December 2012 - discuss - mailman3.cels.anl.gov

Re: [mpich-discuss] Account Request / Bug report MPI_Type_create_f90_real
by Dave Goodell 19 Dec '12

19 Dec '12

A note (mainly for other MPICH devs): I have created a trac account for Harald and sent him his account information off-list. -Dave On Dec 19, 2012, at 9:23 AM CST, Harald Klimach wrote: > Hi there, > > I installed MPICH2 3.0rc1 using GCC 4.7.2 and found a problem with MPI_Type_create_f90_real. > It is very similar to the problem I reported in OpenMPI under: https://svn.open-mpi.org/trac/ompi/ticket/3432 > > Running the following program: > > program test_mpi > use mpi > implicit none > > integer, parameter :: rk_prec = 15 > integer, parameter :: rk = selected_real_kind(rk_prec) > > integer :: rk_mpi > integer :: iError > real(kind=rk) :: a_real > real(kind=rk) :: res > > call MPI_Init(iError) > call mpi_type_create_f90_real(rk_prec, MPI_UNDEFINED, rk_mpi, iError) > write(*,*) 'MPI_REAL8:', MPI_REAL8 > write(*,*) 'MPI_DOUBLE_PRECISION:', MPI_DOUBLE_PRECISION > write(*,*) 'type_create_f90_real:', rk_mpi > a_real = 1.0 > call MPI_Reduce(a_real, res, 1, rk_mpi, MPI_MIN, 0, MPI_COMM_WORLD, iError) > call MPI_Finalize(iError) > > end program test_mpi > > > Results in the following output and error: > > MPI_REAL8: 1275070505 > MPI_DOUBLE_PRECISION: 1275070495 > type_create_f90_real: -1946157049 > Fatal error in PMPI_Reduce: Invalid MPI_Op, error stack: > PMPI_Reduce(1217)........: MPI_Reduce(sbuf=0x7fff43637108, rbuf=0x7fff436370f8, count=1, dtype=USER<f90_real>, MPI_MIN, root=0, MPI_COMM_WORLD) failed > MPIR_MINF_check_dtype(71): MPI_Op MPI_MIN operation not defined for this datatype > > > Further the quadruple precision data type is not properly supported: > MPI_Type_create_f90_real will complain about wrong exponent, regardless of its value, > while MPI_REAL16 is available and the Reduction operation does not complain, it produces > wrong results. Again there is a similar problem with OpenMPI, as reported under: > https://svn.open-mpi.org/trac/ompi/ticket/3433 > > Thanks a lot, > Harald Klimach > _______________________________________________ > 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] Create account for bug tracker
by Dave Goodell 19 Dec '12

19 Dec '12

A tree-based scan pattern can cause a problem, for example. Basically, the MPI library is allowed to play with the associativity of the operations in order to provide better performance. The operation provided in the MPI standard example is not associative: (w,j) = (u,i)*(v,j), with w={ u+v if i=j, v if i!=j } i,j are booleans (0 or 1) If this operator is associative, then it must hold that: ((a,x)*(b,y))*(c,z) = (a,x)*((b,y)*(c,z)), for all (a,x),(b,y),(c,z) So here's a counterexample proving non-associativity: (a,x) = (a,0) (b,y) = (b,1) (c,z) = (c,0) The left-hand side of the equation yields "(c,0)", while the right-hand side yields "(a+c,0)", which are clearly not equal. The scheme that I worked out a while ago to deal with this problem involves adding a third variable to the pair, turning it into a triple. Unfortunately I left my notes containing the scheme at home, so I'll have to send an explanation later. In the meantime, you may wish to use the traditional edge-triggered version of the operator (cribbed from Jesper's proposed ticket https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/28) (val0 , index0)+(val1, index1) = [if index1=0 then (value0+value1 , index) else (value1 , index), where index is set to 1 if index0=1 and set to index1 otherwise] -Dave On Dec 18, 2012, at 9:27 PM CST, n n wrote: > I do not actually need this functionality, but can you show example of robust code, which will work for mpi_scan in all cases? > And can you clarify, why the order of output is not preserved, as in openmpi version? > > On Wed, Dec 19, 2012 at 1:52 AM, Dave Goodell <goodell(a)mcs.anl.gov> wrote: >> This scan operator is known to be buggy, unfortunately. It will not do the correct thing for many (entirely valid) scan implementation patterns. It's something that ought to be fixed in the next version of the MPI standard (it fell through the cracks in the MPI-3 process). >> >> Do you actually need this functionality? If so, I can describe how to implement it in a way that will be robust. Alternatively, most segmented scan algorithms assume an "edge triggered" segmented scan, rather than a "level" triggered one like what is described in the MPI standard. >> >> -Dave

1 0

Account Request / Bug report MPI_Type_create_f90_real
by Harald Klimach 19 Dec '12

19 Dec '12

Hi there, I installed MPICH2 3.0rc1 using GCC 4.7.2 and found a problem with MPI_Type_create_f90_real. It is very similar to the problem I reported in OpenMPI under: https://svn.open-mpi.org/trac/ompi/ticket/3432 Running the following program: program test_mpi use mpi implicit none integer, parameter :: rk_prec = 15 integer, parameter :: rk = selected_real_kind(rk_prec) integer :: rk_mpi integer :: iError real(kind=rk) :: a_real real(kind=rk) :: res call MPI_Init(iError) call mpi_type_create_f90_real(rk_prec, MPI_UNDEFINED, rk_mpi, iError) write(*,*) 'MPI_REAL8:', MPI_REAL8 write(*,*) 'MPI_DOUBLE_PRECISION:', MPI_DOUBLE_PRECISION write(*,*) 'type_create_f90_real:', rk_mpi a_real = 1.0 call MPI_Reduce(a_real, res, 1, rk_mpi, MPI_MIN, 0, MPI_COMM_WORLD, iError) call MPI_Finalize(iError) end program test_mpi Results in the following output and error: MPI_REAL8: 1275070505 MPI_DOUBLE_PRECISION: 1275070495 type_create_f90_real: -1946157049 Fatal error in PMPI_Reduce: Invalid MPI_Op, error stack: PMPI_Reduce(1217)........: MPI_Reduce(sbuf=0x7fff43637108, rbuf=0x7fff436370f8, count=1, dtype=USER<f90_real>, MPI_MIN, root=0, MPI_COMM_WORLD) failed MPIR_MINF_check_dtype(71): MPI_Op MPI_MIN operation not defined for this datatype Further the quadruple precision data type is not properly supported: MPI_Type_create_f90_real will complain about wrong exponent, regardless of its value, while MPI_REAL16 is available and the Reduction operation does not complain, it produces wrong results. Again there is a similar problem with OpenMPI, as reported under: https://svn.open-mpi.org/trac/ompi/ticket/3433 Thanks a lot, Harald Klimach

1 0

building MPICH2-1.5
by Иващук Роман 19 Dec '12

19 Dec '12

Hi all. I need some help in building MPICH2-1.5, if possible, please. I use the latest version of the source code, Windows 7, MS VC 2010, Intel Fortran 2011, cygwin and etc. I have many mistakes such as undeclared identifiers, missing headers (ex. glue_romio.h), syntax mistakes in fortran and c++ files. Is there more detailed instruction for building instead of "load script winconfigure.wsf, load and buld solution..." In other words, what to do with the source code of MPICH2-1.5 to have executable files. Thanks.

1 0

Re: [mpich-discuss] mpich2 make check problems
by Stephen Zheng 18 Dec '12

18 Dec '12

I thought there must be something wrong ,so I ceased it ... I just wonder whether there's any problem with my mpi. I want to find out all the errors to ensure my mpich2 work well. 2012/12/19 Dave Goodell <goodell(a)mcs.anl.gov> > How long did you wait after "coll/testlist" was printed? That directory > can take a very long time (>30 mins on some machines) to complete all the > tests contained within. If you set (and export) the RUNTESTS_VERBOSE=1 > environment variable before running "make testing", you'll get test-by-test > output as they are executed. > > Also, do you actually need to run the MPICH test suite? Is your > application working OK? > > -Dave > > On Dec 18, 2012, at 2:14 AM CST, Stephen Zheng wrote: > > > The libcr.so.0 problem does have gone,but there are still some problems. > > > > I changed the BLCR directory to $HOME and reconfiged & remade it . I > found the test stoped and then there's not any response after outputed the > below when I entered "make testing". It's weird. > > "./runtests -srcdir=. -tests=testlist \ > > -mpiexec=/usr/mpich2/bin/mpiexec -xmlfile=summary.xml > > Looking in ./testlist > > Processing directory attr > > Looking in ./attr/testlist > > Processing directory coll > > Looking in ./coll/testlist" > > > > The files "config.log" "src/pm/hydra/config.log" "c.txt" "m.txt" > "maketesting.txt" are attached below. > > _______________________________________________ > 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] Create account for bug tracker
by n n 18 Dec '12

18 Dec '12

I do not actually need this functionality, but can you show example of robust code, which will work for mpi_scan in all cases? And can you clarify, why the order of output is not preserved, as in openmpi version? On Wed, Dec 19, 2012 at 1:52 AM, Dave Goodell <goodell(a)mcs.anl.gov> wrote: > This scan operator is known to be buggy, unfortunately. It will not do > the correct thing for many (entirely valid) scan implementation patterns. > It's something that ought to be fixed in the next version of the MPI > standard (it fell through the cracks in the MPI-3 process). > > Do you actually need this functionality? If so, I can describe how to > implement it in a way that will be robust. Alternatively, most segmented > scan algorithms assume an "edge triggered" segmented scan, rather than a > "level" triggered one like what is described in the MPI standard. > > -Dave > > On Dec 18, 2012, at 2:06 PM CST, n n wrote: > > > Ok. > > Here is detailed explanation. > > There is example 5.22 on page 182 in mpi22-report.pdf, which can be > found here http://www.mpi-forum.org/docs/docs.html > > > > Here is test code http://pastebin.com/ZhYfhDVK (same code here > http://www.everfall.com/paste/id.php?0b7mgk2zh0xg, just in case) > > > > If i run this code using openmpi in archlinux, i get the following > result (as expected, in accordance with info from mpi22-report.pdf). > > > > mpicc code.c -o openmpi_version (mpicc from openmpi package, mpich is > not installed in archlinux) > > mpirun -n 8 ./openmpi_version > > 0 0 1.000000 > > 1 0 2.000000 > > 2 1 1.000000 > > 3 1 2.000000 > > 4 1 3.000000 > > 5 0 1.000000 > > 6 0 2.000000 > > 7 1 1.000000 <===== 1, as expected > > > > But if i run this code using mpich2 in ubuntu 10.04, (mpich package > version 1.2.1.1), i get unexpected result. It was also tested in > mpich2-1.4.1p1 and mpich2-1.5 in Windows 7 and Ubuntu 10.04.2 respectively. > > mpicc code.c -o mpich_version (mpicc from mpich2 package, openmpi is not > installed in ubuntu) > > mpirun -n 8 ./mpich_version > > 0 0 1.000000 > > 1 0 2.000000 > > 2 1 1.000000 > > 5 0 1.000000 > > 3 1 2.000000 > > 4 1 3.000000 > > 6 0 2.000000 > > 7 1 3.000000 <===== expected to be 1 > > > > > > Also, in openmpi version output is sorted, in accordance with code in > lines 60-73 (see pastebin), but in mpich2 this does not work as expected > too. > > > > On Tue, Dec 18, 2012 at 11:25 PM, Dave Goodell <goodell(a)mcs.anl.gov> > wrote: > > On Dec 18, 2012, at 4:47 AM CST, n n wrote: > > > > > I think, i've found a bug in function mpi_scan > > > > Would you care to share any details of the bug on the mailing list? > Forgive my caution, but you have a somewhat spammy looking email address > and email name and wrote a rather terse email... > > > > -Dave > > > > _______________________________________________ > > 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] Create account for bug tracker
by Pavan Balaji 18 Dec '12

18 Dec '12

On 12/18/2012 01:25 PM US Central Time, Dave Goodell wrote: > On Dec 18, 2012, at 4:47 AM CST, n n wrote: >> I think, i've found a bug in function mpi_scan > > Would you care to share any details of the bug on the mailing list? > Forgive my caution, but you have a somewhat spammy looking email > address and email name and wrote a rather terse email... LoL :-). Next time we could say, "tell me how much 2+4 is, so we know you are not a robot". :-) -- Pavan -- Pavan Balaji http://www.mcs.anl.gov/~balaji

1 0

Re: [mpich-discuss] Fwd: Unresolved symbol for all MPI functions
by Jim Dinan 18 Dec '12

18 Dec '12

Hi, There's no attachment; it's hard to provide an answer without more information. A few generic suggestions: Are you compiling your program with the MPI compiler wrappers? Is it a mixed Fortran/C program? If so, this could be a function underscoring issue. Make sure you use the Fortran compiler that you used to compile MPICH, and that you link with the Fortran compiler (not the C compiler). ~Jim. On 12/18/12 3:56 PM, Dave Goodell wrote: > Forwarding to discuss(a)mpich.org <mailto:[email protected]>, since that > is the actual mailing list address. Please do not send list questions > to the *-owner@ or *-request@ aliases. > > -Dave > > Begin forwarded message: > >> *From: *Sivasankaran P <sivasankaran.p(a)quest-global.com >> <mailto:[email protected]>> >> *Subject: **Unresolved symbol for all MPI functions* >> *Date: *December 18, 2012 5:20:13 AM CST >> *To: *"discuss-owner(a)mpich.org <mailto:[email protected]>" >> <discuss-owner(a)mpich.org <mailto:[email protected]>>, >> "discuss-request(a)mpich.org:" <discuss-request(a)mpich.org:> >> >> Hi, >> I hae installed MPICH2 based on the attachment. I am unable to compile the >> FORTRAN / C source codes with the MPI function calls. >> I am getting the following error: >> Unresolved external symbol _/MPI/_INIT referenced function main >> Thanks, >> *Sivasankaran Palanivelu* >> Technical Managerï¿½Embedded and Engineering Software CoE** >> *QuEST Global* >> #55, Quest Towers, Mahadevpura, >> Bangalore, Karnataka - 560 048, India >> *T:*+91 80 41190900*X:*1256*F:*+91 80 41190900*M:*+91-9880269291 >> *E:*[email protected] >> <mailto:[email protected]> *Follow us >> on:****cid:[email protected]* >> <http://www.facebook.com/QuESTGlobal>*cid:[email protected]* >> <http://www.linkedin.com/company/quest_2>*cid:[email protected]* >> <http://twitter.com/#!/QuEST_Global>** >> *W:*http://engineering.quest-global.com >> >> ---Disclaimer------------------------------ >> This e-mail contains PRIVILEGED AND CONFIDENTIAL INFORMATION intended >> solely for the use of the addressee(s). If you are not the intended >> recipient, please notify the sender by e-mail and delete the original >> message. Opinions, conclusions and other information in this >> transmission that do not relate to the official business of QuEST >> Global and/or its subsidiaries, shall be understood as neither given >> nor endorsed by it. Any statements made herein that are tantamount to >> contractual obligations, promises, claims or commitments shall not be >> binding on the Company unless followed by written confirmation by an >> authorized signatory of the Company. >> ------------------------------------------------------------------------------------ > > > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss >

1 0

Fwd: Unresolved symbol for all MPI functions
by Dave Goodell 18 Dec '12

18 Dec '12

Forwarding to discuss(a)mpich.org, since that is the actual mailing list address. Please do not send list questions to the *-owner@ or *-request@ aliases. -Dave Begin forwarded message: > From: Sivasankaran P <sivasankaran.p(a)quest-global.com> > Subject: Unresolved symbol for all MPI functions > Date: December 18, 2012 5:20:13 AM CST > To: "discuss-owner(a)mpich.org" <discuss-owner(a)mpich.org>, "discuss-request(a)mpich.org:" <discuss-request(a)mpich.org:> > > Hi, > > I hae installed MPICH2 based on the attachment. I am unable to compile the > FORTRAN / C source codes with the MPI function calls. > > I am getting the following error: > Unresolved external symbol _MPI_INIT referenced function main > > Thanks, > Sivasankaran Palanivelu > Technical Manager – Embedded and Engineering Software CoE > QuEST Global > #55, Quest Towers, Mahadevpura, > Bangalore, Karnataka - 560 048, India > T: +91 80 41190900 X: 1256 F: +91 80 41190900 M: +91-9880269291 > E: sivasankaran.p(a)quest-global.com Follow us on: > W: http://engineering.quest-global.com > > > ---Disclaimer------------------------------ > This e-mail contains PRIVILEGED AND CONFIDENTIAL INFORMATION intended solely for the use of the addressee(s). If you are not the intended recipient, please notify the sender by e-mail and delete the original message. Opinions, conclusions and other information in this transmission that do not relate to the official business of QuEST Global and/or its subsidiaries, shall be understood as neither given nor endorsed by it. Any statements made herein that are tantamount to contractual obligations, promises, claims or commitments shall not be binding on the Company unless followed by written confirmation by an authorized signatory of the Company. > ------------------------------------------------------------------------------------

1 0

Re: [mpich-discuss] Create account for bug tracker
by Dave Goodell 18 Dec '12

18 Dec '12

This scan operator is known to be buggy, unfortunately. It will not do the correct thing for many (entirely valid) scan implementation patterns. It's something that ought to be fixed in the next version of the MPI standard (it fell through the cracks in the MPI-3 process). Do you actually need this functionality? If so, I can describe how to implement it in a way that will be robust. Alternatively, most segmented scan algorithms assume an "edge triggered" segmented scan, rather than a "level" triggered one like what is described in the MPI standard. -Dave On Dec 18, 2012, at 2:06 PM CST, n n wrote: > Ok. > Here is detailed explanation. > There is example 5.22 on page 182 in mpi22-report.pdf, which can be found here http://www.mpi-forum.org/docs/docs.html > > Here is test code http://pastebin.com/ZhYfhDVK (same code here http://www.everfall.com/paste/id.php?0b7mgk2zh0xg, just in case) > > If i run this code using openmpi in archlinux, i get the following result (as expected, in accordance with info from mpi22-report.pdf). > > mpicc code.c -o openmpi_version (mpicc from openmpi package, mpich is not installed in archlinux) > mpirun -n 8 ./openmpi_version > 0 0 1.000000 > 1 0 2.000000 > 2 1 1.000000 > 3 1 2.000000 > 4 1 3.000000 > 5 0 1.000000 > 6 0 2.000000 > 7 1 1.000000 <===== 1, as expected > > But if i run this code using mpich2 in ubuntu 10.04, (mpich package version 1.2.1.1), i get unexpected result. It was also tested in mpich2-1.4.1p1 and mpich2-1.5 in Windows 7 and Ubuntu 10.04.2 respectively. > mpicc code.c -o mpich_version (mpicc from mpich2 package, openmpi is not installed in ubuntu) > mpirun -n 8 ./mpich_version > 0 0 1.000000 > 1 0 2.000000 > 2 1 1.000000 > 5 0 1.000000 > 3 1 2.000000 > 4 1 3.000000 > 6 0 2.000000 > 7 1 3.000000 <===== expected to be 1 > > > Also, in openmpi version output is sorted, in accordance with code in lines 60-73 (see pastebin), but in mpich2 this does not work as expected too. > > On Tue, Dec 18, 2012 at 11:25 PM, Dave Goodell <goodell(a)mcs.anl.gov> wrote: > On Dec 18, 2012, at 4:47 AM CST, n n wrote: > > > I think, i've found a bug in function mpi_scan > > Would you care to share any details of the bug on the mailing list? Forgive my caution, but you have a somewhat spammy looking email address and email name and wrote a rather terse email... > > -Dave > > _______________________________________________ > 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