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

Re: [mpich-discuss] Error in compiling MPICH on Mac desktop
by Balaji, Pavan 30 Oct '15

30 Oct '15

Eric, Documenting this is a good idea and should go into the RELEASE_NOTES file. Would you mind sending us a patch with the right description, so we get it right? We might copy it into our FAQ page as well. -- Pavan > On Oct 30, 2015, at 1:53 PM, Eric A. Borisch <eborisch(a)gmail.com> wrote: > > <off topic> > With respect to adding your own links, I would suggest that might > confuse Homebrew (don't know for sure, but they use /usr/local/bin)... > MacPorts uses /opt/local, and with it, at least, the intention is that > you don't go around adding your own files (or links) inside of > /opt/local (with the possible exception of config files in etc/ which > you need to modify; frequently as copies of foo.conf.example) .. so > I'm guessing the same is true for Homebrew. > </off topic> > > This (-Wa,-q for this case of compilers/OS) is something MPICH can > choose to document for this case if they are interested. Or (sorry if > I'm a broken record here) you can install MacPorts and use 'sudo port > install mpich-gcc5' (most of the dependencies are available as > pre-built packages, so it goes fairly quick.) I'm not sure how the > Homebrew MPICH recipe is configured and how to build this flavor with > it; I suppose you could contact its maintainer. > > - Eric > > On Fri, Oct 30, 2015 at 1:29 PM, Ghosh, Debojyoti <ghosh5(a)llnl.gov> wrote: >> Eric, adding those flags worked. Thanks! >> Rob, yes, bug report from libressl matched with what I was getting. >> >> Can you please explain what was wrong? I installed the GNU compilers using >> Homebrew. I realize I should have used “brew link gcc5” or something, but >> instead, I went to /usr/local/bin and manually created the symlinks (gcc, >> gfortran, g++, etc) pointing to gcc-5, gfortran-5, etc. Did I miss >> something? >> >> I did not face this issue when I did the same thing (Homebrew gcc + compile >> MPICH + manual symlinks) on a Macbook. The only difference I see there is >> that the build is apple-darwin14.4.0. >> >> >> >> On Oct 30, 2015, at 8:57 AM, Eric A. Borisch <eborisch(a)gmail.com> wrote: >> >> You will need to add '-Wa,-q' for CFLAGS and CXXFLAGS for gcc* on >> Yosemite and above. This will use the clang assembler. (See man as) >> >> <plug>The MacPorts mpich package does this automatically when needed.</plug> >> >> - Eric >> >> On Fri, Oct 30, 2015 at 10:13 AM, Rob Latham <robl(a)mcs.anl.gov> wrote: >> >> >> >> On 10/28/2015 12:46 PM, Ghosh, Debojyoti wrote: >> >> >> Hi, >> >> I am getting a compilation error while trying to compile MPICH on my Mac >> desktop. I installed the latest GNU compilers using Homebrew. >> >> gcc -v >> >> Using built-in specs. >> COLLECT_GCC=gcc >> >> COLLECT_LTO_WRAPPER=/usr/local/Cellar/gcc/5.2.0/libexec/gcc/x86_64-apple-darwin14.5.0/5.2.0/lto-wrapper >> Target: x86_64-apple-darwin14.5.0 >> Configured with: ../configure --build=x86_64-apple-darwin14.5.0 >> --prefix=/usr/local/Cellar/gcc/5.2.0 >> --libdir=/usr/local/Cellar/gcc/5.2.0/lib/gcc/5 >> --enable-languages=c,c++,objc,obj-c++,fortran --program-suffix=-5 >> --with-gmp=/usr/local/opt/gmp --with-mpfr=/usr/local/opt/mpfr >> --with-mpc=/usr/local/opt/libmpc --with-isl=/usr/local/opt/isl >> --with-system-zlib --enable-libstdcxx-time=yes --enable-stage1-checking >> --enable-checking=release --enable-lto >> --with-build-config=bootstrap-debug --disable-werror >> --with-pkgversion='Homebrew gcc 5.2.0' >> --with-bugurl=https://github.com/Homebrew/homebrew/issues >> --enable-plugin --disable-nls --enable-multilib >> Thread model: posix >> gcc version 5.2.0 (Homebrew gcc 5.2.0) >> >> >> >> We test with gcc-5.2 on our mac build bots, so it's definitely possible. >> >> I am getting the following error while compiling MPICH: >> CC src/mpltrmem.lo >> /var/folders/qf/q3tp7kgn0gz6qj5_c2nr4qqm001qq3/T//ccyirdXM.s:491:suffix >> or operands invalid for `movq' >> /var/folders/qf/q3tp7kgn0gz6qj5_c2nr4qqm001qq3/T//ccyirdXM.s:495:suffix >> or operands invalid for `movq' >> make[2]: *** [src/mpltrmem.lo] Error 1 >> make[1]: *** [all-recursive] Error 1 >> make: *** [all] Error 2 >> >> >> >> When the assembler is suspicous ('movq') then it sure seems like there's a >> toolchain mismatch. >> >> Does this bug report from libressl sound like your environment? >> https://github.com/libressl-portable/portable/issues/121 >> >> ==rob >> >> >> -- >> Rob Latham >> Mathematics and Computer Science Division >> Argonne National Lab, IL USA >> _______________________________________________ >> 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 _______________________________________________ discuss mailing list discuss(a)mpich.org To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss

1 0

Bug (?) report: potential division by zero in ADIOI_LUSTRE_Docollect
by Constantine Khroulev 30 Oct '15

30 Oct '15

Dear MPICH developers, I am writing to you to report what I think is a bug in ADIO (which, if I understand it correctly, is a part of ROMIO, which is a part of MPICH). The function int ADIOI_LUSTRE_Docollect(ADIO_File, int, ADIO_Offset *, int) defined in src/mpi/romio/adio/ad_lustre/ad_lustre_aggregate.c (MPICH version 3.1.4 and several earlier versions) contains an unprotected division: /* estimate average req_size */ avg_req_size = (int)(total_req_size / total_access_count); I suggest adding an if statement protecting from division by zero and stopping (if appropriate). I hope that this may save somebody a chunk of time later. Some context: I am debugging a failure of PISM [1] on NASA Pleiades [2], which uses the SGI MPI implementation, which also uses ROMIO. PISM crashed with SIGFPE in ADIOI_LUSTRE_Docollect deep inside HDF5 and NetCDF. I am not asking for help with this [3] here; I will contact NASA's support as soon as I have a way of reproducing the issue outside of PISM. Thank you! * Footnotes [1] http://pism-docs.org/wiki/doku.php [2] http://www.nas.nasa.gov/hecc/resources/pleiades.html [3] Any advice is greatly appreciated, though. -- Constantine _______________________________________________ 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] Error in compiling MPICH on Mac desktop
by Eric A. Borisch 30 Oct '15

30 Oct '15

<off topic> With respect to adding your own links, I would suggest that might confuse Homebrew (don't know for sure, but they use /usr/local/bin)... MacPorts uses /opt/local, and with it, at least, the intention is that you don't go around adding your own files (or links) inside of /opt/local (with the possible exception of config files in etc/ which you need to modify; frequently as copies of foo.conf.example) .. so I'm guessing the same is true for Homebrew. </off topic> This (-Wa,-q for this case of compilers/OS) is something MPICH can choose to document for this case if they are interested. Or (sorry if I'm a broken record here) you can install MacPorts and use 'sudo port install mpich-gcc5' (most of the dependencies are available as pre-built packages, so it goes fairly quick.) I'm not sure how the Homebrew MPICH recipe is configured and how to build this flavor with it; I suppose you could contact its maintainer. - Eric On Fri, Oct 30, 2015 at 1:29 PM, Ghosh, Debojyoti <ghosh5(a)llnl.gov> wrote: > Eric, adding those flags worked. Thanks! > Rob, yes, bug report from libressl matched with what I was getting. > > Can you please explain what was wrong? I installed the GNU compilers using > Homebrew. I realize I should have used “brew link gcc5” or something, but > instead, I went to /usr/local/bin and manually created the symlinks (gcc, > gfortran, g++, etc) pointing to gcc-5, gfortran-5, etc. Did I miss > something? > > I did not face this issue when I did the same thing (Homebrew gcc + compile > MPICH + manual symlinks) on a Macbook. The only difference I see there is > that the build is apple-darwin14.4.0. > > > > On Oct 30, 2015, at 8:57 AM, Eric A. Borisch <eborisch(a)gmail.com> wrote: > > You will need to add '-Wa,-q' for CFLAGS and CXXFLAGS for gcc* on > Yosemite and above. This will use the clang assembler. (See man as) > > <plug>The MacPorts mpich package does this automatically when needed.</plug> > > - Eric > > On Fri, Oct 30, 2015 at 10:13 AM, Rob Latham <robl(a)mcs.anl.gov> wrote: > > > > On 10/28/2015 12:46 PM, Ghosh, Debojyoti wrote: > > > Hi, > > I am getting a compilation error while trying to compile MPICH on my Mac > desktop. I installed the latest GNU compilers using Homebrew. > > gcc -v > > Using built-in specs. > COLLECT_GCC=gcc > > COLLECT_LTO_WRAPPER=/usr/local/Cellar/gcc/5.2.0/libexec/gcc/x86_64-apple-darwin14.5.0/5.2.0/lto-wrapper > Target: x86_64-apple-darwin14.5.0 > Configured with: ../configure --build=x86_64-apple-darwin14.5.0 > --prefix=/usr/local/Cellar/gcc/5.2.0 > --libdir=/usr/local/Cellar/gcc/5.2.0/lib/gcc/5 > --enable-languages=c,c++,objc,obj-c++,fortran --program-suffix=-5 > --with-gmp=/usr/local/opt/gmp --with-mpfr=/usr/local/opt/mpfr > --with-mpc=/usr/local/opt/libmpc --with-isl=/usr/local/opt/isl > --with-system-zlib --enable-libstdcxx-time=yes --enable-stage1-checking > --enable-checking=release --enable-lto > --with-build-config=bootstrap-debug --disable-werror > --with-pkgversion='Homebrew gcc 5.2.0' > --with-bugurl=https://github.com/Homebrew/homebrew/issues > --enable-plugin --disable-nls --enable-multilib > Thread model: posix > gcc version 5.2.0 (Homebrew gcc 5.2.0) > > > > We test with gcc-5.2 on our mac build bots, so it's definitely possible. > > I am getting the following error while compiling MPICH: > CC src/mpltrmem.lo > /var/folders/qf/q3tp7kgn0gz6qj5_c2nr4qqm001qq3/T//ccyirdXM.s:491:suffix > or operands invalid for `movq' > /var/folders/qf/q3tp7kgn0gz6qj5_c2nr4qqm001qq3/T//ccyirdXM.s:495:suffix > or operands invalid for `movq' > make[2]: *** [src/mpltrmem.lo] Error 1 > make[1]: *** [all-recursive] Error 1 > make: *** [all] Error 2 > > > > When the assembler is suspicous ('movq') then it sure seems like there's a > toolchain mismatch. > > Does this bug report from libressl sound like your environment? > https://github.com/libressl-portable/portable/issues/121 > > ==rob > > > -- > Rob Latham > Mathematics and Computer Science Division > Argonne National Lab, IL USA > _______________________________________________ > 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] Error in compiling MPICH on Mac desktop
by Ghosh, Debojyoti 30 Oct '15

30 Oct '15

Eric, adding those flags worked. Thanks! Rob, yes, bug report from libressl matched with what I was getting. Can you please explain what was wrong? I installed the GNU compilers using Homebrew. I realize I should have used “brew link gcc5” or something, but instead, I went to /usr/local/bin and manually created the symlinks (gcc, gfortran, g++, etc) pointing to gcc-5, gfortran-5, etc. Did I miss something? I did not face this issue when I did the same thing (Homebrew gcc + compile MPICH + manual symlinks) on a Macbook. The only difference I see there is that the build is apple-darwin14.4.0. On Oct 30, 2015, at 8:57 AM, Eric A. Borisch <eborisch(a)gmail.com<mailto:[email protected]>> wrote: You will need to add '-Wa,-q' for CFLAGS and CXXFLAGS for gcc* on Yosemite and above. This will use the clang assembler. (See man as) <plug>The MacPorts mpich package does this automatically when needed.</plug> - Eric On Fri, Oct 30, 2015 at 10:13 AM, Rob Latham <robl(a)mcs.anl.gov<mailto:[email protected]>> wrote: On 10/28/2015 12:46 PM, Ghosh, Debojyoti wrote: Hi, I am getting a compilation error while trying to compile MPICH on my Mac desktop. I installed the latest GNU compilers using Homebrew. gcc -v Using built-in specs. COLLECT_GCC=gcc COLLECT_LTO_WRAPPER=/usr/local/Cellar/gcc/5.2.0/libexec/gcc/x86_64-apple-darwin14.5.0/5.2.0/lto-wrapper Target: x86_64-apple-darwin14.5.0 Configured with: ../configure --build=x86_64-apple-darwin14.5.0 --prefix=/usr/local/Cellar/gcc/5.2.0 --libdir=/usr/local/Cellar/gcc/5.2.0/lib/gcc/5 --enable-languages=c,c++,objc,obj-c++,fortran --program-suffix=-5 --with-gmp=/usr/local/opt/gmp --with-mpfr=/usr/local/opt/mpfr --with-mpc=/usr/local/opt/libmpc --with-isl=/usr/local/opt/isl --with-system-zlib --enable-libstdcxx-time=yes --enable-stage1-checking --enable-checking=release --enable-lto --with-build-config=bootstrap-debug --disable-werror --with-pkgversion='Homebrew gcc 5.2.0' --with-bugurl=https://github.com/Homebrew/homebrew/issues --enable-plugin --disable-nls --enable-multilib Thread model: posix gcc version 5.2.0 (Homebrew gcc 5.2.0) We test with gcc-5.2 on our mac build bots, so it's definitely possible. I am getting the following error while compiling MPICH: CC src/mpltrmem.lo /var/folders/qf/q3tp7kgn0gz6qj5_c2nr4qqm001qq3/T//ccyirdXM.s:491:suffix or operands invalid for `movq' /var/folders/qf/q3tp7kgn0gz6qj5_c2nr4qqm001qq3/T//ccyirdXM.s:495:suffix or operands invalid for `movq' make[2]: *** [src/mpltrmem.lo] Error 1 make[1]: *** [all-recursive] Error 1 make: *** [all] Error 2 When the assembler is suspicous ('movq') then it sure seems like there's a toolchain mismatch. Does this bug report from libressl sound like your environment? https://github.com/libressl-portable/portable/issues/121 ==rob -- Rob Latham Mathematics and Computer Science Division Argonne National Lab, IL USA _______________________________________________ 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<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

long message testing in MPICH test suite
by Rusty Lusk 30 Oct '15

30 Oct '15

In a recent exchange with the OSU folks, it was discovered that there is a gap in the MPICH test suite in that the code that tests for the correctness of handling long messages forgets to test that count is handled correctly. This led to the escape of a bug in the MVAPICH code, since corrected. Hari tracked this down and has sent a patch. Could the MPICH team look at this? Thanks, Rusty Begin forwarded message: > From: Hari Subramoni <subramoni.1(a)osu.edu> > Subject: Re: [lcrc-support #279383] Re: Long messages > Date: October 26, 2015 at 12:19:42 PM CDT > To: Rusty Lusk <lusk(a)mcs.anl.gov> > Cc: "Low, John J." <jlow(a)mcs.anl.gov>, Steve Pieper <spieper(a)anl.gov>, "Wiringa, Robert B." <wiringa(a)anl.gov>, "D.K. Panda" <panda(a)cse.ohio-state.edu>, "support(a)lcrc.anl.gov" <support(a)lcrc.anl.gov>, "Balaji, Pavan" <balaji(a)anl.gov> > > Hello All, > > I was investigating to see how this was missed earlier. In our MPICH large-message test (test/mpi/pt2pt/large_message.c) which should've caught this, the code which tests the result of MPI_Get_count was missing. > > I was checking the large_message.c test that is shipped with MPICH-3.1.4 (latest stable) and MPICH-3.2rc1 (preview release) available from the MPICH site. Both have the same issue. The large-message-test.patch should fix this. > > If the patch looks good, it would be good if this gets applied to MPICH itself so that all MPICH derivatives can benefit from it. > > Best Regards, > Hari. >

1 0

Re: [mpich-discuss] Error in compiling MPICH on Mac desktop
by Eric A. Borisch 30 Oct '15

30 Oct '15

You will need to add '-Wa,-q' for CFLAGS and CXXFLAGS for gcc* on Yosemite and above. This will use the clang assembler. (See man as) <plug>The MacPorts mpich package does this automatically when needed.</plug> - Eric On Fri, Oct 30, 2015 at 10:13 AM, Rob Latham <robl(a)mcs.anl.gov> wrote: > > > On 10/28/2015 12:46 PM, Ghosh, Debojyoti wrote: >> >> Hi, >> >> I am getting a compilation error while trying to compile MPICH on my Mac >> desktop. I installed the latest GNU compilers using Homebrew. >> >> > gcc -v >> Using built-in specs. >> COLLECT_GCC=gcc >> >> COLLECT_LTO_WRAPPER=/usr/local/Cellar/gcc/5.2.0/libexec/gcc/x86_64-apple-darwin14.5.0/5.2.0/lto-wrapper >> Target: x86_64-apple-darwin14.5.0 >> Configured with: ../configure --build=x86_64-apple-darwin14.5.0 >> --prefix=/usr/local/Cellar/gcc/5.2.0 >> --libdir=/usr/local/Cellar/gcc/5.2.0/lib/gcc/5 >> --enable-languages=c,c++,objc,obj-c++,fortran --program-suffix=-5 >> --with-gmp=/usr/local/opt/gmp --with-mpfr=/usr/local/opt/mpfr >> --with-mpc=/usr/local/opt/libmpc --with-isl=/usr/local/opt/isl >> --with-system-zlib --enable-libstdcxx-time=yes --enable-stage1-checking >> --enable-checking=release --enable-lto >> --with-build-config=bootstrap-debug --disable-werror >> --with-pkgversion='Homebrew gcc 5.2.0' >> --with-bugurl=https://github.com/Homebrew/homebrew/issues >> --enable-plugin --disable-nls --enable-multilib >> Thread model: posix >> gcc version 5.2.0 (Homebrew gcc 5.2.0) > > > We test with gcc-5.2 on our mac build bots, so it's definitely possible. > >> I am getting the following error while compiling MPICH: >> CC src/mpltrmem.lo >> /var/folders/qf/q3tp7kgn0gz6qj5_c2nr4qqm001qq3/T//ccyirdXM.s:491:suffix >> or operands invalid for `movq' >> /var/folders/qf/q3tp7kgn0gz6qj5_c2nr4qqm001qq3/T//ccyirdXM.s:495:suffix >> or operands invalid for `movq' >> make[2]: *** [src/mpltrmem.lo] Error 1 >> make[1]: *** [all-recursive] Error 1 >> make: *** [all] Error 2 > > > When the assembler is suspicous ('movq') then it sure seems like there's a > toolchain mismatch. > > Does this bug report from libressl sound like your environment? > https://github.com/libressl-portable/portable/issues/121 > > ==rob > > > -- > Rob Latham > Mathematics and Computer Science Division > Argonne National Lab, IL USA > _______________________________________________ > 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] Error in compiling MPICH on Mac desktop
by Rob Latham 30 Oct '15

30 Oct '15

On 10/28/2015 12:46 PM, Ghosh, Debojyoti wrote: > Hi, > > I am getting a compilation error while trying to compile MPICH on my Mac > desktop. I installed the latest GNU compilers using Homebrew. > > > gcc -v > Using built-in specs. > COLLECT_GCC=gcc > COLLECT_LTO_WRAPPER=/usr/local/Cellar/gcc/5.2.0/libexec/gcc/x86_64-apple-darwin14.5.0/5.2.0/lto-wrapper > Target: x86_64-apple-darwin14.5.0 > Configured with: ../configure --build=x86_64-apple-darwin14.5.0 > --prefix=/usr/local/Cellar/gcc/5.2.0 > --libdir=/usr/local/Cellar/gcc/5.2.0/lib/gcc/5 > --enable-languages=c,c++,objc,obj-c++,fortran --program-suffix=-5 > --with-gmp=/usr/local/opt/gmp --with-mpfr=/usr/local/opt/mpfr > --with-mpc=/usr/local/opt/libmpc --with-isl=/usr/local/opt/isl > --with-system-zlib --enable-libstdcxx-time=yes --enable-stage1-checking > --enable-checking=release --enable-lto > --with-build-config=bootstrap-debug --disable-werror > --with-pkgversion='Homebrew gcc 5.2.0' > --with-bugurl=https://github.com/Homebrew/homebrew/issues > --enable-plugin --disable-nls --enable-multilib > Thread model: posix > gcc version 5.2.0 (Homebrew gcc 5.2.0) We test with gcc-5.2 on our mac build bots, so it's definitely possible. > I am getting the following error while compiling MPICH: > CC src/mpltrmem.lo > /var/folders/qf/q3tp7kgn0gz6qj5_c2nr4qqm001qq3/T//ccyirdXM.s:491:suffix > or operands invalid for `movq' > /var/folders/qf/q3tp7kgn0gz6qj5_c2nr4qqm001qq3/T//ccyirdXM.s:495:suffix > or operands invalid for `movq' > make[2]: *** [src/mpltrmem.lo] Error 1 > make[1]: *** [all-recursive] Error 1 > make: *** [all] Error 2 When the assembler is suspicous ('movq') then it sure seems like there's a toolchain mismatch. Does this bug report from libressl sound like your environment? https://github.com/libressl-portable/portable/issues/121 ==rob -- Rob Latham Mathematics and Computer Science Division Argonne National Lab, IL USA _______________________________________________ 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] location of MPI_IBCAST
by Min Si 29 Oct '15

29 Oct '15

Hi Jianhua, Non-blocking collective (e.g., MPI_IBCAST) is added from MPI-3.0. Min On 10/29/15 10:39 AM, Balaji, Pavan wrote: > Sorry, windows is not supported. > > https://wiki.mpich.org/mpich/index.php/Frequently_Asked_Questions#Q:_Why_ca… > >> On Oct 29, 2015, at 10:35 AM, Jianhua Jiang <jjiang(a)breault.com> wrote: >> >> Hi, all: >> >> I’m trying to use MPI_IBCAST (the non-blocking version) instead of MPI_BCAST for MPI broadcasting . However I got a link error of unresolved external symbol for MPi_IBCAST. I put cxx.lib, mpi.lib, fmpich2.lib and fmpich2g.lib in the linker input list. Does MPICH2 implement MPI_IBCAST? If so, which library shall I use to access it? >> >> I’m using the binary package from Jayesh Krishna on a 64-bit Windows 7 PC. It’s based on mpich2 1.4.1p1. >> >> Thanks, >> >> >> Jianhua(John) Jiang >> _______________________________________________ >> 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] location of MPI_IBCAST
by Balaji, Pavan 29 Oct '15

29 Oct '15

Sorry, windows is not supported. https://wiki.mpich.org/mpich/index.php/Frequently_Asked_Questions#Q:_Why_ca… > On Oct 29, 2015, at 10:35 AM, Jianhua Jiang <jjiang(a)breault.com> wrote: > > Hi, all: > > I’m trying to use MPI_IBCAST (the non-blocking version) instead of MPI_BCAST for MPI broadcasting . However I got a link error of unresolved external symbol for MPi_IBCAST. I put cxx.lib, mpi.lib, fmpich2.lib and fmpich2g.lib in the linker input list. Does MPICH2 implement MPI_IBCAST? If so, which library shall I use to access it? > > I’m using the binary package from Jayesh Krishna on a 64-bit Windows 7 PC. It’s based on mpich2 1.4.1p1. > > Thanks, > > > Jianhua(John) Jiang > _______________________________________________ > 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

location of MPI_IBCAST
by Jianhua Jiang 29 Oct '15

29 Oct '15

Hi, all: I'm trying to use MPI_IBCAST (the non-blocking version) instead of MPI_BCAST for MPI broadcasting . However I got a link error of unresolved external symbol for MPi_IBCAST. I put cxx.lib, mpi.lib, fmpich2.lib and fmpich2g.lib in the linker input list. Does MPICH2 implement MPI_IBCAST? If so, which library shall I use to access it? I'm using the binary package from Jayesh Krishna<mailto:[email protected]> on a 64-bit Windows 7 PC. It's based on mpich2 1.4.1p1. Thanks, Jianhua(John) Jiang _______________________________________________ discuss mailing list discuss(a)mpich.org To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss

1 0