March 2016 - discuss - mailman3.cels.anl.gov

Re: [mpich-discuss] unable to build 3.2
by sanjesh pant 14 Mar '16

14 Mar '16

Configuring mpich-3.2can be done as ./configure -prefix=/home/software/mpich3.2-install |& tee c.txt To build mpich Make |& tee m.txt I think it will work. Sanjesh On 14 Mar 2016 19:48, "Rob Latham" <robl(a)mcs.anl.gov> wrote: > > > On 03/14/2016 07:06 AM, Peter Steinbach wrote: > >> Hi - >> >> I just downloaded mpich 3.2 and wanted to build it on a centos 6.3 based >> system (gcc 4.9.2). The configure call looked like this: >> $ ./configure --with-ibverbs --with-ibverbs-include=/usr/include >> --with-ibverbs-lib=/usr/lib64 --enable-fast --enable-shared --with-java >> --prefix=$HOME/software/mpich/3.2 >> >> the configure runs through, but the make call finishes with: >> $ make >> ... >> ( cd . && \ >> ./maint/extracterrmsgs -careful=unusederr.txt \ >> -skip=src/util/multichannel/mpi.c \ >> `cat maint/errmsgdirs` > .tmp 2>.err ) >> ( cd . && if test -s .err ; then rm -f .tmp ; cat .err ; exit 1 ; fi ) >> Warning: attempt to redefine **fail. Duplicate ignored. >> Warning: **fail . . >> src/mpid/ch3/channels/nemesis/netmod/llc/errnames.txt : 15 >> make: *** [src/mpi/errhan/defmsg.h] Error 1 >> >> The problem appears to be coming from: >> $ egrep '^\*\*fail:' `find -name "errnames.txt"` >> ./src/mpi/errhan/errnames.txt:**fail: >> ./src/mpid/ch3/channels/nemesis/netmod/llc/errnames.txt:**fail:fail failed >> > > I thought we fixed this? Maybe it went in just after the mpich-3.2 > release. > > >> Any idea how to resolve it? >> > > Ah, it sure did. 4 commits after 3.2 was tagged in fact. Simply delete > the **fail line from > src/mpid/ch3/channels/nemesis/netmod/llc/errnames.txt. I've attached a > patch if that's easier for you: > > ==rob > > > > Best, >> P >> >> _______________________________________________ >> 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] unable to build 3.2
by Rob Latham 14 Mar '16

14 Mar '16

On 03/14/2016 07:06 AM, Peter Steinbach wrote: > Hi - > > I just downloaded mpich 3.2 and wanted to build it on a centos 6.3 based > system (gcc 4.9.2). The configure call looked like this: > $ ./configure --with-ibverbs --with-ibverbs-include=/usr/include > --with-ibverbs-lib=/usr/lib64 --enable-fast --enable-shared --with-java > --prefix=$HOME/software/mpich/3.2 > > the configure runs through, but the make call finishes with: > $ make > ... > ( cd . && \ > ./maint/extracterrmsgs -careful=unusederr.txt \ > -skip=src/util/multichannel/mpi.c \ > `cat maint/errmsgdirs` > .tmp 2>.err ) > ( cd . && if test -s .err ; then rm -f .tmp ; cat .err ; exit 1 ; fi ) > Warning: attempt to redefine **fail. Duplicate ignored. > Warning: **fail . . > src/mpid/ch3/channels/nemesis/netmod/llc/errnames.txt : 15 > make: *** [src/mpi/errhan/defmsg.h] Error 1 > > The problem appears to be coming from: > $ egrep '^\*\*fail:' `find -name "errnames.txt"` > ./src/mpi/errhan/errnames.txt:**fail: > ./src/mpid/ch3/channels/nemesis/netmod/llc/errnames.txt:**fail:fail failed I thought we fixed this? Maybe it went in just after the mpich-3.2 release. > > Any idea how to resolve it? Ah, it sure did. 4 commits after 3.2 was tagged in fact. Simply delete the **fail line from src/mpid/ch3/channels/nemesis/netmod/llc/errnames.txt. I've attached a patch if that's easier for you: ==rob > Best, > P > > _______________________________________________ > 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

unable to build 3.2
by Peter Steinbach 14 Mar '16

14 Mar '16

Hi - I just downloaded mpich 3.2 and wanted to build it on a centos 6.3 based system (gcc 4.9.2). The configure call looked like this: $ ./configure --with-ibverbs --with-ibverbs-include=/usr/include --with-ibverbs-lib=/usr/lib64 --enable-fast --enable-shared --with-java --prefix=$HOME/software/mpich/3.2 the configure runs through, but the make call finishes with: $ make ... ( cd . && \ ./maint/extracterrmsgs -careful=unusederr.txt \ -skip=src/util/multichannel/mpi.c \ `cat maint/errmsgdirs` > .tmp 2>.err ) ( cd . && if test -s .err ; then rm -f .tmp ; cat .err ; exit 1 ; fi ) Warning: attempt to redefine **fail. Duplicate ignored. Warning: **fail . . src/mpid/ch3/channels/nemesis/netmod/llc/errnames.txt : 15 make: *** [src/mpi/errhan/defmsg.h] Error 1 The problem appears to be coming from: $ egrep '^\*\*fail:' `find -name "errnames.txt"` ./src/mpi/errhan/errnames.txt:**fail: ./src/mpid/ch3/channels/nemesis/netmod/llc/errnames.txt:**fail:fail failed Any idea how to resolve it? Best, P _______________________________________________ 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] about the makefile
by Hardwood Bao 13 Mar '16

13 Mar '16

Hey Halim, The makefile is produced after configure. It's very complex, so i don't know how to modify it. by Hardwood. 2016-03-14 8:00 GMT+08:00 Halim Amer <aamer(a)anl.gov>: > I don't understand what you are trying to do. Are you trying to extend > MPICH to support a different transport layer than what it already supports? > > Which Makefile are you referring to? > > --Halim > www.mcs.anl.gov/~aamer > > > On 3/12/16 9:30 PM, Hardwood Bao wrote: > >> hello,everyone. >> I want to invoke some third-party libraries to send and receive the >> message. But i don't know how to modify the makefile and how to include >> the >> >> this is a example when using the library. >> g++ -o normFileRecv normFileRecv.cpp -D_FILE_OFFSET_BITS=64 -I../common/ \ >> -I../protolib/include -I../include/ ../lib/libnorm.a >> ../protolib/lib/libprotokit.a \ >> -lpthread >> >> the third-party libraries include libnorm.a, libprotokit.a, and lpthread. >> and the include files are under the directory of common and >> protolib/include. >> >> If i want to involve the API in the file: >> src\mpid\ch3\channels\nemesis\src\Ch3-istartmsg.c >> >> how should i modify the makefile? >> I am really appreciate your help. >> >> Best Regards. >> >> >> >> _______________________________________________ >> 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] about the makefile
by Hardwood Bao 13 Mar '16

13 Mar '16

Hey, Halim. I want to involve a nack-based reliable multicast protocol to send and receive data, which is based on udp protocol. What do you mean that it already supports? by Hardwood. 2016-03-14 8:00 GMT+08:00 Halim Amer <aamer(a)anl.gov>: > I don't understand what you are trying to do. Are you trying to extend > MPICH to support a different transport layer than what it already supports? > > Which Makefile are you referring to? > > --Halim > www.mcs.anl.gov/~aamer > > > On 3/12/16 9:30 PM, Hardwood Bao wrote: > >> hello,everyone. >> I want to invoke some third-party libraries to send and receive the >> message. But i don't know how to modify the makefile and how to include >> the >> >> this is a example when using the library. >> g++ -o normFileRecv normFileRecv.cpp -D_FILE_OFFSET_BITS=64 -I../common/ \ >> -I../protolib/include -I../include/ ../lib/libnorm.a >> ../protolib/lib/libprotokit.a \ >> -lpthread >> >> the third-party libraries include libnorm.a, libprotokit.a, and lpthread. >> and the include files are under the directory of common and >> protolib/include. >> >> If i want to involve the API in the file: >> src\mpid\ch3\channels\nemesis\src\Ch3-istartmsg.c >> >> how should i modify the makefile? >> I am really appreciate your help. >> >> Best Regards. >> >> >> >> _______________________________________________ >> 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] about the makefile
by Halim Amer 13 Mar '16

13 Mar '16

I don't understand what you are trying to do. Are you trying to extend MPICH to support a different transport layer than what it already supports? Which Makefile are you referring to? --Halim www.mcs.anl.gov/~aamer On 3/12/16 9:30 PM, Hardwood Bao wrote: > hello,everyone. > I want to invoke some third-party libraries to send and receive the > message. But i don't know how to modify the makefile and how to include the > > this is a example when using the library. > g++ -o normFileRecv normFileRecv.cpp -D_FILE_OFFSET_BITS=64 -I../common/ \ > -I../protolib/include -I../include/ ../lib/libnorm.a > ../protolib/lib/libprotokit.a \ > -lpthread > > the third-party libraries include libnorm.a, libprotokit.a, and lpthread. > and the include files are under the directory of common and > protolib/include. > > If i want to involve the API in the file: > src\mpid\ch3\channels\nemesis\src\Ch3-istartmsg.c > > how should i modify the makefile? > I am really appreciate your help. > > Best Regards. > > > > _______________________________________________ > 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 with MPI_Register_datarep
by Thakur, Rajeev 13 Mar '16

13 Mar '16

To get file portability, most users use some higher-level library such as PnetCDF or HDF-5. Maybe you could write a small utility for the 32-bit users to do the data conversion before using the data. > Maybe I can contribute? :) Code contributions are always welcome :-). I will note that MPICH didn’t support the external32 file format for many years until Intel contributed code for it. Rajeev > On Mar 13, 2016, at 4:49 PM, Eric Chamberland <Eric.Chamberland(a)giref.ulaval.ca> wrote: > > Le 2016-03-13 16:22, Thakur, Rajeev a écrit : >> In the 19 years since the MPI-2 standard was released, you may be the first one requesting this functionality :-). I am curious what are you using it for? > Here is the reason.... > > In 2012, when I read your book about MPI-2, I wrote all our I/O functions to use collective MPI I/O calls with "native" as datarep. I knew it would not be portable "as is" but the main usage is 64bits native format... then I told myself "It will be ok when I will want to read/write from 32 bits architectures or different endian style, since there is that user defined datarep possibility, as illustrated in chapet 9.1 of the book...", but unfortunately for me, I didn't _tested_ the functionality in 2012... :/ > > So we moved all our file formats (meshes, results, etc) to be aligned with the output of "native" datarep into a 64bits standard linux x86 architecture... That means we have produced, for example, 1200 validations tests with input/output files into that format... > > That is, of course, excluding all user data since 4 years.. > > No I am in the phase of porting the code to fully support 32bits architecture (we have an 32bit ubuntu running here, and working on windows port). > > So, I did test for the first time the datarep functionality to be able to convert 64 bits "long int" from/to 32 bits "long int" for writing/reading in respect to existing file format... > > It was all an easy task in my mind with the user datarep functionality just as it is in the standard.. > > But it appears I am the only one in the world (wow...) to request a way to have his files portable with MPI... without performance losses for the "main" usage which is 64bits... > > Maybe there is another simple way to do this? > > Anyway, I am now expecting that you will tell me it will never be implemented _just for me_... :/ > > Maybe I can contribute? :) > > Thanks, > > Eric > >> >> Rajeev >> >> >>> On Mar 13, 2016, at 1:51 PM, Eric Chamberland <Eric.Chamberland(a)giref.ulaval.ca> wrote: >>> >>> Le 2016-03-10 18:04, Thakur, Rajeev a écrit : >>>> Support for user-defined data representations has not been added to MPICH yet. There has not been much demand for it. >>> Thanks for your answer... >>> >>> Is it a "big" task? >>> >>> Should it be scheduled so it may appear in a forthcoming version? >>> >>> Thanks, >>> >>> Eric >>> >>> _______________________________________________ >>> 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 with MPI_Register_datarep
by Jeff Hammond 13 Mar '16

13 Mar '16

Just tell your 32b Ubuntu users to upgrade to 64b. That's the right solution here. As for Windows, that is irrelevant, because MPICH does not support Windows. Jeff On Sunday, March 13, 2016, Eric Chamberland < Eric.Chamberland(a)giref.ulaval.ca> wrote: > Le 2016-03-13 16:22, Thakur, Rajeev a écrit : > >> In the 19 years since the MPI-2 standard was released, you may be the >> first one requesting this functionality :-). I am curious what are you >> using it for? >> > Here is the reason.... > > In 2012, when I read your book about MPI-2, I wrote all our I/O functions > to use collective MPI I/O calls with "native" as datarep. I knew it would > not be portable "as is" but the main usage is 64bits native format... then > I told myself "It will be ok when I will want to read/write from 32 bits > architectures or different endian style, since there is that user defined > datarep possibility, as illustrated in chapet 9.1 of the book...", but > unfortunately for me, I didn't _tested_ the functionality in 2012... :/ > > So we moved all our file formats (meshes, results, etc) to be aligned with > the output of "native" datarep into a 64bits standard linux x86 > architecture... That means we have produced, for example, 1200 validations > tests with input/output files into that format... > > That is, of course, excluding all user data since 4 years.. > > No I am in the phase of porting the code to fully support 32bits > architecture (we have an 32bit ubuntu running here, and working on windows > port). > > So, I did test for the first time the datarep functionality to be able to > convert 64 bits "long int" from/to 32 bits "long int" for writing/reading > in respect to existing file format... > > It was all an easy task in my mind with the user datarep functionality > just as it is in the standard.. > > But it appears I am the only one in the world (wow...) to request a way to > have his files portable with MPI... without performance losses for the > "main" usage which is 64bits... > > Maybe there is another simple way to do this? > > Anyway, I am now expecting that you will tell me it will never be > implemented _just for me_... :/ > > Maybe I can contribute? :) > > Thanks, > > Eric > > >> Rajeev >> >> >> On Mar 13, 2016, at 1:51 PM, Eric Chamberland < >>> Eric.Chamberland(a)giref.ulaval.ca> wrote: >>> >>> Le 2016-03-10 18:04, Thakur, Rajeev a écrit : >>> >>>> Support for user-defined data representations has not been added to >>>> MPICH yet. There has not been much demand for it. >>>> >>> Thanks for your answer... >>> >>> Is it a "big" task? >>> >>> Should it be scheduled so it may appear in a forthcoming version? >>> >>> Thanks, >>> >>> Eric >>> >>> _______________________________________________ >>> 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 > -- Jeff Hammond jeff.science(a)gmail.com http://jeffhammond.github.io/ _______________________________________________ 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 with MPI_Register_datarep
by Eric Chamberland 13 Mar '16

13 Mar '16

Le 2016-03-13 16:22, Thakur, Rajeev a écrit : > In the 19 years since the MPI-2 standard was released, you may be the first one requesting this functionality :-). I am curious what are you using it for? Here is the reason.... In 2012, when I read your book about MPI-2, I wrote all our I/O functions to use collective MPI I/O calls with "native" as datarep. I knew it would not be portable "as is" but the main usage is 64bits native format... then I told myself "It will be ok when I will want to read/write from 32 bits architectures or different endian style, since there is that user defined datarep possibility, as illustrated in chapet 9.1 of the book...", but unfortunately for me, I didn't _tested_ the functionality in 2012... :/ So we moved all our file formats (meshes, results, etc) to be aligned with the output of "native" datarep into a 64bits standard linux x86 architecture... That means we have produced, for example, 1200 validations tests with input/output files into that format... That is, of course, excluding all user data since 4 years.. No I am in the phase of porting the code to fully support 32bits architecture (we have an 32bit ubuntu running here, and working on windows port). So, I did test for the first time the datarep functionality to be able to convert 64 bits "long int" from/to 32 bits "long int" for writing/reading in respect to existing file format... It was all an easy task in my mind with the user datarep functionality just as it is in the standard.. But it appears I am the only one in the world (wow...) to request a way to have his files portable with MPI... without performance losses for the "main" usage which is 64bits... Maybe there is another simple way to do this? Anyway, I am now expecting that you will tell me it will never be implemented _just for me_... :/ Maybe I can contribute? :) Thanks, Eric > > Rajeev > > >> On Mar 13, 2016, at 1:51 PM, Eric Chamberland <Eric.Chamberland(a)giref.ulaval.ca> wrote: >> >> Le 2016-03-10 18:04, Thakur, Rajeev a écrit : >>> Support for user-defined data representations has not been added to MPICH yet. There has not been much demand for it. >> Thanks for your answer... >> >> Is it a "big" task? >> >> Should it be scheduled so it may appear in a forthcoming version? >> >> Thanks, >> >> Eric >> >> _______________________________________________ >> 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 with MPI_Register_datarep
by Thakur, Rajeev 13 Mar '16

13 Mar '16

In the 19 years since the MPI-2 standard was released, you may be the first one requesting this functionality :-). I am curious what are you using it for? Rajeev > On Mar 13, 2016, at 1:51 PM, Eric Chamberland <Eric.Chamberland(a)giref.ulaval.ca> wrote: > > Le 2016-03-10 18:04, Thakur, Rajeev a écrit : >> Support for user-defined data representations has not been added to MPICH yet. There has not been much demand for it. > Thanks for your answer... > > Is it a "big" task? > > Should it be scheduled so it may appear in a forthcoming version? > > Thanks, > > Eric > > _______________________________________________ > 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