So are you saying that the parallel build uses files created in the serial
build? It was ambiguous in the tutorial why they want you to do the serial
build first. I assumed it was just so you know what errors happen in each
version.
On Fri, Jan 13, 2017 at 3:26 PM, Hai Nguyen <nhai.qn.gmail.com> wrote:
> You told "I have the serial build in a completely different folder"
>
> What do you mean here?
>
> Hai
>
> On Fri, Jan 13, 2017 at 3:25 PM, Ada Sedova <ada.a.sedova.gmail.com>
> wrote:
>
> > I already did the serial build first, as in my previous email.
> >
> > On Fri, Jan 13, 2017 at 3:23 PM, Hai Nguyen <nhai.qn.gmail.com> wrote:
> >
> > > Hi
> > >
> > > > I have the serial build in a completely different folder, and I ran
> the
> > > parallel build complete anew, from a different login, even.
> > >
> > > Try to do the serial build first, then parallel build. Both should
> happen
> > > in the same folder.
> > >
> > > *> page 23 of amber16 manual says: *
> > > *> "Do not choose any parallel options at this step! You will need to
> > > install *
> > > *> the serial version first; options for parallel builds are described
> > > below *
> > >
> > >
> > > *> at Step 8. " *
> > > Hai
> > >
> > > On Fri, Jan 13, 2017 at 3:12 PM, Ada Sedova <ada.a.sedova.gmail.com>
> > > wrote:
> > >
> > > > Thanks Scott,
> > > >
> > > > To answer a few of your questions:
> > > >
> > > > No, ORNL does not have a supported amber installation. We have a site
> > > > license for amber14 only, which is why no upgrade. If we want to use
> it
> > > we
> > > > have to download it ourselves and build.
> > > >
> > > > Yes, I am very familiar with the compiling procedure at ORNL and I
> have
> > > > submitted a help ticket, but currently no one with a good amount of
> > > > compiling experience here can figure out how to deal with this
> problem.
> > > >
> > > > I have the serial build in a completely different folder, and I ran
> the
> > > > parallel build complete anew, from a different login, even.
> > > >
> > > > Yes, I do need parallel executable, as the reason we are trying to
> use
> > > > amber is for the DFT-MD for a larger small system.
> > > >
> > > > I will check the serial tests. That is strange.
> > > >
> > > >
> > > >
> > > > Thanks again,
> > > >
> > > > Ada
> > > >
> > > >
> > > >
> > > > On Fri, Jan 13, 2017 at 2:35 PM, Scott Brozell <
> > sbrozell.rci.rutgers.edu
> > > >
> > > > wrote:
> > > >
> > > > > Hi,
> > > > >
> > > > > The most likely possibilities are probably:
> > > > > 1. broken parallel tools;
> > > > > 2. conflicting serial and parallel tools;
> > > > > 3. residual conflicting amber files between the serial and parallel
> > > > builds.
> > > > >
> > > > > The apparent first error:
> > > > > > /usr/bin/ld: reader.o(.text+0x5a): unresolvable R_X86_64_PC32
> > > > relocation
> > > > > might be fixed with compiling options like -fpic.
> > > > >
> > > > > But my advice is to take a step back and tackle these bigger
> picture
> > > > items:
> > > > >
> > > > > Doesn't ORNL have supported amber installations ?
> > > > > Did you read the ORNL documentation on installing software, etc ?
> > > > > Did you ask them for help ?
> > > > > Do you need parallel amber executables ?
> > > > > When will you upgrade to amber 16 ?
> > > > > Double check your serial tests - i haven't seen all passing tests
> in
> > > > > amber for at least a decade.
> > > > >
> > > > > If you still need help from us after investigating those then send
> > > > > verbose and verbatim details: compiler names, versions, -show
> > output,
> > > > > your config.h s, the redhat-release, the complete build logs, etc.
> > > > >
> > > > > scott
> > > > >
> > > > > On Fri, Jan 13, 2017 at 12:52:40PM -0500, Ada Sedova wrote:
> > > > > > I am trying to build amber14/amberTools on RHEA, an OLCF (ORNL)
> > > > commodity
> > > > > > cluster, with dual Intel® Xeon® E5-2650 CPUs, which uses modules.
> > > > > >
> > > > > > The build was successful on the same system in serial, and all
> > tests
> > > > > passed
> > > > > > after building.
> > > > > >
> > > > > > However, when trying to build with mpi, I am exiting the make
> > install
> > > > > from
> > > > > > the following folder:
> > > > > >
> > > > > > make[2]: Entering directory
> > > > > > `/autofs/nccs-svm1_home2/adaa/AMBER/amber14/AmberTools/src/
> > cifparse'
> > > > > >
> > > > > > flex -t cifparse.l | sed 's/yy/cifp/g' > lex.cif.c
> > > > > >
> > > > > > /ccs/home/adaa/AMBER/amber14/bin/yacc -v -d cifparse.y
> > > > > >
> > > > > > make[2]: Leaving directory
> > > > > > `/autofs/nccs-svm1_home2/adaa/AMBER/amber14/AmberTools/src/
> > cifparse'
> > > > > >
> > > > > > make[1]: Leaving directory
> > > > > > `/autofs/nccs-svm1_home2/adaa/AMBER/amber14/AmberTools/src'
> > > > > >
> > > > > > with the following error:
> > > > > >
> > > > > > main.o: In function `create_file_names':
> > > > > >
> > > > > > main.c:(.text+0x607): warning: the use of `mktemp' is dangerous,
> > > better
> > > > > use
> > > > > > `mkstemp'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x5a): unresolvable R_X86_64_PC32
> > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x6c): unresolvable R_X86_64_PC32
> > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x73): unresolvable R_X86_64_PC32
> > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x84): unresolvable R_X86_64_PC32
> > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x5b1): unresolvable R_X86_64_PC32
> > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x5c9): unresolvable R_X86_64_PC32
> > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x5eb): unresolvable R_X86_64_PC32
> > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x60d): unresolvable R_X86_64_PC32
> > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x62f): unresolvable R_X86_64_PC32
> > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x651): unresolvable R_X86_64_PC32
> > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x669): unresolvable R_X86_64_PC32
> > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x68b): unresolvable R_X86_64_PC32
> > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x6ad): unresolvable R_X86_64_PC32
> > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x6cf): unresolvable R_X86_64_PC32
> > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x1590): unresolvable R_X86_64_PC32
> > > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x15f7): unresolvable R_X86_64_PC32
> > > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x1785): unresolvable R_X86_64_PC32
> > > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x194f): unresolvable R_X86_64_PC32
> > > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x1967): unresolvable R_X86_64_PC32
> > > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x1976): unresolvable R_X86_64_PC32
> > > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x1b70): unresolvable R_X86_64_PC32
> > > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x1c41): unresolvable R_X86_64_PC32
> > > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x1ffb): unresolvable R_X86_64_PC32
> > > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x2002): unresolvable R_X86_64_PC32
> > > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x2659): unresolvable R_X86_64_PC32
> > > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x2693): unresolvable R_X86_64_PC32
> > > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > > /usr/bin/ld: reader.o(.text+0x26ac): unresolvable R_X86_64_PC32
> > > > > relocation
> > > > > > against symbol `cache'
> > > > > >
> > > > > >
> > > > > > yacc:3249 terminated with signal 11 at PC=40abac SP=7fff87123600.
> > > > > > Backtrace:
> > > > > >
> > > > > > /ccs/home/adaa/AMBER/amber14/bin/yacc[0x40abac]
> > > > > >
> > > > > > /ccs/home/adaa/AMBER/amber14/bin/yacc[0x40ce6b]
> > > > > >
> > > > > > /ccs/home/adaa/AMBER/amber14/bin/yacc[0x404dfd]
> > > > > >
> > > > > > /lib64/libc.so.6(__libc_start_main+0xfd)[0x7eff172bed1d]
> > > > > >
> > > > > > /ccs/home/adaa/AMBER/amber14/bin/yacc[0x400ea9]
> > > > > >
> > > > > > make[2]: *** [cifp.tab.c] Error 1
> > > > > >
> > > > > > make[1]: *** [parallel] Error 2
> > > > > >
> > > > > > make: *** [install] Error 2
> > > > >
> > > > > _______________________________________________
> > > > > AMBER mailing list
> > > > > AMBER.ambermd.org
> > > > > http://lists.ambermd.org/mailman/listinfo/amber
> > > > >
> > > > _______________________________________________
> > > > AMBER mailing list
> > > > AMBER.ambermd.org
> > > > http://lists.ambermd.org/mailman/listinfo/amber
> > > >
> > > _______________________________________________
> > > AMBER mailing list
> > > AMBER.ambermd.org
> > > http://lists.ambermd.org/mailman/listinfo/amber
> > >
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> >
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Jan 13 2017 - 12:30:05 PST
