Package Details: gromacs 2020.4-1

Git Clone URL: https://aur.archlinux.org/gromacs.git (read-only, click to copy)
Package Base: gromacs
Description: A versatile package to perform molecular dynamics, i.e. simulate the Newtonian equations of motion for systems with hundreds to millions of particles.
Upstream URL: http://www.gromacs.org/
Licenses: LGPL
Submitter: xyproto
Maintainer: hseara
Last Packager: hseara
Votes: 20
Popularity: 0.000912
First Submitted: 2011-12-14 17:03
Last Updated: 2020-10-14 19:35

Dependencies (12)

Required by (1)

Sources (1)

Latest Comments

« First ‹ Previous 1 2 3 4 5 6 7 ... Next › Last »

mefistofeles commented on 2017-12-03 22:53

@malinke I was also getting avx2 errors. Specifically this: error: #error "compiling simd-avx2.h without avx2 support", which some guys at #archlinux think it's an error from upstream. The complete error output is at: https://bpaste.net/show/3bbef4217034 (it may have a short lifetime, sorry)

There one can see that the problem is building the fftw, so I tried removing the -DGMX_BUILD_OWN_FFTW=ON \ lines in the PKGBUILD (lines 51 and 62) and got better results but still errors on testing it. The output I'm getting doing what I said is here: https://bpaste.net/show/01d9d68140bd

BTW, when compiling directly the source code by using cmake and make it builds, with not special options, just as it's shown in: http://web.archive.org/web/20170714101612/http://www.gromacs.org/Developer_Zone/Building_and_Testing/Cmake in the "Configuring with default settings" section. Hope this helps to give ideas on how to make gromacs build using this PKGBUILD/AUR?

hseara commented on 2017-10-01 11:00

@malink "(I didn't care if I had a avx build or not)"

You should care. mdrun will be 2x to 4x faster if compiled with avx (depending on the version) activated.

My fix will activate the highest level of avx by default. My fix will allow also CUDA support if available. This is why we need gcc5. CUDA 9 does not seem to solve this issue.

In any case, thanks for your comment as the fix simplifies the installation.

malinke commented on 2017-10-01 07:33

I read the avx instructions. My build problem was that the default pkgbuild would fail since gromacs cmake config tried to enable avx and the global flags in `/etc/makepkg.conf` deactivated it later resulting in a build error (I didn't care if I had a avx build or not). The changed cflags in the current version fix that.

hseara commented on 2017-09-24 15:39

@mailkinke please read the content of the PKGBUILD file. There it is clearly stated:
###### CMAKE OPTIONS DISABLE BY DEFAULT ###########
# If you are using an AVX2 capable CPU, you will #
# not have AVX2 binaries unless you set -march to #
# 'native', your respective architecture flag: #
# https://gcc.gnu.org/onlinedocs/gcc-5.3.0/gcc/x86-Options.html#x86-Options #
# or just include '-mavx2' to the default compiler#
# flags in the /etc/makepkg.conf: #
# https://wiki.archlinux.org/index.php/Makepkg#Architecture.2C_compile_flagsAdd #
###################################################

hseara commented on 2017-09-24 15:38

@malinke: Currently compiling CUDA against gcc7 is not possible. If you are not using Cuda, feel free to comment the lines where we change the compiler and their options in the PKGBUILD file.

malinke commented on 2017-09-21 13:37

One problem is that in `/etc/makepkg.conf` the cflags and cxxflags contain `-march=x86_64 -mtune=generic`. These flags are not used by gromacs to determine if avx should be activated during the configuration stage of cmake. Therefore gromacs tries to build with avx instructions (on my machine). This results in the following flags to the c/c++ compiler '-march=cpu-avx -march=x86_64 -mtune=generic'. This breaks the builds as only the last usage of `march/tune` is using deactivating the avx build. The build then stops complaining it doesn't know about the avx functions. Overwritting the flags in the PKGBUILD lets the compilation finish.

These cflag I have are the standard for every user.

https://git.archlinux.org/svntogit/packages.git/tree/trunk/makepkg.conf?h=packages/pacman

malinke commented on 2017-09-21 12:31

Gromacs doesn't build with gcc5. Using gcc5 I get the following error.

```
gcc-5: error: unrecognized command line option ‘-fno-plt’
```

I had no problem building and running gromacs with gcc7

hseara commented on 2017-09-13 17:23

Compiling gromacs with CUDA support will fail after the update to glibc-2.26. This is incompatible with CUDA 8 and it is not clear whether the soon to come CUDA 9 will solve this problem. (NEWS: A dirty fix implemented in the CUDA package allows again the compilation).

hseara commented on 2017-04-28 16:42

At least for Nvidia, Gromacs detects CUDA capable cards. If you have CUDA installed, it then includes CUDA capability without the need of using any flag. This default behavior makes the current Gromacs installation platform agnostic.
I do not wish to enable any flag that can potentially change this platform independent behavior. If Gromacs does not by default activate in your system OpenCL, please write a bug report to Gromacs people, so they provide similar functionality as for CUDA.
You can always edit the PKGBUILD before installing to include your flags. Besides, you will need to tell gromacs to use your compilers.

jaw8621 commented on 2017-04-27 11:44

Now that LLVM and Clang 4.0 is out, gromacs should have the cmake OpenCL compile flags turned on by default. Works since gromacs 5.1 and with both, amd and nvidia as soon as a suitable gpu and driver is present.

Flags: -DGMX_GPU=ON -DGMX_USE_OPENCL=ON