|
|
本帖最后由 雾雨魔理沙 于 2024-1-10 15:32 编辑
求助：环境Rocky linux9.3，AMD 7950x，GLIBC：ldd (GNU libc) 2.34，gcc 版本 11.4.1，AOCC-4.1.0
参考的贴子http://bbs.keinsci.com/thread-40792-1-1.html
which clang、clang++、flang都有路径显示，但是在编译openmpi-5.0.0和openmpi-4.1.6 时均报错。错误显示：
*** C++ compiler and preprocessor
checking whether the compiler supports GNU C++... yes
checking whether /opt/aocc-compiler-4.1.0/bin/clang++ accepts -g... yes
checking for /opt/aocc-compiler-4.1.0/bin/clang++ option to enable C++11 features... none needed
checking dependency style of /opt/aocc-compiler-4.1.0/bin/clang++... gcc3
checking how to run the C++ preprocessor... /opt/aocc-compiler-4.1.0/bin/clang++ -E
checking if C and C++ are link compatible... no
**********************************************************************
* It appears that your C++ compiler is unable to link against object
* files created by your C compiler. This generally indicates either
* a conflict between the options specified in CFLAGS and CXXFLAGS
* or a problem with the local compiler installation. More
* information (including exactly what command was given to the
* compilers and what error resulted when the commands were executed) is
* available in the config.log file in this directory.
**********************************************************************
configure: error: C and C++ compilers are not link compatible. Can not continue.
此报错已解决：
在此操作前需要验证下which c++和gfortran是否都有路径返回。没有则安装。
-------------------------分割线----------------------------------------------------
现在遇到了第二个问题，在对openMPI make时，报错信息如下：
make[2]: 进入目录“/opt/openmpi-5.0.0/ompi/tools/ompi_info”
CC ompi_info.o
CC param.o
CCLD ompi_info
ld.lld: error: undefined reference due to --no-allow-shlib-undefined: __extendhfsf2
>>> referenced by ../../../ompi/.libs/libmpi.so
ld.lld: error: undefined reference due to --no-allow-shlib-undefined: __truncsfhf2
>>> referenced by ../../../ompi/.libs/libmpi.so
clang-16: error: linker command failed with exit code 1 (use -v to see invocation)
make[2]: *** [Makefile:1472：ompi_info] 错误 1
make[2]: 离开目录“/opt/openmpi-5.0.0/ompi/tools/ompi_info”
make[1]: *** [Makefile:2798：all-recursive] 错误 1
make[1]: 离开目录“/opt/openmpi-5.0.0/ompi”
make: *** [Makefile:1532：all-recursive] 错误 1
-------------------------分割线----------------------------------------------------
用openMPI 4.1.6的版本上面的报错不会出现。
但是继续编译vasp时又出现了warnning，
（用的是makefile.include.aocc_ompi_aocl，BLAS、LAPACK等修改了相应路径，修改了VASP_TARGET_CPU ?= -march=znver4）
warning如下：
ld.lld: warning: parser/libparser.a: archive member 'locproj.tab.h' is neither ET_REL nor LLVM bitcode
make[2]: 离开目录“/opt/vasp/vasp.6.4.2/build/std”
make[1]: 离开目录“/opt/vasp/vasp.6.4.2/build/std”
make test 和试跑都出现了error，如下：
running 4 mpi-ranks, on 1 nodes
distrk: each k-point on 4 cores, 1 groups
distr: one band on 4 cores, 1 groups
vasp.6.4.2 20Jul23 (build Dec 28 2023 17:05:26) complex
POSCAR found type information on POSCAR C
POSCAR found : 1 types and 2 ions
scaLAPACK will be used
LDA part: xc-table for Pade appr. of Perdew
-----------------------------------------------------------------------------
| |
| EEEEEEE RRRRRR RRRRRR OOOOOOO RRRRRR ### ### ### |
| E R R R R O O R R ### ### ### |
| E R R R R O O R R ### ### ### |
| EEEEE RRRRRR RRRRRR O O RRRRRR # # # |
| E R R R R O O R R |
| E R R R R O O R R ### ### ### |
| EEEEEEE R R R R OOOOOOO R R ### ### ### |
| |
| VERY BAD NEWS! internal error in subroutine SGRGEN: Too many |
| elements 49 |
| |
| ----> I REFUSE TO CONTINUE WITH THIS SICK JOB ... BYE!!! <---- |
| |
-----------------------------------------------------------------------------
Warning: ieee_inexact is signaling
1
Warning: ieee_inexact is signaling
1
--------------------------------------------------------------------------
Primary job terminated normally, but 1 process returned
a non-zero exit code. Per user-direction, the job has been aborted.
--------------------------------------------------------------------------
--------------------------------------------------------------------------
mpirun detected that one or more processes exited with non-zero status, thus causing
the job to be terminated. The first process to do so was:
Process name: [[40421,1],2]
Exit code: 1
--------------------------------------------------------------------------
请问是什么原因导致的？不知道是不是和openmpi有关，该如何解决呢?
——————更新————————
第三个问题，也就是vasp运行报错，发现只要把POSCAR晶格坐标改成——小数点后只保留3位，就可以正常跑任务！但是多一位都会报错
有老师同学知道是什么原因导致的吗？
|
|
发表于 Post on 2023-12-20 22:15:39
|
只看该作者 Only view this author
|倒序浏览 Reverse view
|阅读模式 Reading model
收藏 Add to favorites
楼主 Author
