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

Re: [mpich-discuss] Specifying a timeout for MPI_Comm_Accept
by Balaji, Pavan 08 Sep '14

08 Sep '14

Thanks, Matthieu. — Pavan On Sep 8, 2014, at 3:17 AM, Matthieu Dorier <matthieu.dorier(a)irisa.fr> wrote: > Hi Pavan, > > With all the work on in situ analysis/visualization, non-blocking versions of MPI_Comm_accept/connect could become very useful in the future to easily couple simulations and visualization codes. > > Here is an example: the VisIt software can use in situ, but right now the connection between VisIt and the simulation is done through a normal socket. The simulation will periodically call VisItDetectInput to check (in a non-blocking manner) if VisIt connected to the simulation. > > I recently had to use VisIt with a master-worker type simulation, that is, the master has an MPI_Recv(MPI_ANY_SOURCE...) to wait for workers to finish pieces of work. When using VisIt, this simple MPI_Recv becomes a non-blocking receive an active loop alternating between an MPI_Testany and VisItDetectInput. If VisIt could leverage non-blocking accept/connect, this active loop could be replaced with an MPI_Waitany on a set of MPI_Requests, one of them being VisIt connecting (the result of an MPI_Comm_iaccept), the other being workers finishing some work. > > I'm not a VisIt developer, though. This is just a thought. > > Matthieu Dorier > PhD student at ENS Rennes > http://people.irisa.fr/Matthieu.Dorier > ----- Mail original ----- >> De: "Pavan Balaji" <balaji(a)anl.gov> >> À: discuss(a)mpich.org >> Envoyé: Dimanche 7 Septembre 2014 00:53:58 >> Objet: Re: [mpich-discuss] Specifying a timeout for MPI_Comm_Accept >> >> Hirak, >> >> The function definitions for MPI_Comm_accept, etc., are a part of the MPI >> standard. We cannot change them in MPICH without changing them in the MPI >> standard first. Changes to the MPI standard go through the MPI Forum, and >> through a formal proposal and voting process before they get in. >> >> FWIW, both a timeout model and a nonblocking connect/accept have been >> proposed in the past, but they were both voted down. The nonblocking >> connect/accept proposal was originally done by Josh Hursey >> (http://www.cs.uwlax.edu/~jjhursey/) but I’m planning to revive the ticket >> but more broadly asking for nonblocking variants for many other operations >> as well. It’s unclear if/when this will get in, but we can try. >> >> The timeout proposal was put together by Jeff Squyres @ Cisco and Fab Tillier >> @ Microsoft (they are both on this list). I personally thought it was a >> very elegant proposal, but it was voted down because there was no use case >> for it at the time, particularly given that there was no standardized fault >> model in MPI. Once the Fault Tolerance working group gets its proposal in, >> there might be room to revisit this. But you’ll need to talk to the above >> mentioned guys to see if they are planning to revive it. >> >> Hope that helps. >> >> — Pavan >> >> On Sep 6, 2014, at 12:35 PM, Roy, Hirak <Hirak_Roy(a)mentor.com> wrote: >> >>> Hi, >>> >>> The thread at the end of my email, shows that there is no way we can >>> specify a timeout in MPI_Comm_accept/connect. >>> Since the thread is pretty old (2007), I would like to know if there is any >>> development related to this or not? >>> >>> If we still can not specify a timeout, is there any provision of >>> non-blocking accept/connect ? >>> >>> Thanks, >>> Hirak >>> >>> >>> https://lists.mcs.anl.gov/mailman/htdig/mpich-discuss/2007-April/002159.html >>> _______________________________________________ >>> 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 > _______________________________________________ > 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] Specifying a timeout for MPI_Comm_Accept
by Matthieu Dorier 08 Sep '14

08 Sep '14

Hi Pavan, With all the work on in situ analysis/visualization, non-blocking versions of MPI_Comm_accept/connect could become very useful in the future to easily couple simulations and visualization codes. Here is an example: the VisIt software can use in situ, but right now the connection between VisIt and the simulation is done through a normal socket. The simulation will periodically call VisItDetectInput to check (in a non-blocking manner) if VisIt connected to the simulation. I recently had to use VisIt with a master-worker type simulation, that is, the master has an MPI_Recv(MPI_ANY_SOURCE...) to wait for workers to finish pieces of work. When using VisIt, this simple MPI_Recv becomes a non-blocking receive an active loop alternating between an MPI_Testany and VisItDetectInput. If VisIt could leverage non-blocking accept/connect, this active loop could be replaced with an MPI_Waitany on a set of MPI_Requests, one of them being VisIt connecting (the result of an MPI_Comm_iaccept), the other being workers finishing some work. I'm not a VisIt developer, though. This is just a thought. Matthieu Dorier PhD student at ENS Rennes http://people.irisa.fr/Matthieu.Dorier ----- Mail original ----- > De: "Pavan Balaji" <balaji(a)anl.gov> > À: discuss(a)mpich.org > Envoyé: Dimanche 7 Septembre 2014 00:53:58 > Objet: Re: [mpich-discuss] Specifying a timeout for MPI_Comm_Accept > > Hirak, > > The function definitions for MPI_Comm_accept, etc., are a part of the MPI > standard. We cannot change them in MPICH without changing them in the MPI > standard first. Changes to the MPI standard go through the MPI Forum, and > through a formal proposal and voting process before they get in. > > FWIW, both a timeout model and a nonblocking connect/accept have been > proposed in the past, but they were both voted down. The nonblocking > connect/accept proposal was originally done by Josh Hursey > (http://www.cs.uwlax.edu/~jjhursey/) but I’m planning to revive the ticket > but more broadly asking for nonblocking variants for many other operations > as well. It’s unclear if/when this will get in, but we can try. > > The timeout proposal was put together by Jeff Squyres @ Cisco and Fab Tillier > @ Microsoft (they are both on this list). I personally thought it was a > very elegant proposal, but it was voted down because there was no use case > for it at the time, particularly given that there was no standardized fault > model in MPI. Once the Fault Tolerance working group gets its proposal in, > there might be room to revisit this. But you’ll need to talk to the above > mentioned guys to see if they are planning to revive it. > > Hope that helps. > > — Pavan > > On Sep 6, 2014, at 12:35 PM, Roy, Hirak <Hirak_Roy(a)mentor.com> wrote: > > > Hi, > > > > The thread at the end of my email, shows that there is no way we can > > specify a timeout in MPI_Comm_accept/connect. > > Since the thread is pretty old (2007), I would like to know if there is any > > development related to this or not? > > > > If we still can not specify a timeout, is there any provision of > > non-blocking accept/connect ? > > > > Thanks, > > Hirak > > > > > > https://lists.mcs.anl.gov/mailman/htdig/mpich-discuss/2007-April/002159.html > > _______________________________________________ > > 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

Re: Specifying a timeout for MPI_Comm_Accept
by Roy, Hirak 07 Sep '14

07 Sep '14

Thanks Pavan for the detailed explanation. Regards, Hirak

1 0

Re: [mpich-discuss] Error building mpich-3.1.2 on Solaris 10 withgcc-4.9.0
by Siegmar Gross 07 Sep '14

07 Sep '14

Hi, > >> Thanks, Siegmar. Are you seeing some errors with Ã¢?omake t > >> testingÃ¢?Â or are all tests failing? On some platforms, s > >> some tests take longer and hence show up as Ã¢?otimeoutsÃ¢?Â, but > >> they are not really hanging. But the number of such tests > >> should be small. > > > > I see some bus errors on Solaris 10 Sparc. I attached the log > > files for both Solaris 10 Sparc and x86_64, so that you can > > see yourself the results. > > > > Kind regards > > > > Siegmar > > > > Can you send the output of: > > gcc -v > g++ -v > gfortran -v > > Just want to double check the build configuration so I can match it on > my test setup. Solaris 10 x86_64: ================== sunpc1 root 2 gcc -v Using built-in specs. COLLECT_GCC=gcc COLLECT_LTO_WRAPPER=/export2/prog/SunOS_x86_64/gcc-4.9.0/bin/../libexec/gcc/i386-pc-solaris2.10/4.9.0/lto-wrapper Target: i386-pc-solaris2.10 Configured with: ../gcc-4.9.0/configure --prefix=/usr/local/gcc-4.9.0 --with-gmp=/usr/local --with-mpfr=/usr/local --with-mpc=/usr/local --with-isl=/usr/local --with-cloog=/usr/local --without-gnu-ld --with-ld=/usr/ccs/bin/ld ABI=32 --enable-languages=c,c++,objc,obj-c++,fortran --disable-nls --enable-threads=posix --disable-libgcj --with-system-zlib --enable-shared --enable-bootstrap Thread model: posix gcc version 4.9.0 (GCC) sunpc1 root 3 g++ -v Using built-in specs. COLLECT_GCC=g++ COLLECT_LTO_WRAPPER=/export2/prog/SunOS_x86_64/gcc-4.9.0/bin/../libexec/gcc/i386-pc-solaris2.10/4.9.0/lto-wrapper Target: i386-pc-solaris2.10 Configured with: ../gcc-4.9.0/configure --prefix=/usr/local/gcc-4.9.0 --with-gmp=/usr/local --with-mpfr=/usr/local --with-mpc=/usr/local --with-isl=/usr/local --with-cloog=/usr/local --without-gnu-ld --with-ld=/usr/ccs/bin/ld ABI=32 --enable-languages=c,c++,objc,obj-c++,fortran --disable-nls --enable-threads=posix --disable-libgcj --with-system-zlib --enable-shared --enable-bootstrap Thread model: posix gcc version 4.9.0 (GCC) sunpc1 root 4 gfortran -v Using built-in specs. COLLECT_GCC=gfortran COLLECT_LTO_WRAPPER=/export2/prog/SunOS_x86_64/gcc-4.9.0/bin/../libexec/gcc/i386-pc-solaris2.10/4.9.0/lto-wrapper Target: i386-pc-solaris2.10 Configured with: ../gcc-4.9.0/configure --prefix=/usr/local/gcc-4.9.0 --with-gmp=/usr/local --with-mpfr=/usr/local --with-mpc=/usr/local --with-isl=/usr/local --with-cloog=/usr/local --without-gnu-ld --with-ld=/usr/ccs/bin/ld ABI=32 --enable-languages=c,c++,objc,obj-c++,fortran --disable-nls --enable-threads=posix --disable-libgcj --with-system-zlib --enable-shared --enable-bootstrap Thread model: posix gcc version 4.9.0 (GCC) sunpc1 root 5 Solaris 10 Sparc: ================= tyr root 2 gcc -v Using built-in specs. COLLECT_GCC=gcc COLLECT_LTO_WRAPPER=/export2/prog/SunOS_sparc/gcc-4.9.0/bin/../libexec/gcc/sparc-sun-solaris2.10/4.9.0/lto-wrapper Target: sparc-sun-solaris2.10 Configured with: ../gcc-4.9.0/configure --prefix=/usr/local/gcc-4.9.0 --with-gnu-as --with-as=/usr/local/bin/as --without-gnu-ld --with-ld=/usr/ccs/bin/ld --enable-nls --with-system-zlib --enable-languages=c,c++,objc,fortran,java --enable-java-gc=boehm --enable-multilib --disable-libgcj --enable-bootstrap Thread model: posix gcc version 4.9.0 (GCC) tyr root 3 g++ -v Using built-in specs. COLLECT_GCC=g++ COLLECT_LTO_WRAPPER=/export2/prog/SunOS_sparc/gcc-4.9.0/bin/../libexec/gcc/sparc-sun-solaris2.10/4.9.0/lto-wrapper Target: sparc-sun-solaris2.10 Configured with: ../gcc-4.9.0/configure --prefix=/usr/local/gcc-4.9.0 --with-gnu-as --with-as=/usr/local/bin/as --without-gnu-ld --with-ld=/usr/ccs/bin/ld --enable-nls --with-system-zlib --enable-languages=c,c++,objc,fortran,java --enable-java-gc=boehm --enable-multilib --disable-libgcj --enable-bootstrap Thread model: posix gcc version 4.9.0 (GCC) tyr root 4 gfortran -v Using built-in specs. COLLECT_GCC=gfortran COLLECT_LTO_WRAPPER=/export2/prog/SunOS_sparc/gcc-4.9.0/bin/../libexec/gcc/sparc-sun-solaris2.10/4.9.0/lto-wrapper Target: sparc-sun-solaris2.10 Configured with: ../gcc-4.9.0/configure --prefix=/usr/local/gcc-4.9.0 --with-gnu-as --with-as=/usr/local/bin/as --without-gnu-ld --with-ld=/usr/ccs/bin/ld --enable-nls --with-system-zlib --enable-languages=c,c++,objc,fortran,java --enable-java-gc=boehm --enable-multilib --disable-libgcj --enable-bootstrap Thread model: posix gcc version 4.9.0 (GCC) tyr root 5 Kind regards Siegmar

1 0

Re: [mpich-discuss] Error building mpich-3.1.2 on Solaris 10 withgcc-4.9.0
by Kenneth Raffenetti 06 Sep '14

06 Sep '14

On 09/06/2014 05:35 PM, Siegmar Gross wrote: > Hi, > >> Thanks, Siegmar. Are you seeing some errors with â?omake t >> testingâ? or are all tests failing? On some platforms, s >> some tests take longer and hence show up as â?otimeoutsâ?, but >> they are not really hanging. But the number of such tests >> should be small. > > I see some bus errors on Solaris 10 Sparc. I attached the log > files for both Solaris 10 Sparc and x86_64, so that you can > see yourself the results. > > Kind regards > > Siegmar > Can you send the output of: gcc -v g++ -v gfortran -v Just want to double check the build configuration so I can match it on my test setup. Thanks, Ken

1 0

Re: [mpich-discuss] Specifying a timeout for MPI_Comm_Accept
by Balaji, Pavan 06 Sep '14

06 Sep '14

Hirak, The function definitions for MPI_Comm_accept, etc., are a part of the MPI standard. We cannot change them in MPICH without changing them in the MPI standard first. Changes to the MPI standard go through the MPI Forum, and through a formal proposal and voting process before they get in. FWIW, both a timeout model and a nonblocking connect/accept have been proposed in the past, but they were both voted down. The nonblocking connect/accept proposal was originally done by Josh Hursey (http://www.cs.uwlax.edu/~jjhursey/) but I’m planning to revive the ticket but more broadly asking for nonblocking variants for many other operations as well. It’s unclear if/when this will get in, but we can try. The timeout proposal was put together by Jeff Squyres @ Cisco and Fab Tillier @ Microsoft (they are both on this list). I personally thought it was a very elegant proposal, but it was voted down because there was no use case for it at the time, particularly given that there was no standardized fault model in MPI. Once the Fault Tolerance working group gets its proposal in, there might be room to revisit this. But you’ll need to talk to the above mentioned guys to see if they are planning to revive it. Hope that helps. — Pavan On Sep 6, 2014, at 12:35 PM, Roy, Hirak <Hirak_Roy(a)mentor.com> wrote: > Hi, > > The thread at the end of my email, shows that there is no way we can specify a timeout in MPI_Comm_accept/connect. > Since the thread is pretty old (2007), I would like to know if there is any development related to this or not? > > If we still can not specify a timeout, is there any provision of non-blocking accept/connect ? > > Thanks, > Hirak > > > https://lists.mcs.anl.gov/mailman/htdig/mpich-discuss/2007-April/002159.html > _______________________________________________ > 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] [mpich-trac] #2168: various Solaris build issues
by Balaji, Pavan 06 Sep '14

06 Sep '14

Siegmar, If you are interested, we can create a trac account for you, so you can add comments to the trac ticket directly. Regards, — Pavan On Sep 6, 2014, at 5:41 PM, Siegmar Gross <Siegmar.Gross(a)informatik.hs-fulda.de> wrote: > Hi, > >> #2168: various Solaris build issues >> ----------------------+-------------------------- >> Reporter: raffenet | Owner: raffenet >> Type: bug | Status: new >> Priority: major | Milestone: mpich-3.1.3 >> Component: mpich | Resolution: >> Keywords: | >> ----------------------+-------------------------- >> >> Comment (by balaji): >> >> Did you try the same configure options as Siegmar? He was seeing a build >> failure even with x86_64, as per one of his emails. > > Yes, on my machine building MPICH for Solaris 10 x86_64 breaks for > gcc-4.9.0 as you can see below. > > sunpc1 mpich-3.2-SunOS.x86_64.64_gcc_only-m64 132 head config.log | grep mpich > $ ../mpich-master-v3.2a1/configure --prefix=/usr/local/mpich-3.2_64_gcc > --libdir=/usr/local/mpich-3.2_64_gcc/lib64 > --includedir=/usr/local/mpich-3.2_64_gcc/include64 > CC=gcc CXX=g++ F77=gfortran FC=gfortran CFLAGS=-m64 CXXFLAGS=-m64 > FFLAGS=-m64 FCFLAGS=-m64 LDFLAGS=-m64 > -L/usr/lib/amd64 -Wl,-rpath /usr/lib/amd64 --enable-f77 --enable-fc > --enable-cxx --enable-romio --enable-debuginfo --enable-smpcoll > --enable-threads=runtime --with-thread-package=posix --enable-shared > > > sunpc1 mpich-3.2-SunOS.x86_64.64_gcc_only-m64 133 tail log.make.SunOS.x86_64.64_gcc > CXX src/binding/cxx/initcxx.lo > CXXLD lib/libmpicxx.la > ld: fatal: file libmpicxx.so.12: open failed: No such file or directory > ld: fatal: file processing errors. No output written to lib/.libs/libmpicxx.so.12.0.3 > collect2: error: ld returned 1 exit status > make[2]: *** [lib/libmpicxx.la] Error 1 > make[2]: Leaving directory `/export2/src/mpich-3.2/mpich-3.2-SunOS.x86_64.64_gcc_only-m64' > make[1]: *** [all-recursive] Error 1 > make[1]: Leaving directory `/export2/src/mpich-3.2/mpich-3.2-SunOS.x86_64.64_gcc_only-m64' > make: *** [all] Error 2 > sunpc1 mpich-3.2-SunOS.x86_64.64_gcc_only-m64 134 > > > sunpc1 mpich-3.2-SunOS.x86_64.64_gcc_only-m64 134 uname -a > SunOS sunpc1 5.10 Generic_147441-21 i86pc i386 i86pc Solaris > > sunpc1 mpich-3.2-SunOS.x86_64.64_gcc_only-m64 135 isainfo -k > amd64 > > > Kind regards > > Siegmar > > _______________________________________________ > 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] Error building mpich-3.1.2 on Solaris 10 withgcc-4.9.0
by Balaji, Pavan 06 Sep '14

06 Sep '14

On Sep 6, 2014, at 5:35 PM, Siegmar Gross <Siegmar.Gross(a)informatik.hs-fulda.de> wrote: >> Thanks, Siegmar. Are you seeing some errors with â?omake t >> testingâ? or are all tests failing? On some platforms, s >> some tests take longer and hence show up as â?otimeoutsâ?, but >> they are not really hanging. But the number of such tests >> should be small. > > I see some bus errors on Solaris 10 Sparc. I attached the log > files for both Solaris 10 Sparc and x86_64, so that you can > see yourself the results. Ah, yes. The bus errors on SPARC are a known issue: https://trac.mpich.org/projects/mpich/ticket/1159 The reason is that SPARC is very strict with respect to datatype alignment restrictions. The compiler is supposed to align integers at 4-byte boundaries, doubles at 8-byte boundaries, etc. But in MPICH we sometimes use a char array and later typecast it to a structure that might contain ints and pointers, etc. Unfortunately, this means that the compiler cannot do the appropriate alignment of the char buffer, cause bus errors when we try to access the data. We should certainly fix this, but it’s not going to be easy (at least not without adding performance overhead). — Pavan -- Pavan Balaji ✉️ http://www.mcs.anl.gov/~balaji

1 0

Re: [mpich-discuss] [mpich-trac] #2168: various Solaris build issues
by Siegmar Gross 06 Sep '14

06 Sep '14

Hi, > #2168: various Solaris build issues > ----------------------+-------------------------- > Reporter: raffenet | Owner: raffenet > Type: bug | Status: new > Priority: major | Milestone: mpich-3.1.3 > Component: mpich | Resolution: > Keywords: | > ----------------------+-------------------------- > > Comment (by balaji): > > Did you try the same configure options as Siegmar? He was seeing a build > failure even with x86_64, as per one of his emails. Yes, on my machine building MPICH for Solaris 10 x86_64 breaks for gcc-4.9.0 as you can see below. sunpc1 mpich-3.2-SunOS.x86_64.64_gcc_only-m64 132 head config.log | grep mpich $ ../mpich-master-v3.2a1/configure --prefix=/usr/local/mpich-3.2_64_gcc --libdir=/usr/local/mpich-3.2_64_gcc/lib64 --includedir=/usr/local/mpich-3.2_64_gcc/include64 CC=gcc CXX=g++ F77=gfortran FC=gfortran CFLAGS=-m64 CXXFLAGS=-m64 FFLAGS=-m64 FCFLAGS=-m64 LDFLAGS=-m64 -L/usr/lib/amd64 -Wl,-rpath /usr/lib/amd64 --enable-f77 --enable-fc --enable-cxx --enable-romio --enable-debuginfo --enable-smpcoll --enable-threads=runtime --with-thread-package=posix --enable-shared sunpc1 mpich-3.2-SunOS.x86_64.64_gcc_only-m64 133 tail log.make.SunOS.x86_64.64_gcc CXX src/binding/cxx/initcxx.lo CXXLD lib/libmpicxx.la ld: fatal: file libmpicxx.so.12: open failed: No such file or directory ld: fatal: file processing errors. No output written to lib/.libs/libmpicxx.so.12.0.3 collect2: error: ld returned 1 exit status make[2]: *** [lib/libmpicxx.la] Error 1 make[2]: Leaving directory `/export2/src/mpich-3.2/mpich-3.2-SunOS.x86_64.64_gcc_only-m64' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/export2/src/mpich-3.2/mpich-3.2-SunOS.x86_64.64_gcc_only-m64' make: *** [all] Error 2 sunpc1 mpich-3.2-SunOS.x86_64.64_gcc_only-m64 134 sunpc1 mpich-3.2-SunOS.x86_64.64_gcc_only-m64 134 uname -a SunOS sunpc1 5.10 Generic_147441-21 i86pc i386 i86pc Solaris sunpc1 mpich-3.2-SunOS.x86_64.64_gcc_only-m64 135 isainfo -k amd64 Kind regards Siegmar

1 0

Re: [mpich-discuss] Error building mpich-3.1.2 on Solaris 10 withgcc-4.9.0
by Siegmar Gross 06 Sep '14

06 Sep '14

Hi, > Thanks, Siegmar. Are you seeing some errors with âmake t > testingâ or are all tests failing? On some platforms, s > some tests take longer and hence show up as âtimeoutsâ, but > they are not really hanging. But the number of such tests > should be small. I see some bus errors on Solaris 10 Sparc. I attached the log files for both Solaris 10 Sparc and x86_64, so that you can see yourself the results. Kind regards Siegmar > Weâll look into the C++ issues. We are trying to get a m > machine setup locally to test this configuration out. > > Regards, > > â Pavan > > On Sep 6, 2014, at 2:24 AM, Siegmar Gross <Siegmar.Gross(a)informatik.hs-fulda.de> wrote: > > > Hi, > > > >> Could you configure with --disable-cxx to see if you can continue? > > > > I can build MPICH with gcc-4.9.0 without C++ bindings on Solaris 10 > > Sparc and x86_64. > > > > tyr small_prog 114 mpichversion > > MPICH Version: 3.2a1 > > MPICH Release date: Fri Sep 5 00:00:35 CDT 2014 > > MPICH Device: ch3:nemesis > > MPICH configure: --prefix=/usr/local/mpich-3.2_64_gcc > > --libdir=/usr/local/mpich-3.2_64_gcc/lib64 > > --includedir=/usr/local/mpich-3.2_64_gcc/include64 > > CC=gcc CXX=g++ F77=gfortran FC=gfortran CFLAGS=-m64 CXXFLAGS=-m64 > > FFLAGS=-m64 FCFLAGS=-m64 LDFLAGS=-m64 > > -L/usr/lib/sparcv9 -Wl,-rpath /usr/lib/sparcv9 --enable-f77 --enable-fc > > --disable-cxx --enable-romio --enable-debuginfo --enable-smpcoll > > --enable-threads=runtime --with-thread-package=posix --enable-shared > > MPICH CC: gcc -m64 -O2 > > MPICH CXX: no -m64 > > MPICH F77: gfortran -m64 -O2 > > MPICH FC: gfortran -m64 -O2 > > > > tyr small_prog 115 mpicc init_finalize.c > > tyr small_prog 116 mpiexec -np 1 a.out > > Hello! > > > > tyr small_prog 117 mpicc thread_support.c > > tyr small_prog 118 mpiexec -np 1 a.out > > I have requested MPI_THREAD_MULTIPLE in "MPI_Init_thread ()" and > > it provides MPI_THREAD_MULTIPLE > > "MPI_Query_thread ()" returned MPI_THREAD_MULTIPLE: > > Many threads are supported and any thread may call > > MPI functions at any time. > > "MPI_Is_thread_main ()" returned: "true". > > tyr small_prog 119 > > > > > > I have attached my config and log files for my Solaris Sparc version > > (even "make check" works fine). However, I get errors with "make > > testing" (APPLICATION TIMED OUT). Generally, I only use "make check", > > so perhaps I have to do it in a different way (e.g., running the test > > as normal user and not as root). I stopped this test. > > > > By the way, you can avoid the warning about X11/Xutil.h, if you include > > X11/Xlib.h before this header file. > > > > tyr mpich-3.2-SunOS.sparc.64_gcc_disable-cxx 135 more /usr/include/X11/Xutil.h > > /* $TOG: Xutil.h /main/45 1998/02/06 18:03:35 kaleb $ */ > > ... > > /* You must include <X11/Xlib.h> before including this file */ > > ... > > > > > > I would be grateful, if you can fix the problem for C++ bindings. > > C++ bindings are available in my Sun C 5.12 version of MPICH. > > > > tyr fd1026 101 mpichversion > > MPICH Version: 3.2a1 > > MPICH Release date: Fri Sep 5 00:00:35 CDT 2014 > > MPICH Device: ch3:nemesis > > MPICH configure: --prefix=/usr/local/mpich-3.2_64_cc > > --libdir=/usr/local/mpich-3.2_64_cc/lib64 > > --includedir=/usr/local/mpich-3.2_64_cc/include64 > > CC=cc CXX=CC F77=f77 FC=f95 CFLAGS=-m64 CXXFLAGS=-m64 > > FFLAGS=-m64 FCFLAGS=-m64 LDFLAGS=-m64 > > -L/usr/lib/sparcv9 -R/usr/lib/sparcv9 --enable-f77 --enable-fc > > --enable-cxx --enable-romio --enable-debuginfo --enable-smpcoll > > --enable-threads=runtime --with-thread-package=posix --enable-shared > > MPICH CC: cc -m64 -O2 > > MPICH CXX: CC -m64 -O2 > > MPICH F77: f77 -m64 > > MPICH FC: f95 -m64 -O2 > > tyr fd1026 102 > > > > > > Kind regards and thank you very much for any help in advance > > > > Siegmar > > > > > >> --Junchao Zhang > >> > >> > >> On Fri, Sep 5, 2014 at 2:26 PM, Siegmar Gross < > >> Siegmar.Gross(a)informatik.hs-fulda.de> wrote: > >> > >>> Hi, > >>> > >>> thank you very much for all answers to my problems building > >>> MPICH on Solaris with gcc-4.9.0. Today I tried to build > >>> mpich-master-v3.2a1. Unfortunately I have had the same problems. > >>> > >>>> 1) For the linking errors you should provide the full build log but it > >>>> could just be an issue with the order of arguments passed to Solaris ld, > >>> or > >>>> if you compiled gcc to use GNU ld, the same issue might occur if the > >>>> configure script assumes Solaris => Solaris ld. > >>> > >>> I have attached the files in the archive > >>> mpich-3.2-SunOS.sparc.64_gcc_only-m64.tar.gz > >>> > >>> > >>>> 2) I keep this sticky note on my wiki: > >>>> > >>>> "Compiler or options invalid for pre-UNIX 03 X/Open applications and > >>>> pre-2001 POSIX applications" > >>>> > >>>> > >>> /usr/gcc/4.6/lib/gcc/i386-pc-solaris2.11/4.6.3/include-fixed/sys/feature_tests.h :362:2: > >>>> error: #error "Compiler or options invalid for pre-UNIX 03 X/Open > >>> applications > >>>> and pre-2001 POSIX applications" > >>>> > >>>> Conflict of standards between _XOPEN_SOURCE and C99 extensions. > >>>> > >>>> #if defined(_STDC_C99) && (defined(__XOPEN_OR_POSIX) && !defined(_XPG6)) > >>>> > >>>> As GCC standards do not match ANSI/ISO/POSIX standards strictly you need > >>> to > >>>> specify the standard you want to comply with. > >>>> Usually adding -D_XOPEN_SOURCE=500 (or 600) and/or -D__EXTENSIONS__ to > >>> your > >>>> compiler flags is a good start. > >>> > >>> It was a little bit more complicated to get this one as far as > >>> the build with just "CFLAGS=-m64". Adding "-D__EXTENSIONS__" broke > >>> the build process. Adding "-D_XOPEN_SOURCE=500" didn't solve the > >>> problem, because Solaris expects "-D_XOPEN_SOURCE=600". Using this > >>> flag wasn't helpful, because you redefined it to "500" in file > >>> mpich-master-v3.2a1/src/mpi/romio/adio/common/ad_read.c, so that I > >>> had to add "-D_STDC_C99" to keep "600". Another problem came from > >>> file /usr/include/sys/statvfs.h ("FSTYPSZ" wasn't defined). Looking > >>> into the file showed the following. > >>> > >>> ... > >>> #define _FSTYPSZ 16 > >>> #if !defined(_XPG4_2) || defined(__EXTENSIONS__) > >>> #ifndef FSTYPSZ > >>> #define FSTYPSZ _FSTYPSZ > >>> #endif > >>> ... > >>> > >>> Finally I could build MPICH with "CFLAGS=-m64 -std=c99 -D_STDC_C99 > >>> -D_POSIX_C_SOURCE=200112L -D_XOPEN_SOURCE=600 -DFSTYPSZ=_FSTYPSZ" > >>> to the same point as with "CFLAGS=-m64". In both cases libmpicxx.so > >>> wasn't available. I have attached the files for this build process > >>> in the archive mpich-3.2-SunOS.sparc.64_gcc.tar.gz. Please let me > >>> know if you need anything else. > >>> > >>> Kind regards > >>> > >>> Siegmar > >>> > >>> _______________________________________________ > >>> discuss mailing list discuss(a)mpich.org > >>> To manage subscription options or unsubscribe: > >>> https://lists.mpich.org/mailman/listinfo/discuss > >>> > > <mpich-3.2-SunOS.sparc.64_gcc_disable-cxx.tar.gz>_______________________________ ________________ > > discuss mailing list discuss(a)mpich.org > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > -- > Pavan Balaji âï¸ > http://www.mcs.anl.gov/~balaji > >

1 0