- discuss - mailman3.cels.anl.gov

sizeof() usage inconsistencies in pm/hydra/utils/sock/sock.c
by Yauheni Zelenko 14 May '13

14 May '13

Hi! sizeof() is used inconsistently in pm/hydra/utils/sock/sock.c. In some places this is size of type, in some - size of variable. Sure, it's not major problem, but I think will be good idea to improve code. Eugene.

1 0

Re: [mpich-discuss] assert (!closed) failed
by Jeff Hammond 13 May '13

13 May '13

If cpi works then I suspect MPICH is not your problem. You should consult the HPCC developers for more help. >From http://icl.cs.utk.edu/hpcc/faq/index.html#269: "Is there a mailing list for asking questions? Yes, hpcc(a)cs.utk.edu." Best, Jeff On Mon, May 13, 2013 at 2:34 PM, Reza Bakhshayeshi <reza.b2008(a)gmail.com> wrote: > Yes, it was successful: > > localadmin@ubuntu-benchmark01:~$ mpiexec -f ./hosts2 -n 2 ./cpi > Process 0 of 2 is on ubuntu-benchmark01 > Process 1 of 2 is on ubuntu-benchmark02 > pi is approximately 3.1415926544231318, Error is 0.0000000008333387 > wall clock time = 0.050358 > > > On 13 May 2013 23:34, Jeff Hammond <jhammond(a)alcf.anl.gov> wrote: >> >> Can you run MPICH cpi on the cloud setup? That's a prerequisite for >> running something nontrivial like HPCC. >> >> Jeff >> >> On Mon, May 13, 2013 at 1:51 PM, Reza Bakhshayeshi <reza.b2008(a)gmail.com> >> wrote: >> > Thanks Wesley, >> > but not only I have followed the READMEs but also all the tutorials on >> > the >> > web! >> > I have rebuilt everything and this time with mpich. >> > Two nodes are the instance of a private cloud (similar to amazon), have >> > you >> > (or other members) have similar experience? >> > I mean can the cloud infrastructure cause the problem? and is there a >> > log >> > file for mpich activities? >> > >> > >> > On 13 May 2013 22:57, Wesley Bland <wbland(a)mcs.anl.gov> wrote: >> >> >> >> You might try looking at the HPCC documentation in the README that's >> >> distributed with the package, along with the website: >> >> >> >> http://icl.cs.utk.edu/hpcc/software/index.html >> >> >> >> If you have trouble with the specific benchmark, I would suggest >> >> contacting the HPCC people at the address provided on their website. >> >> >> >> Wesley >> >> >> >> On May 13, 2013, at 1:15 PM, Pavan Balaji <balaji(a)mcs.anl.gov> wrote: >> >> >> >> > >> >> > Please keep discuss(a)mpich.org cc'ed. We tend to ignore emails sent >> >> > to >> >> > developers directly. >> >> > >> >> > On 05/13/2013 01:13 PM US Central Time, Reza Bakhshayeshi wrote: >> >> >> Thanks for replying, but I can run hpcc on a single node >> >> >> successfully. >> >> >> Isn't it mpich responsibility to run it distributive? >> >> >> Can you suggest how I can find the main problem? >> >> > >> >> > Is it an MPI program? >> >> > >> >> > From the error message, all I can tell is that the application >> >> > process >> >> > died for some reason (e.g., it might have segfaulted), and mpich >> >> > cleaned >> >> > it up for your convenience. >> >> > >> >> > -- Pavan >> >> > >> >> > -- >> >> > 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 >> >> >> > >> > >> > _______________________________________________ >> > discuss mailing list discuss(a)mpich.org >> > To manage subscription options or unsubscribe: >> > https://lists.mpich.org/mailman/listinfo/discuss >> >> >> >> -- >> Jeff Hammond >> Argonne Leadership Computing Facility >> University of Chicago Computation Institute >> jhammond(a)alcf.anl.gov / (630) 252-5381 >> http://www.linkedin.com/in/jeffhammond >> https://wiki.alcf.anl.gov/parts/index.php/User:Jhammond >> ALCF docs: http://www.alcf.anl.gov/user-guides >> _______________________________________________ >> 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 -- Jeff Hammond Argonne Leadership Computing Facility University of Chicago Computation Institute jhammond(a)alcf.anl.gov / (630) 252-5381 http://www.linkedin.com/in/jeffhammond https://wiki.alcf.anl.gov/parts/index.php/User:Jhammond ALCF docs: http://www.alcf.anl.gov/user-guides

1 0

Re: [mpich-discuss] GCC 4.8.0 warnings
by Jeff Hammond 13 May '13

13 May '13

The vast majority of these errors are of no concern. I suggest that you do not set --enable-strict unless you intend to carefully inspect the resulting warnings for significance. If size_t is only typedef'd to 'unsigned int' since this is a 32-bit system, then the implicit cast to 'long unsigned int' is really not a big deal. There is no need for an MPI implementation to print anything for a correct (i.e. standard-compliant) program so those printf calls are probably warnings or errors. I recall that %zu is C99 so MPICH would need an explicit up-cast to "long unsigned" there but I fail to see how it matters. I do not see how any possible harm can come from "unused-but-set-variable". The "empty-body" warning is similarly of no interest. There is no reason to care about "unused-function". The compiler or linker should remove it from the object/binary anyways. I didn't look at the MPICH source but every time I have seen "maybe-uninitialized" warnings, it is not actually possible for the variable to be uninitialized, but rather that the compiler cannot figure this out for certain (maybe it's equivalent to the halting problem to do so...). It is just pedantic to tell the user that a comparison to NULL will always be true. The compiler will probably optimize this away anyways if it is certain about the branch. The only error that has me even the slightest bit worried is "nonnull" but my guess is that it really doesn't matter. If you inspect the source and find that my analysis is misguided and these warnings correspond to real problems that need addressing, I'm sure that the MPICH team will welcome your contribution of a patch for them. Best, Jeff $ grep warning Log src/mpltrmem.c:352:17: warning: format '%lu' expects argument of type 'long unsigned int', but argument 5 has type 'size_t' [-Wformat=] src/mpltrmem.c:372:17: warning: format '%lu' expects argument of type 'long unsigned int', but argument 5 has type 'size_t' [-Wformat=] src/mpltrmem.c:414:9: warning: format '%lu' expects argument of type 'long unsigned int', but argument 4 has type 'size_t' [-Wformat=] src/mpltrmem.c:531:17: warning: format '%lu' expects argument of type 'long unsigned int', but argument 5 has type 'size_t' [-Wformat=] src/mpltrmem.c:597:13: warning: format '%lu' expects argument of type 'long unsigned int', but argument 4 has type 'size_t' [-Wformat=] src/mpltrmem.c:583:14: warning: variable 'old_head' set but not used [-Wunused-but-set-variable] mpi-io/mpiu_external32.c:123:14: warning: unused variable 'extent' [-Wunused-variable] ./pm/pmiserv/pmip_cb.c:811:9: warning: null argument where non-null required (argument 1) [-Wnonnull] ./pm/pmiserv/pmip_cb.c:811:9: warning: null argument where non-null required (argument 2) [-Wnonnull] src/mpi/coll/opminloc.c:85:24: warning: variable 'flen' set but not used [-Wunused-but-set-variable] src/mpi/coll/opmaxloc.c:86:24: warning: variable 'flen' set but not used [-Wunused-but-set-variable] src/mpid/ch3/src/ch3u_handle_connection.c:316:194: warning: suggest braces around empty body in an 'if' statement [-Wempty-body] src/mpid/ch3/src/ch3u_rma_sync.c:2884:12: warning: 'MPIDI_CH3I_Send_flush_msg' defined but not used [-Wunused-function] ./src/include/mpimem.h:267:31: warning: 'mpiu_chkpmem_stk_[0]' may be used uninitialized in this function [-Wmaybe-uninitialized] ./src/include/mpimem.h:267:31: warning: 'mpiu_chkpmem_stk_[0]' may be used uninitialized in this function [-Wmaybe-uninitialized] ./src/include/mpimem.h:480:32: warning: the comparison will always evaluate as 'true' for the address of 'MPID_nem_tcp_g_lstn_sc' will never be NULL [-Waddress] ./src/include/mpimem.h:480:32: warning: the comparison will always evaluate as 'true' for the address of 'MPID_nem_tcp_g_lstn_plfd' will never be NULL [-Waddress] src/mpid/common/datatype/mpid_type_debug.c:238:59: warning: variable 'sticky_ub' set but not used [-Wunused-but-set-variable] src/mpid/common/datatype/mpid_type_debug.c:238:48: warning: variable 'sticky_lb' set but not used [-Wunused-but-set-variable] src/mpid/common/datatype/mpid_type_debug.c:238:44: warning: variable 'ub' set but not used [-Wunused-but-set-variable] src/mpid/common/datatype/mpid_type_debug.c:238:31: warning: variable 'true_ub' set but not used [-Wunused-but-set-variable] src/mpid/common/datatype/mpid_type_debug.c:236:11: warning: variable 'string' set but not used [-Wunused-but-set-variable] ./src/include/mpir_nbc.h:151:23: warning: 'mpir_sched_chkpmem_stk_[0]' may be used uninitialized in this function [-Wmaybe-uninitialized] ./src/include/mpimem.h:267:31: warning: 'mpir_sched_chkpmem_stk_[0]' may be used uninitialized in this function [-Wmaybe-uninitialized] ./src/include/mpir_nbc.h:151:23: warning: 'mpir_sched_chkpmem_stk_[0]' may be used uninitialized in this function [-Wmaybe-uninitialized] ./src/include/mpir_nbc.h:151:23: warning: 'mpir_sched_chkpmem_stk_[0]' may be used uninitialized in this function [-Wmaybe-uninitialized] ./src/include/mpimem.h:267:31: warning: 'mpir_sched_chkpmem_stk_[0]' may be used uninitialized in this function [-Wmaybe-uninitialized] ./src/include/mpimem.h:267:31: warning: 'mpir_sched_chkpmem_stk_[0]' may be used uninitialized in this function [-Wmaybe-uninitialized] ./src/include/mpir_nbc.h:151:23: warning: 'mpir_sched_chkpmem_stk_[0]' may be used uninitialized in this function [-Wmaybe-uninitialized] ./src/include/mpimem.h:267:31: warning: 'mpir_sched_chkpmem_stk_[0]' may be used uninitialized in this function [-Wmaybe-uninitialized] On Mon, May 13, 2013 at 2:46 PM, Yauheni Zelenko <zelenko(a)cadence.com> wrote: > Hi, Jeff! > > See attached file in original e-mail. > > Eugene. > ________________________________________ > From: discuss-bounces(a)mpich.org [discuss-bounces(a)mpich.org] On Behalf Of Jeff Hammond [jhammond(a)alcf.anl.gov] > Sent: Monday, May 13, 2013 11:59 AM > To: discuss(a)mpich.org > Subject: Re: [mpich-discuss] GCC 4.8.0 warnings > > Care to share make.log with the warnings to which you refer? > > Jeff > > On Mon, May 13, 2013 at 1:51 PM, Yauheni Zelenko <zelenko(a)cadence.com> wrote: >> Hi, Antonio! >> >> I compiled on RHEL 6.0. >> >> My configuration is: >> >> ./configure --prefix=${InstallDir} \ >> --with-pm=hydra --with-device=ch3:nemesis \ >> --disable-chekpointing \ >> --enable-strict --enable-fast=O3 \ >> --disable-f77 --disable-fc \ >> CC=${GCCDir}/bin/gcc CFLAGS="-m32 -fPIC" \ >> CPP=${GCCDir}/bin/cpp \ >> CXX=${GCCDir}/bin/g++ CXXFLAGS="-m32 -fPIC" \ >> CXXCPP=${GCCDir}/bin/cpp \ >> >> Eugene. >> ________________________________________ >> From: Antonio J. Peña [apenya(a)mcs.anl.gov] >> Sent: Monday, May 13, 2013 11:37 AM >> To: discuss(a)mpich.org >> Cc: Yauheni Zelenko >> Subject: Re: [mpich-discuss] GCC 4.8.0 warnings >> >> Hi Eugene, >> >> Could you please provide us with the whole details of your platform? (i.e., >> Linux distribution and version, and HW details). >> >> Thanks, >> Antonio >> >> >> On Monday, May 13, 2013 11:25:38 AM Yauheni Zelenko wrote: >>> Hi! >>> >>> I compiled recent nightly build (3.0.4 as of May 13) with GCC 4.8.0. >>> >>> There are some warnings which will be good idea to fix. >>> >>> Eugene. >> >> _______________________________________________ >> discuss mailing list discuss(a)mpich.org >> To manage subscription options or unsubscribe: >> https://lists.mpich.org/mailman/listinfo/discuss > > > > -- > Jeff Hammond > Argonne Leadership Computing Facility > University of Chicago Computation Institute > jhammond(a)alcf.anl.gov / (630) 252-5381 > http://www.linkedin.com/in/jeffhammond > https://wiki.alcf.anl.gov/parts/index.php/User:Jhammond > ALCF docs: http://www.alcf.anl.gov/user-guides > _______________________________________________ > 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 -- Jeff Hammond Argonne Leadership Computing Facility University of Chicago Computation Institute jhammond(a)alcf.anl.gov / (630) 252-5381 http://www.linkedin.com/in/jeffhammond https://wiki.alcf.anl.gov/parts/index.php/User:Jhammond ALCF docs: http://www.alcf.anl.gov/user-guides

1 0

Re: [mpich-discuss] GCC 4.8.0 warnings
by Yauheni Zelenko 13 May '13

13 May '13

Hi, Jeff! See attached file in original e-mail. Eugene. ________________________________________ From: discuss-bounces(a)mpich.org [discuss-bounces(a)mpich.org] On Behalf Of Jeff Hammond [jhammond(a)alcf.anl.gov] Sent: Monday, May 13, 2013 11:59 AM To: discuss(a)mpich.org Subject: Re: [mpich-discuss] GCC 4.8.0 warnings Care to share make.log with the warnings to which you refer? Jeff On Mon, May 13, 2013 at 1:51 PM, Yauheni Zelenko <zelenko(a)cadence.com> wrote: > Hi, Antonio! > > I compiled on RHEL 6.0. > > My configuration is: > > ./configure --prefix=${InstallDir} \ > --with-pm=hydra --with-device=ch3:nemesis \ > --disable-chekpointing \ > --enable-strict --enable-fast=O3 \ > --disable-f77 --disable-fc \ > CC=${GCCDir}/bin/gcc CFLAGS="-m32 -fPIC" \ > CPP=${GCCDir}/bin/cpp \ > CXX=${GCCDir}/bin/g++ CXXFLAGS="-m32 -fPIC" \ > CXXCPP=${GCCDir}/bin/cpp \ > > Eugene. > ________________________________________ > From: Antonio J. Peña [apenya(a)mcs.anl.gov] > Sent: Monday, May 13, 2013 11:37 AM > To: discuss(a)mpich.org > Cc: Yauheni Zelenko > Subject: Re: [mpich-discuss] GCC 4.8.0 warnings > > Hi Eugene, > > Could you please provide us with the whole details of your platform? (i.e., > Linux distribution and version, and HW details). > > Thanks, > Antonio > > > On Monday, May 13, 2013 11:25:38 AM Yauheni Zelenko wrote: >> Hi! >> >> I compiled recent nightly build (3.0.4 as of May 13) with GCC 4.8.0. >> >> There are some warnings which will be good idea to fix. >> >> Eugene. > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss -- Jeff Hammond Argonne Leadership Computing Facility University of Chicago Computation Institute jhammond(a)alcf.anl.gov / (630) 252-5381 http://www.linkedin.com/in/jeffhammond https://wiki.alcf.anl.gov/parts/index.php/User:Jhammond ALCF docs: http://www.alcf.anl.gov/user-guides _______________________________________________ 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] assert (!closed) failed
by Reza Bakhshayeshi 13 May '13

13 May '13

Yes, it was successful: localadmin@ubuntu-benchmark01:~$ mpiexec -f ./hosts2 -n 2 ./cpi Process 0 of 2 is on ubuntu-benchmark01 Process 1 of 2 is on ubuntu-benchmark02 pi is approximately 3.1415926544231318, Error is 0.0000000008333387 wall clock time = 0.050358 On 13 May 2013 23:34, Jeff Hammond <jhammond(a)alcf.anl.gov> wrote: > Can you run MPICH cpi on the cloud setup? That's a prerequisite for > running something nontrivial like HPCC. > > Jeff > > On Mon, May 13, 2013 at 1:51 PM, Reza Bakhshayeshi <reza.b2008(a)gmail.com> > wrote: > > Thanks Wesley, > > but not only I have followed the READMEs but also all the tutorials on > the > > web! > > I have rebuilt everything and this time with mpich. > > Two nodes are the instance of a private cloud (similar to amazon), have > you > > (or other members) have similar experience? > > I mean can the cloud infrastructure cause the problem? and is there a log > > file for mpich activities? > > > > > > On 13 May 2013 22:57, Wesley Bland <wbland(a)mcs.anl.gov> wrote: > >> > >> You might try looking at the HPCC documentation in the README that's > >> distributed with the package, along with the website: > >> > >> http://icl.cs.utk.edu/hpcc/software/index.html > >> > >> If you have trouble with the specific benchmark, I would suggest > >> contacting the HPCC people at the address provided on their website. > >> > >> Wesley > >> > >> On May 13, 2013, at 1:15 PM, Pavan Balaji <balaji(a)mcs.anl.gov> wrote: > >> > >> > > >> > Please keep discuss(a)mpich.org cc'ed. We tend to ignore emails sent > to > >> > developers directly. > >> > > >> > On 05/13/2013 01:13 PM US Central Time, Reza Bakhshayeshi wrote: > >> >> Thanks for replying, but I can run hpcc on a single node > successfully. > >> >> Isn't it mpich responsibility to run it distributive? > >> >> Can you suggest how I can find the main problem? > >> > > >> > Is it an MPI program? > >> > > >> > From the error message, all I can tell is that the application process > >> > died for some reason (e.g., it might have segfaulted), and mpich > cleaned > >> > it up for your convenience. > >> > > >> > -- Pavan > >> > > >> > -- > >> > 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 > >> > > > > > > _______________________________________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > > > -- > Jeff Hammond > Argonne Leadership Computing Facility > University of Chicago Computation Institute > jhammond(a)alcf.anl.gov / (630) 252-5381 > http://www.linkedin.com/in/jeffhammond > https://wiki.alcf.anl.gov/parts/index.php/User:Jhammond > ALCF docs: http://www.alcf.anl.gov/user-guides > _______________________________________________ > 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] assert (!closed) failed
by Jeff Hammond 13 May '13

13 May '13

Can you run MPICH cpi on the cloud setup? That's a prerequisite for running something nontrivial like HPCC. Jeff On Mon, May 13, 2013 at 1:51 PM, Reza Bakhshayeshi <reza.b2008(a)gmail.com> wrote: > Thanks Wesley, > but not only I have followed the READMEs but also all the tutorials on the > web! > I have rebuilt everything and this time with mpich. > Two nodes are the instance of a private cloud (similar to amazon), have you > (or other members) have similar experience? > I mean can the cloud infrastructure cause the problem? and is there a log > file for mpich activities? > > > On 13 May 2013 22:57, Wesley Bland <wbland(a)mcs.anl.gov> wrote: >> >> You might try looking at the HPCC documentation in the README that's >> distributed with the package, along with the website: >> >> http://icl.cs.utk.edu/hpcc/software/index.html >> >> If you have trouble with the specific benchmark, I would suggest >> contacting the HPCC people at the address provided on their website. >> >> Wesley >> >> On May 13, 2013, at 1:15 PM, Pavan Balaji <balaji(a)mcs.anl.gov> wrote: >> >> > >> > Please keep discuss(a)mpich.org cc'ed. We tend to ignore emails sent to >> > developers directly. >> > >> > On 05/13/2013 01:13 PM US Central Time, Reza Bakhshayeshi wrote: >> >> Thanks for replying, but I can run hpcc on a single node successfully. >> >> Isn't it mpich responsibility to run it distributive? >> >> Can you suggest how I can find the main problem? >> > >> > Is it an MPI program? >> > >> > From the error message, all I can tell is that the application process >> > died for some reason (e.g., it might have segfaulted), and mpich cleaned >> > it up for your convenience. >> > >> > -- Pavan >> > >> > -- >> > 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 >> > > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss -- Jeff Hammond Argonne Leadership Computing Facility University of Chicago Computation Institute jhammond(a)alcf.anl.gov / (630) 252-5381 http://www.linkedin.com/in/jeffhammond https://wiki.alcf.anl.gov/parts/index.php/User:Jhammond ALCF docs: http://www.alcf.anl.gov/user-guides

1 0

Re: [mpich-discuss] GCC 4.8.0 warnings
by Jeff Hammond 13 May '13

13 May '13

Care to share make.log with the warnings to which you refer? Jeff On Mon, May 13, 2013 at 1:51 PM, Yauheni Zelenko <zelenko(a)cadence.com> wrote: > Hi, Antonio! > > I compiled on RHEL 6.0. > > My configuration is: > > ./configure --prefix=${InstallDir} \ > --with-pm=hydra --with-device=ch3:nemesis \ > --disable-chekpointing \ > --enable-strict --enable-fast=O3 \ > --disable-f77 --disable-fc \ > CC=${GCCDir}/bin/gcc CFLAGS="-m32 -fPIC" \ > CPP=${GCCDir}/bin/cpp \ > CXX=${GCCDir}/bin/g++ CXXFLAGS="-m32 -fPIC" \ > CXXCPP=${GCCDir}/bin/cpp \ > > Eugene. > ________________________________________ > From: Antonio J. Peña [apenya(a)mcs.anl.gov] > Sent: Monday, May 13, 2013 11:37 AM > To: discuss(a)mpich.org > Cc: Yauheni Zelenko > Subject: Re: [mpich-discuss] GCC 4.8.0 warnings > > Hi Eugene, > > Could you please provide us with the whole details of your platform? (i.e., > Linux distribution and version, and HW details). > > Thanks, > Antonio > > > On Monday, May 13, 2013 11:25:38 AM Yauheni Zelenko wrote: >> Hi! >> >> I compiled recent nightly build (3.0.4 as of May 13) with GCC 4.8.0. >> >> There are some warnings which will be good idea to fix. >> >> Eugene. > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss -- Jeff Hammond Argonne Leadership Computing Facility University of Chicago Computation Institute jhammond(a)alcf.anl.gov / (630) 252-5381 http://www.linkedin.com/in/jeffhammond https://wiki.alcf.anl.gov/parts/index.php/User:Jhammond ALCF docs: http://www.alcf.anl.gov/user-guides

1 0

Re: [mpich-discuss] assert (!closed) failed
by Reza Bakhshayeshi 13 May '13

13 May '13

Thanks Wesley, but not only I have followed the READMEs but also all the tutorials on the web! I have rebuilt everything and this time with mpich. Two nodes are the instance of a private cloud (similar to amazon), have you (or other members) have similar experience? I mean can the cloud infrastructure cause the problem? and is there a log file for mpich activities? On 13 May 2013 22:57, Wesley Bland <wbland(a)mcs.anl.gov> wrote: > You might try looking at the HPCC documentation in the README that's > distributed with the package, along with the website: > > http://icl.cs.utk.edu/hpcc/software/index.html > > If you have trouble with the specific benchmark, I would suggest > contacting the HPCC people at the address provided on their website. > > Wesley > > On May 13, 2013, at 1:15 PM, Pavan Balaji <balaji(a)mcs.anl.gov> wrote: > > > > > Please keep discuss(a)mpich.org cc'ed. We tend to ignore emails sent to > > developers directly. > > > > On 05/13/2013 01:13 PM US Central Time, Reza Bakhshayeshi wrote: > >> Thanks for replying, but I can run hpcc on a single node successfully. > >> Isn't it mpich responsibility to run it distributive? > >> Can you suggest how I can find the main problem? > > > > Is it an MPI program? > > > > From the error message, all I can tell is that the application process > > died for some reason (e.g., it might have segfaulted), and mpich cleaned > > it up for your convenience. > > > > -- Pavan > > > > -- > > 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

Re: [mpich-discuss] GCC 4.8.0 warnings
by Yauheni Zelenko 13 May '13

13 May '13

Hi, Antonio! I compiled on RHEL 6.0. My configuration is: ./configure --prefix=${InstallDir} \ --with-pm=hydra --with-device=ch3:nemesis \ --disable-chekpointing \ --enable-strict --enable-fast=O3 \ --disable-f77 --disable-fc \ CC=${GCCDir}/bin/gcc CFLAGS="-m32 -fPIC" \ CPP=${GCCDir}/bin/cpp \ CXX=${GCCDir}/bin/g++ CXXFLAGS="-m32 -fPIC" \ CXXCPP=${GCCDir}/bin/cpp \ Eugene. ________________________________________ From: Antonio J. Peña [apenya(a)mcs.anl.gov] Sent: Monday, May 13, 2013 11:37 AM To: discuss(a)mpich.org Cc: Yauheni Zelenko Subject: Re: [mpich-discuss] GCC 4.8.0 warnings Hi Eugene, Could you please provide us with the whole details of your platform? (i.e., Linux distribution and version, and HW details). Thanks, Antonio On Monday, May 13, 2013 11:25:38 AM Yauheni Zelenko wrote: > Hi! > > I compiled recent nightly build (3.0.4 as of May 13) with GCC 4.8.0. > > There are some warnings which will be good idea to fix. > > Eugene.

1 0

Re: [mpich-discuss] GCC 4.8.0 warnings
by Antonio J. Peña 13 May '13

13 May '13

Hi Eugene, Could you please provide us with the whole details of your platform? (i.e., Linux distribution and version, and HW details). Thanks, Antonio On Monday, May 13, 2013 11:25:38 AM Yauheni Zelenko wrote: > Hi! > > I compiled recent nightly build (3.0.4 as of May 13) with GCC 4.8.0. > > There are some warnings which will be good idea to fix. > > Eugene.

1 0