v1.2.0 version, make error under arm architecture. Does it not support the ARM architecture?

[Ruanjq98]

make error：

mpic++ -std=c++14 -fno-stack-protector -Wall -g3 -O0 -DDEBUG -fPIC -I../../dmtcp/include -I../../dmtcp/jalib -I.. -I../../dmtcp/src -I../lower-half -c -o mpi_op_wrappers.o mpi_op_wrappers.cpp
In file included from ../lower-half/lower-half-api.h:30,
                 from ../mpi_plugin.h:29,
                 from mpi_type_wrappers.cpp:22:
../lower-half/switch-context.h:24:10: fatal error: asm/prctl.h: No such file or directory
   24 | #include <asm/prctl.h>
      |          ^~~~~~~~~~~~~
compilation terminated.
In file included from ../lower-half/lower-half-api.h:30,
                 from ../mpi_plugin.h:29,
                 from mpi_group_wrappers.cpp:22:
../lower-half/switch-context.h:24:10: fatal error: asm/prctl.h: No such file or directory
   24 | #include <asm/prctl.h>
      |          ^~~~~~~~~~~~~
compilation terminated.
In file included from ../lower-half/lower-half-api.h:30,
                 from ../mpi_plugin.h:29,
                 from mpi_op_wrappers.cpp:22:
../lower-half/switch-context.h:24:10: fatal error: asm/prctl.h: No such file or directory
   24 | #include <asm/prctl.h>
      |          ^~~~~~~~~~~~~
In file included from ../lower-half/lower-half-api.h:30,
                 from ../mpi_plugin.h:29,
                 from mpi_cart_wrappers.cpp:22:
../lower-half/switch-context.h:24:10: fatal error: asm/prctl.h: No such file or directory
   24 | #include <asm/prctl.h>
      |          ^~~~~~~~~~~~~
compilation terminated.
compilation terminated.
make[3]: *** [Makefile:65: mpi_type_wrappers.o] Error 1
make[3]: *** Waiting for unfinished jobs....
make[3]: *** [Makefile:65: mpi_group_wrappers.o] Error 1
make[3]: *** [Makefile:65: mpi_cart_wrappers.o] Error 1
make[3]: *** [Makefile:65: mpi_op_wrappers.o] Error 1
In file included from ../lower-half/lower-half-api.h:30,
                 from ../mpi_plugin.h:29,
                 from mpi_comm_wrappers.cpp:28:
../lower-half/switch-context.h:24:10: fatal error: asm/prctl.h: No such file or directory
   24 | #include <asm/prctl.h>
      |          ^~~~~~~~~~~~~
compilation terminated.
make[3]: *** [Makefile:65: mpi_comm_wrappers.o] Error 1
In file included from ../lower-half/lower-half-api.h:30,
                 from ../mpi_plugin.h:29,
                 from mpi_p2p_wrappers.cpp:28:
../lower-half/switch-context.h:24:10: fatal error: asm/prctl.h: No such file or directory
   24 | #include <asm/prctl.h>
      |          ^~~~~~~~~~~~~
compilation terminated.
make[3]: *** [Makefile:74: mpi_p2p_wrappers.o] Error 1
In file included from ../lower-half/lower-half-api.h:30,
                 from ../mpi_plugin.h:29,
                 from mpi_collective_wrappers.cpp:29:
../lower-half/switch-context.h:24:10: fatal error: asm/prctl.h: No such file or directory
   24 | #include <asm/prctl.h>
      |          ^~~~~~~~~~~~~
compilation terminated.
make[3]: *** [Makefile:86: mpi_collective_wrappers.o] Error 1
In file included from ../lower-half/lower-half-api.h:30,
                 from ../record-replay.h:36,
                 from mpi_request_wrappers.cpp:31:
../lower-half/switch-context.h:24:10: fatal error: asm/prctl.h: No such file or directory
   24 | #include <asm/prctl.h>
      |          ^~~~~~~~~~~~~
compilation terminated.

Please reply, thank you.

[xuyao0127]

MANA requires the ability to change the fs/gs register (x86) at runtime. ARM and RISC-V does not have these registers but have something equivalent for thread local storage. We are working on fixing the compatibility issue on ARM and RISC-V in the next release.

[Ruanjq98]

I tested multiple versions of MPICH according to https://github.com/mpickpt/mana/releases/tag/v1.2.0 and none of them could successfully restart. What specific version can successfully restart? Also, does it support OPENMPI?

Please reply, thank you.

CentOS7 + MANA v1.2.0 + MPICH 3.3.2 (or 4.2.0)
MPICH 3.3.2

mana_coordinator
mpirun -np 1 mana_launch ./mpi-proxy-split/test/mpi_hello_world.mana.exe
mana_status --checkpoint

mpirun -np 1 mana_restart
===================================================================================
=   BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
=   PID 16918 RUNNING AT dfee-agent10
=   EXIT CODE: 9
=   CLEANING UP REMAINING PROCESSES
=   YOU CAN IGNORE THE BELOW CLEANUP MESSAGES
===================================================================================
YOUR APPLICATION TERMINATED WITH THE EXIT STRING: Killed (signal 9)
This typically refers to a problem with your application.
Please see the FAQ page for debugging suggestions

MPICH 4.2.0

mana_coordinator
mpirun -np 1 mana_launch --use-shadowlibs ./mpi-proxy-split/test/mpi_hello_world.mana.exe
mana_status --checkpoint
mana_coordinator
mpirun -np 1 mana_restart

[master:16121:0:16121] Caught signal 11 (Segmentation fault: address not mapped to object at address 0x400000)
==== backtrace (tid:  16121) ====
 0 0x000000000016f8c1 __strlen_sse2_pminub()  :0
 1 0x0000000000123c9a std::char_traits<char>::length()  /home/ShareWRF/gcc_install/gcc-10.3.0/build/x86_64-pc-linux-gnu/libstdc++-v3/include/bits/char_traits.h:371
 2 0x0000000010016620 jassert_internal::JAssert::Print<char*>()  /home/rjq/mana-1.2.0/mpi-proxy-split/lower-half/../../dmtcp/src/../jalib/jassert.h:226
 3 0x0000000010013cc5 restoreMemoryArea()  /home/rjq/mana-1.2.0/mpi-proxy-split/lower-half/lower-half.cpp:800
 4 0x000000001001342a restore_memory_data()  /home/rjq/mana-1.2.0/mpi-proxy-split/lower-half/lower-half.cpp:675
 5 0x000000001001185f main()  /home/rjq/mana-1.2.0/mpi-proxy-split/lower-half/lower-half.cpp:166
 6 0x0000000000022555 __libc_start_main()  ???:0
 7 0x000000001000bd99 _start()  ???:0
=================================

===================================================================================
=   BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
=   PID 16121 RUNNING AT master
=   EXIT CODE: 139
=   CLEANING UP REMAINING PROCESSES
=   YOU CAN IGNORE THE BELOW CLEANUP MESSAGES
===================================================================================
YOUR APPLICATION TERMINATED WITH THE EXIT STRING: Segmentation fault (signal 11)
This typically refers to a problem with your application.
Please see the FAQ page for debugging suggestions

MANA requires the ability to change the fs/gs register (x86) at runtime. ARM and RISC-V does not have these registers but have something equivalent for thread local storage. We are working on fixing the compatibility issue on ARM and RISC-V in the next release.