Charm++ parallel programming system

[Phil Miller <mille121 AT illinois.edu>]

Nah, just drop the 'gcc' option and let it use the default mpicc/mpicxx.

Charm++ folks: this is a pretty classic build error that's recurred for quite a while. Maybe make it less likely that people will get dumb results from what they think are sensible options?

On Tue, Jun 4, 2019 at 11:58 PM Jeff Hammond <jeff.science AT gmail.com> wrote:

The build system can’t find MPI. I’m sure the option to point the build system to the location of MPI is documented. Please look for it.

checking "whether build on MPI"... "no"

Error: Cannot compile an MPI program

*** Please find detailed output in tmp/charmconfig.out ***

Jeff

On Tue, Jun 4, 2019 at 8:19 AM Abram, Kaleb <KZAbram AT uams.edu> wrote:

Hi Charm team,

I work at University of Arkansas for Medical Sciences and I’m trying to get NAMD running on our HPC. We’re running a mix of Xeons and KNL nodes that have OmniPath running between them. 

I think our issue is that we need an omnipath version of Charm to allow NAMD to communicate between nodes. We have modules for fftw and openmpi so that shouldn’t be the issue. 

I found this email thread and tried using that flag combination to compile Charm. However, this led to a NOT BUILT status and I’ve copied the log below. 

Any help or advice would be much appreciated!

Best,

Kaleb Abram

#########################

####### LOG BELOW #######

#########################

[kabram@login1 charm]$ ./build charm++ mpi-linux-x86_64 gcc smp --with-production -j8

Selected Compiler: gcc

Selected Options:  smp

Creating dir: mpi-linux-x86_64-smp-gcc

Creating dir: mpi-linux-x86_64-smp-gcc/tmp

Soft-linking over bin

Soft-linking over lib

Soft-linking over include

Soft-linking over tmp

Copying src/scripts/Makefile to mpi-linux-x86_64-smp-gcc/tmp

Generating mpi-linux-x86_64-smp-gcc/tmp/conv-mach-pre.sh

Generating mpi-linux-x86_64-smp-gcc/tmp/conv-mach-opt.h, conv-mach-opt.sh, conv-mach-opt.mak

Performing '/usr/bin/gmake -j 8 basics OPTS=-optimize -production QUIET= CONFIG_OPTS=--disable-controlpoint --disable-tracing --disable-tracing-commthread --disable-charmdebug --disable-replay --disable-error-checking --disable-stats ' in mpi-linux-x86_64-smp-gcc/tmp

pwd

/home/kabram/charm/mpi-linux-x86_64-smp-gcc/tmp

find . -type l -exec rm {} \;

rm -rf QuickThreads

rm -rf hwloc

rm -rf libs

rm -rf ../bin ; mkdir ../bin

if case mpi-linux-x86_64 in *win64*) true;; *-win-*) true;; *) false;; esac; then \

cp -f ../../src/arch/win/unix2nt* ../bin; \

cp -f ../../src/arch/win/createlink.exe ../bin; \

fi

rm -rf ../lib ; mkdir ../lib

rm -rf ../benchmarks

rm -rf ../examples

rm -rf ../tests

../../src/scripts/gatherflat ../../src/scripts .

./gatherflat ../../src/conv-core        .

./gatherflat ../../src/conv-ldb         .

./gatherflat ../../src/conv-ccs         .

./gatherflat ../../src/conv-perf .

./gatherflat ../../src/ck-core          .

./gatherflat ../../src/ck-perf          .

./gatherflat ../../src/ck-pics          .

./gatherflat ../../src/ck-ldb           .

./gatherflat ../../src/ck-cp            .

./gatherflat ../../src/langs/simplemsg  .

./gatherflat ../../src/langs/pvmc       .

./gatherflat ../../src/langs/bluegene   .

./gatherflat ../../src/langs/f90charm   .

./gatherflat ../../src/xlatcpm          .

./gathertree ../../src/QuickThreads     QuickThreads

./gathertree ../../src/../contrib/hwloc hwloc

./gathertree ../../src/libs             libs

./gathertree ../../src/arch/util        .

./gathertree ../../src/util             .

./gathertree ../../src/langs            langs

./gathertree ../../src/langs/jade       langs/jade

./gathertree ../../src/arch/common      .

./gathertree ../../src/arch/mpi .

test -f ../../src/arch/mpi/gdir_link && cat ../../src/arch/mpi/gdir_link > .gdir.new && ./gathertree ../../src/arch/`cat .gdir.new`      . || true

./gatherflat ../../src/arch/mpi-linux-x86_64 .

./gathertree ../../src/../benchmarks           ../benchmarks

./gathertree ../../src/../examples             ../examples

./gathertree ../../src/../tests                ../tests

rm -f ../bin/dep.pl ; cp dep.pl ../bin/

chmod +x charmc

./system_ln  ../tmp/charmc ../bin/

./system_ln  ../../src/scripts/testrun ../bin/

./system_ln  ../tmp/tree.txt ../bin/

./system_ln  ../tmp/fuzzytree.txt ../bin/

rm -rf ../include ; mkdir ../include

./system_ln  ../tmp/conv-*.*h ../include

./system_ln  ../tmp/cc-*.*h ../include

./system_ln  ../tmp/conv-mach-opt.sh ../include

./system_ln  ../tmp/conv-mach-opt.mak ../include

if [ -x ./special.sh ] ; then SRCBASE=../../src ./special.sh ; fi

if [ ! -f conv-common.h ] ; then ( touch conv-common.h ) ; fi

touch dirs+sources

/usr/bin/gmake -C topomanager -f Makefile_charm headers

gmake[1]: Entering directory `/home/kabram/charm/mpi-linux-x86_64-smp-gcc/tmp/topomanager'

/bin/cp TopoManager.h ../../include

/bin/cp BGQTorus.h ../../include

/bin/cp XTTorus.h ../../include

echo "//empty" > topomanager_config.h

/bin/cp topomanager_config.h ../../include

gmake[1]: Leaving directory `/home/kabram/charm/mpi-linux-x86_64-smp-gcc/tmp/topomanager'

/usr/bin/gmake charmxi

gmake[1]: Entering directory `/home/kabram/charm/mpi-linux-x86_64-smp-gcc/tmp'

autoreconf

autoheader

automake: warning: am_libhwloc_embedded_la_rpath was already defined in condition TRUE, which includes condition !HWLOC_BUILD_STANDALONE ...

configure.ac:2932: ... 'am_libhwloc_embedded_la_rpath' previously defined here

./configure --disable-controlpoint --disable-tracing --disable-tracing-commthread --disable-charmdebug --disable-replay --disable-error-checking --disable-stats

Error checking is disabled

AMPI error checking is enabled

Statistics collection is disabled

Charm tracing is disabled

Charm tracing communication thread is disabled

Task Queue is disabled

Drone mode is disabled

CharmDebug is disabled

Charm record/replay is disabled

CCS is enabled

Charm control point is disabled

Charm LB user data is disabled

Lockless queue for PE local and node queue is disabled

Charm shrink expand is disabled

charm4py support is disabled

Setting load balancing timer type as 'double'

checking machine name... mpi-linux-x86_64-smp-gcc

checking "cp command as"... cp -p

checking "C++ compiler as"... "g++  "

checking "whether C++ compiler works"... "ok"

checking "C++ linker as"... "g++  "

checking "whether linker works"... "ok"

checking "Native C++ compiler as"... "g++"

checking "Sequential C++ compiler as"... "g++"

checking "whether compiler accept -fno-stack-protector"... "ok"

checking "whether C++ compiler supports C++11 without flags"... "no"

checking "whether C++ compiler supports C++11 with '-std=c++11'"... "yes"

checking "whether native C++ compiler supports C++11 without flags"... "yes"

checking "whether C++ compiler accepts -fno-lifetime-dse"... "yes"

checking "whether C++ compiler accepts -mno-tls-direct-seg-refs"... "yes"

Setting charm++ envelope refnum field to unsigned short

Configuring support for message priorities of arbitrary size (bitvectors)

Using mempool cutoff... 2^26

checking "whether compiler generates code for 64-bit"... "yes"

checking "whether compiler supports std::is_constructible"... "yes"

checking "whether compiler supports std::alignment_of"... "yes"

checking "whether compiler implements regex"... "yes"

checking "whether has values.h "... "yes"

checking "whether has stdint.h "... "yes"

checking "whether has malloc.h "... "yes"

checking "whether has alloca.h "... "yes"

checking "whether has regex.h "... "yes"

checking "whether long long works"... "yes"

checking "whether __int64 works"... "no"

checking "whether __int128 (128-bit integer) works"... "yes"

checking "whether __int128_t (128-bit integer) works"... "yes"

checking "whether C++ library has <atomic> "... "yes"

checking "whether long double works"... "yes"

checking "whether ucontext has FPU pointer"... "yes"

checking "whether ucontext uses uc_regs"... "no"

checking "whether ucontext has pointer (v_regs) of vector type"... "no"

checking "whether ibverbs ibv_port_attr has link_layer field"... "yes"

checking "whether inline works in C"... "yes"

checking "whether C++ signed char and char differ"... "yes"

checking "whether typeinfo/typeid works"... "ok"

checking "whether std::iterator_traits is defined"... "ok"

checking "whether std::distance is defined"... "ok"

checking "whether std::inserter is defined"... "ok"

checking "whether std::void_t is defined"... "no"

checking "whether operator delete can be overloaded in same class"... "ok"

checking "whether offsetof is defined"... "yes"

checking "whether GCC x86 assembly works"... "yes"

checking "whether GCC x86 assembly for atomic increment works"... "yes"

checking "whether asm eieio assembly works"... "no"

checking "whether __thread (Thread Local Storage) is supported"... "yes"

checking "whether synchronization primitives (__sync_add_and_fetch) works in C"... "yes"

checking "whether synchronization primitives (__sync_synchronize) works in C"... "yes"

checking "whether has __executable_start"... "yes"

checking "whether switching TLS register (64-bit) is supported"... "yes"

checking "whether has dl_iterate_phdr"... "yes"

checking "whether has __malloc_hook"... "yes"

checking "whether build on MPI"... "no"

Error: Cannot compile an MPI program

*** Please find detailed output in tmp/charmconfig.out ***

gmake[1]: *** No rule to make target `conv-autoconfig.h', needed by `xi-main.o'.  Stop.

gmake[1]: Leaving directory `/home/kabram/charm/mpi-linux-x86_64-smp-gcc/tmp'

gmake: *** [headers] Error 2

-------------------------------------------------

Charm++ NOT BUILT. Either cd into mpi-linux-x86_64-smp-gcc/tmp and try

to resolve the problems yourself, visit

http://charm.cs.illinois.edu/

for more information. Otherwise, email the developers at charm AT cs.illinois.edu

---

Confidentiality Notice: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message.

--

Jeff Hammond
jeff.science AT gmail.com
http://jeffhammond.github.io/