Dear Amber community,
I'm trying to install Amber22 on a machine with 3 Titan Blacks, with cuda
10.1 and I'm having some problems. I do not have problems when I try to
install Amber without cuda.
So I downloaded and extracted the tar files, and then I edited the
run_cmake file to something like
....
cmake $AMBER_PREFIX/amber22_src \
-DCMAKE_INSTALL_PREFIX=$AMBER_PREFIX/amber22 \
-DCOMPILER=GNU \
-DMPI=TRUE -DOPENMP=TRUE -DCUDA=FALSE -DINSTALL_TESTS=TRUE \
-DCUDA_TOOLKIT_ROOT_DIR=/usr/lib/cuda/ \
-DDOWNLOAD_MINICONDA=TRUE \
2>&1 | tee cmake.log
....
When I run bash run_cmake everything went smooth and I was able to install
Amber.
But I want to use my GPUs and I got the error about incompatible CUDA
version. So now some details about my system:
nvidia-smi
Wed Aug 17 17:01:49 2022
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.141.03 Driver Version: 470.141.03 CUDA Version: 11.4
|
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr.
ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute
M. |
| | | MIG
M. |
|===============================+======================+======================|
| 0 NVIDIA GeForce ... Off | 00000000:02:00.0 Off |
N/A |
| 26% 40C P8 14W / 250W | 12MiB / 6083MiB | 0%
Default |
| | |
N/A |
+-------------------------------+----------------------+----------------------+
| 1 NVIDIA GeForce ... Off | 00000000:05:00.0 Off |
N/A |
| 28% 43C P8 27W / 189W | 166MiB / 6080MiB | 0%
Default |
| | |
N/A |
+-------------------------------+----------------------+----------------------+
| 2 NVIDIA GeForce ... Off | 00000000:06:00.0 Off |
N/A |
| 28% 43C P8 32W / 189W | 10MiB / 6083MiB | 0%
Default |
| | |
N/A |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes:
|
| GPU GI CI PID Type Process name GPU
Memory |
| ID ID Usage
|
|=============================================================================|
| 0 N/A N/A 2441 G /usr/lib/xorg/Xorg
4MiB |
| 0 N/A N/A 3057 G /usr/lib/xorg/Xorg
4MiB |
| 1 N/A N/A 2441 G /usr/lib/xorg/Xorg
30MiB |
| 1 N/A N/A 3057 G /usr/lib/xorg/Xorg
58MiB |
| 1 N/A N/A 3197 G /usr/bin/gnome-shell
63MiB |
| 2 N/A N/A 2441 G /usr/lib/xorg/Xorg
4MiB |
| 2 N/A N/A 3057 G /usr/lib/xorg/Xorg
4MiB |
+-----------------------------------------------------------------------------+
nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2019 NVIDIA Corporation
Built on Sun_Jul_28_19:07:16_PDT_2019
Cuda compilation tools, release 10.1, V10.1.243
gcc --version
gcc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0
In my /usr/lib/cuda/bin I got the soft links for lower version of gcc
compilers
ls -l /usr/lib/cuda/bin/
total 0
lrwxrwxrwx 1 root root 14 sie 17 16:48 g++-8 -> /usr/bin/g++-8
lrwxrwxrwx 1 root root 14 sie 17 16:48 gcc-8 -> /usr/bin/gcc-8
lrwxrwxrwx 1 root root 13 lis 22 2018 nvcc -> /usr/bin/nvcc
I also tried adding -DCMAKE_C_COMPILER=gcc-8 -DCMAKE_CXX_COMPILER=g++-8 to
the run_cmake file but that didn't work. I got the following error and when
I rerun it I got more errors:
Error: Incompatible CUDA and GNU versions
9.4.0
See
https://stackoverflow.com/questions/6622454/cuda-incompatible-with-my-gcc-version
************************************************************
--
CMake Error at cmake/CudaConfig.cmake:125 (message):
Call Stack (most recent call first):
CMakeLists.txt:120 (include)
-- Configuring incomplete, errors occurred!
See also "/usr/local/src/amber22_src/build/CMakeFiles/CMakeOutput.log".
See also "/usr/local/src/amber22_src/build/CMakeFiles/CMakeError.log".
You have changed variables that require your cache to be deleted.
Configure will be re-run and you may have to reset some variables.
The following variables have changed:
CMAKE_C_COMPILER= gcc-8
CMAKE_CXX_COMPILER= g++-8
-- Generating done
CMake Warning:
Manually-specified variables were not used by the project:
DOWNLOAD_MINICONDA
CMake Generate step failed. Build files cannot be regenerated correctly.
If the cmake build report looks OK, you should now do the following:
make install
source /usr/local/src/amber22/amber.sh
Consider adding the last line to your login startup script, e.g. ~/.bashrc
So for now I'm stuck. Can someone help me?
Looking forward to some tips :)
Best regards,
Karolina Mitusińska
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Aug 18 2022 - 04:30:05 PDT
