Skip to content

enhance CUDA support in CP2K easyblock (WIP) #2349

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
boegel wants to merge 5 commits into
base: develop
Choose a base branch
from
Open

enhance CUDA support in CP2K easyblock (WIP) #2349

Changes from 1 commit
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
b2b9dc6
enhance CUDA support in CP2K easyblock
boegel Feb 26, 2021
9b53c58
Merge branch 'develop' into cp2k_cuda
boegel Mar 30, 2021
f71f992
use -arch in NVFLAGS to compile for specific CUDA compute capabilitie…
boegel Mar 30, 2021
664d91b
require a single CUDA compute capability when building CP2K with CUDA…
boegel Mar 30, 2021
68e15c5
wrap compiler command and flags in single quotes in NVFLAGS value
boegel Mar 30, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
26 changes: 22 additions & 4 deletions easybuild/easyblocks/c/cp2k.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -228,12 +228,30 @@ def configure_step(self):
else:
options['DFLAGS'] += ' -D__ELPA3'

# CUDA
# CUDA support
# see https://github.com/cp2k/cp2k/blob/master/INSTALL.md#2j-cuda-optional-improved-performance-on-gpu-systems
cuda = get_software_root('CUDA')
if cuda:
options['DFLAGS'] += ' -D__ACC -D__DBCSR_ACC'
options['LIBS'] += ' -lcudart -lcublas -lcufft -lrt'
options['NVCC'] = ' nvcc'
# check GPU architecture(s) to build for
cuda_cc_sm_vals = self.cfg.template_values.get('cuda_sm_comma_sep')
if cuda_cc_sm_vals:
gpu_targets = '--gpu-code=%s' % cuda_cc_sm_vals
else:
gpu_targets = ''

@boegel boegel Feb 26, 2021

Copy link
Copy Markdown
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This needs a better default, and/or a warning if no CUDA compute capabilities were specified?


options['DFLAGS'] += ' -D__ACC -D__DBCSR_ACC -D__PW_CUDA -D__GRID_CUDA'

@boegel boegel Feb 26, 2021

Copy link
Copy Markdown
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Do we need easyconfig parameters to opt-out of using -D__PW_CUDA and -D__GRID_CUDA?

options['LIBS'] += ' -lcudart -lnvrtc -lcuda -lcublas -lcufft -lrt'
options['NVCC'] = 'nvcc'
options['NVFLAGS'] = ' '.join([
options['DFLAGS'],
'-O3',
'--std=c++11',
gpu_targets,
# control host compilers + options
'-ccbin="%s"' % os.getenv('CXX'),
'-Xcompiler="%s"' % os.getenv('CXXFLAGS'),
])
options['GPUVER'] = 'V100'

@boegel boegel Feb 26, 2021

Copy link
Copy Markdown
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is obviously hardcoded, should we auto-detect what the correct value should be for this?

Not sure what the significance of this is, actually...

@Flamefire Flamefire Mar 31, 2021 •
edited
Loading

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is this the point where you said in the call that only a single GPU arch is supported?

I just searched the source of CP2K and didn't find any place where this is actually used: https://github.com/cp2k/cp2k/search?q=GPUVER
All uses are either in the autotools build chain, which we don't use (I'd guess we could?) So I guess we don't need it at all. Or did I miss anything?

I'd say only the NVFLAGS are required

@boegel boegel Apr 4, 2021

Copy link
Copy Markdown
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No, this is really required, I tried without it. It's used by the DBCSR component.

I have an updated version of this, which I apparently haven't pushed yet, doing that right now (see 68e15c5)


# avoid group nesting
options['LIBS'] = options['LIBS'].replace('-Wl,--start-group', '').replace('-Wl,--end-group', '')
Expand Down