February 2015 - discuss - mailman3.cels.anl.gov

Re: [mpich-discuss] MPICH3 Problem
by Seo, Sangmin 03 Feb '15

03 Feb '15

Hi Abhishek, As the error message says, it looks that the node running the application doesn’t have libpgmp.so. Can you confirm whether the node has libpgmp.so and whether LD_LIBRARY_PATH is correctly set on the node in case that it has libpgmp.so? Best regards, Sangmin On Feb 3, 2015, at 5:31 PM, Abhishek Bhat <abhat(a)trinityconsultants.com<mailto:[email protected]>> wrote: HI All, I installed MPICH3 on a master terminal along with PGI Fortran. Then used MPICH3 and PGI to compile my software. When I run the program on the master terminal only I do not have any error messages but when I am trying to run them on one of the nodes, I am getting following message error – /home/Earth/MODELS/camx/src_611/CAMx.v6.11.MPICH3.pgfomp: error while loading shared libraries: libpgmp.so: cannot open shared object file: No such file or directory /home/Earth/MODELS/camx/src_611/CAMx.v6.11.MPICH3.pgfomp: error while loading shared libraries: libpgmp.so: cannot open shared object file: No such file or directory =================================================================================== = BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES = PID 5200 RUNNING AT node3 = EXIT CODE: 127 = CLEANING UP REMAINING PROCESSES = YOU CAN IGNORE THE BELOW CLEANUP MESSAGES =================================================================================== [proxy:0:0@Earth] HYD_pmcd_pmip_control_cmd_cb (/home/Earth/MODELS/mpi/mpich-3.1.3/src/pm/hydra/pm/pmiserv/pmip_cb.c:885): assert (!closed) failed [proxy:0:0@Earth] HYDT_dmxu_poll_wait_for_event (/home/Earth/MODELS/mpi/mpich-3.1.3/src/pm/hydra/tools/demux/demux_poll.c:76): callback returned error status [proxy:0:0@Earth] main (/home/Earth/MODELS/mpi/mpich-3.1.3/src/pm/hydra/pm/pmiserv/pmip.c:206): demux engine error waiting for event [mpiexec@Earth] HYDT_bscu_wait_for_completion (/home/Earth/MODELS/mpi/mpich-3.1.3/src/pm/hydra/tools/bootstrap/utils/bscu_wait.c:76): one of the processes terminated badly; aborting [mpiexec@Earth] HYDT_bsci_wait_for_completion (/home/Earth/MODELS/mpi/mpich-3.1.3/src/pm/hydra/tools/bootstrap/src/bsci_wait.c:23): launcher returned error waiting for completion [mpiexec@Earth] HYD_pmci_wait_for_completion (/home/Earth/MODELS/mpi/mpich-3.1.3/src/pm/hydra/pm/pmiserv/pmiserv_pmci.c:218): launcher returned error waiting for completion [mpiexec@Earth] main (/home/Earth/MODELS/mpi/mpich-3.1.3/src/pm/hydra/ui/mpich/mpiexec.c:344): process manager error waiting for completion The /home/Earth is shared and mapped on all nodes Any help is much appreciated. Thank You Abhishek …………………………………………………………………………………………………. Abhishek Bhat, PhD, EPI, Senior Consultant _________________________________________________________________________ The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. _________________________________________________________________________ _______________________________________________ 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 To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] question in MPI_Info_get
by Thakur, Rajeev 03 Feb '15

03 Feb '15

This bug is tracked in ticket #2225. https://trac.mpich.org/projects/mpich/ticket/2225 Rajeev On Feb 3, 2015, at 4:26 PM, William Gropp <wgropp(a)illinois.edu> wrote: > Thanks. Yes, I believe this is a bug. The implementation of MPI_Info_get in MPICH passes valuelen to the internal copy routine, yet it is clear that the definition in the standard for MPI_Info_get, it should pass valuelen+1 (it should also check for the error return code from the internal routine and return a value MPI error code if the buffer is too small. > > Bill > > On Feb 1, 2015, at 12:13 AM, Hajime Fujita <hfujita(a)uchicago.edu> wrote: > >> Hello, >> >> I came across some question about a behavior of MPI_Info_get in MPICH. >> >> If I knew the size of the value, what should I pass to MPI_Info_get as `valuelen`? >> Suppose I have 3-byte value in the info object, and I allocate 4-byte buffer including a space for null-termination. Then I think I should specify 3 as valuelen to retrieve the whole value, as the MPI-3 specifications says >> "In C, valuelen should be one less than the amount of allocated space to allow for the null terminator.” >> >> However, when I run the attached program in MPICH 3.1.1, I get: >> $ ./test >> key found: vaule=va # The last character ‘l’ is missing >> >> If I specify `sizeof(buff)` as valuelen, it shows “val” correctly. >> >> So apparently MPICH thinks valuelen includes null terminator. >> Am I missing anything, or any misunderstanding here? >> >> >> Thank you in advance, >> Hajime >> >> -- >> Hajime Fujita >> Postdoctoral Scholar, Large-Scale Systems Group >> Department of Computer Science, The University of Chicago >> http://www.cs.uchicago.edu/people/hfujita >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss >> <test.c> > > _______________________________________________ > 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] question in MPI_Info_get
by Thakur, Rajeev 03 Feb '15

03 Feb '15

I guess it may have been defined that way because in Fortran strings are not null-terminated. The definition in the standard allows the same value to be returned/passed in valuelen in both C and Fortran. Rajeev On Feb 3, 2015, at 5:28 PM, Junchao Zhang <jczhang(a)mcs.anl.gov> wrote: > OK, I guess MPI standard wants to have this kind of code. > > int valuelen; > MPI_Info_get_valuelen(..., &valuelen,...); > int buflen = valuelen + 1; > char* value = (char *)malloc(buflen); > MPI_Info_get(..., valuelen, value, ...); > > Though, from a user's point of view, I fell better if it was MPI_Info_get(..., buflen, value, ...), that is, using buflen to specify the length of buffer I pass to MPI. > > > --Junchao Zhang > > On Tue, Feb 3, 2015 at 4:26 PM, William Gropp <wgropp(a)illinois.edu> wrote: > Thanks. Yes, I believe this is a bug. The implementation of MPI_Info_get in MPICH passes valuelen to the internal copy routine, yet it is clear that the definition in the standard for MPI_Info_get, it should pass valuelen+1 (it should also check for the error return code from the internal routine and return a value MPI error code if the buffer is too small. > > Bill > > On Feb 1, 2015, at 12:13 AM, Hajime Fujita <hfujita(a)uchicago.edu> wrote: > >> Hello, >> >> I came across some question about a behavior of MPI_Info_get in MPICH. >> >> If I knew the size of the value, what should I pass to MPI_Info_get as `valuelen`? >> Suppose I have 3-byte value in the info object, and I allocate 4-byte buffer including a space for null-termination. Then I think I should specify 3 as valuelen to retrieve the whole value, as the MPI-3 specifications says >> "In C, valuelen should be one less than the amount of allocated space to allow for the null terminator.” >> >> However, when I run the attached program in MPICH 3.1.1, I get: >> $ ./test >> key found: vaule=va # The last character ‘l’ is missing >> >> If I specify `sizeof(buff)` as valuelen, it shows “val” correctly. >> >> So apparently MPICH thinks valuelen includes null terminator. >> Am I missing anything, or any misunderstanding here? >> >> >> Thank you in advance, >> Hajime >> >> -- >> Hajime Fujita >> Postdoctoral Scholar, Large-Scale Systems Group >> Department of Computer Science, The University of Chicago >> http://www.cs.uchicago.edu/people/hfujita >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss >> <test.c> > > > _______________________________________________ > 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

MPICH3 Problem
by Abhishek Bhat 03 Feb '15

03 Feb '15

HI All, I installed MPICH3 on a master terminal along with PGI Fortran. Then used MPICH3 and PGI to compile my software. When I run the program on the master terminal only I do not have any error messages but when I am trying to run them on one of the nodes, I am getting following message error - /home/Earth/MODELS/camx/src_611/CAMx.v6.11.MPICH3.pgfomp: error while loading shared libraries: libpgmp.so: cannot open shared object file: No such file or directory /home/Earth/MODELS/camx/src_611/CAMx.v6.11.MPICH3.pgfomp: error while loading shared libraries: libpgmp.so: cannot open shared object file: No such file or directory =================================================================================== = BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES = PID 5200 RUNNING AT node3 = EXIT CODE: 127 = CLEANING UP REMAINING PROCESSES = YOU CAN IGNORE THE BELOW CLEANUP MESSAGES =================================================================================== [proxy:0:0@Earth] HYD_pmcd_pmip_control_cmd_cb (/home/Earth/MODELS/mpi/mpich-3.1.3/src/pm/hydra/pm/pmiserv/pmip_cb.c:885): assert (!closed) failed [proxy:0:0@Earth] HYDT_dmxu_poll_wait_for_event (/home/Earth/MODELS/mpi/mpich-3.1.3/src/pm/hydra/tools/demux/demux_poll.c:76): callback returned error status [proxy:0:0@Earth] main (/home/Earth/MODELS/mpi/mpich-3.1.3/src/pm/hydra/pm/pmiserv/pmip.c:206): demux engine error waiting for event [mpiexec@Earth] HYDT_bscu_wait_for_completion (/home/Earth/MODELS/mpi/mpich-3.1.3/src/pm/hydra/tools/bootstrap/utils/bscu_wait.c:76): one of the processes terminated badly; aborting [mpiexec@Earth] HYDT_bsci_wait_for_completion (/home/Earth/MODELS/mpi/mpich-3.1.3/src/pm/hydra/tools/bootstrap/src/bsci_wait.c:23): launcher returned error waiting for completion [mpiexec@Earth] HYD_pmci_wait_for_completion (/home/Earth/MODELS/mpi/mpich-3.1.3/src/pm/hydra/pm/pmiserv/pmiserv_pmci.c:218): launcher returned error waiting for completion [mpiexec@Earth] main (/home/Earth/MODELS/mpi/mpich-3.1.3/src/pm/hydra/ui/mpich/mpiexec.c:344): process manager error waiting for completion The /home/Earth is shared and mapped on all nodes Any help is much appreciated. Thank You Abhishek ................................................................................................................ Abhishek Bhat, PhD, EPI, Senior Consultant -- _________________________________________________________________________ The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. _________________________________________________________________________ _______________________________________________ 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] question in MPI_Info_get
by Junchao Zhang 03 Feb '15

03 Feb '15

OK, I guess MPI standard wants to have this kind of code. int valuelen; MPI_Info_get_valuelen(..., &valuelen,...); int buflen = valuelen + 1; char* value = (char *)malloc(buflen); MPI_Info_get(..., valuelen, value, ...); Though, from a user's point of view, I fell better if it was MPI_Info_get(..., buflen, value, ...), that is, using buflen to specify the length of buffer I pass to MPI. --Junchao Zhang On Tue, Feb 3, 2015 at 4:26 PM, William Gropp <wgropp(a)illinois.edu> wrote: > Thanks. Yes, I believe this is a bug. The implementation of MPI_Info_get > in MPICH passes valuelen to the internal copy routine, yet it is clear that > the definition in the standard for MPI_Info_get, it should pass valuelen+1 > (it should also check for the error return code from the internal routine > and return a value MPI error code if the buffer is too small. > > Bill > > On Feb 1, 2015, at 12:13 AM, Hajime Fujita <hfujita(a)uchicago.edu> wrote: > > Hello, > > I came across some question about a behavior of MPI_Info_get in MPICH. > > If I knew the size of the value, what should I pass to MPI_Info_get as > `valuelen`? > Suppose I have 3-byte value in the info object, and I allocate 4-byte > buffer including a space for null-termination. Then I think I should > specify 3 as valuelen to retrieve the whole value, as the MPI-3 > specifications says > "In C, valuelen should be one less than the amount of allocated space to > allow for the null terminator.” > > However, when I run the attached program in MPICH 3.1.1, I get: > $ ./test > key found: vaule=va # The last character ‘l’ is missing > > If I specify `sizeof(buff)` as valuelen, it shows “val” correctly. > > So apparently MPICH thinks valuelen includes null terminator. > Am I missing anything, or any misunderstanding here? > > > Thank you in advance, > Hajime > > -- > Hajime Fujita > Postdoctoral Scholar, Large-Scale Systems Group > Department of Computer Science, The University of Chicago > http://www.cs.uchicago.edu/people/hfujita > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss > <test.c> > > > > _______________________________________________ > 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] question in MPI_Info_get
by William Gropp 03 Feb '15

03 Feb '15

Thanks. Yes, I believe this is a bug. The implementation of MPI_Info_get in MPICH passes valuelen to the internal copy routine, yet it is clear that the definition in the standard for MPI_Info_get, it should pass valuelen+1 (it should also check for the error return code from the internal routine and return a value MPI error code if the buffer is too small. Bill On Feb 1, 2015, at 12:13 AM, Hajime Fujita <hfujita(a)uchicago.edu> wrote: > Hello, > > I came across some question about a behavior of MPI_Info_get in MPICH. > > If I knew the size of the value, what should I pass to MPI_Info_get as `valuelen`? > Suppose I have 3-byte value in the info object, and I allocate 4-byte buffer including a space for null-termination. Then I think I should specify 3 as valuelen to retrieve the whole value, as the MPI-3 specifications says > "In C, valuelen should be one less than the amount of allocated space to allow for the null terminator.” > > However, when I run the attached program in MPICH 3.1.1, I get: > $ ./test > key found: vaule=va # The last character ‘l’ is missing > > If I specify `sizeof(buff)` as valuelen, it shows “val” correctly. > > So apparently MPICH thinks valuelen includes null terminator. > Am I missing anything, or any misunderstanding here? > > > Thank you in advance, > Hajime > > -- > Hajime Fujita > Postdoctoral Scholar, Large-Scale Systems Group > Department of Computer Science, The University of Chicago > http://www.cs.uchicago.edu/people/hfujita > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss > <test.c> _______________________________________________ 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] question in MPI_Info_get
by Hajime Fujita 03 Feb '15

03 Feb '15

Hi Junchao, Thank you for your reply. Okay, so it sounds like in the current dominant implementation I can treat valuelen as “length of the buffer, including a space for the null terminator”. If anyone could provide a good explanation of the spec it would be great. Best, Hajime > On Feb 2, 2015, at 2:54 PM, Junchao Zhang <jczhang(a)mcs.anl.gov> wrote: > > Hi, Hajime, > MPI spec also says "valuelen is the number of characters available in value". So it seems if you should pass valuelen = 4 if value is a buffer of 4 bytes. > I am also confused by "In C, valuelen should be one less than the amount of allocated space to allow for the null terminator". So I leave the question for others. > Interestingly, for MPI_Info_get_valuelen(), MPI spec says "The length returned in C does not include the end-of-string". > > BTW, I found MPICH and OpenMPI have the same behavior. > > --Junchao Zhang > > On Sun, Feb 1, 2015 at 12:13 AM, Hajime Fujita <hfujita(a)uchicago.edu> wrote: > Hello, > > I came across some question about a behavior of MPI_Info_get in MPICH. > > If I knew the size of the value, what should I pass to MPI_Info_get as `valuelen`? > Suppose I have 3-byte value in the info object, and I allocate 4-byte buffer including a space for null-termination. Then I think I should specify 3 as valuelen to retrieve the whole value, as the MPI-3 specifications says > "In C, valuelen should be one less than the amount of allocated space to allow for the null terminator.” > > However, when I run the attached program in MPICH 3.1.1, I get: > $ ./test > key found: vaule=va # The last character ‘l’ is missing > > If I specify `sizeof(buff)` as valuelen, it shows “val” correctly. > > So apparently MPICH thinks valuelen includes null terminator. > Am I missing anything, or any misunderstanding here? > > > > Thank you in advance, > Hajime > > -- > Hajime Fujita > Postdoctoral Scholar, Large-Scale Systems Group > Department of Computer Science, The University of Chicago > http://www.cs.uchicago.edu/people/hfujita > _______________________________________________ > 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] Problem in running parallel if two programs merged
by Huiwei Lu 03 Feb '15

03 Feb '15

Can you change the following to use mpi fortran compiler and try again? FC = mpifort CLMF = mpifort Also, what do you mean by "merge" these two programs? I didn't see any interaction between program_off.F90 and aadmn.F. -- Huiwei On Tue, Feb 3, 2015 at 7:42 AM, Muhammad Ashfaqur Rahman < ashfaq226(a)gmail.com> wrote: > Dear Huiwei Lu, > Many thanks for your email. I have attached here the source code of my > programs. These are two models. > The Makefile should be looked at. > > Regards > Ashfaq > > On Tue, Feb 3, 2015 at 4:32 PM, Muhammad Ashfaqur Rahman < > ashfaq226(a)gmail.com> wrote: > >> Dear Huiwei Lu, >> Many thanks for your email. I have attached here the source code of my >> programs. The folder contain two models. >> The Makefile should be looked at. >> >> Regards >> Ashfaq >> >> On Tue, Feb 3, 2015 at 3:20 PM, Huiwei Lu <huiweilu(a)mcs.anl.gov> wrote: >> >>> Hi Ashfaq, >>> >>> If you program already has started with separate parallel instances, how >>> can they becomes serial again? >>> >>> Can you give us a sample of minimum code to explain your problem? >>> >>> -- >>> Huiwei Lu >>> Postdoc Appointee >>> Mathematics and Computer Science Division >>> Argonne National Laboratory >>> http://www.mcs.anl.gov/~huiweilu/ >>> >>> On Tue, Feb 3, 2015 at 1:22 AM, Muhammad Ashfaqur Rahman < >>> ashfaq226(a)gmail.com> wrote: >>> >>>> Dear All, >>>> Take my greetings. I am new in mpich usage. I have problems in parallel >>>> run, when two fortran mpi programs are merged to one executable. If these >>>> two are separate, then they are running parallel. >>>> >>>> One program has used spmd and another one has used mpich header >>>> directly. >>>> >>>> Other issue is that while trying to run the above mentioned merged >>>> program in mpich, it's first started with separate parallel instances of >>>> same step and then after some steps it becomes serial. >>>> >>>> Please help me in this regards >>>> >>>> Ashfaq >>>> Ph.D Student >>>> Dept. of Meteorology >>>> >>>> _______________________________________________ >>>> 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 > _______________________________________________ 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] Problem in running parallel if two programs merged
by Muhammad Ashfaqur Rahman 03 Feb '15

03 Feb '15

Dear Huiwei Lu, Many thanks for your email. I have attached here the source code of my programs. These are two models. The Makefile should be looked at. Regards Ashfaq On Tue, Feb 3, 2015 at 4:32 PM, Muhammad Ashfaqur Rahman < ashfaq226(a)gmail.com> wrote: > Dear Huiwei Lu, > Many thanks for your email. I have attached here the source code of my > programs. The folder contain two models. > The Makefile should be looked at. > > Regards > Ashfaq > > On Tue, Feb 3, 2015 at 3:20 PM, Huiwei Lu <huiweilu(a)mcs.anl.gov> wrote: > >> Hi Ashfaq, >> >> If you program already has started with separate parallel instances, how >> can they becomes serial again? >> >> Can you give us a sample of minimum code to explain your problem? >> >> -- >> Huiwei Lu >> Postdoc Appointee >> Mathematics and Computer Science Division >> Argonne National Laboratory >> http://www.mcs.anl.gov/~huiweilu/ >> >> On Tue, Feb 3, 2015 at 1:22 AM, Muhammad Ashfaqur Rahman < >> ashfaq226(a)gmail.com> wrote: >> >>> Dear All, >>> Take my greetings. I am new in mpich usage. I have problems in parallel >>> run, when two fortran mpi programs are merged to one executable. If these >>> two are separate, then they are running parallel. >>> >>> One program has used spmd and another one has used mpich header >>> directly. >>> >>> Other issue is that while trying to run the above mentioned merged >>> program in mpich, it's first started with separate parallel instances of >>> same step and then after some steps it becomes serial. >>> >>> Please help me in this regards >>> >>> Ashfaq >>> Ph.D Student >>> Dept. of Meteorology >>> >>> _______________________________________________ >>> 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] Problem in running parallel if two programs merged
by Huiwei Lu 03 Feb '15

03 Feb '15

Hi Ashfaq, If you program already has started with separate parallel instances, how can they becomes serial again? Can you give us a sample of minimum code to explain your problem? -- Huiwei Lu Postdoc Appointee Mathematics and Computer Science Division Argonne National Laboratory http://www.mcs.anl.gov/~huiweilu/ On Tue, Feb 3, 2015 at 1:22 AM, Muhammad Ashfaqur Rahman < ashfaq226(a)gmail.com> wrote: > Dear All, > Take my greetings. I am new in mpich usage. I have problems in parallel > run, when two fortran mpi programs are merged to one executable. If these > two are separate, then they are running parallel. > > One program has used spmd and another one has used mpich header directly. > > Other issue is that while trying to run the above mentioned merged program > in mpich, it's first started with separate parallel instances of same step > and then after some steps it becomes serial. > > Please help me in this regards > > Ashfaq > Ph.D Student > Dept. of Meteorology > > _______________________________________________ > 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