Get_address(&collected_solution, &base_addr); > MPI_Get_address(&collected_solution.x1, &x1_addr); > MPI_Get_address(&collected_solution.max_sum, &max_sum_addr); > > MPI_Aint displacements[] = > { > x1_addr - base_addr, > max_sum_addr - base_addr > }; > > MPI_Type_create_struct(2, block_lengths, displacements, types, &MPI_msp_solution_t); > MPI_Type_commit(&MPI_msp_solution_t); > > // max reduction function > MPI_Op_create((MPI_User_function *) max_msp_solution, 1, &max_msp_solution_op); > > ... > > msp_solution solution, received_solution; > MPI_Comm comm, > ... > // comm is created using MPI_Comm_split > // solution is initialized > MPI_Reduce(&solution, &received_solution, 1, MPI_msp_solution_t, max_msp_solution_op , 0, MPI_COMM_WORLD); > // ERROR above!!! > } > > > Is there some error in this? How can I make it run? > P.S. MPI_Send and MPI_Recv on my struct MPI_msp_solution_t seems to work fine > > Thanks in advance! > Best regards, > Kuba > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss

1 0

Internal Error: invalid error code 409e10 (Ring ids do not match)
by Jakub Łuczyński 02 Jun '14

02 Jun '14

I wrote my assignment using MPI, and tested it both locally on Open MPI (1.6.5) and on IBM Blue Gene/P (with mpi implementation provided by IBM). Everything worked fine. Turns out that our solutions are tested also in our labs where MPICH, is installed: $ mpich2version MPICH2 Version: 1.4.1p1 And when I run my solution there I get this strange error: $ mpirun -n 2 msp-par.exe 10 10 1 Internal Error: invalid error code 409e10 (Ring ids do not match) in MPIR_Reduce_impl:1087 Fatal error in PMPI_Reduce: Other MPI error, error stack: PMPI_Reduce(1270).....: MPI_Reduce(sbuf=0x7fff693a92e8, rbuf=0x7fff693a9300, count=1, dtype=USER<struct>, op=0x98000000, root=0, MPI_COMM_WORLD) failed MPIR_Reduce_impl(1087): I am literally out of ideas what is wrong! Below source code fragments (c++): struct msp_solution { int x1, y1, x2, y2; m_entry_t max_sum; msp_solution(); msp_solution(const pair<int, int> &c1, const pair<int, int> &c2, int max_sum); friend bool operator<(const msp_solution &s1, const msp_solution &s2); }; void max_msp_solution(msp_solution *in, msp_solution *inout, int, MPI_Datatype*) { *inout = max(*in, *inout); } // somewhere in code { MPI_Datatype MPI_msp_solution_t; MPI_Op max_msp_solution_op; // create MPI struct from msp_solution MPI_Datatype types[] = { MPI_INT, MPI_LONG_LONG_INT }; int block_lengths[] = { 4, 2 }; MPI_Aint base_addr, x1_addr, max_sum_addr; MPI_Get_address(&collected_solution, &base_addr); MPI_Get_address(&collected_solution.x1, &x1_addr); MPI_Get_address(&collected_solution.max_sum, &max_sum_addr); MPI_Aint displacements[] = { x1_addr - base_addr, max_sum_addr - base_addr }; MPI_Type_create_struct(2, block_lengths, displacements, types, &MPI_msp_solution_t); MPI_Type_commit(&MPI_msp_solution_t); // max reduction function MPI_Op_create((MPI_User_function *) max_msp_solution, 1, &max_msp_solution_op); ... msp_solution solution, received_solution; MPI_Comm comm, ... // comm is created using MPI_Comm_split // solution is initialized MPI_Reduce(&solution, &received_solution, 1, MPI_msp_solution_t, max_msp_solution_op , 0, MPI_COMM_WORLD); // ERROR above!!! } Is there some error in this? How can I make it run? P.S. MPI_Send and MPI_Recv on my struct MPI_msp_solution_t seems to work fine Thanks in advance! Best regards, Kuba

1 0

Re: [mpich-discuss] How does MPICH spawn processes, and is there a way to configure it?
by Lu, Huiwei 02 Jun '14

02 Jun '14

Thanks for reporting this, Scott. — Huiwei On Jun 2, 2014, at 4:29 PM, Scott O'Malley <scottyomalley(a)gmail.com> wrote: > After extensive research it turns out it was my slave nodes that weren't configured correctly. Only the Master Node was able to access the Slaves so when mpiexec told a slave to spawn a new process and error was thrown. This wasn't being thrown with the original param I was passing as it forced the Master Node to distribute all tasks. It was an odd one to track down but hopefully if anyone else has this problem they can see this information here > > > On 2 June 2014 22:09, Lu, Huiwei <huiweilu(a)mcs.anl.gov> wrote: > I am not sure what tree mode is and why you need to disable it. In general users does not need to specify which mode hydra should use to run MPI. You can simply run it using: mpiexec -f hosts -n 4 ./app, as explained in http://wiki.mpich.org/mpich/index.php/Using_the_Hydra_Process_Manager#Quick… > > In order to understand you failed run, can you give us your host file and the command line you launch you application? > > — > Huiwei > > On May 29, 2014, at 9:16 AM, Scott O'Malley <scottyomalley(a)gmail.com> wrote: > > > I've looked through the hydra docs but I haven't found a flag or anything else that would let me disable it spawning in tree mode, if MPICH even does that. > > > > I was able to kind of rectify the error using OpenMPI and specifying the "-mca plm_rsh_no_tree_spawn" flag > > > > > > On 29 May 2014 14:54, Lu, Huiwei <huiweilu(a)mcs.anl.gov> wrote: > > Hi Scott, > > > > Check https://wiki.mpich.org/mpich/index.php/Using_the_Hydra_Process_Manager to see if it is what you want. > > > > Best, > > — > > Huiwei > > > > On May 29, 2014, at 6:37 AM, Scott O'Malley <scottyomalley(a)gmail.com> wrote: > > > > > I have recently built a cluster of Raspberry Pi computers. When this was made up of a master node and 2 slave nodes there was no issue calling mpiexec across the cluster. When I add an additional node to the cluster bringing the size to 1 master and 3 slave nodes I get an error of "Host File Verification Failed". I've tested password-less SSH on all nodes and it's working correctly. > > > > > > I've done some googling of the error and set up and the only mention I've gotten to it was on this link here relating to OpenMPI - http://www.open-mpi.org/community/lists/users/2013/11/22940.php > > > > > > Is there a similar setup for MPICH and if so can someone link me to the correct documentation page? > > > > > > Cheers > > > > > > Scott > > > _______________________________________________ > > > discuss mailing list discuss(a)mpich.org > > > To manage subscription options or unsubscribe: > > > https://lists.mpich.org/mailman/listinfo/discuss > > > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > > > > > > > -- > > - Scott > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss > > > > -- > - Scott > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] How does MPICH spawn processes, and is there a way to configure it?
by Scott O'Malley 02 Jun '14

02 Jun '14

After extensive research it turns out it was my slave nodes that weren't configured correctly. Only the Master Node was able to access the Slaves so when mpiexec told a slave to spawn a new process and error was thrown. This wasn't being thrown with the original param I was passing as it forced the Master Node to distribute all tasks. It was an odd one to track down but hopefully if anyone else has this problem they can see this information here On 2 June 2014 22:09, Lu, Huiwei <huiweilu(a)mcs.anl.gov> wrote: > I am not sure what tree mode is and why you need to disable it. In general > users does not need to specify which mode hydra should use to run MPI. You > can simply run it using: mpiexec -f hosts -n 4 ./app, as explained in > http://wiki.mpich.org/mpich/index.php/Using_the_Hydra_Process_Manager#Quick… > > In order to understand you failed run, can you give us your host file and > the command line you launch you application? > > — > Huiwei > > On May 29, 2014, at 9:16 AM, Scott O'Malley <scottyomalley(a)gmail.com> > wrote: > > > I've looked through the hydra docs but I haven't found a flag or > anything else that would let me disable it spawning in tree mode, if MPICH > even does that. > > > > I was able to kind of rectify the error using OpenMPI and specifying the > "-mca plm_rsh_no_tree_spawn" flag > > > > > > On 29 May 2014 14:54, Lu, Huiwei <huiweilu(a)mcs.anl.gov> wrote: > > Hi Scott, > > > > Check > https://wiki.mpich.org/mpich/index.php/Using_the_Hydra_Process_Manager to > see if it is what you want. > > > > Best, > > — > > Huiwei > > > > On May 29, 2014, at 6:37 AM, Scott O'Malley <scottyomalley(a)gmail.com> > wrote: > > > > > I have recently built a cluster of Raspberry Pi computers. When this > was made up of a master node and 2 slave nodes there was no issue calling > mpiexec across the cluster. When I add an additional node to the cluster > bringing the size to 1 master and 3 slave nodes I get an error of "Host > File Verification Failed". I've tested password-less SSH on all nodes and > it's working correctly. > > > > > > I've done some googling of the error and set up and the only mention > I've gotten to it was on this link here relating to OpenMPI - > http://www.open-mpi.org/community/lists/users/2013/11/22940.php > > > > > > Is there a similar setup for MPICH and if so can someone link me to > the correct documentation page? > > > > > > Cheers > > > > > > Scott > > > _______________________________________________ > > > discuss mailing list discuss(a)mpich.org > > > To manage subscription options or unsubscribe: > > > https://lists.mpich.org/mailman/listinfo/discuss > > > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > > > > > > > -- > > - Scott > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss > -- - Scott

1 0

Re: [mpich-discuss] How does MPICH spawn processes, and is there a way to configure it?
by Lu, Huiwei 02 Jun '14

02 Jun '14

I am not sure what tree mode is and why you need to disable it. In general users does not need to specify which mode hydra should use to run MPI. You can simply run it using: mpiexec -f hosts -n 4 ./app, as explained in http://wiki.mpich.org/mpich/index.php/Using_the_Hydra_Process_Manager#Quick… In order to understand you failed run, can you give us your host file and the command line you launch you application? — Huiwei On May 29, 2014, at 9:16 AM, Scott O'Malley <scottyomalley(a)gmail.com> wrote: > I've looked through the hydra docs but I haven't found a flag or anything else that would let me disable it spawning in tree mode, if MPICH even does that. > > I was able to kind of rectify the error using OpenMPI and specifying the "-mca plm_rsh_no_tree_spawn" flag > > > On 29 May 2014 14:54, Lu, Huiwei <huiweilu(a)mcs.anl.gov> wrote: > Hi Scott, > > Check https://wiki.mpich.org/mpich/index.php/Using_the_Hydra_Process_Manager to see if it is what you want. > > Best, > — > Huiwei > > On May 29, 2014, at 6:37 AM, Scott O'Malley <scottyomalley(a)gmail.com> wrote: > > > I have recently built a cluster of Raspberry Pi computers. When this was made up of a master node and 2 slave nodes there was no issue calling mpiexec across the cluster. When I add an additional node to the cluster bringing the size to 1 master and 3 slave nodes I get an error of "Host File Verification Failed". I've tested password-less SSH on all nodes and it's working correctly. > > > > I've done some googling of the error and set up and the only mention I've gotten to it was on this link here relating to OpenMPI - http://www.open-mpi.org/community/lists/users/2013/11/22940.php > > > > Is there a similar setup for MPICH and if so can someone link me to the correct documentation page? > > > > Cheers > > > > Scott > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss > > > > -- > - Scott > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] What is the difference between MPI_THREAD_SINGLE and MPI_THREAD_FUNNELED
by Balaji, Pavan 01 Jun '14

01 Jun '14

On May 30, 2014, at 12:30 PM, Boisvert, Sebastien <boisvert(a)anl.gov> wrote: > On Friday, May 30, 2014 12:26 PM, discuss-bounces(a)mpich.org [discuss-bounces(a)mpich.org] on behalf of Rajeev Thakur [thakur(a)mcs.anl.gov] wrote: >> To: discuss(a)mpich.org >> Subject: Re: [mpich-discuss] What is the difference between MPI_THREAD_SINGLE and MPI_THREAD_FUNNELED >> >> It shouldn't, but what does it return in your program? > > I am currently using MPI_Init (I thought it was OK if only the main thread calls MPI functions). > > Code-> https://github.com/GeneAssembly/biosal/blob/master/engine/node.c#L33 If you initialize with MPI_INIT, the MPI implementation can choose to provide whatever thread-level it likes. There are two options for you: Option 1: MPI_INIT /* tell MPI to give you what it can */ MPI_QUERY_THREAD /* ask for what you got */ if (less than FUNNELED) abort; Option 2: MPI_INIT_THREAD(FUNNELED, &provided); /* tell MPI to give you FUNNELED */ if (provided < FUNNELED) abort; Option 2 is cleaner since you gave MPI a hint that you would like to have FUNNELED level. — Pavan

1 0

Re: [mpich-discuss] question on correct use of mpi_isend/mpi_irecv
by Rob Latham 01 Jun '14

01 Jun '14

On 05/27/2014 11:15 AM, Grindeanu, Iulian R. wrote: > Hello, > We are trying to track down a memory leak, and we are not sure if we are > using correctly MPI_ISend and MPI_Irecv compiling mpich with --enable-g=all will report (in a manner not nearly as refined as valgrind) if you are leaking any MPI resources. Things like memory, communicators, datatypes, attributes, INFO objects... ==rob > > We would like to use non-blocking communication, between pairs of processes. > > (it happens during a computation, at every time step; every processor > has a list of processors it needs to communicate with; what we do not > know, is the size of the messages, in advance) > > Let's assume A needs to send to B > Because we do not know the size in advance, we send first a fixed size > message from A to B; this first message has info about how big the total > message needs to be; A knows it has to send more. When B receives the > first part, it sends an ack (small size, int 4) to A, acknowledging, and > also resizing the local buffer, to fill more data; then A sends the rest. > > All these sends / receives use Isend and Irecv, and we try to match the > messages, using proper tags. We use different tags for fixed size, ack, > and rest of the message. > > Before sending the code, I would like to know is what we try is feasible > and doable? > > We notice that the memory use on processor A keeps increasing, when we > do this in a loop, on the order of about 1000-2000 bytes per iteration. > > I think we are matching correctly isends and ireceives, and that our > buffers are not leaking (we may be wrong) > > So again, an iteration is like this: > A sends a fixed size message to B; B sends back an ack, and when A > receives it, sends the rest, because B has now the proper buffer to receive. > > Next iteration, the size of the message is again unknown, so we do again > the dance. > > Do you have an example like that we can use? Should we use other types > of sends/receives? > > Thanks, > Iulian > > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss > -- Rob Latham Mathematics and Computer Science Division Argonne National Lab, IL USA

1 0