July 2013 - discuss - mailman3.cels.anl.gov

Re: [mpich-discuss] mpich on Mac os x
by Pavan Balaji 08 Jul '13

08 Jul '13

This is a problem with the application. Can you contact the application team about it? -- Pavan On 07/08/2013 11:32 AM, Reem Alraddadi wrote: > > Hi Wesley, > > How could I do that? as I read in the internet. I have to set: > include "mpif.h" > Call MPI_COMM_FREE > > I found where I can set that in FLASH code ( I have attached the file) > So I set: > integer, parameter :: FLASH_COMM = MPI_COMM_FREE > but I got an error when I compile the code: > > Included at Driver_abortFlash.F90:34: > > integer, parameter :: FLASH_COMM = MPI_COMM_FREE > 1 > Error: Symbol 'flash_comm' at (1) already has basic type of INTEGER > make: *** [Driver_abortFlash.o] Error 1 > > the code used MPI_COMM_WORLD. > So is there a way to fix that? > > Thanks, > Reem > > > > Date: Mon, 8 Jul 2013 10:25:38 -0500 > From: Wesley Bland <wbland(a)mcs.anl.gov <mailto:[email protected]>> > To: discuss(a)mpich.org <mailto:[email protected]> > Subject: Re: [mpich-discuss] discuss Digest, Vol 9, Issue 13 > Message-ID: <9A8EEBB2-72BB-429B-9368-150983B1C283(a)mcs.anl.gov > <mailto:[email protected]>> > Content-Type: text/plain; charset="iso-8859-1" > > That's the correct way to set that environment variable, but I would > have to agree with Pavan here. If you're running into the cap on > context id's (especially with so few processes) it would seem that > you're creating a lot of communicators that you probably don't need. > Are you sure that you're freeing them correctly after use? > > Wesley > > On Jul 8, 2013, at 10:05 AM, Reem Alraddadi <raba500(a)york.ac.uk > <mailto:[email protected]>> wrote: > > > Hi Wesley. > > I wrote the following: > > mpirun --np 4 --env MPIR_PARAM_CTXID_EAGER_SIZE 1 ./flash4 > > but the error still the same. Did I do it in the wrong way?? > > > > Thanks, > > Reem > > > > Message: 5 > > Date: Mon, 8 Jul 2013 08:14:48 -0500 > > From: Wesley Bland <wbland(a)mcs.anl.gov <mailto:[email protected]>> > > To: discuss(a)mpich.org <mailto:[email protected]> > > Subject: Re: [mpich-discuss] mpich on Mac os x > > Message-ID: <8DC984B2-4E4B-4BFE-806E-203463A7A4E4(a)mcs.anl.gov > <mailto:[email protected]>> > > Content-Type: text/plain; charset=iso-8859-1 > > > > It seems that you're creating more communicators than MPICH can > handle. It's possible that you might be able to get around this by > setting the environment variable MPIR_PARAM_CTXID_EAGER_SIZE to > something smaller than its default (which is 2). That frees up a few > more communicators, but there is a pathological case where even with > fewer communicators than the max, MPICH won't be able to agree on a > new communicator id when needed. Try changing that environment > variable and see if that fixes things. > > > > Wesley > > > > On Jul 8, 2013, at 5:33 AM, Reem Alraddadi <raba500(a)york.ac.uk > <mailto:[email protected]>> wrote: > > > > > Hi all, > > > I am using mpich-3.0.4 on Mac os x version 10.7.5 to run FLASH > code. It works fine in the beginning of the run and then I got the > following error: > > > > > > Fatal error in MPI_Comm_create: Other MPI error, error stack: > > > MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000002, group=0xc8001349, > new_comm=0x7fff606a8614) failed > > > MPI_Comm_create(577).................: > > > MPIR_Comm_create_intra(241)..........: > > > MPIR_Get_contextid(799)..............: > > > MPIR_Get_contextid_sparse_group(1146): Cannot allocate context > ID because of fragmentation (169/2048 free on this process; ignore_id=0) > > > Fatal error in MPI_Comm_create: Other MPI error, error stack: > > > MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000002, group=0xc80012b6, > new_comm=0x7fff670cc614) failed > > > MPI_Comm_create(577).................: > > > MPIR_Comm_create_intra(241)..........: > > > MPIR_Get_contextid(799)..............: > > > MPIR_Get_contextid_sparse_group(1146): Cannot allocate context > ID because of fragmentation (316/2048 free on this process; ignore_id=0) > > > Fatal error in MPI_Comm_create: Other MPI error, error stack: > > > MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000004, group=0xc800000e, > new_comm=0x7fff629d5614) failed > > > MPI_Comm_create(577).................: > > > MPIR_Comm_create_intra(241)..........: > > > MPIR_Get_contextid(799)..............: > > > MPIR_Get_contextid_sparse_group(1146): Cannot allocate context > ID because of fragmentation (2020/2048 free on this process; > ignore_id=0) > > > Fatal error in MPI_Comm_create: Other MPI error, error stack: > > > MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000002, group=0xc8000020, > new_comm=0x7fff639ae614) failed > > > MPI_Comm_create(577).................: > > > MPIR_Comm_create_intra(241)..........: > > > MPIR_Get_contextid(799)..............: > > > MPIR_Get_contextid_sparse_group(1146): Cannot allocate context > ID because of fragmentation (2002/2048 free on this process; ignore_id=0 > > > > > > Is there a way to fix that ? > > > > > > Thanks, > > > Reem > > On Mon, Jul 8, 2013 at 9:17 AM, Pavan Balaji <balaji(a)mcs.anl.gov > <mailto:[email protected]>> wrote: > > > > Also make sure you are freeing unused communicators. > > > > -- Pavan > > > > On 07/08/2013 08:14 AM, Wesley Bland wrote: > >> > >> It seems that you're creating more communicators than MPICH can > handle. > >> It's possible that you might be able to get around this by > setting the > >> environment variable MPIR_PARAM_CTXID_EAGER_SIZE to something > smaller than > >> its default (which is 2). That frees up a few more > communicators, but there > >> is a pathological case where even with fewer communicators than > the max, > >> MPICH won't be able to agree on a new communicator id when > needed. Try > >> changing that environment variable and see if that fixes things. > >> > >> Wesley > >> > >> On Jul 8, 2013, at 5:33 AM, Reem Alraddadi <raba500(a)york.ac.uk > <mailto:[email protected]>> wrote: > >> > >>> Hi all, > >>> I am using mpich-3.0.4 on Mac os x version 10.7.5 to run FLASH > code. It > >>> works fine in the beginning of the run and then I got the > following error: > >>> > >>> Fatal error in MPI_Comm_create: Other MPI error, error stack: > >>> MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000002, > >>> group=0xc8001349, new_comm=0x7fff606a8614) failed > >>> MPI_Comm_create(577).................: > >>> MPIR_Comm_create_intra(241)..........: > >>> MPIR_Get_contextid(799)..............: > >>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > >>> because of fragmentation (169/2048 free on this process; > ignore_id=0) > >>> Fatal error in MPI_Comm_create: Other MPI error, error stack: > >>> MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000002, > >>> group=0xc80012b6, new_comm=0x7fff670cc614) failed > >>> MPI_Comm_create(577).................: > >>> MPIR_Comm_create_intra(241)..........: > >>> MPIR_Get_contextid(799)..............: > >>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > >>> because of fragmentation (316/2048 free on this process; > ignore_id=0) > >>> Fatal error in MPI_Comm_create: Other MPI error, error stack: > >>> MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000004, > >>> group=0xc800000e, new_comm=0x7fff629d5614) failed > >>> MPI_Comm_create(577).................: > >>> MPIR_Comm_create_intra(241)..........: > >>> MPIR_Get_contextid(799)..............: > >>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > >>> because of fragmentation (2020/2048 free on this process; > ignore_id=0) > >>> Fatal error in MPI_Comm_create: Other MPI error, error stack: > >>> MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000002, > >>> group=0xc8000020, new_comm=0x7fff639ae614) failed > >>> MPI_Comm_create(577).................: > >>> MPIR_Comm_create_intra(241)..........: > >>> MPIR_Get_contextid(799)..............: > >>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > >>> because of fragmentation (2002/2048 free on this process; > ignore_id=0 > >>> > >>> Is there a way to fix that ? > >>> > >>> Thanks, > >>> Reem > >>> _______________________________________________ > >>> discuss mailing list discuss(a)mpich.org <mailto:[email protected]> > >>> To manage subscription options or unsubscribe: > >>> https://lists.mpich.org/mailman/listinfo/discuss > >> > >> > >> _______________________________________________ > >> discuss mailing list discuss(a)mpich.org <mailto:[email protected]> > >> To manage subscription options or unsubscribe: > >> https://lists.mpich.org/mailman/listinfo/discuss > >> > > > > -- > > Pavan Balaji > > http://www.mcs.anl.gov/~balaji > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org <mailto:[email protected]> > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > > > -- > Jeff Hammond > jeff.science(a)gmail.com <mailto:[email protected]> > > > ------------------------------ > > Message: 3 > Date: Mon, 8 Jul 2013 16:39:03 +0100 > From: Reem Alraddadi <raba500(a)york.ac.uk <mailto:[email protected]>> > To: discuss(a)mpich.org <mailto:[email protected]> > Subject: [mpich-discuss] mpich on Mac os x > Message-ID: > > <CANKsntEJR0b_TA-tbdBO3fOV2O8PkwTG82a718qZPd3FnEBT3w(a)mail.gmail.com > <mailto:CANKsntEJR0b_TA-tbdBO3fOV2O8PkwTG82a718qZPd3FnEBT3w@mail.gmail.com>> > Content-Type: text/plain; charset="iso-8859-1" > > Hi Pavan, > > Could you explain more what do you mean by that? I am sorry I am still > beginner in MPICH. > > Hi Wesley, > apologize about my first confused reply I have forgotten to change the > subject. However, as I mentioned I wrote the following line: > > mpirun --np 4 --env MPIR_PARAM_CTXID_EAGER_SIZE 1 ./flash4 > > I am wondering if I made any mistake here as the problem is still > the same > > Thanks, > Reem > > > > Message: 1 > > Date: Mon, 08 Jul 2013 09:17:00 -0500 > > From: Pavan Balaji <balaji(a)mcs.anl.gov <mailto:[email protected]>> > > To: discuss(a)mpich.org <mailto:[email protected]> > > Subject: Re: [mpich-discuss] mpich on Mac os x > > Message-ID: <51DAC9DC.5000502(a)mcs.anl.gov > <mailto:[email protected]>> > > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > > > > > Also make sure you are freeing unused communicators. > > > > -- Pavan > > > > On 07/08/2013 08:14 AM, Wesley Bland wrote: > > > It seems that you're creating more communicators than MPICH can > handle. > > It's possible that you might be able to get around this by > setting the > > environment variable MPIR_PARAM_CTXID_EAGER_SIZE to something > smaller than > > its default (which is 2). That frees up a few more communicators, > but there > > is a pathological case where even with fewer communicators than > the max, > > MPICH won't be able to agree on a new communicator id when > needed. Try > > changing that environment variable and see if that fixes things. > > > > > > Wesley > > > > > > On Jul 8, 2013, at 5:33 AM, Reem Alraddadi <raba500(a)york.ac.uk > <mailto:[email protected]>> wrote: > > > > > >> Hi all, > > >> I am using mpich-3.0.4 on Mac os x version 10.7.5 to run FLASH > code. It > > works fine in the beginning of the run and then I got the > following error: > > >> > > >> Fatal error in MPI_Comm_create: Other MPI error, error stack: > > >> MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000002, > > group=0xc8001349, new_comm=0x7fff606a8614) failed > > >> MPI_Comm_create(577).................: > > >> MPIR_Comm_create_intra(241)..........: > > >> MPIR_Get_contextid(799)..............: > > >> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > > because of fragmentation (169/2048 free on this process; ignore_id=0) > > >> Fatal error in MPI_Comm_create: Other MPI error, error stack: > > >> MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000002, > > group=0xc80012b6, new_comm=0x7fff670cc614) failed > > >> MPI_Comm_create(577).................: > > >> MPIR_Comm_create_intra(241)..........: > > >> MPIR_Get_contextid(799)..............: > > >> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > > because of fragmentation (316/2048 free on this process; ignore_id=0) > > >> Fatal error in MPI_Comm_create: Other MPI error, error stack: > > >> MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000004, > > group=0xc800000e, new_comm=0x7fff629d5614) failed > > >> MPI_Comm_create(577).................: > > >> MPIR_Comm_create_intra(241)..........: > > >> MPIR_Get_contextid(799)..............: > > >> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > > because of fragmentation (2020/2048 free on this process; > ignore_id=0) > > >> Fatal error in MPI_Comm_create: Other MPI error, error stack: > > >> MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000002, > > group=0xc8000020, new_comm=0x7fff639ae614) failed > > >> MPI_Comm_create(577).................: > > >> MPIR_Comm_create_intra(241)..........: > > >> MPIR_Get_contextid(799)..............: > > >> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > > because of fragmentation (2002/2048 free on this process; ignore_id=0 > > >> > > >> Is there a way to fix that ? > > >> > > >> Thanks, > > >> Reem > > >> _______________________________________________ > > >> discuss mailing list discuss(a)mpich.org <mailto:[email protected]> > > >> To manage subscription options or unsubscribe: > > >> https://lists.mpich.org/mailman/listinfo/discuss > > > > > > _______________________________________________ > > > discuss mailing list discuss(a)mpich.org <mailto:[email protected]> > > > To manage subscription options or unsubscribe: > > > https://lists.mpich.org/mailman/listinfo/discuss > > > > > > > -- > > Pavan Balaji > > http://www.mcs.anl.gov/~balaji > > >

1 0

mpich on Mac os x
by Reem Alraddadi 08 Jul '13

08 Jul '13

Hi Wesley, How could I do that? as I read in the internet. I have to set: include "mpif.h" Call MPI_COMM_FREE I found where I can set that in FLASH code ( I have attached the file) So I set: integer, parameter :: FLASH_COMM = MPI_COMM_FREE but I got an error when I compile the code: Included at Driver_abortFlash.F90:34: integer, parameter :: FLASH_COMM = MPI_COMM_FREE 1 Error: Symbol 'flash_comm' at (1) already has basic type of INTEGER make: *** [Driver_abortFlash.o] Error 1 the code used MPI_COMM_WORLD. So is there a way to fix that? Thanks, Reem Date: Mon, 8 Jul 2013 10:25:38 -0500 > From: Wesley Bland <wbland(a)mcs.anl.gov> > To: discuss(a)mpich.org > Subject: Re: [mpich-discuss] discuss Digest, Vol 9, Issue 13 > Message-ID: <9A8EEBB2-72BB-429B-9368-150983B1C283(a)mcs.anl.gov> > Content-Type: text/plain; charset="iso-8859-1" > > That's the correct way to set that environment variable, but I would have > to agree with Pavan here. If you're running into the cap on context id's > (especially with so few processes) it would seem that you're creating a lot > of communicators that you probably don't need. Are you sure that you're > freeing them correctly after use? > > Wesley > > On Jul 8, 2013, at 10:05 AM, Reem Alraddadi <raba500(a)york.ac.uk> wrote: > > > Hi Wesley. > > I wrote the following: > > mpirun --np 4 --env MPIR_PARAM_CTXID_EAGER_SIZE 1 ./flash4 > > but the error still the same. Did I do it in the wrong way?? > > > > Thanks, > > Reem > > > > Message: 5 > > Date: Mon, 8 Jul 2013 08:14:48 -0500 > > From: Wesley Bland <wbland(a)mcs.anl.gov> > > To: discuss(a)mpich.org > > Subject: Re: [mpich-discuss] mpich on Mac os x > > Message-ID: <8DC984B2-4E4B-4BFE-806E-203463A7A4E4(a)mcs.anl.gov> > > Content-Type: text/plain; charset=iso-8859-1 > > > > It seems that you're creating more communicators than MPICH can handle. > It's possible that you might be able to get around this by setting the > environment variable MPIR_PARAM_CTXID_EAGER_SIZE to something smaller than > its default (which is 2). That frees up a few more communicators, but there > is a pathological case where even with fewer communicators than the max, > MPICH won't be able to agree on a new communicator id when needed. Try > changing that environment variable and see if that fixes things. > > > > Wesley > > > > On Jul 8, 2013, at 5:33 AM, Reem Alraddadi <raba500(a)york.ac.uk> wrote: > > > > > Hi all, > > > I am using mpich-3.0.4 on Mac os x version 10.7.5 to run FLASH code. > It works fine in the beginning of the run and then I got the following > error: > > > > > > Fatal error in MPI_Comm_create: Other MPI error, error stack: > > > MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000002, group=0xc8001349, new_comm=0x7fff606a8614) > failed > > > MPI_Comm_create(577).................: > > > MPIR_Comm_create_intra(241)..........: > > > MPIR_Get_contextid(799)..............: > > > MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > because of fragmentation (169/2048 free on this process; ignore_id=0) > > > Fatal error in MPI_Comm_create: Other MPI error, error stack: > > > MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000002, group=0xc80012b6, new_comm=0x7fff670cc614) > failed > > > MPI_Comm_create(577).................: > > > MPIR_Comm_create_intra(241)..........: > > > MPIR_Get_contextid(799)..............: > > > MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > because of fragmentation (316/2048 free on this process; ignore_id=0) > > > Fatal error in MPI_Comm_create: Other MPI error, error stack: > > > MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000004, group=0xc800000e, new_comm=0x7fff629d5614) > failed > > > MPI_Comm_create(577).................: > > > MPIR_Comm_create_intra(241)..........: > > > MPIR_Get_contextid(799)..............: > > > MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > because of fragmentation (2020/2048 free on this process; ignore_id=0) > > > Fatal error in MPI_Comm_create: Other MPI error, error stack: > > > MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000002, group=0xc8000020, new_comm=0x7fff639ae614) > failed > > > MPI_Comm_create(577).................: > > > MPIR_Comm_create_intra(241)..........: > > > MPIR_Get_contextid(799)..............: > > > MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > because of fragmentation (2002/2048 free on this process; ignore_id=0 > > > > > > Is there a way to fix that ? > > > > > > Thanks, > > > Reem > > On Mon, Jul 8, 2013 at 9:17 AM, Pavan Balaji <balaji(a)mcs.anl.gov> wrote: > > > > Also make sure you are freeing unused communicators. > > > > -- Pavan > > > > On 07/08/2013 08:14 AM, Wesley Bland wrote: > >> > >> It seems that you're creating more communicators than MPICH can handle. > >> It's possible that you might be able to get around this by setting the > >> environment variable MPIR_PARAM_CTXID_EAGER_SIZE to something smaller > than > >> its default (which is 2). That frees up a few more communicators, but > there > >> is a pathological case where even with fewer communicators than the max, > >> MPICH won't be able to agree on a new communicator id when needed. Try > >> changing that environment variable and see if that fixes things. > >> > >> Wesley > >> > >> On Jul 8, 2013, at 5:33 AM, Reem Alraddadi <raba500(a)york.ac.uk> wrote: > >> > >>> Hi all, > >>> I am using mpich-3.0.4 on Mac os x version 10.7.5 to run FLASH code. It > >>> works fine in the beginning of the run and then I got the following > error: > >>> > >>> Fatal error in MPI_Comm_create: Other MPI error, error stack: > >>> MPI_Comm_create(600).................: MPI_Comm_create(comm=0x84000002, > >>> group=0xc8001349, new_comm=0x7fff606a8614) failed > >>> MPI_Comm_create(577).................: > >>> MPIR_Comm_create_intra(241)..........: > >>> MPIR_Get_contextid(799)..............: > >>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > >>> because of fragmentation (169/2048 free on this process; ignore_id=0) > >>> Fatal error in MPI_Comm_create: Other MPI error, error stack: > >>> MPI_Comm_create(600).................: MPI_Comm_create(comm=0x84000002, > >>> group=0xc80012b6, new_comm=0x7fff670cc614) failed > >>> MPI_Comm_create(577).................: > >>> MPIR_Comm_create_intra(241)..........: > >>> MPIR_Get_contextid(799)..............: > >>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > >>> because of fragmentation (316/2048 free on this process; ignore_id=0) > >>> Fatal error in MPI_Comm_create: Other MPI error, error stack: > >>> MPI_Comm_create(600).................: MPI_Comm_create(comm=0x84000004, > >>> group=0xc800000e, new_comm=0x7fff629d5614) failed > >>> MPI_Comm_create(577).................: > >>> MPIR_Comm_create_intra(241)..........: > >>> MPIR_Get_contextid(799)..............: > >>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > >>> because of fragmentation (2020/2048 free on this process; ignore_id=0) > >>> Fatal error in MPI_Comm_create: Other MPI error, error stack: > >>> MPI_Comm_create(600).................: MPI_Comm_create(comm=0x84000002, > >>> group=0xc8000020, new_comm=0x7fff639ae614) failed > >>> MPI_Comm_create(577).................: > >>> MPIR_Comm_create_intra(241)..........: > >>> MPIR_Get_contextid(799)..............: > >>> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > >>> because of fragmentation (2002/2048 free on this process; ignore_id=0 > >>> > >>> Is there a way to fix that ? > >>> > >>> Thanks, > >>> Reem > >>> _______________________________________________ > >>> 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 > >> > > > > -- > > Pavan Balaji > > http://www.mcs.anl.gov/~balaji > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > > > -- > Jeff Hammond > jeff.science(a)gmail.com > > > ------------------------------ > > Message: 3 > Date: Mon, 8 Jul 2013 16:39:03 +0100 > From: Reem Alraddadi <raba500(a)york.ac.uk> > To: discuss(a)mpich.org > Subject: [mpich-discuss] mpich on Mac os x > Message-ID: > < > CANKsntEJR0b_TA-tbdBO3fOV2O8PkwTG82a718qZPd3FnEBT3w(a)mail.gmail.com> > Content-Type: text/plain; charset="iso-8859-1" > > Hi Pavan, > > Could you explain more what do you mean by that? I am sorry I am still > beginner in MPICH. > > Hi Wesley, > apologize about my first confused reply I have forgotten to change the > subject. However, as I mentioned I wrote the following line: > > mpirun --np 4 --env MPIR_PARAM_CTXID_EAGER_SIZE 1 ./flash4 > > I am wondering if I made any mistake here as the problem is still the same > > Thanks, > Reem > > > > Message: 1 > > Date: Mon, 08 Jul 2013 09:17:00 -0500 > > From: Pavan Balaji <balaji(a)mcs.anl.gov> > > To: discuss(a)mpich.org > > Subject: Re: [mpich-discuss] mpich on Mac os x > > Message-ID: <51DAC9DC.5000502(a)mcs.anl.gov> > > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > > > > > Also make sure you are freeing unused communicators. > > > > -- Pavan > > > > On 07/08/2013 08:14 AM, Wesley Bland wrote: > > > It seems that you're creating more communicators than MPICH can handle. > > It's possible that you might be able to get around this by setting the > > environment variable MPIR_PARAM_CTXID_EAGER_SIZE to something smaller > than > > its default (which is 2). That frees up a few more communicators, but > there > > is a pathological case where even with fewer communicators than the max, > > MPICH won't be able to agree on a new communicator id when needed. Try > > changing that environment variable and see if that fixes things. > > > > > > Wesley > > > > > > On Jul 8, 2013, at 5:33 AM, Reem Alraddadi <raba500(a)york.ac.uk> wrote: > > > > > >> Hi all, > > >> I am using mpich-3.0.4 on Mac os x version 10.7.5 to run FLASH code. > It > > works fine in the beginning of the run and then I got the following > error: > > >> > > >> Fatal error in MPI_Comm_create: Other MPI error, error stack: > > >> MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000002, > > group=0xc8001349, new_comm=0x7fff606a8614) failed > > >> MPI_Comm_create(577).................: > > >> MPIR_Comm_create_intra(241)..........: > > >> MPIR_Get_contextid(799)..............: > > >> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > > because of fragmentation (169/2048 free on this process; ignore_id=0) > > >> Fatal error in MPI_Comm_create: Other MPI error, error stack: > > >> MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000002, > > group=0xc80012b6, new_comm=0x7fff670cc614) failed > > >> MPI_Comm_create(577).................: > > >> MPIR_Comm_create_intra(241)..........: > > >> MPIR_Get_contextid(799)..............: > > >> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > > because of fragmentation (316/2048 free on this process; ignore_id=0) > > >> Fatal error in MPI_Comm_create: Other MPI error, error stack: > > >> MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000004, > > group=0xc800000e, new_comm=0x7fff629d5614) failed > > >> MPI_Comm_create(577).................: > > >> MPIR_Comm_create_intra(241)..........: > > >> MPIR_Get_contextid(799)..............: > > >> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > > because of fragmentation (2020/2048 free on this process; ignore_id=0) > > >> Fatal error in MPI_Comm_create: Other MPI error, error stack: > > >> MPI_Comm_create(600).................: > MPI_Comm_create(comm=0x84000002, > > group=0xc8000020, new_comm=0x7fff639ae614) failed > > >> MPI_Comm_create(577).................: > > >> MPIR_Comm_create_intra(241)..........: > > >> MPIR_Get_contextid(799)..............: > > >> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > > because of fragmentation (2002/2048 free on this process; ignore_id=0 > > >> > > >> Is there a way to fix that ? > > >> > > >> Thanks, > > >> Reem > > >> _______________________________________________ > > >> 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 > > > > > > > -- > > Pavan Balaji > > http://www.mcs.anl.gov/~balaji > > >

1 0

Re: [mpich-discuss] "undefined references"
by Jeff Hammond 08 Jul '13

08 Jul '13

I looked through the code and suspect Siesta is at fault but since they rolled their own MPI F90 wrappers and I know very little about F90, I cannot debug it. You will see that there are no wrappers for send, recv, bcast or reduce (the functions for which you get errors) in ./Src/MPI/mpi.F. There are references to these in ./Src/MPI/mpi__*.f90 but I don't understand what's going on very well. It also appears that these files are from a prototype implementation of the F90 module for MPI 1.1 written in 1996. I don't know if that was before the MPI standard defined F90 bindings, but it surely does now and applications should use those instead. The other thing you can do is turn off MPI in the utility code since it might not be necessary there. Jeff On Mon, Jul 8, 2013 at 11:09 AM, Hana Milani <hana.milani(a)yahoo.com> wrote: > yes, apologies > > ________________________________ > From: Hana Milani <hana.milani(a)yahoo.com> > To: Pavan Balaji <balaji(a)mcs.anl.gov> > Sent: Monday, July 8, 2013 7:53 PM > > Subject: Re: [mpich-discuss] "undefined references" > > yes Pavan, But it's a long time I have not received any response > > > ________________________________ > From: Pavan Balaji <balaji(a)mcs.anl.gov> > To: Hana Milani <hana.milani(a)yahoo.com>; discuss(a)mpich.org > Sent: Monday, July 8, 2013 7:03 PM > > Subject: Re: [mpich-discuss] "undefined references" > > > This doesn't look like an MPI error. Did you contact the authors of > whatever package you are trying to compile? > > -- Pavan > > On 07/08/2013 08:25 AM, Hana Milani wrote: >> Hello, >> >> I have compiled a code (siesta) by mpich2, blacs, lapack, blas and >> scalapack. It works fine under grid computation. But when I want to >> "make" a utility, I receive the following errors: >> >> [Home@localhost Gen-basis]$ make >> /usr/local/bin/mpif90 -o gen-basis \ >> alloc.o basis_types.o precision.o recipes.o parallel.o >> parsing.o xcmod.o basis_io.o chemical.o atm_transfer.o atm_types.o >> atmparams.o old_atmfuncs.o radial.o io.o paste.o flib_spline.o >> bessph.o m_filter.o basis_specs.o atom.o memoryinfo.o memory.o >> periodic_table.o pseudopotential.o pxf.o dot.o xc.o atom_options.o >> arw.o sys.o timer.o xml.o m_walltime.o gen-basis.o nag.o libfdf.a >> /home/Home/Packages/scalapack-2.0.2/libscalapack.a >> /home/Home/Packages/BLACS/LIB/blacsCinit_MPI-LINUX-0.a >> /home/Home/Packages/BLACS/LIB/blacsF77init_MPI-LINUX-0.a >> /home/Home/Packages/BLACS/LIB/blacs_MPI-LINUX-0.a >> /home/Home/Packages/lapack-3.4.2/liblapack.a >> /home/Home/Packages/BLAS/blas_LINUX.a >> alloc.o: In function `print_report': >> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1705: >> undefined reference to `__mpi__r8_s_MOD_mpi_reduce_t' >> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1707: >> undefined reference to `__mpi__r8_v_MOD_mpi_reduce_t' >> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1711: >> undefined reference to `__mpi__integer_s_MOD_mpi_bcast_t' >> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1723: >> undefined reference to `__mpi__r8_s_MOD_mpi_recv_t' >> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1718: >> undefined reference to `__mpi__r8_s_MOD_mpi_send_t' >> alloc.o: In function `__alloc_MOD_alloc_report': >> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:318: >> undefined reference to `__mpi__integer_s_MOD_mpi_bcast_t' >> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:319: >> undefined reference to `__mpi__integer_s_MOD_mpi_bcast_t' >> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:320: >> undefined reference to `__mpi__character_s_MOD_mpi_bcast_t' >> xcmod.o: In function `__xcmod_MOD_setxc': >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:136: >> undefined >> reference to `__mpi__integer_s_MOD_mpi_bcast_t' >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:138: >> undefined >> reference to `__mpi__character_v_MOD_mpi_bcast_t' >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:140: >> undefined >> reference to `__mpi__character_v_MOD_mpi_bcast_t' >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:142: >> undefined >> reference to `__mpi__r8_v_MOD_mpi_bcast_t' >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:144: >> undefined >> reference to `__mpi__r8_v_MOD_mpi_bcast_t' >> memory.o: In function `printmemory_': >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/memory.F:144: >> undefined reference to `__mpi__integer_s_MOD_mpi_allreduce_t' >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/memory.F:167: >> undefined reference to `__mpi__integer_sv_MOD_mpi_gather_t' >> timer.o: In function `timer1_': >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/timer.F:117: >> undefined >> reference to `__mpi__r8_s_MOD_mpi_allreduce_t' >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/timer.F:164: >> undefined >> reference to `__mpi__r8_v_MOD_mpi_reduce_t' >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/timer.F:139: >> undefined >> reference to `__mpi__r8_v_MOD_mpi_reduce_t' >> collect2: error: ld returned 1 exit status >> make: *** [gen-basis] Error 1 >> >> I will be really grateful if you kindly help me to solve this problem. >> >> >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss >> > > -- > Pavan Balaji > http://www.mcs.anl.gov/~balaji > > > > > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss -- Jeff Hammond jeff.science(a)gmail.com

1 0

Re: [mpich-discuss] first exit problems
by Antonio J. Peña 08 Jul '13

08 Jul '13

Maybe you could use a root process to drive others'. Let's say your root is rank 0. All ranks just communicate with rank 0. If any process finds the solution, it notifies rank 0, who broadcasts the exit message. The root process may just discard a second notify from other process who found a solution before receiving the exit message. This way the root process can handle ties and avoid race conditions. Antonio On Monday, July 08, 2013 11:02:48 AM Gideon Simpson wrote: > Hi, I was wondering if someone had a good MPI solution to the following > first passage problem. Suppose I have N workers, each using a Monte Carlo > method to solve a problem. An example would be if I wanted to find the > exit distribution in time and space of a stochastic differential equation > from some compact set in space. > > Assuming that each worker is properly receiving an independent stream of > pseudo random numbers, and they are computing asynchronously, when one > worker finds a solution, he must notify the others that they can cease > working, and the program can continue. I have in mind something like: > > while (!local_exit & !global_exit){ > > /*Check if a message has been received*/ > MPI_Iprobe(...,&global_exit,...); > > /*If no one else finished, continue the algorithm */ > if(!global_exit){ > /* Step algorithm */ > local_exit = exit_test(...); > } > } > > if(local_exit){ > > /* Alert all other workers */ > for(...){ > MPI_Isend(...); > } > > } > else{ > > /* Receive the message that was sent from the other worker */ > > MPI_Recv(...); > } > > However, I believe this leads to a race condition because if processor 0 > finds and exit, processor 1 might also find one before processor 0 has a > chance to do an MPI_Isend() to 1. Assuming that, for the moment, I am not > too concerned about tie breaking, I would really like to find a robust > solution to this problem, as I have a lot of first exit/first passage type > problems that could be solved asynchronously. > > Thanks for any suggestions, > > -gideon > > _______________________________________________ > 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] "undefined references"
by Hana Milani 08 Jul '13

08 Jul '13

yes, apologies ________________________________ From: Hana Milani <hana.milani(a)yahoo.com> To: Pavan Balaji <balaji(a)mcs.anl.gov> Sent: Monday, July 8, 2013 7:53 PM Subject: Re: [mpich-discuss] "undefined references" yes Pavan, But it's a long time I have not received any response ________________________________ From: Pavan Balaji <balaji(a)mcs.anl.gov> To: Hana Milani <hana.milani(a)yahoo.com>; discuss(a)mpich.org Sent: Monday, July 8, 2013 7:03 PM Subject: Re: [mpich-discuss] "undefined references" This doesn't look like an MPI error. Did you contact the authors of whatever package you are trying to compile? -- Pavan On 07/08/2013 08:25 AM, Hana Milani wrote: > Hello, > > I have compiled a code (siesta) by mpich2, blacs, lapack, blas and > scalapack. It works fine under grid computation. But when I want to > "make" a utility, I receive the following errors: > > [Home@localhost Gen-basis]$ make > /usr/local/bin/mpif90 -o gen-basis \ > alloc.o basis_types.o precision.o recipes.o parallel.o > parsing.o xcmod.o basis_io.o chemical.o atm_transfer.o atm_types.o > atmparams.o old_atmfuncs.o radial.o io.o paste.o flib_spline.o > bessph.o m_filter.o basis_specs.o atom.o memoryinfo.o memory.o > periodic_table.o pseudopotential.o pxf.o dot.o xc.o atom_options.o > arw.o sys.o timer.o xml.o m_walltime.o gen-basis.o nag.o libfdf.a > /home/Home/Packages/scalapack-2.0.2/libscalapack.a > /home/Home/Packages/BLACS/LIB/blacsCinit_MPI-LINUX-0.a > /home/Home/Packages/BLACS/LIB/blacsF77init_MPI-LINUX-0.a > /home/Home/Packages/BLACS/LIB/blacs_MPI-LINUX-0.a > /home/Home/Packages/lapack-3.4.2/liblapack.a > /home/Home/Packages/BLAS/blas_LINUX.a > alloc.o: In function `print_report': > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1705: > undefined reference to `__mpi__r8_s_MOD_mpi_reduce_t' > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1707: > undefined reference to `__mpi__r8_v_MOD_mpi_reduce_t' > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1711: > undefined reference to `__mpi__integer_s_MOD_mpi_bcast_t' > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1723: > undefined reference to `__mpi__r8_s_MOD_mpi_recv_t' > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1718: > undefined reference to `__mpi__r8_s_MOD_mpi_send_t' > alloc.o: In function `__alloc_MOD_alloc_report': > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:318: > undefined reference to `__mpi__integer_s_MOD_mpi_bcast_t' > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:319: > undefined reference to `__mpi__integer_s_MOD_mpi_bcast_t' > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:320: > undefined reference to `__mpi__character_s_MOD_mpi_bcast_t' > xcmod.o: In function `__xcmod_MOD_setxc': > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:136: undefined > reference to `__mpi__integer_s_MOD_mpi_bcast_t' > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:138: undefined > reference to `__mpi__character_v_MOD_mpi_bcast_t' > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:140: undefined > reference to `__mpi__character_v_MOD_mpi_bcast_t' > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:142: undefined > reference to `__mpi__r8_v_MOD_mpi_bcast_t' > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:144: undefined > reference to `__mpi__r8_v_MOD_mpi_bcast_t' > memory.o: In function `printmemory_': > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/memory.F:144: > undefined reference to `__mpi__integer_s_MOD_mpi_allreduce_t' > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/memory.F:167: > undefined reference to `__mpi__integer_sv_MOD_mpi_gather_t' > timer.o: In function `timer1_': > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/timer.F:117: undefined > reference to `__mpi__r8_s_MOD_mpi_allreduce_t' > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/timer.F:164: undefined > reference to `__mpi__r8_v_MOD_mpi_reduce_t' > /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/timer.F:139: undefined > reference to `__mpi__r8_v_MOD_mpi_reduce_t' > collect2: error: ld returned 1 exit status > make: *** [gen-basis] Error 1 > > I will be really grateful if you kindly help me to solve this problem. > > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss > -- Pavan Balaji http://www.mcs.anl.gov/~balaji

1 0

Fw: "undefined references"
by Hana Milani 08 Jul '13

08 Jul '13

Dear Jeff, I have compiled siesta with no problem. It is the Gen-Basis in the Utils that is making a mess for me. The mpif90 is working fine executing the code. Regards, ________________________________ From: Jeff Hammond <jeff.science(a)gmail.com> To: discuss(a)mpich.org Cc: Hana Milani <hana.milani(a)yahoo.com> Sent: Monday, July 8, 2013 7:52 PM Subject: Re: [mpich-discuss] "undefined references" I built Siesta against MPICH a week or two ago, albeit on BGQ (https://wiki.alcf.anl.gov/parts/index.php/SIESTA) and had no issues. What is the result of "mpif90 -show"? I wonder if you are mixing MPI builds or otherwise have a problem with your F90 configuration. Jeff On Mon, Jul 8, 2013 at 9:33 AM, Pavan Balaji <balaji(a)mcs.anl.gov> wrote: > > This doesn't look like an MPI error. Did you contact the authors of > whatever package you are trying to compile? > > -- Pavan > > On 07/08/2013 08:25 AM, Hana Milani wrote: >> >> Hello, >> >> I have compiled a code (siesta) by mpich2, blacs, lapack, blas and >> scalapack. It works fine under grid computation. But when I want to >> "make" a utility, I receive the following errors: >> >> [Home@localhost Gen-basis]$ make >> /usr/local/bin/mpif90 -o gen-basis \ >> alloc.o basis_types.o precision.o recipes.o parallel.o >> parsing.o xcmod.o basis_io.o chemical.o atm_transfer.o atm_types.o >> atmparams.o old_atmfuncs.o radial.o io.o paste.o flib_spline.o >> bessph.o m_filter.o basis_specs.o atom.o memoryinfo.o memory.o >> periodic_table.o pseudopotential.o pxf.o dot.o xc.o atom_options.o >> arw.o sys.o timer.o xml.o m_walltime.o gen-basis.o nag.o libfdf.a >> /home/Home/Packages/scalapack-2.0.2/libscalapack.a >> /home/Home/Packages/BLACS/LIB/blacsCinit_MPI-LINUX-0.a >> /home/Home/Packages/BLACS/LIB/blacsF77init_MPI-LINUX-0.a >> /home/Home/Packages/BLACS/LIB/blacs_MPI-LINUX-0.a >> /home/Home/Packages/lapack-3.4.2/liblapack.a >> /home/Home/Packages/BLAS/blas_LINUX.a >> alloc.o: In function `print_report': >> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1705: >> undefined reference to `__mpi__r8_s_MOD_mpi_reduce_t' >> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1707: >> undefined reference to `__mpi__r8_v_MOD_mpi_reduce_t' >> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1711: >> undefined reference to `__mpi__integer_s_MOD_mpi_bcast_t' >> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1723: >> undefined reference to `__mpi__r8_s_MOD_mpi_recv_t' >> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:1718: >> undefined reference to `__mpi__r8_s_MOD_mpi_send_t' >> alloc.o: In function `__alloc_MOD_alloc_report': >> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:318: >> undefined reference to `__mpi__integer_s_MOD_mpi_bcast_t' >> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:319: >> undefined reference to `__mpi__integer_s_MOD_mpi_bcast_t' >> >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/alloc.F90:320: >> undefined reference to `__mpi__character_s_MOD_mpi_bcast_t' >> xcmod.o: In function `__xcmod_MOD_setxc': >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:136: >> undefined >> reference to `__mpi__integer_s_MOD_mpi_bcast_t' >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:138: >> undefined >> reference to `__mpi__character_v_MOD_mpi_bcast_t' >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:140: >> undefined >> reference to `__mpi__character_v_MOD_mpi_bcast_t' >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:142: >> undefined >> reference to `__mpi__r8_v_MOD_mpi_bcast_t' >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/xcmod.F:144: >> undefined >> reference to `__mpi__r8_v_MOD_mpi_bcast_t' >> memory.o: In function `printmemory_': >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/memory.F:144: >> undefined reference to `__mpi__integer_s_MOD_mpi_allreduce_t' >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/memory.F:167: >> undefined reference to `__mpi__integer_sv_MOD_mpi_gather_t' >> timer.o: In function `timer1_': >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/timer.F:117: >> undefined >> reference to `__mpi__r8_s_MOD_mpi_allreduce_t' >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/timer.F:164: >> undefined >> reference to `__mpi__r8_v_MOD_mpi_reduce_t' >> /home/Home/Packages/siesta-TEST-3.2/Util/Gen-basis/../../Src/timer.F:139: >> undefined >> reference to `__mpi__r8_v_MOD_mpi_reduce_t' >> collect2: error: ld returned 1 exit status >> make: *** [gen-basis] Error 1 >> >> I will be really grateful if you kindly help me to solve this problem. >> >> >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss >> > > -- > Pavan Balaji > http://www.mcs.anl.gov/~balaji > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss -- Jeff Hammond jeff.science(a)gmail.com

1 0

first exit problems
by Gideon Simpson 08 Jul '13

08 Jul '13

Hi, I was wondering if someone had a good MPI solution to the following first passage problem. Suppose I have N workers, each using a Monte Carlo method to solve a problem. An example would be if I wanted to find the exit distribution in time and space of a stochastic differential equation from some compact set in space. Assuming that each worker is properly receiving an independent stream of pseudo random numbers, and they are computing asynchronously, when one worker finds a solution, he must notify the others that they can cease working, and the program can continue. I have in mind something like: while (!local_exit & !global_exit){ /*Check if a message has been received*/ MPI_Iprobe(...,&global_exit,...); /*If no one else finished, continue the algorithm */ if(!global_exit){ /* Step algorithm */ local_exit = exit_test(...); } } if(local_exit){ /* Alert all other workers */ for(...){ MPI_Isend(...); } } else{ /* Receive the message that was sent from the other worker */ MPI_Recv(...); } However, I believe this leads to a race condition because if processor 0 finds and exit, processor 1 might also find one before processor 0 has a chance to do an MPI_Isend() to 1. Assuming that, for the moment, I am not too concerned about tie breaking, I would really like to find a robust solution to this problem, as I have a lot of first exit/first passage type problems that could be solved asynchronously. Thanks for any suggestions, -gideon

1 0

Re: [mpich-discuss] mpich on Mac os x
by Wesley Bland 08 Jul '13

08 Jul '13

On Jul 8, 2013, at 10:39 AM, Reem Alraddadi <raba500(a)york.ac.uk> wrote: > Hi Pavan, > > Could you explain more what do you mean by that? I am sorry I am still beginner in MPICH. This isn't specific to MPICH. This would be the case with any MPI implementation. What he means is that you need to make sure that when you are done using a communicator, you call MPI_Comm_free(communicator) so MPICH can free the internal resources used by the communicator. If you never free your communicators, then you will eventually run out. There is a limited number of communicators available internally. > > Hi Wesley, > apologize about my first confused reply I have forgotten to change the subject. However, as I mentioned I wrote the following line: > > mpirun --np 4 --env MPIR_PARAM_CTXID_EAGER_SIZE 1 ./flash4 > > I am wondering if I made any mistake here as the problem is still the same Yes, that's the correct way to use it, so that means your problem is probably what we described above. > > Thanks, > Reem > > > > Message: 1 > Date: Mon, 08 Jul 2013 09:17:00 -0500 > From: Pavan Balaji <balaji(a)mcs.anl.gov> > To: discuss(a)mpich.org > Subject: Re: [mpich-discuss] mpich on Mac os x > Message-ID: <51DAC9DC.5000502(a)mcs.anl.gov> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > > Also make sure you are freeing unused communicators. > > -- Pavan > > On 07/08/2013 08:14 AM, Wesley Bland wrote: > > It seems that you're creating more communicators than MPICH can handle. It's possible that you might be able to get around this by setting the environment variable MPIR_PARAM_CTXID_EAGER_SIZE to something smaller than its default (which is 2). That frees up a few more communicators, but there is a pathological case where even with fewer communicators than the max, MPICH won't be able to agree on a new communicator id when needed. Try changing that environment variable and see if that fixes things. > > > > Wesley > > > > On Jul 8, 2013, at 5:33 AM, Reem Alraddadi <raba500(a)york.ac.uk> wrote: > > > >> Hi all, > >> I am using mpich-3.0.4 on Mac os x version 10.7.5 to run FLASH code. It works fine in the beginning of the run and then I got the following error: > >> > >> Fatal error in MPI_Comm_create: Other MPI error, error stack: > >> MPI_Comm_create(600).................: MPI_Comm_create(comm=0x84000002, group=0xc8001349, new_comm=0x7fff606a8614) failed > >> MPI_Comm_create(577).................: > >> MPIR_Comm_create_intra(241)..........: > >> MPIR_Get_contextid(799)..............: > >> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID because of fragmentation (169/2048 free on this process; ignore_id=0) > >> Fatal error in MPI_Comm_create: Other MPI error, error stack: > >> MPI_Comm_create(600).................: MPI_Comm_create(comm=0x84000002, group=0xc80012b6, new_comm=0x7fff670cc614) failed > >> MPI_Comm_create(577).................: > >> MPIR_Comm_create_intra(241)..........: > >> MPIR_Get_contextid(799)..............: > >> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID because of fragmentation (316/2048 free on this process; ignore_id=0) > >> Fatal error in MPI_Comm_create: Other MPI error, error stack: > >> MPI_Comm_create(600).................: MPI_Comm_create(comm=0x84000004, group=0xc800000e, new_comm=0x7fff629d5614) failed > >> MPI_Comm_create(577).................: > >> MPIR_Comm_create_intra(241)..........: > >> MPIR_Get_contextid(799)..............: > >> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID because of fragmentation (2020/2048 free on this process; ignore_id=0) > >> Fatal error in MPI_Comm_create: Other MPI error, error stack: > >> MPI_Comm_create(600).................: MPI_Comm_create(comm=0x84000002, group=0xc8000020, new_comm=0x7fff639ae614) failed > >> MPI_Comm_create(577).................: > >> MPIR_Comm_create_intra(241)..........: > >> MPIR_Get_contextid(799)..............: > >> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID because of fragmentation (2002/2048 free on this process; ignore_id=0 > >> > >> Is there a way to fix that ? > >> > >> Thanks, > >> Reem > >> _______________________________________________ > >> 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 > > > > -- > Pavan Balaji > http://www.mcs.anl.gov/~balaji > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss

1 0

mpich on Mac os x
by Reem Alraddadi 08 Jul '13

08 Jul '13

Hi Pavan, Could you explain more what do you mean by that? I am sorry I am still beginner in MPICH. Hi Wesley, apologize about my first confused reply I have forgotten to change the subject. However, as I mentioned I wrote the following line: mpirun --np 4 --env MPIR_PARAM_CTXID_EAGER_SIZE 1 ./flash4 I am wondering if I made any mistake here as the problem is still the same Thanks, Reem Message: 1 > Date: Mon, 08 Jul 2013 09:17:00 -0500 > From: Pavan Balaji <balaji(a)mcs.anl.gov> > To: discuss(a)mpich.org > Subject: Re: [mpich-discuss] mpich on Mac os x > Message-ID: <51DAC9DC.5000502(a)mcs.anl.gov> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > > Also make sure you are freeing unused communicators. > > -- Pavan > > On 07/08/2013 08:14 AM, Wesley Bland wrote: > > It seems that you're creating more communicators than MPICH can handle. > It's possible that you might be able to get around this by setting the > environment variable MPIR_PARAM_CTXID_EAGER_SIZE to something smaller than > its default (which is 2). That frees up a few more communicators, but there > is a pathological case where even with fewer communicators than the max, > MPICH won't be able to agree on a new communicator id when needed. Try > changing that environment variable and see if that fixes things. > > > > Wesley > > > > On Jul 8, 2013, at 5:33 AM, Reem Alraddadi <raba500(a)york.ac.uk> wrote: > > > >> Hi all, > >> I am using mpich-3.0.4 on Mac os x version 10.7.5 to run FLASH code. It > works fine in the beginning of the run and then I got the following error: > >> > >> Fatal error in MPI_Comm_create: Other MPI error, error stack: > >> MPI_Comm_create(600).................: MPI_Comm_create(comm=0x84000002, > group=0xc8001349, new_comm=0x7fff606a8614) failed > >> MPI_Comm_create(577).................: > >> MPIR_Comm_create_intra(241)..........: > >> MPIR_Get_contextid(799)..............: > >> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > because of fragmentation (169/2048 free on this process; ignore_id=0) > >> Fatal error in MPI_Comm_create: Other MPI error, error stack: > >> MPI_Comm_create(600).................: MPI_Comm_create(comm=0x84000002, > group=0xc80012b6, new_comm=0x7fff670cc614) failed > >> MPI_Comm_create(577).................: > >> MPIR_Comm_create_intra(241)..........: > >> MPIR_Get_contextid(799)..............: > >> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > because of fragmentation (316/2048 free on this process; ignore_id=0) > >> Fatal error in MPI_Comm_create: Other MPI error, error stack: > >> MPI_Comm_create(600).................: MPI_Comm_create(comm=0x84000004, > group=0xc800000e, new_comm=0x7fff629d5614) failed > >> MPI_Comm_create(577).................: > >> MPIR_Comm_create_intra(241)..........: > >> MPIR_Get_contextid(799)..............: > >> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > because of fragmentation (2020/2048 free on this process; ignore_id=0) > >> Fatal error in MPI_Comm_create: Other MPI error, error stack: > >> MPI_Comm_create(600).................: MPI_Comm_create(comm=0x84000002, > group=0xc8000020, new_comm=0x7fff639ae614) failed > >> MPI_Comm_create(577).................: > >> MPIR_Comm_create_intra(241)..........: > >> MPIR_Get_contextid(799)..............: > >> MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID > because of fragmentation (2002/2048 free on this process; ignore_id=0 > >> > >> Is there a way to fix that ? > >> > >> Thanks, > >> Reem > >> _______________________________________________ > >> 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 > > > > -- > Pavan Balaji > http://www.mcs.anl.gov/~balaji >

1 0

Re: [mpich-discuss] discuss Digest, Vol 9, Issue 13
by Wesley Bland 08 Jul '13

08 Jul '13

That's the correct way to set that environment variable, but I would have to agree with Pavan here. If you're running into the cap on context id's (especially with so few processes) it would seem that you're creating a lot of communicators that you probably don't need. Are you sure that you're freeing them correctly after use? Wesley On Jul 8, 2013, at 10:05 AM, Reem Alraddadi <raba500(a)york.ac.uk> wrote: > Hi Wesley. > I wrote the following: > mpirun --np 4 --env MPIR_PARAM_CTXID_EAGER_SIZE 1 ./flash4 > but the error still the same. Did I do it in the wrong way?? > > Thanks, > Reem > > Message: 5 > Date: Mon, 8 Jul 2013 08:14:48 -0500 > From: Wesley Bland <wbland(a)mcs.anl.gov> > To: discuss(a)mpich.org > Subject: Re: [mpich-discuss] mpich on Mac os x > Message-ID: <8DC984B2-4E4B-4BFE-806E-203463A7A4E4(a)mcs.anl.gov> > Content-Type: text/plain; charset=iso-8859-1 > > It seems that you're creating more communicators than MPICH can handle. It's possible that you might be able to get around this by setting the environment variable MPIR_PARAM_CTXID_EAGER_SIZE to something smaller than its default (which is 2). That frees up a few more communicators, but there is a pathological case where even with fewer communicators than the max, MPICH won't be able to agree on a new communicator id when needed. Try changing that environment variable and see if that fixes things. > > Wesley > > On Jul 8, 2013, at 5:33 AM, Reem Alraddadi <raba500(a)york.ac.uk> wrote: > > > Hi all, > > I am using mpich-3.0.4 on Mac os x version 10.7.5 to run FLASH code. It works fine in the beginning of the run and then I got the following error: > > > > Fatal error in MPI_Comm_create: Other MPI error, error stack: > > MPI_Comm_create(600).................: MPI_Comm_create(comm=0x84000002, group=0xc8001349, new_comm=0x7fff606a8614) failed > > MPI_Comm_create(577).................: > > MPIR_Comm_create_intra(241)..........: > > MPIR_Get_contextid(799)..............: > > MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID because of fragmentation (169/2048 free on this process; ignore_id=0) > > Fatal error in MPI_Comm_create: Other MPI error, error stack: > > MPI_Comm_create(600).................: MPI_Comm_create(comm=0x84000002, group=0xc80012b6, new_comm=0x7fff670cc614) failed > > MPI_Comm_create(577).................: > > MPIR_Comm_create_intra(241)..........: > > MPIR_Get_contextid(799)..............: > > MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID because of fragmentation (316/2048 free on this process; ignore_id=0) > > Fatal error in MPI_Comm_create: Other MPI error, error stack: > > MPI_Comm_create(600).................: MPI_Comm_create(comm=0x84000004, group=0xc800000e, new_comm=0x7fff629d5614) failed > > MPI_Comm_create(577).................: > > MPIR_Comm_create_intra(241)..........: > > MPIR_Get_contextid(799)..............: > > MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID because of fragmentation (2020/2048 free on this process; ignore_id=0) > > Fatal error in MPI_Comm_create: Other MPI error, error stack: > > MPI_Comm_create(600).................: MPI_Comm_create(comm=0x84000002, group=0xc8000020, new_comm=0x7fff639ae614) failed > > MPI_Comm_create(577).................: > > MPIR_Comm_create_intra(241)..........: > > MPIR_Get_contextid(799)..............: > > MPIR_Get_contextid_sparse_group(1146): Cannot allocate context ID because of fragmentation (2002/2048 free on this process; ignore_id=0 > > > > Is there a way to fix that ? > > > > Thanks, > > Reem > > _______________________________________________ > > 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