Error When Compiling LAMMPS with USER-DEEPMD package:

[Taixi-Feng]

Bug summary

Hi, when I do the final compilation of LAMMPS the error report occurred.
](https://user-images.githubusercontent.com/104333661/178149918-a2a93c13-1417-4df7-bbfa-6ff9e2f9dfe4.png)
An error will be reported when make final compilation of LAMMPS using the file of user-deepmd package (such as pair_deepmd.h).

DeePMD-kit Version

2.1.3

TensorFlow Version

2.8.2

How did you download the software?

Built from source

Input Files, Running Commands, Error Log, etc.

Gathering installed package information (may take a little while)
make[1]: 进入目录“/root/softwares/lammps-stable_23Jun2022/src”
Gathering git version information
make[1]: 离开目录“/root/softwares/lammps-stable_23Jun2022/src”
Compiling LAMMPS for machine mpi
make[1]: 进入目录“/root/softwares/lammps-stable_23Jun2022/src/Obj_mpi”
make[1]: 离开目录“/root/softwares/lammps-stable_23Jun2022/src/Obj_mpi”
make[1]: 进入目录“/root/softwares/lammps-stable_23Jun2022/src/Obj_mpi”
mpicxx -g -O3 -std=c++11 -DLAMMPS_GZIP -DLAMMPS_MEMALIGN=64 -DLMP_GPU -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -c ../fft3d_wrap.cpp
mpicxx -g -O3 -std=c++11 -DLAMMPS_GZIP -DLAMMPS_MEMALIGN=64 -DLMP_GPU -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -c ../pppm_cg.cpp
mpicxx -g -O3 -std=c++11 -DLAMMPS_GZIP -DLAMMPS_MEMALIGN=64 -DLMP_GPU -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -c ../remap_wrap.cpp
mpicxx -g -O3 -std=c++11 -DLAMMPS_GZIP -DLAMMPS_MEMALIGN=64 -DLMP_GPU -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -c ../pair_deepmd.cpp
mpicxx -g -O3 -std=c++11 -DLAMMPS_GZIP -DLAMMPS_MEMALIGN=64 -DLMP_GPU -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -c ../remap.cpp
mpicxx -g -O3 -std=c++11 -DLAMMPS_GZIP -DLAMMPS_MEMALIGN=64 -DLMP_GPU -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -c ../pppm_gpu.cpp
mpicxx -g -O3 -std=c++11 -DLAMMPS_GZIP -DLAMMPS_MEMALIGN=64 -DLMP_GPU -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -c ../pppm_tip4p.cpp
mpicxx -g -O3 -std=c++11 -DLAMMPS_GZIP -DLAMMPS_MEMALIGN=64 -DLMP_GPU -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -c ../pppm_stagger.cpp
mpicxx -g -O3 -std=c++11 -DLAMMPS_GZIP -DLAMMPS_MEMALIGN=64 -DLMP_GPU -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -c ../force.cpp
mpicxx -g -O3 -std=c++11 -DLAMMPS_GZIP -DLAMMPS_MEMALIGN=64 -DLMP_GPU -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -c ../pppm_dipole.cpp
mpicxx -g -O3 -std=c++11 -DLAMMPS_GZIP -DLAMMPS_MEMALIGN=64 -DLMP_GPU -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -c ../fft3d.cpp
mpicxx -g -O3 -std=c++11 -DLAMMPS_GZIP -DLAMMPS_MEMALIGN=64 -DLMP_GPU -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -c ../pppm_disp.cpp
mpicxx -g -O3 -std=c++11 -DLAMMPS_GZIP -DLAMMPS_MEMALIGN=64 -DLMP_GPU -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -c ../pppm.cpp
mpicxx -g -O3 -std=c++11 -DLAMMPS_GZIP -DLAMMPS_MEMALIGN=64 -DLMP_GPU -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -c ../modify.cpp
mpicxx -g -O3 -std=c++11 -DLAMMPS_GZIP -DLAMMPS_MEMALIGN=64 -DLMP_GPU -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -c ../compute_deeptensor_atom.cpp
mpicxx -g -O3 -std=c++11 -DLAMMPS_GZIP -DLAMMPS_MEMALIGN=64 -DLMP_GPU -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -c ../fix_dplr.cpp
mpicxx -g -O3 -std=c++11 -DLAMMPS_GZIP -DLAMMPS_MEMALIGN=64 -DLMP_GPU -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -std=c++14 -DHIGH_PREC -DLAMMPS_VERSION_NUMBER=20220623 -I/root/softwares/include/ -DMPICH_SKIP_MPICXX -DOMPI_SKIP_MPICXX=1 -c ../lammps.cpp
In file included from ../fft3d_wrap.h:17:0,
from ../fft3d_wrap.cpp:15:
../fft3d.h:14:10: 致命错误：mpi.h：没有那个文件或目录
#include <mpi.h>
^~~~~~~
编译中断。
make[1]: *** [fft3d_wrap.o] 错误 1
make[1]: *** 正在等待未完成的任务....
In file included from ../remap.cpp:15:0:
../remap.h:14:10: 致命错误：mpi.h：没有那个文件或目录
#include <mpi.h>
^~~~~~~
编译中断。
make[1]: *** [remap.o] 错误 1
In file included from ../fft3d.cpp:23:0:
../fft3d.h:14:10: 致命错误：mpi.h：没有那个文件或目录
#include <mpi.h>
^~~~~~~
编译中断。
make[1]: *** [fft3d.o] 错误 1
In file included from ../pppm_cg.cpp:28:0:
../remap.h:14:10: 致命错误：mpi.h：没有那个文件或目录
#include <mpi.h>
^~~~~~~
编译中断。
../pppm_stagger.cpp:20:10: 致命错误：mpi.h：没有那个文件或目录
#include <mpi.h>
^~~~~~~
编译中断。
make[1]: *** [pppm_cg.o] 错误 1
make[1]: *** [pppm_stagger.o] 错误 1
In file included from ../remap_wrap.h:18:0,
from ../remap_wrap.cpp:15:
../remap.h:14:10: 致命错误：mpi.h：没有那个文件或目录
#include <mpi.h>
^~~~~~~
编译中断。
make[1]: *** [remap_wrap.o] 错误 1
../pppm_tip4p.cpp:20:10: 致命错误：mpi.h：没有那个文件或目录
#include <mpi.h>
^~~~~~~
编译中断。
make[1]: *** [pppm_tip4p.o] 错误 1
In file included from ../fft3d_wrap.h:17:0,
from ../pppm_gpu.cpp:24:
../fft3d.h:14:10: 致命错误：mpi.h：没有那个文件或目录
#include <mpi.h>
^~~~~~~
编译中断。
make[1]: *** [pppm_gpu.o] 错误 1
In file included from ../pair_deepmd.cpp:23:0:
../pair_deepmd.h:14:10: 致命错误：deepmd/DeepPot.h：没有那个文件或目录
#include "deepmd/DeepPot.h"
^~~~~~~~~~~~~~~~~~
编译中断。
In file included from ../compute_deeptensor_atom.h:11:0,
from ../compute_deeptensor_atom.cpp:1:
../pair_deepmd.h:14:10: 致命错误：deepmd/DeepPot.h：没有那个文件或目录
#include "deepmd/DeepPot.h"
^~~~~~~~~~~~~~~~~~
编译中断。
make[1]: *** [pair_deepmd.o] 错误 1
make[1]: *** [compute_deeptensor_atom.o] 错误 1
In file included from ../style_pair.h:26:0,
from ../force.cpp:21:
../pair_deepmd.h:14:10: 致命错误：deepmd/DeepPot.h：没有那个文件或目录
#include "deepmd/DeepPot.h"
^~~~~~~~~~~~~~~~~~
编译中断。
In file included from ../fft3d_wrap.h:17:0,
from ../pppm_disp.cpp:28:
../fft3d.h:14:10: 致命错误：mpi.h：没有那个文件或目录
#include <mpi.h>
^~~~~~~
In file included from ../compute_deeptensor_atom.h:11:0,
from ../style_compute.h:17,
from ../modify.cpp:15:
../pair_deepmd.h:14:10: 致命错误：deepmd/DeepPot.h：没有那个文件或目录
#include "deepmd/DeepPot.h"
^~~~~~~~~~~~~~~~~~
编译中断。
编译中断。
In file included from ../fft3d_wrap.h:17:0,
from ../pppm.cpp:30:
../fft3d.h:14:10: 致命错误：mpi.h：没有那个文件或目录
#include <mpi.h>
^~~~~~~
编译中断。
make[1]: *** [force.o] 错误 1
make[1]: *** [pppm_disp.o] 错误 1
make[1]: *** [modify.o] 错误 1
In file included from ../fft3d_wrap.h:17:0,
from ../pppm_dipole.cpp:25:
../fft3d.h:14:10: 致命错误：mpi.h：没有那个文件或目录
#include <mpi.h>
^~~~~~~
编译中断。
make[1]: *** [pppm.o] 错误 1
make[1]: *** [pppm_dipole.o] 错误 1
In file included from ../fix_dplr.h:13:0,
from ../fix_dplr.cpp:13:
../pair_deepmd.h:14:10: 致命错误：deepmd/DeepPot.h：没有那个文件或目录
#include "deepmd/DeepPot.h"
^~~~~~~~~~~~~~~~~~
编译中断。
make[1]: *** [fix_dplr.o] 错误 1
In file included from ../compute_deeptensor_atom.h:11:0,
from ../style_compute.h:17,
from ../lammps.cpp:21:
../pair_deepmd.h:14:10: 致命错误：deepmd/DeepPot.h：没有那个文件或目录
#include "deepmd/DeepPot.h"
^~~~~~~~~~~~~~~~~~
编译中断。
make[1]: *** [lammps.o] 错误 1
make[1]: 离开目录“/root/softwares/lammps-stable_23Jun2022/src/Obj_mpi”
make: *** [mpi] 错误 2

Steps to Reproduce

make mpi -j4
Follow all the instructions given on the manual. Use the Built-in mode to compile LAMMPS. At the final completion of LAMMPS. The bug will show itself

Further Information, Files, and Links

No response

[njzjz]

-I/root/softwares/include/

This path is in your compile command. Could you check if it is correct and /root/softwares/include/deepmd/DeepPot.h exists?

[Taixi-Feng]

Thanks for your help.
I have made the changes according to your advice. But a new error occured. I don't know how to deal it.

mpicxx -g -O3 -std=c++11 -lrt main.o -L../../lib/gpu -L/root/tensorflow/lib -L/root/tensorflow/lib -L/root/softwares/lib -L/root/tensorflow/lib -L/root/tensorflow/lib -L/root/softwares/lib -L/root/tensorflow/lib -L/root/tensorflow/lib -L/root/softwares/lib -L/root/tensorflow/lib -L/root/tensorflow/lib -L/root/softwares/lib -L/root/tensorflow/lib -L/root/tensorflow/lib -L/root/softwares/lib -L/usr/local/mpich-3.3/lib -L/usr/local/cuda/lib64 -L/usr/local/cuda/lib64/stubs -L. -llammps_mpi -lgpu -Wl,--no-as-needed -ldeepmd_cc -ltensorflow_cc -ltensorflow_framework -Wl,-rpath=/root/tensorflow/lib -Wl,-rpath=/root/tensorflow/lib -Wl,-rpath=/root/softwares/lib -Wl,--no-as-needed -ldeepmd_cc -ltensorflow_cc -ltensorflow_framework -Wl,-rpath=/root/tensorflow/lib -Wl,-rpath=/root/tensorflow/lib -Wl,-rpath=/root/softwares/lib -Wl,--no-as-needed -ldeepmd_cc -ltensorflow_cc -ltensorflow_framework -Wl,-rpath=/root/tensorflow/lib -Wl,-rpath=/root/tensorflow/lib -Wl,-rpath=/root/softwares/lib -Wl,--no-as-needed -ldeepmd_cc -ltensorflow_cc -ltensorflow_framework -Wl,-rpath=/root/tensorflow/lib -Wl,-rpath=/root/tensorflow/lib -Wl,-rpath=/root/softwares/lib -Wl,--no-as-needed -ldeepmd_cc -ltensorflow_cc -ltensorflow_framework -Wl,-rpath=/root/tensorflow/lib -Wl,-rpath=/root/tensorflow/lib -Wl,-rpath=/root/softwares/lib /usr/local/mpich-3.3/lib/libmpi.a -lmpich -lmpl -lpthread -lcudart -lcuda -ldl -o ../lmp_mpi
/usr/bin/ld: warning: libdeepmd.so, needed by //lib/libdeepmd_cc.so, not found (try using -rpath or -rpath-link)
/usr/bin/ld: warning: libdeepmd_op_cuda.so, needed by //lib/libdeepmd_cc.so, not found (try using -rpath or -rpath-link)
./liblammps_mpi.a(info.o)：在函数‘LAMMPS_NS::Info::get_gpu_device_infoabi:cxx11’中：
/root/softwares/lammps-stable_23Jun2022/src/Obj_mpi/../info.cpp:1152：对‘lmp_gpu_device_infoabi:cxx11’未定义的引用
./liblammps_mpi.a(info.o)：在函数‘LAMMPS_NS::Info::has_accelerator_feature(std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&, std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&, std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&)’中：
/root/softwares/lammps-stable_23Jun2022/src/Obj_mpi/../info.cpp:1203：对‘lmp_gpu_config(std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&, std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&)’未定义的引用
/root/softwares/lammps-stable_23Jun2022/src/Obj_mpi/../info.cpp:1203：对‘lmp_gpu_config(std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&, std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&)’未定义的引用
./liblammps_mpi.a(info.o)：在函数‘LAMMPS_NS::Info::get_gpu_device_infoabi:cxx11’中：
/root/softwares/lammps-stable_23Jun2022/src/Obj_mpi/../info.cpp:1152：对‘lmp_gpu_device_infoabi:cxx11’未定义的引用
./liblammps_mpi.a(info.o)：在函数‘LAMMPS_NS::Info::has_accelerator_feature(std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&, std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&, std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&)’中：
/root/softwares/lammps-stable_23Jun2022/src/Obj_mpi/../info.cpp:1203：对‘lmp_gpu_config(std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&, std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&)’未定义的引用
collect2: 错误：ld 返回 1
make[1]: *** [../lmp_mpi] 错误 1
make[1]: 离开目录“/root/softwares/lammps-stable_23Jun2022/src/Obj_mpi”
make: *** [mpi] 错误 2

[njzjz]

I do not quite understand where do you install the DeePMD-kit? The log shows //lib/libdeepmd_cc.so but the command also shows -L/root/softwares/lib.

[Taixi-Feng]

Thank you for your help. The directory for installing DEEPMD-kit is /root/softwares/deepmd-kit.

[njzjz]

Your new error looks unrelated with DeePMD-kit. Did you successfully compile without DeePMD-kit?

[Taixi-Feng]

Thanks. I'm going to try to apply this method.